Fujitsu DISK DRIVES MHM215OAT User Manual

C141-E104-02EN  
MHL2300AT, MHM2200AT,  
MHM2150AT, MHM2100AT  
DISK DRIVES  
PRODUCT MANUAL  
This page is intentionally left blank.  
Revision History  
(1/1)  
Edition  
Date  
Revised section (*1)  
(Added/Deleted/Altered)  
Details  
01  
02  
2000-02-15  
2000-09-20 -Table 1.1  
- Specification (Number of Sections for  
MHL2300AT) was altered.  
- Table 1.2  
- Order No. was changed.  
- (16) SET MAX in Section  
- SET MAX commands are added.  
5.3.2  
- Table 5.17  
- Values of host termination for DIOR-,  
DIOW- and DMACK- signals are  
changed.  
*1 Section(s) with asterisk (*) refer to the previous edition when those were deleted.  
C141-E104-02EN  
This page is intentionally left blank.  
Preface  
This manual describes the MHL Series and MHM Series, 2.5-inch hard disk drives.  
These drives have a built-in controller that is compatible with the ATA interface.  
This manual describes the specifications and functions of the drives and explains in  
detail how to incorporate the drives into user systems. This manual assumes that  
the reader has a basic knowledge of hard disk drives and their implementations in  
computer systems.  
This manual consists of seven chapters and sections explaining the special  
terminology and abbreviations used in this manual:  
Overview of Manual  
CHAPTER 1  
Device Overview  
This chapter gives an overview of the MHL Series and MHM Series and describes  
their features.  
CHAPTER 2  
Device Configuration  
This chapter describes the internal configurations of the MHL Series and MHM  
Series and the configuration of the systems in which they operate.  
CHAPTER 3  
Installation Conditions  
This chapter describes the external dimensions, installation conditions, and switch  
settings of the MHL Series and MHM Series.  
CHAPTER 4  
This chapter describes the operation theory of the MHL Series and MHM Series.  
CHAPTER 5 Interface  
Theory of Device Operation  
This chapter describes the interface specifications of the MHL Series and MHM  
Series.  
CHAPTER 6  
Operations  
This chapter describes the operations of the MHL Series and MHM Series.  
Terminology  
This section explains the special terminology used in this manual.  
Abbreviation  
This section gives the meanings of the definitions used in this manual.  
C141-E104-02EN  
i
Preface  
Conventions for Alert Messages  
This manual uses the following conventions to show the alert messages. An alert  
message consists of an alert signal and alert statements. The alert signal consists of  
an alert symbol and a signal word or just a signal word.  
The following are the alert signals and their meanings:  
This indicates a hazardous situation could result in  
minor or moderate personal injury if the user does  
not perform the procedure correctly. This alert  
signal also indicates that damages to the product or  
other property may occur if the user does not perform  
the procedure correctly.  
CAUTION  
This indicates information that could help the user  
use the product more efficiently.  
IMPORTANT  
In the text, the alert signal is centered, followed below by the indented message. A  
wider line space precedes and follows the alert message to show where the alert  
message begins and ends. The following is an example:  
(Example)  
CAUTION  
Data corruption: Avoid mounting the disk drive near strong  
magnetic sources such as loud speakers. Ensure that the disk drive is  
not affected by external magnetic fields.  
The main alert messages in the text are also listed in the “Important Alert Items.”  
Operating Environment  
This product is designed to be used in offices or computer rooms.  
For details regarding the operating environment of use, refer to the  
(Cnnn-Xnnn) and the  
(Cnnn-Xnnn).  
Attention  
Please forward any comments you may have regarding this manual.  
To make this manual easier for users to understand, opinions from readers are  
needed. Please write your opinions or requests on the Comment at the back of this  
manual and forward it to the address described in the sheet.  
ii  
C141-E104-02EN  
Preface  
Liability Exception  
“Disk drive defects” refers to defects that involve adjustment, repair, or  
replacement.  
Fujitsu is not liable for any other disk drive defects, such as those caused by user  
misoperation or mishandling, inappropriate operating environments, defects in the  
power supply or cable, problems of the host system, or other causes outside the  
disk drive.  
C141-E104-02EN  
iii  
This page is intentionally left blank.  
Important Alert Items  
Important Alert Messages  
The important alert messages in this manual are as follows:  
A hazardous situation could result in minor or moderate personal  
CAUTION  
injury if the user does not perform the procedure correctly. Also,  
damage to the predate or other property, may occur if the user does not  
perform the procedure correctly.  
Task  
Alert message  
Page  
3-8  
Normal Operation  
Data corruption: Avoid mounting the disk near strong  
magnetic sources such as loud speakers. Ensure that the disk  
drive is not affected by external magnetic fields.  
Damage: Do not press the cover of the disk drive. Pressing  
it too hard, the cover and the spindle motor contact, which  
may cause damage to the disk drive.  
Static: When handling the device, disconnect the body  
ground (500 kor greater). Do not touch the printed circuit  
board, but hold it by the edges.  
C141-E104-02EN  
v
This page is intentionally left blank.  
Manual Organization  
MHL2300AT, MHM2200AT,  
MHM2150AT, MHM2100AT  
• Device Overview  
• Device Configuration  
• Installation Conditions  
• Theory of Device Operation  
• Interface  
DISK DRIVES  
PRODUCT MANUAL  
(C141-E104)  
• Operations  
<This manual>  
MHL2300AT, MHM2200AT,  
MHM2150AT, MHM2100AT  
• Maintenance and Diagnosis  
• Removal and Replacement Procedure  
DISK DRIVES  
MAINTENANCE MANUAL  
(C141-F043)  
C141-E104-02EN  
vii  
This page is intentionally left blank.  
Contents  
CHAPTER 1 Device Overview ....................................................................... 1-1  
1.1 Features 1-2  
1.1.1 Functions and performance 1-2  
1.1.2 Adaptability 1-2  
1.1.3 Interface 1-3  
1.2 Device Specifications 1-4  
1.2.1 Specifications summary 1-4  
1.2.2 Model and product number 1-5  
1.3 Power Requirements 1-5  
1.4 Environmental Specifications 1-7  
1.5 Acoustic Noise 1-8  
1.6 Shock and Vibration 1-8  
1.7 Reliability 1-9  
1.8 Error Rate 1-10  
1.9 Media Defects 1-10  
CHAPTER 2 Device Configuration................................................................ 2-1  
2.1 Device Configuration 2-2  
2.2 System Configuration 2-4  
2.2.1 ATA interface 2-4  
2.2.2 1 drive connection 2-4  
2.2.3 2 drives connection 2-4  
C141-E104-02EN  
ix  
Contents  
CHAPTER 3 Installation Conditions..............................................................3-1  
3.1 Dimensions 3-2  
3.2 Mounting 3-4  
3.3 Cable Connections 3-10  
3.3.1 Device connector 3-10  
3.3.2 Cable connector specifications 3-11  
3.3.3 Device connection 3-11  
3.3.4 Power supply connector (CN1) 3-12  
3.4 Jumper Settings 3-12  
3.4.1 Location of setting jumpers 3-12  
3.4.2 Factory default setting 3-13  
3.4.3 Master drive-slave drive setting 3-13  
3.4.4 CSEL setting 3-14  
CHAPTER 4 Theory of Device Operation......................................................4-1  
4.1 Outline 4-2  
4.2 Subassemblies 4-2  
4.2.1 Disk 4-2  
4.2.2 Head 4-2  
4.2.3 Spindle 4-3  
4.2.4 Actuator 4-3  
4.2.5 Air filter 4-3  
4.3 Circuit Configuration 4-4  
4.4 Power-on Sequence 4-7  
4.5 Self-calibration 4-8  
4.5.1 Self-calibration contents 4-8  
4.5.2 Execution timing of self-calibration 4-9  
4.5.3 Command processing during self-calibration 4-10  
4.6 Read/write Circuit 4-10  
4.6.1 Read/write preamplifier (HDIC) 4-10  
x
C141-E104-02EN  
Contents  
4.6.2 Write circuit 4-11  
4.6.3 Read circuit 4-13  
4.6.4 Digital PLL circuit 4-14  
4.7 Servo Control 4-15  
4.7.1 Servo control circuit 4-15  
4.7.2 Data-surface servo format 4-18  
4.7.3 Servo frame format 4-20  
4.7.4 Actuator motor control 4-21  
4.7.5 Spindle motor control 4-22  
CHAPTER 5 Interface..................................................................................... 5-1  
5.1 Physical Interface 5-2  
5.1.1 Interface signals 5-2  
5.1.2 Signal assignment on the connector 5-3  
5.2 Logical Interface 5-6  
5.2.1 I/O registers 5-7  
5.2.2 Command block registers 5-8  
5.2.3 Control block registers 5-13  
5.3 Host Commands 5-13  
5.3.1 Command code and parameters 5-14  
5.3.2 Command descriptions 5-16  
5.3.3 Error posting 5-83  
5.4 Command Protocol 5-85  
5.4.1 Data transferring commands from device to host 5-85  
5.4.2 Data transferring commands from host to device 5-87  
5.4.3 Commands without data transfer 5-89  
5.4.4 Other commands 5-90  
5.4.5 DMA data transfer commands 5-90  
5.5 Ultra DMA Feature Set 5-92  
5.5.1 Overview 5-92  
5.5.2 Phases of operation 5-93  
C141-E104-02EN  
xi  
Contents  
5.5.2.1 Ultra DMA burst initiation phase 5-93  
5.5.2.2 Data transfer phase 5-94  
5.5.2.3 Ultra DMA burst termination phase 5-94  
5.5.3 Ultra DMA data in commands 5-95  
5.5.3.1 Initiating an Ultra DMA data in burst 5-95  
5.5.3.2 The data in transfer 5-96  
5.5.3.3 Pausing an Ultra DMA data in burst 5-96  
5.5.3.4 Terminating an Ultra DMA data in burst 5-97  
5.5.4 Ultra DMA data out commands 5-100  
5.5.4.1 Initiating an Ultra DMA data out burst 5-100  
5.5.4.2 The data out transfer 5-100  
5.5.4.3 Pausing an Ultra DMA data out burst 5-101  
5.5.4.4 Terminating an Ultra DMA data out burst 5-102  
5.5.5 Ultra DMA CRC rules 5-104  
5.5.6 Series termination required for Ultra DMA 5-106  
5.6 Timing 5-107  
5.6.1 PIO data transfer 5-107  
5.6.2 Multiword DMA data transfer 5-109  
5.6.3 Transfer of Ultra DMA data 5-110  
5.6.3.1 Starting of Ultra DMA data In Burst 5-110  
5.6.3.2 Ultra DMA data burst timing requirements 5-111  
5.6.3.3 Sustained Ultra DMA data in burst 5-113  
5.6.3.4 Host pausing an Ultra DMA data in burst 5-114  
5.6.3.5 Device terminating an Ultra DMA data in burst 5-115  
5.6.3.6 Host terminating an Ultra DMA data in burst 5-116  
5.6.3.7 Initiating an Ultra DMA data out burst 5-117  
5.6.3.8 Sustained Ultra DMA data out burst 5-118  
5.6.3.9 Device pausing an Ultra DMA data out burst 5-119  
5.6.3.10 Host terminating an Ultra DMA data out burst 5-120  
5.6.3.11 Device terminating an Ultra DMA data in burst 5-121  
5.6.4 Power-on and reset 5-122  
CHAPTER 6 Operations .................................................................................6-1  
6.1 Device Response to the Reset 6-2  
xii  
C141-E104-02EN  
Contents  
6.1.1 Response to power-on 6-2  
6.1.2 Response to hardware reset 6-4  
6.1.3 Response to software reset 6-5  
6.1.4 Response to diagnostic command 6-6  
6.2 Address Translation 6-7  
6.2.1 Default parameters 6-7  
6.2.2 Logical address 6-8  
6.3 Power Save 6-9  
6.3.1 Power save mode 6-9  
6.3.2 Power commands 6-11  
6.4 Defect Management 6-11  
6.4.1 Spare area 6-12  
6.4.2 Alternating defective sectors 6-12  
6.5 Read-Ahead Cache 6-14  
6.5.1 Data buffer configuration 6-14  
6.5.2 Caching operation 6-14  
6.5.3 Usage of read segment 6-16  
6.5.3.1 Mis-hit (no hit) 6-16  
6.5.3.2 Sequential read 6-17  
6.5.3.3 Full hit (hit all) 6-20  
6.5.3.4 Partially hit 6-21  
6.6 Write Cache 6-22  
Glossary  
..................................................................................................GL-1  
Acronyms and Abbreviations......................................................................... AB-1  
C141-E104-02EN  
xiii  
Contents  
Illustrations  
Figures  
Figure 1.1 Current fluctuation (Typ.) at +5V when power is turned on 1-7  
Figure 2.1 Disk drive outerview (the MHL Series and MHM Series) 2-2  
Figure 2.2 Configuration of disk media heads 2-3  
Figure 2.3 1 drive system configuration 2-4  
Figure 2.4 2 drives configuration 2-4  
Figure 3.1 Dimensions (MHL/MHM series) 3-2  
Figure 3.2 Orientation (Sample: MHL2300AT) 3-4  
Figure 3.3 Mounting frame structure 3-5  
Figure 3.4 Location of breather 3-6  
Figure 3.5 Surface temperature measurement points  
(Sample: MHL2300AT) 3-7  
Figure 3.6 Service area (Sample: MHL2300AT) 3-8  
Figure 3.7 Handling cautions 3-9  
Figure 3.8 Connector locations (Sample: MHL2300AT) 3-10  
Figure 3.9 Cable connections 3-11  
Figure 3.10 Power supply connector pins (CN1) 3-12  
Figure 3.11 Jumper location 3-12  
Figure 3.12 Factory default setting 3-13  
Figure 3.13 Jumper setting of master or slave device 3-13  
Figure 3.14 CSEL setting 3-14  
Figure 3.15 Example (1) of Cable Select 3-14  
Figure 3.16 Example (2) of Cable Select 3-15  
Figure 4.1 Head structure 4-3  
Figure 4.2 Power Supply Configuration 4-5  
Figure 4.3 Circuit Configuration 4-6  
Figure 4.4 Power-on operation sequence 4-8  
Figure 4.5 Read/write circuit block diagram 4-12  
Figure 4.6 Frequency characteristic of programmable filter 4-13  
Figure 4.7 Block diagram of servo control circuit 4-15  
Figure 4.8 Physical sector servo configuration on disk surface 4-19  
Figure 4.9 Servo frame format 4-20  
Figure 5.1 Interface signals 5-2  
Figure 5.2 Execution example of READ MULTIPLE command 5-20  
xiv  
C141-E104-02EN  
Contents  
Figure 5.3 Read Sector(s) command protocol 5-86  
Figure 5.4 Protocol for command abort 5-87  
Figure 5.5 WRITE SECTOR(S) command protocol 5-88  
Figure 5.6 Protocol for the command execution without data transfer 5-90  
Figure 5.7 Normal DMA data transfer 5-91  
Figure 5.8 An example of generation of parallel CRC 5-105  
Figure 5.9 Ultra DMA termination with pull-up or pull-down 5-106  
Figure 5.10 Data transfer timing 5-108  
Figure 5.11 Multiword DMA data transfer timing (mode 2) 5-109  
Figure 5.12 Starting of Ultra DMA data In Burst transfer 5-110  
Figure 5.13 Sustained Ultra DMA data in burst 5-113  
Figure 5.14 Host pausing an Ultra DMA data in burst 5-114  
Figure 5.15 Device terminating an Ultra DMA data in burst 5-115  
Figure 5.16 Host terminating an Ultra DMA data in burst 5-116  
Figure 5.17 Initiating an Ultra DMA data out burst 5-117  
Figure 5.18 Sustained Ultra DMA data out burst 5-118  
Figure 5.19 Device pausing an Ultra DMA data out burst 5-119  
Figure 5.20 Host terminating an Ultra DMA data out burst 5-120  
Figure 5.21 Device terminating an Ultra DMA data out burst 5-121  
Figure 5.22 Power on Reset Timing 5-122  
Figure 6.1 Response to power-on 6-3  
Figure 6.2 Response to hardware reset 6-4  
Figure 6.3 Response to software reset 6-5  
Figure 6.4 Response to diagnostic command 6-6  
Figure 6.5 Address translation (example in CHS mode) 6-8  
Figure 6.6 Address translation (example in LBA mode) 6-9  
Figure 6.7 Sector slip processing 6-12  
Figure 6.8 Alternate cylinder assignment 6-13  
Figure 6.9 Data buffer configuration 6-14  
Tables  
Table 1.1 Specifications 1-4  
Table 1.2 Model names and product numbers 1-5  
Table 1.3 Current and power dissipation 1-6  
Table 1.4 Environmental specifications 1-7  
Table 1.5 Acoustic noise specification 1-8  
Table 1.6 Shock and vibration specification 1-8  
Table 3.1 Surface temperature measurement points and standard values 3-7  
Table 3.2 Cable connector specifications 3-11  
C141-E104-02EN  
xv  
Contents  
Table 4.1 Self-calibration execution timechart 4-10  
Table 4.2 Write precompensation algorithm 4-11  
Table 5.1 Signal assignment on the interface connector 5-3  
Table 5.2 I/O registers 5-7  
Table 5.3 Command code and parameters 5-14  
Table 5.4 Information to be read by IDENTIFY DEVICE command 5-32  
Table 5.5 Features register values and settable modes 5-41  
Table 5.6 Diagnostic code 5-52  
Table 5.7 Features Register values (subcommands) and functions 5-64  
Table 5.8 Format of device attribute value data 5-68  
Table 5.9 Format of insurance failure threshold value data 5-68  
Table 5.10 SMART error log data format 5-72  
Table 5.11 SMART self test log data format 5-74  
Table 5.12 Contents of security password 5-76  
Table 5.13 Contents of SECURITY SET PASSWORD data 5-80  
Table 5.14 Relationship between combination of Identifier and Security level,  
and operation of the lock function 5-80  
Table 5.15 Command code and parameters 5-83  
Table 5.16 Parallel generation equation of CRC polynomial 5-105  
Table 5.17 Recommended series termination for Ultra DMA 5-106  
Table 5.18 Ultra DMA data burst timing requirements 5-111  
Table 6.1 Default parameters 6-7  
xvi  
C141-E104-02EN  
CHAPTER 1 Device Overview  
1.1  
1.2  
1.3  
1.4  
1.5  
1.6  
1.7  
1.8  
1.9  
Features  
Device Specifications  
Power Requirements  
Environmental Specifications  
Acoustic Noise  
Shock and Vibration  
Reliability  
Error Rate  
Media Defects  
Overview and features are described in this chapter, and specifications and power  
requirement are described.  
The MHL Series and MHM Series are 2.5-inch hard disk drives with built-in disk  
controllers. These disk drives use the AT-bus hard disk interface protocol and are  
compact and reliable.  
C141-E104-02EN  
1-1  
Device Overview  
1.1 Features  
1.1.1 Functions and performance  
The following features of the MHL Series and MHM Series are described.  
(1) Compact  
The MHL2300AT has 3 built-in disks (the diameter is 65mm[2.5inch]), and its  
height is 12.5 mm (0.492 inch). The MHM2200AT, MHM2150AT and  
MHM2100AT have 1 disk or 2 disks of 65 mm (2.5 inches) diameter, and its  
height is 9.5 mm (0.374 inch).  
(2) Large capacity  
The disk drive can record up to 10 GB (formatted) on one disk using the 16/17  
MTR recording method and 15 recording zone technology. The MHL Series and  
MHM Series have a formatted capacity of 30 GB (MHL2300AT), 20 GB  
(MHM2200AT), 15 GB (MHM2150AT) and 10 GB (MHM2100AT) respectively.  
(3) High-speed Transfer rate  
The disk drives (the MHL Series and MHM Series) have an internal data rate up  
to 28.7 MB/s. The disk drive supports an external data rate up to 66.6 MB/s (U-  
DMA mode 4).  
(4) Average positioning time  
Use of a rotary voice coil motor in the head positioning mechanism greatly  
increases the positioning speed. The average positioning time is 12 ms (at read).  
1.1.2 Adaptability  
(1) Power save mode  
The power save mode feature for idle operation, stand by and sleep modes makes  
The disk drives (the MHL Series and MHM Series) ideal for applications where  
power consumption is a factor.  
(2) Wide temperature range  
The disk drives (the MHL Series and MHM Series) can be used over a wide  
temperature range (5°C to 55°C).  
(3) Low noise and vibration  
In Ready status, the noise of the disk drives (the MHL Series and MHM Series) is  
only about 30 dBA (measured at 1 m apart from the drive under the idle mode).  
1-2  
C141-E104-02EN  
1.1 Features  
1.1.3 Interface  
(1) Connection to interface  
With the built-in ATA interface controller, the disk drives (the MHL Series and  
MHM Series) can be connected to an ATA interface of a personal computer.  
(2) 2 MB data buffer  
The disk drives (the MHL Series and MHM Series) uses a 2 MB data buffer to  
transfer data between the host and the disk media.  
In combination with the read-ahead cache system described in item (3) and the  
write cache described in item (7), the buffer contributes to efficient I/O  
processing.  
(3) Read-ahead cache system  
After the execution of a disk read command, the disk drive automatically reads the  
subsequent data block and writes it to the data buffer (read ahead operation). This  
cache system enables fast data access. The next disk read command would  
normally cause another disk access. But, if the read ahead data corresponds to the  
data requested by the next read command, the data in the buffer can be transferred  
instead.  
(4) Master/slave  
The disk drives (the MHL Series and MHM Series) can be connected to ATA  
interface as daisy chain configuration. Drive 0 is a master device, drive 1 is a  
slave device.  
(5) Error correction and retry by ECC  
If a recoverable error occurs, the disk drives (the MHL Series and MHM Series)  
themselves attempt error recovery. The ECC has improved buffer error correction  
for correctable data errors.  
(6) Self-diagnosis  
The disk drives (the MHL Series and MHM Series) have a diagnostic function to  
check operation of the controller and disk drives. Executing the diagnostic  
command invokes self-diagnosis.  
(7) Write cache  
When the disk drives (the MHL Series and MHM Series) receive a write  
command, the disk drives post the command completion at completion of  
transferring data to the data buffer completion of writing to the disk media. This  
feature reduces the access time at writing.  
C141-E104-02EN  
1-3  
Device Overview  
1.2 Device Specifications  
1.2.1 Specifications summary  
Table 1.1 shows the specifications of the disk drives (MHL Series and MHM Series).  
Table 1.1 Specifications (1/2)  
MHL2300AT MHM2200AT MHM2150AT MHM2100AT  
Format Capacity (*1)  
Number of Heads  
30 GB  
6
20 GB  
4
15 GB  
3
10 GB  
2
Number of Cylinders (User)  
Number of Sectors (User)  
Bytes per Sector  
19,904  
512  
58,605,120  
39,070,080  
29,498,112  
19,640,880  
Recording Method  
Track Density  
16/17 MTR  
32,300 TPI (1271 track/mm)  
499.7 Kbpi (19.67 k bit/mm)  
4,200 rpm ± 1%  
Bit Density  
Rotational Speed  
Average Latency  
7.14 ms  
Positioning time (read and seek)  
Minimum (Track to Track)  
Average  
1.5 ms (typ.)  
Read: 12 ms (typ.)  
22 ms (typ.)  
Maximum (Full)  
Start/Stop time  
Start (0 rpm to Drive Read)  
Stop (at Power Down)  
Typ.: 5 sec  
Typ.: 5 sec  
Interface  
ATA-5 (Max. Cable length: 0.46 m)  
Data Transfer Rate  
To/From Media  
To/From Host  
16.4 to 28.7 MB/s  
66.6 MB/s Max.  
(U-DMA mode 4)  
Data Buffer Size  
2 MB  
Physical Dimensions  
(Height × Width × Depth)  
12.5 mm ×  
100.0 mm ×  
70.0 mm  
9.5 mm × 100.0 mm ×70.0 mm  
Weight  
134 g  
98 g  
*1: Capacity under the LBA mode.  
1-4  
C141-E104-02EN  
1.3 Power Requirements  
Under the CHS mode (normal BIOS specification), formatted capacity,  
number of cylinders, number of heads, and number of sectors are as  
follows.  
Table 1.1 Specifications (2/2)  
Model  
Capacity  
No. of Cylinder  
No. of Heads  
No. of Sectors  
MHL2300AT  
MHM2200AT  
MHM2150AT  
MHM2100AT  
8.45 GB  
8.45 GB  
8.45 GB  
8.45 GB  
16,383  
16,383  
16,383  
16,383  
16  
16  
16  
16  
63  
63  
63  
63  
1.2.2 Model and product number  
Table 1.2 lists the model names and product numbers of the MHL Series and  
MHM Series.  
Table 1.2 Model names and product numbers  
Model Name  
Capacity  
Mounting screw  
Order No.  
(user area)  
MHL2300AT  
MHM2200AT  
MHM2150AT  
MHM2100AT  
30 GB  
20 GB  
15 GB  
10 GB  
M3, depth 3  
M3, depth 3  
M3, depth 3  
M3, depth 3  
CA05428-B061  
CA05429-B041  
CA05429-B031  
CA05429-B021  
1.3 Power Requirements  
(1) Input Voltage  
+ 5 V ± 5 %  
(2) Ripple  
+5 V  
Maximum  
Frequency  
100 mV (peak to peak)  
DC to 1 MHz  
C141-E104-02EN  
1-5  
Device Overview  
(3) Current Requirements and Power Dissipation  
Table 1.3 lists the current and power dissipation.  
Table 1.3 Current and power dissipation  
Typical RMS Current  
Typical Power (*3)  
MHL Series MHM Series MHL Series  
MHM Series  
4.5 W  
Spin up (*1)  
Idle  
0.9 A  
190 mA  
520 mA  
490 mA  
50 mA  
20 mA  
0.9 A  
160 mA  
500 mA  
460 mA  
50 mA  
20 mA  
4.5 W  
0.95 W  
2.6 W  
0.8 W  
R/W (on track) (*2)  
Seek (*5)  
Standby  
2.5 W  
2.45 W  
0.25 W  
0.1 W  
2.3 W  
0.25 W  
0.1 W  
Sleep  
Energy  
Efficiency (*4)  
0.032 W/GB  
(rank E)  
0.040 W/GB  
(rank E / MHM2200AT)  
0.040 W/GB  
(rank E / MHM2150AT)  
0.080 W/GB  
(rank D / MHM2100AT)  
*1  
*2  
*3  
*4  
Current at starting spindle motor.  
At 30% disk accessing.  
Power requirements reflect nominal values for +5V power.  
Energy efficiency based on the Law concerning the Rational Use of Energy  
indicates the value obtained by dividing power consumption by the storage  
capacity. (Japan only)  
*5  
The seek average current is specified based on three operations per 100  
msec.  
(4) Current fluctuation (Typ.) at +5V when power is turned on  
1-6  
C141-E104-02EN  
1.4 Environmental Specifications  
Figure 1.1 Current fluctuation (Typ.) at +5V when power is turned on  
(5) Power on/off sequence  
The voltage detector circuits (the MHL Series and MHM Series) monitor +5 V.  
The circuits do not allow a write signal if either voltage is abnormal. These  
prevent data from being destroyed and eliminates the need to be concerned with  
the power on/off sequence.  
1.4 Environmental Specifications  
Table 1.4 lists the environmental specifications.  
Table 1.4 Environmental specifications  
Item  
Specification  
Temperature  
• Operating  
5°C to 55°C (ambient)  
5°C to 60°C (disk enclosure surface)  
–40°C to 65°C  
• Non-operating  
• Thermal Gradient  
Humidity  
20°C/h or less  
• Operating  
8% to 90% RH (Non-condensing)  
5% to 95% RH (Non-condensing)  
• Non-operating  
• Maximum Wet Bulb  
29°C (Operating)  
40°C (Non-operating)  
Altitude (relative to sea level)  
• Operating  
–300 to 3,000 m  
–300 to 12,000 m  
• Non-operating  
C141-E104-02EN  
1-7  
Device Overview  
1.5 Acoustic Noise  
Table 1.5 lists the acoustic noise specification.  
Table 1.5 Acoustic noise specification  
Item  
Specification  
Sound Pressure  
• Idle mode (DRIVE READY)  
30 dBA typical at 1 m  
Note:  
Measure the noise from the cover top surface.  
1.6 Shock and Vibration  
Table 1.6 lists the shock and vibration specification.  
Table 1.6 Shock and vibration specification  
Item  
Specification  
Vibration (swept sine, one octave per minute)  
• Operating  
5 to 500 Hz, 1.0G 0-peak (MHL series)  
5 to 400 Hz, 1.0G 0-peak (MHM series)  
(without non-recovered errors) (9.8 m/s2 0-peak)  
• Non-operating  
5 to 500 Hz, 5G 0-peak (MHL series)  
5 to 400 Hz, 5G 0-peak (MHM series)  
(no damage) (49 m/s2 0-peak)  
Shock (half-sine pulse)  
• Operating  
175G 0-peak (1,715 m/s2 0-peak)  
2 ms duration (without non-recovered errors)  
• Non-operating  
600G 0-peak (5,880 m/s2 0-peak)  
2 ms duration (no damage)  
MHL series  
700G 0-peak (6,860 m/s2 0-peak)  
2 ms duration (no damage)  
MHM series  
120G 0-peak (1,176 m/s2 0-peak)  
11 ms duration (no damage)  
1-8  
C141-E104-02EN  
1.7 Reliability  
1.7  
Reliability  
(1) Mean time between failures (MTBF)  
Conditions of 300,000 h Power-on time 250H/month or less 3000H/years  
or less  
Operating time 20% or less of power-on time  
CSS operations 50/day or less  
Total 50,000 or less  
Power on/off  
Environment  
1/day or more needed.  
5 to 55°C/8 to 90%  
But humidity bulb temperature  
29°C or less  
MTBF is defined as follows:  
Total operation time in all fields  
MTBF=  
(H)  
number of device failure in all fields (*1)  
*1 “Disk drive defects” refers to defects that involve repair, readjustment, or  
replacement. Disk drive defects do not include failures caused by external  
factors, such as damage caused by handling, inappropriate operating  
environments, defects in the power supply host system, or interface cable.  
(2) Mean time to repair (MTTR)  
The mean time to repair (MTTR) is 30 minutes or less, if repaired by a specialist  
maintenance staff member.  
(3) Service life  
In situations where management and handling are correct, the disk drive requires  
no overhaul for five years when the DE surface temperature is less than 48°C.  
When the DE surface temperature exceeds 48°C, the disk drives requires no  
overhaul for five years or 20,000 hours of operation, whichever occurs first.  
Refer to item (3) in Subsection 3.2 for the measurement point of the DE surface  
temperature. Also the operating conditions except the environment temperature  
are based on the MTBF conditions.  
(4) Data assurance in the event of power failure  
Except for the data block being written to, the data on the disk media is assured in  
the event of any power supply abnormalities. This does not include power supply  
abnormalities during disk media initialization (formatting) or processing of  
defects (alternative block assignment).  
C141-E104-02EN  
1-9  
Device Overview  
1.8 Error Rate  
Known defects, for which alternative blocks can be assigned, are not included in  
the error rate count below. It is assumed that the data blocks to be accessed are  
evenly distributed on the disk media.  
(1) Unrecoverable read error  
Read errors that cannot be recovered by maximum read retries of drive without  
user’s retry and ECC corrections shall occur no more than 10 times when reading  
data of 1014 bits. Read retries are executed according to the disk drive’s error  
recovery procedure, and include read retries accompanying head offset operations.  
(2) Positioning error  
Positioning (seek) errors that can be recovered by one retry shall occur no more  
than 10 times in 107 seek operations.  
1.9 Media Defects  
Defective sectors are replaced with alternates when the disk (the MHL Series and  
MHM Series) are formatted prior to shipment from the factory (low level format).  
Thus, the hosts see a defect-free devices.  
Alternate sectors are automatically accessed by the disk drive. The user need not  
be concerned with access to alternate sectors.  
1-10  
C141-E104-02EN  
CHAPTER 2 Device Configuration  
2.1  
2.2  
Device Configuration  
System Configuration  
This chapter describes the internal configurations of the hard disk drives and the  
configuration of the systems in which they operate.  
C141-E104-02EN  
2-1  
Device Configuration  
2.1 Device Configuration  
Figure 2.1 shows the disk drive. The disk drive consists of a disk enclosure (DE),  
read/write preamplifier, and controller PCA. The disk enclosure contains the disk  
media, heads, spindle motors, actuators, and a circulating air filter.  
MHL Series  
MHM Series  
Figure 2.1 Disk drive outerview  
(1) Disk  
The outer diameter of the disk is 65 mm. The inner diameter is 20 mm. The  
number of disks used varies with the model, as described below. The disks are  
rated at over 50,000 start/stop operations.  
MHL2300AT: 3 disks  
MHM2200AT: 2 disks  
MHM2150AT: 2 disks  
MHM2100AT: 1 disk  
(2) Head  
The heads are of the contact start/stop (CSS) type. The head touches the disk  
surface while the disk is not rotating and automatically lifts when the disk starts.  
Figure 2.2 illustrates the configuration of the disks and heads of each model. In  
the disk surface, servo information necessary for controlling positioning and  
read/write and user data are written. Numerals 0 to 5 indicate read/write heads.  
2-2  
C141-E104-02EN  
2.1 Device Configuration  
Head  
5
4
Head  
3
Head  
3
3
2
2
2
Head  
1
1
0
1
0
1
0
0
MHL2300AT  
MHM2200AT  
MHM2150AT  
(Either of head 0 or  
MHM2100AT  
head 3 is mounted.)  
Figure 2.2 Configuration of disk media heads  
(3) Spindle motor  
The disks are rotated by a direct drive Hall-less DC motor.  
(4) Actuator  
The actuator uses a revolving voice coil motor (VCM) structure which consumes  
low power and generates very little heat. The head assembly at the edge of the  
actuator arm is controlled and positioned by feedback of the servo information  
read by the read/write head. If the power is not on or if the spindle motor is  
stopped, the head assembly stays in the specific CSS zone on the disk and is fixed  
by a mechanical lock.  
(5) Air circulation system  
The disk enclosure (DE) is sealed to prevent dust and dirt from entering. The disk  
enclosure features a closed loop air circulation system that relies on the blower  
effect of the rotating disk. This system continuously circulates the air through the  
circulation filter to maintain the cleanliness of the air within the disk enclosure.  
(6) Read/write circuit  
The read/write circuit uses a LSI chip for the read/write preamplifier. It improves  
data reliability by preventing errors caused by external noise.  
(7) Controller circuit  
The controller circuit consists of an LSI chip to improve reliability. The high-  
speed microprocessor unit (MPU) achieves a high-performance AT controller.  
C141-E104-02EN  
2-3  
Device Configuration  
2.2 System Configuration  
2.2.1 ATA interface  
Figures 2.3 and 2.4 show the ATA interface system configuration. The drive has a  
44pin PC AT interface connector and supports PIO mode 4 transfer at 16.6 MB/s,  
Multiword DMA mode 2 transfer at 16.6 MB/s and also U-DMA mode 4 transfer  
at 66.6 MB/s.  
2.2.2 1 drive connection  
MHL2300AT  
MHM2200AT  
MHM2150AT  
MHM2100AT  
Figure 2.3 1 drive system configuration  
2.2.3 2 drives connection  
MHL2300AT  
MHM2200AT  
(Host adaptor)  
MHM2150AT  
MHM2100AT  
MHL2300AT  
MHM2200AT  
MHM2150AT  
MHM2100AT  
Note:  
When the drive that is not conformed to ATA is connected to the disk drive above  
configuration, the operation is not guaranteed.  
Figure 2.4 2 drives configuration  
2-4  
C141-E104-02EN  
2.2 System Configuration  
IMPORTANT  
HA (host adaptor) consists of address decoder, driver, and receiver.  
ATA is an abbreviation of “AT attachment”. The disk drive is  
conformed to the ATA-4 interface.  
At high speed data transfer (PIO mode 3, mode 4, or DMA mode 2  
U-DMA mode 4), occurrence of ringing or crosstalk of the signal  
lines (AT bus) between the HA and the disk drive may be a great  
cause of the obstruction of system reliability. Thus, it is necessary  
that the capacitance of the signal lines including the HA and cable  
does not exceed the ATA-5 standard, and the cable length between  
the HA and the disk drive should be as short as possible.  
No need to push the top cover of the disk drive. If the over-power  
worked, the cover could be contacted with the spindle motor. Thus,  
that could be made it the cause of failure.  
C141-E104-02EN  
2-5  
This page is intentionally left blank.  
CHAPTER 3 Installation Conditions  
3.1  
3.2  
3.3  
3.4  
Dimensions  
Mounting  
Cable Connections  
Jumper Settings  
This chapter gives the external dimensions, installation conditions, surface  
temperature conditions, cable connections, and switch settings of the hard disk  
drives.  
C141-E104-02EN  
3-1  
Installation Conditions  
3.1 Dimensions  
Figure 3.1 illustrates the dimensions of the disk drive and positions of the  
mounting screw holes. All dimensions are in mm.  
Figure 3.1 Dimensions (MHL series) (1/2)  
3-2  
C141-E104-02EN  
3.1 Dimensions  
Figure 3.1 Dimensions (MHM series) (2/2)  
C141-E104-02EN  
3-3  
Installation Conditions  
3.2 Mounting  
(1) Orientation  
Figure 3.2 illustrates the allowable orientations for the disk drive.  
gravity  
(a) Horizontal –1  
(b) Horizontal –1  
gravity  
(c) Vertical –1  
(d) Vertical –2  
gravity  
(e) Vertical –3  
(f) Vertical –4  
Figure 3.2 Orientation (Sample: MHL2300AT)  
3-4  
C141-E104-02EN  
3.2 Mounting  
(2) Frame  
The MR head bias of the HDD disk enclosure (DE) is zero. The mounting frame  
is connected to SG.  
IMPORTANT  
Use M3 screw for the mounting screw and the screw length should  
satisfy the specification in Figure 3.3.  
The tightening torque must be 0.49N·m(5kgf·cm).  
When attaching the HDD to the system frame, do not allow the  
system frame to touch parts (cover and base) other than parts to  
which the HDD is attached.  
(3) Limitation of mounting  
Do not use the center hole. For screw length, see Figure 3.3.  
Note)  
These dimensions are recommended values; if it is not possible to  
satisfy them, contact us.  
Side surface  
mounting  
2.5  
2.5  
Bottom surface mounting  
DE  
2.5  
2.5  
2
B
PCA  
A
Frame of system  
cabinet  
Frame of system  
cabinet  
3.0 or less  
Details of A  
Screw  
Screw  
3.0 or less  
Details of B  
Figure 3.3 Mounting frame structure  
C141-E104-02EN  
3-5  
Installation Conditions  
IMPORTANT  
Because of breather hole mounted to the HDD, do not allow this to  
close during mounting.  
Locating of breather hole is shown as Figure 3.4 in both MHL  
series and MHM series.  
For breather hole of Figure 3.4, at least, do not allow its around φ3  
to block.  
Figure 3.4 Location of breather  
3-6  
C141-E104-02EN  
3.2 Mounting  
(4) Ambient temperature  
The temperature conditions for a disk drive mounted in a cabinet refer to the  
ambient temperature at a point 3 cm from the disk drive. The ambient  
temperature must satisfy the temperature conditions described in Section 1.4, and  
the airflow must be considered to prevent the DE surface temperature from  
exceeding 60°C.  
Provide air circulation in the cabinet such that the PCA side, in particular,  
receives sufficient cooling. To check the cooling efficiency, measure the surface  
temperatures of the DE. Regardless of the ambient temperature, this surface  
temperature must meet the standards listed in Table 3.1. Figure 3.5 shows the  
temperature measurement point.  
1
Figure 3.5 Surface temperature measurement points (Sample: MHL2300AT)  
Table 3.1 Surface temperature measurement points and standard values  
No.  
1
Measurement point  
DE cover  
Temperature  
60°C max  
C141-E104-02EN  
3-7  
Installation Conditions  
(5) Service area  
Figure 3.6 shows how the drive must be accessed (service areas) during and after  
installation.  
Mounting screw hole  
Cable connection  
Mounting screw hole  
Figure 3.6 Service area (Sample: MHL2300AT)  
CAUTION  
Data corruption: Avoid mounting the disk drive near strong  
magnetic sources such as loud speakers. Ensure that the disk drive  
is not affected by external magnetic fields.  
Damage: Do not press the cover of the disk drive. Pressing it too  
hard, the cover and the spindle motor contact, which may cause  
damage to the disk drive.  
Static: When handling the device, disconnect the body ground  
(500 kor greater). Do not touch the printed circuit board, but  
hold it by the edges.  
(6) Handling cautions  
Please keep the following cautions, and handle the HDD under the safety  
environment.  
3-8  
C141-E104-02EN  
3.2 Mounting  
- General notes  
ESD mat  
Wrist strap  
Shock absorbing mat  
Use the Wrist strap.  
Place the shock absorbing mat on the  
operation table, and place ESD mat on it.  
Do not hit HDD each other.  
Do not stack when carrying.  
Do not place HDD vertically  
to avoid falling down.  
Do not drop.  
Figure 3.7 Handling cautions  
- Installation  
(1)  
(2)  
Please use the driver of a low impact when you use an electric driver.  
HDD is occasionally damaged by the impact of the driver.  
Please observe the tightening torque of the screw strictly.  
M3 ······· 0.49 N·m (5 Kg·cm)  
- Recommended equipments  
Contents  
Model  
JX-1200-3056-8  
76000DES (ASK7876)  
SS-3000  
Maker  
SUMITOMO 3M  
COMKYLE  
HIOS  
ESD  
Wrist strap  
ESD mat  
Shock  
Low shock driver  
C141-E104-02EN  
3-9  
Installation Conditions  
3.3 Cable Connections  
3.3.1 Device connector  
The disk drive has the connectors and terminals listed below for connecting  
external devices. Figure 3.8 shows the locations of these connectors and  
terminals.  
PCA  
Connector,  
setting pins  
Figure 3.8 Connector locations (Sample: MHL2300AT)  
3-10  
C141-E104-02EN  
3.3 Cable Connections  
3.3.2 Cable connector specifications  
Table 3.2 lists the recommended specifications for the cable connectors.  
Table 3.2 Cable connector specifications  
Name  
Model  
Manufacturer  
BERG  
ATA interface and power  
supply cable (44-pin type)  
Cable socket  
(44-pin type)  
89361-144  
IMPORTANT  
For the host interface cable, use a ribbon cable. A twisted cable or  
a cable with wires that have become separated from the ribbon may  
cause crosstalk between signal lines. This is because the interface  
is designed for ribbon cables and not for cables carrying differential  
signals.  
3.3.3 Device connection  
Figure 3.9 shows how to connect the devices.  
Figure 3.9 Cable connections  
C141-E104-02EN  
3-11  
Installation Conditions  
3.3.4 Power supply connector (CN1)  
Figure 3.10 shows the pin assignment of the power supply connector (CN1).  
Figure 3.10 Power supply connector pins (CN1)  
3.4 Jumper Settings  
3.4.1 Location of setting jumpers  
Figure 3.11 shows the location of the jumpers to select drive configuration and  
functions.  
Figure 3.11 Jumper location  
3-12  
C141-E104-02EN  
3.4 Jumper Settings  
3.4.2 Factory default setting  
Figure 3.12 shows the default setting position at the factory.  
Open  
Figure 3.12 Factory default setting  
3.4.3 Master drive-slave drive setting  
Master drive (disk drive #0) or slave drive (disk drive #1) is selected.  
Open  
1
2
C
D
A
B
A
1
2
C
Short  
Open  
D
B
Open  
(a) Master drive  
(b) Slave drive  
Figure 3.13 Jumper setting of master or slave drive  
Note:  
Pins A and C should be open.  
C141-E104-02EN  
3-13  
Installation Conditions  
3.4.4 CSEL setting  
Figure 3.14 shows the cable select (CSEL) setting.  
Open  
1
2
C
A
D
B
Short  
Note:  
The CSEL setting is not depended on setting between pins Band D.  
Figure 3.14 CSEL setting  
Figure 3.15 and 3.16 show examples of cable selection using unique interface  
cables.  
By connecting the CSEL of the master drive to the CSEL Line (conducer) of the  
cable and connecting it to ground further, the CSEL is set to low level. The drive  
is identified as a master drive. At this time, the CSEL of the slave drive does not  
have a conductor. Thus, since the slave drive is not connected to the CSEL  
conductor, the CSEL is set to high level. The drive is identified as a slave drive.  
drive  
drive  
Figure 3.15 Example (1) of Cable Select  
3-14  
C141-E104-02EN  
3.4 Jumper Settings  
drive  
drive  
Figure 3.16 Example (2) of Cable Select  
C141-E104-02EN  
3-15  
This page is intentionally left blank.  
CHAPTER 4 Theory of Device Operation  
4.1  
4.2  
4.3  
4.4  
4.5  
4.6  
4.7  
Outline  
Subassemblies  
Circuit Configuration  
Power-on Sequence  
Self-calibration  
Read/write Circuit  
Servo Control  
This chapter explains basic design concepts of the disk drive. Also, this chapter  
explains subassemblies of the disk drive, each sequence, servo control, and  
electrical circuit blocks.  
C141-E104-02EN  
4-1  
Theory of Device Operation  
4.1 Outline  
This chapter consists of two parts. First part (Section 4.2) explains mechanical  
assemblies of the disk drive. Second part (Sections 4.3 through 4.7) explains a  
servo information recorded in the disk drive and drive control method.  
4.2 Subassemblies  
The disk drive consists of a disk enclosure (DE) and printed circuit assembly  
(PCA).  
The DE contains all movable parts in the disk drive, including the disk, spindle,  
actuator, read/write head, and air filter. For details, see Subsections 4.2.1 to 4.2.5.  
The PCA contains the control circuits for the disk drive. The disk drive has one  
PCA. For details, see Sections 4.3.  
4.2.1 Disk  
The DE contains disks with an outer diameter of 65 mm and an inner diameter of  
20 mm. The MHL2300AT have three disks and MHM2200AT and  
MHM2150AT have two disks and MHM2100AT have one disk.  
The head contacts the disk each time the disk rotation stops; the disk surface is  
durable at least 50,000 CSS (Contact Start/Stop) or more.  
Servo data is recorded on each cylinder (total 66). Servo data written at factory is  
read out by the read head. For servo data, see Section 4.7.  
4.2.2 Head  
Figure 4.1 shows the head structures. MHL2300AT has 6 heads and  
MHM2200AT has 4 heads and MHM2150AT has 3 heads and MHM2100AT has  
2 heads. These heads are raised from the disk surface as the spindle motor the  
rated rotation speed.  
4-2  
C141-E104-02EN  
4.2 Subassemblies  
Head  
5
4
Head  
3
Head  
3
3
2
2
2
Head  
1
1
0
1
0
1
0
0
MHL2300AT  
MHM2200AT  
MHM2150AT  
(Either of head 0 or  
MHK2100AT  
head 3 is mounted.)  
Figure 4.1 Head structure  
4.2.3 Spindle  
The spindle consists of a disk stack assembly and spindle motor. The disk stack  
assembly is activated by the direct drive sensor-less DC spindle motor, which has  
a speed of 4,200 rpm ±1%. The spindle is controlled with detecting a PHASE  
signal generated by counter electromotive voltage of the spindle motor at starting.  
4.2.4 Actuator  
The actuator consists of a voice coil motor (VCM) and a head carriage. The  
VCM moves the head carriage along the inner or outer edge of the disk. The head  
carriage position is controlled by feeding back the difference of the target position  
that is detected and reproduced from the servo information read by the read/write  
head.  
4.2.5 Air filter  
There are two types of air filters: a breather filter and a circulation filter.  
The breather filter makes an air in and out of the DE to prevent unnecessary  
pressure around the spindle when the disk starts or stops rotating. When disk  
drives are transported under conditions where the air pressure changes a lot,  
filtered air is circulated in the DE.  
The circulation filter cleans out dust and dirt from inside the DE. The disk drive  
cycles air continuously through the circulation filter through an enclosed loop air  
cycle system operated by a blower on the rotating disk.  
C141-E104-02EN  
4-3  
Theory of Device Operation  
4.3 Circuit Configuration  
Figure 4.2 shows the power supply configuration of the disk drive, and Figure 4.3  
shows the disk drive circuit configuration.  
(1) Read/write circuit  
The read/write circuit consists of two LSIs; read/write preamplifier (PreAMP) and  
read channel (RDC).  
The PreAMP consists of the write current switch circuit, that flows the write  
current to the head coil, and the voltage amplifier circuit, that amplitudes the read  
output from the head.  
The RDC is the read demodulation circuit using the Modified Extended Partial  
Response (MEEPR), and contains the Viterbi detector, programmable filter,  
adaptable transversal filter, times base generator, data separator circuits, 16/17  
MTR (Maximum Transitions Limited) encoder Run Length and servo  
demodulation circuit.  
(2) Servo circuit  
The position and speed of the voice coil motor are controlled by 2 closed-loop  
servo using the servo information recorded on the data surface. The servo  
information is an analog signal converted to digital for processing by a MPU and  
then reconverted to an analog signal for control of the voice coil motor.  
The MPU precisely sets each head on the track according on the servo  
information on the media surface.  
(3) Spindle motor driver circuit  
The circuit measures the interval of a PHASE signal generated by counter-  
electromotive voltage of a motor at the MPU and controls the motor speed  
comparing target speed.  
(4) Controller circuit  
Major functions are listed below.  
Data buffer (2 MB) management  
ATA interface control and data transfer control  
Sector format control  
Defect management  
ECC control  
Error recovery and self-diagnosis  
4-4  
C141-E104-02EN  
4.3 Circuit Configuration  
5V  
HDC  
SVC  
HDIC  
MCU  
FROM  
- 3V  
3.3V  
SDRAM  
RDC  
Figure 4.2 Power Supply Configuration  
C141-E104-02EN  
4-5  
Theory of Device Operation  
Printed  
Circuit  
Board  
FROM  
Program  
Memory  
16 bit  
Local Bus  
Host  
ATA  
HDC  
MPU  
Interface  
(Hard Disk  
Controller)  
(Micro  
Processor  
Unit)  
SVC  
Control  
Signal  
RDC  
Control  
Signal  
Read  
and  
Write  
Data  
Data  
Buffer  
Bus  
HDIC  
Control  
Signal  
16 bit  
SDRAM  
RDC  
Read  
Channel  
SVC  
Data  
Motor  
Servo  
Pulse  
and  
Position  
Signal  
Buffer  
RAM  
Controller  
and  
Driver  
Write Data  
HDIC  
Read Data  
Disk  
Enclosure Head  
(Read/Write  
Preamplifier)  
Voice  
Coil  
Motor  
Spindle  
Motor  
Read  
Head  
Read Data  
Write Data  
Write  
Head  
Figure 4.3 Circuit Configuration  
4-6  
C141-E104-02EN  
4.3 Circuit Configuration  
4.4 Power-on Sequence  
Figure 4.4 describes the operation sequence of the disk drive at power-on. The  
outline is described below.  
a) After the power is turned on, the disk drive executes the MPU bus test,  
internal register read/write test, and work RAM read/write test. When the  
self-diagnosis terminates successfully, the disk drive starts the spindle motor.  
b) The disk drive executes self-diagnosis (data buffer read/write test) after  
enabling response to the ATA bus.  
c) After confirming that the spindle motor has reached rated speed, the disk  
drive releases the heads from the actuator magnet lock mechanism by  
applying current to the VCM. This unlocks the heads which are parked at the  
inner circumference of the disks.  
d) The disk drive positions the heads onto the SA area and reads out the system  
information.  
e) The disk drive executes self-seek-calibration. This collects data for VCM  
torque and mechanical external forces applied to the actuator, and updates the  
calibrating value.  
f) The drive becomes ready. The host can issue commands.  
C141-E104-02EN  
4-7  
Theory of Device Operation  
Power-on  
Start  
a)  
Self-diagnosis 1  
- MPU bus test  
- Internal register  
write/read test  
- Work RAM write/read  
test  
The spindle motor starts.  
b)  
c)  
Self-diagnosis 2  
- Data buffer write/read  
test  
d)  
Initial on-track and read  
out of system information  
Confirming spindle motor  
speed  
e)  
f)  
Execute self-calibration  
Releasing heads from  
Actuator lock  
Drive ready state  
(command waiting state)  
End  
Figure 4.4 Power-on operation sequence  
4.5 Self-calibration  
The disk drive occasionally performs self-calibration in order to sense and  
calibrate mechanical external forces on the actuator, and VCM torque. This  
enables precise seek and read/write operations.  
4.5.1 Self-calibration contents  
(1) Sensing and compensating for external forces  
The actuator suffers from torque due to the FPC forces and winds accompanying  
disk revolution. The torque vary with the disk drive and the cylinder where the  
head is positioned. To execute stable fast seek operations, external forces are  
occasionally sensed.  
The firmware of the drive measures and stores the force (value of the actuator  
motor drive current) that balances the torque for stopping head stably. This  
includes the current offset in the power amplifier circuit and DAC system.  
4-8  
C141-E104-02EN  
4.5 Self-calibration  
The forces are compensated by adding the measured value to the specified current  
value to the power amplifier. This makes the stable servo control.  
To compensate torque varying by the cylinder, the disk is divided into 23 areas  
from the innermost to the outermost circumference and the compensating value is  
measured at the measuring cylinder on each area at factory calibration. The  
measured values are stored in the SA cylinder. In the self-calibration, the  
compensating value is updated using the value in the SA cylinder.  
(2) Compensating open loop gain  
Torque constant value of the VCM has a dispersion for each drive, and varies  
depending on the cylinder that the head is positioned. To realize the high speed  
seek operation, the value that compensates torque constant value change and loop  
gain change of the whole servo system due to temperature change is measured and  
stored.  
For sensing, the firmware mixes the disturbance signal to the position signal at the  
state that the head is positioned to any cylinder. The firmware calculates the loop  
gain from the position signal and stores the compensation value against to the  
target gain as ratio.  
For compensating, the direction current value to the power amplifier is multiplied  
by the compensation value. By this compensation, loop gain becomes constant  
value and the stable servo control is realized.  
To compensate torque constant value change depending on cylinder, whole  
cylinders from most inner to most outer cylinder are divided into 16 partitions at  
calibration in the factory, and the compensation data is measured for  
representative cylinder of each partition. This measured value is stored in the SA  
area. The compensation value at self-calibration is calculated using the value in  
the SA area.  
4.5.2 Execution timing of self-calibration  
Self-calibration is executed when:  
The power is turned on.  
The disk drive receives the RECALIBRATE command from the host.  
The self-calibration execution timechart of the disk drive specifies self-  
calibration.  
The disk drive performs self-calibration according to the timechart based on the  
time elapsed from power-on. The timechart is shown in Table 4.1. After power-  
on, self-calibration is performed about every five or ten or fifteen minutes for the  
first 60 minutes or six RECALIBRATE command executions, and about every 30  
minutes after that.  
C141-E104-02EN  
4-9  
Theory of Device Operation  
Table 4.1 Self-calibration execution timechart  
Time elapsed  
Time elapsed  
(accumulated)  
1
2
3
4
5
6
7
At power-on  
Initial calibration  
About 5 minutes  
About 5 minutes  
About 10 minutes  
About 10 minutes  
About 15 minutes  
About 15 minutes  
About 5 minutes  
About 10 minutes  
About 20 minutes  
About 30 minutes  
About 45 minutes  
About 60 minutes  
8
.
.
.
Every about 30  
minutes  
.
4.5.3 Command processing during self-calibration  
If the disk drive receives a command execution request from the host while  
executing self-calibration according to the timechart, the disk drive terminates  
self-calibration and starts executing the command precedingly. In other words, if  
a disk read or write service is necessary, the disk drive positions the head to the  
track requested by the host, reads or writes data, and restarts calibration.  
This enables the host to execute the command without waiting for a long time,  
even when the disk drive is performing self-calibration. The command execution  
wait time is about maximum 40 ms.  
4.6 Read/write Circuit  
The read/write circuit consists of the read/write preamplifier (HDIC), the write  
circuit, the read circuit, and the time base generator in the read channel (RDC).  
Figure 4.4 is a block diagram of the read/write circuit.  
4.6.1 Read/write preamplifier (HDIC)  
HDIC equips a read preamplifier and a write current switch, that sets the bias  
current to the MR device and the current in writing. Each channel is connected to  
each data head, and HDIC switches channel by serial I/O. HDIC generates a write  
unsafe signal (WUS) when a write error occurs due to head short-circuits or head  
disconnection, that avoids error writing.  
4-10  
C141-E104-02EN  
4.6 Read/write Circuit  
4.6.2 Write circuit  
The write data is output from the hard disk controller (HDC) with the NRZ data  
format, and sent to the encoder circuit in the RDC. The NRZ write data is  
converted from 16-bit data to 17-bit data by the encoder circuit then sent to the  
HDIC, and the data is written onto the media.  
(1) 16/17 MTR MEEPRML  
This device converts data using the 16/17 MTR (Maximum Transitions Run  
Length Limited) algorithm.  
This code is converted so that a maximum of three 1’s are placed continuously  
and so that there are two or fewer 1’s in a 17-bit border.  
(2) Write precompensation  
Write precompensation compensates, during a write process, for write non-  
linearity generated at reading. Table 4.2 shows the write precompensation  
algorithm.  
Table 4.2 Write precompensation algorithm  
Bits  
Compensation  
111001  
111010  
:
–7  
–6  
111111  
000000  
000001  
:
–1  
±0  
+1  
010000  
:
+16  
+32  
100000  
C141-E104-02EN  
4-11  
Theory of Device Operation  
HDIC  
WDX/WDY  
RDX/RDY  
SD  
SC  
SE  
RDC  
Serial I/O  
Registers  
AGC  
Amplifier  
Write  
PreCompen-  
sation  
Digital  
PLL  
Programmable  
Filter  
Flash  
Digitizer  
ServoPulse  
Detector  
MEEPR  
Viterbi  
Detect  
16/17  
ENDEC  
Position  
A/B/C/D  
(to reg)  
WTGATE  
REFCLK  
DATA RWCLK SRV_CLK  
[7:0]  
RDGATE  
SRV_OUT[1:0]  
Figure 4.5 Read/write circuit block diagram  
4-12  
C141-E104-02EN  
4.6 Read/write Circuit  
4.6.3 Read circuit  
The head read signal from the PreAMP is regulated by the automatic gain control  
(AGC) circuit. Then the output is converted into the sampled read data pulse by  
the programmable filter circuit and the flash digitizer circuit. This clock signal is  
converted into the NRZ data by the 16/17 GCR decoder circuit based on the read  
data maximum-likelihood-detected by the Viterbi detection circuit, then is sent to  
the HDC.  
(1) AGC circuit  
The AGC circuit automatically regulates the output amplitude to a constant value  
even when the input amplitude level fluctuates. The AGC amplifier output is  
maintained at a constant level even when the head output fluctuates due to the  
head characteristics or outer/inner head positions.  
(2) Programmable filter circuit  
The programmable filter circuit has a low-pass filter function that eliminates  
unnecessary high frequency noise component and a high frequency boost-up  
function that equalizes the waveform of the read signal.  
Cut-off frequency of the low-pass filter and boost-up gain are controlled from the  
register in read channel by an instruction of the serial data signal from MPU  
(M5). The MPU optimizes the cut-off frequency and boost-up gain according to  
the transfer frequency of each zone.  
Figure 4.6 shows the frequency characteristic sample of the programmable filter.  
-3 dB  
Figure 4.6 Frequency characteristic of programmable filter  
C141-E104-02EN  
4-13  
Theory of Device Operation  
(3) Flash digitizer circuit  
This circuit is 10-tap sampled analog transversal filter circuit that cosine-equalizes  
the head read signal to the Modified Extended Partial Response (MEEPR)  
waveform.  
(4) Viterbi detection circuit  
The sample hold waveform output from the flash digitizer circuit is sent to the  
Viterbi detection circuit. The Viterbi detection circuit demodulates data  
according to the survivor path sequence.  
(5) MEEPRM  
This circuit converts the 17-bit read data into the 16-bit NRZ data.  
4.6.4 Digital PLL circuit  
The drive uses constant density recording to increase total capacity. This is  
different from the conventional method of recording data with a fixed data  
transfer rate at all data area. In the constant density recording method, data area is  
divided into zones by radius and the data transfer rate is set so that the recording  
density of the inner cylinder of each zone is nearly constant. The drive divides  
data area into 15 zones to set the data transfer rate.  
The MPU transfers the data transfer rate setup data (SD/SC) to the RDC that  
includes the Digital PLL circuit to change the data transfer rate.  
4-14  
C141-E104-02EN  
4.6 Read/write Circuit  
4.7 Servo Control  
The actuator motor and the spindle motor are submitted to servo control. The  
actuator motor is controlled for moving and positioning the head to the track  
containing the desired data. To turn the disk at a constant velocity, the actuator  
motor is controlled according to the servo data that is written on the data side  
beforehand.  
4.7.1 Servo control circuit  
Figure 4.7 is the block diagram of the servo control circuit. The following  
describes the functions of the blocks:  
(1)  
MPU  
SVC  
(3)  
(4)  
Power  
(2)  
Servo  
burst  
capture  
DAC  
Head  
DSP  
unit  
VCM current  
Amp  
CSR  
(7)  
Position Sense  
VCM  
(5)  
(6)  
Driver  
Spindle  
motor  
control  
Spindle  
motor  
CSR: Current Sense Resister  
VCM: Voice Coil Motor  
Figure 4.7 Block diagram of servo control circuit  
(1) Microprocessor unit (MPU)  
The MPU includes the DSP unit, and the MPU starts the spindle motor, moves the  
heads to the reference cylinders, seeks the specified cylinder, and executes  
calibration according to the internal operations of the MPU. Main internal  
operation of the MPU are shown below.  
C141-E104-02EN  
4-15  
Theory of Device Operation  
The major internal operations are listed below.  
a. Spindle motor start  
Starts the spindle motor and accelerates it to normal speed when power is  
applied.  
b. Move head to reference cylinder  
Drives the VCM to position the head at the any cylinder in the data area. The  
logical initial cylinder is at the outermost circumference (cylinder 0).  
c. Seek to specified cylinder  
Drives the VCM to position the head to the specified cylinder.  
d. Calibration  
Senses and stores the thermal offset between heads and the mechanical forces  
on the actuator, and stores the calibration value.  
4-16  
C141-E104-02EN  
4.7 Servo Control  
(2) Servo burst capture circuit  
The servo burst capture circuit reproduces signals (position signals) that indicate  
the head position from the servo data on the data surface. SERVO A, SERVO B,  
SERVO C and SERVO D burst signals shown in Figure 4.9 followed the servo  
mark, cylinder gray and index information are output from the servo area on the  
data surface via the data head. The servo signals do A/D-convert by Fourier-  
demodulator in the servo burst capture circuit. At that time the AGC circuit is in  
hold mode. The A/D converted data is recognized by the MPU as position  
information with A-B and C-D processed.  
(3) D/A converter (DAC)  
The D/A converter (DAC) converts the VCM drive current value (digital value)  
calculated by the DSP unit into analog values and transfers them to the power  
amplifier.  
(4) Power amplifier  
The power amplifier feeds currents, corresponding to the DAC output signal  
voltage to the VCM.  
(5) Spindle motor control circuit  
The spindle motor control circuit controls the sensor-less spindle motor. This  
circuit detects number of revolution of the motor by the interrupt generated  
periodically, compares with the target revolution speed, then flows the current  
into the motor coil according to the differentiation (aberration).  
(6) Driver circuit  
The driver circuit is a power amplitude circuit that receives signals from the  
spindle motor control circuit and feeds currents to the spindle motor.  
(7) VCM current sense resistor (CSR)  
This resistor controls current at the power amplifier by converting the VCM  
current into voltage and feeding back.  
C141-E104-02EN  
4-17  
Theory of Device Operation  
4.7.2 Data-surface servo format  
Figure 4.8 describes the physical layout of the servo frame. The three areas  
indicated by (1) to (3) in Figure 4.8 are described below.  
(1) Inner guard band  
The head is in contact with the disk in this space when the spindle starts turning  
or stops, and the rotational speed of the spindle can be controlled on this cylinder  
area for head moving.  
(2) Data area  
This area is used as the user data area SA area.  
(3) Outer guard band  
This area is located at outer position of the user data area, and the rotational speed  
of the spindle can be controlled on this cylinder area for head moving.  
4-18  
C141-E104-02EN  
4.7 Servo Control  
Servo frame  
(66 servo frames per revolution)  
OGB  
IGB  
Data area  
expand  
CYLn  
CYLn – 1 (n: even number)  
CYLn + 1  
!" Diameter  
direction  
W/R Recovery  
Servo Mark  
Gray Code  
W/R Recovery  
Servo Mark  
Gray Code  
W/R Recovery  
Servo Mark  
Gray Code  
#
#
Erase  
Servo B  
Servo C  
Erase  
Servo A  
Erase  
Erase  
Servo B  
Servo A  
Erase  
Circumference  
Direction  
Erase  
Servo D  
PAD  
Servo C  
Erase  
Erase: DC erase  
area  
Figure 4.8 Physical sector servo configuration on disk surface  
C141-E104-02EN  
4-19  
Theory of Device Operation  
4.7.3 Servo frame format  
As the servo information, the IDD uses the two-phase servo generated from the  
gray code and servo A to D. This servo information is used for positioning  
operation of radius direction and position detection of circumstance direction.  
The servo frame consists of 6 blocks; write/read recovery, servo mark, gray code,  
servo A to D, and PAD. Figure 4.9 shows the servo frame format.  
Figure 4.9 Servo frame format  
4-20  
C141-E104-02EN  
4.7 Servo Control  
(1) Write/read recovery  
This area is used to absorb the write/read transient and to stabilize the AGC.  
(2) Servo mark  
This area generates a timing for demodulating the gray code and position-  
demodulating the servo A to D by detecting the servo mark.  
(3) Gray code (including index bit)  
This area is used as cylinder address. The data in this area is converted into the  
binary data by the gray code demodulation circuit  
(4) Servo A, servo B, servo C, servo D  
This area is used as position signals between tracks and the IDD control at on-  
track so that servo A level equals to servo B level.  
(5) PAD  
This area is used as a gap between servo and data.  
4.7.4 Actuator motor control  
The voice coil motor (VCM) is controlled by feeding back the servo data recorded  
on the data surface. The MPU fetches the position sense data on the servo frame  
at a constant interval of sampling time, executes calculation, and updates the  
VCM drive current.  
The servo control of the actuator includes the operation to move the head to the  
reference cylinder, the seek operation to move the head to the target cylinder to  
read or write data, and the track-following operation to position the head onto the  
target track.  
(1) Operation to move the head to the reference cylinder  
The MPU moves the head to the reference cylinder when the power is turned.  
The reference cylinder is in the data area.  
When power is applied the heads are moved from the inner circumference shunt  
zone to the normal servo data zone in the following sequence:  
a) Micro current is fed to the VCM to press the head against the inner  
circumference.  
b) Micro current is fed to the VCM to move the head toward the outer  
circumference.  
c) When the servo mark is detected the head is moved slowly toward the outer  
circumference at a constant speed.  
C141-E104-02EN  
4-21  
Theory of Device Operation  
d) If the head is stopped at the reference cylinder from there. Track following  
control starts.  
(2) Seek operation  
Upon a data read/write request from the host, the MPU confirms the necessity of  
access to the disk. If a read/write instruction is issued, the MPU seeks the desired  
track.  
The MPU feeds the VCM current via the D/A converter and power amplifier to  
move the head. The MPU calculates the difference (speed error) between the  
specified target position and the current position for each sampling timing during  
head moving. The MPU then feeds the VCM drive current by setting the  
calculated result into the D/A converter. The calculation is digitally executed by  
the firmware. When the head arrives at the target cylinder, the track is followed.  
(3) Track following operation  
Except during head movement to the reference cylinder and seek operation under  
the spindle rotates in steady speed, the MPU does track following control. To  
position the head at the center of a track, the DSP drives the VCM by feeding  
micro current. For each sampling time, the VCM drive current is determined by  
filtering the position difference between the target position and the position  
clarified by the detected position sense data. The filtering includes servo  
compensation. These are digitally controlled by the firmware.  
4.7.5 Spindle motor control  
Hall-less three-phase twelve-pole motor is used for the spindle motor, and the 3-  
phase full/half-wave analog current control circuit is used as the spindle motor  
driver (called SVC hereafter). The firmware operates on the MPU manufactured  
by Fujitsu. The spindle motor is controlled by sending several signals from the  
MPU to the SVC. There are three modes for the spindle control; start mode,  
acceleration mode, and stable rotation mode.  
(1) Start mode  
When power is supplied, the spindle motor is started in the following sequence:  
a) After the power is turned on, the MPU sends a signal to the SVC to charge  
the charge pump capacitor of the SVC. The charged amount defines the  
current that flows in the spindle motor.  
b) When the charge pump capacitor is charged enough, the MPU sets the SVC  
to the motor start mode. Then, a current (approx. 0.7 A) flows into the  
spindle motor.  
c) The SVC generates a phase switching signal by itself, and changes the phase  
of the current flowed in the motor in the order of (V-phase to U-phase), (W-  
phase to U-phase), (W-phase to V-phase), (U-phase to V-phase), (U-phase to  
W-phase), and (V-phase to W-phase) (after that, repeating this order).  
4-22  
C141-E104-02EN  
4.7 Servo Control  
d) During phase switching, the spindle motor starts rotating in low speed, and  
generates a counter electromotive force. The SVC detects this counter  
electromotive force and reports to the MPU using a PHASE signal for speed  
detection.  
e) The MPU is waiting for a PHASE signal. When no phase signal is sent for a  
specific period, the MPU resets the SVC and starts from the beginning.  
When a PHASE signal is sent, the SVC enters the acceleration mode.  
(2) Acceleration mode  
In this mode, the MPU stops to send the phase switching signal to the SVC. The  
SVC starts a phase switching by itself based on the counter electromotive force.  
Then, rotation of the spindle motor accelerates. The MPU calculates a rotational  
speed of the spindle motor based on the PHASE signal from the SVC, and waits  
till the rotational speed reaches 4,200 rpm. When the rotational speed reaches  
4,200 rpm, the SVC enters the stable rotation mode.  
(3) Stable rotation mode  
The SVC calculates a time for one revolution of the spindle motor based on the  
PHASE signal. The MPU takes a difference between the current time and a time  
for one revolution at 4,200 rpm that the MPU already recognized. Then, the MPU  
keeps the rotational speed to 4,200 rpm by charging or discharging the charge  
pump for the different time. For example, when the actual rotational speed is  
4,000 rpm, the time for one revolution is 15.000 ms. And the time for one  
revolution at 4,200 rpm is 14.286 ms. Therefore, the MPU charges the charge  
pump for 0.714 ms × k (k: constant value). This makes the flowed current into the  
motor higher and the rotational speed up. When the actual rotational speed is  
faster than 4,200 rpm, the MPU discharges the pump the other way. This control  
(charging/discharging) is performed every 1 revolution.  
C141-E104-02EN  
4-23  
This page is intentionally left blank.  
CHAPTER 5 Interface  
5.1  
5.2  
5.3  
5.4  
5.5  
5.6  
Physical Interface  
Logical Interface  
Host Commands  
Command Protocol  
Ultra DMA Feature Set  
Timing  
This chapter gives details about the interface, and the interface commands and  
timings.  
C141-E104-02EN  
5-1  
Interface  
5.1 Physical Interface  
5.1.1 Interface signals  
Figure 5.1 shows the interface signals.  
Host  
IDD  
DATA 0-15: DATA BUS  
DMACK-: DMA ACKNOWLEDGE  
DMARQ: DMA REQUEST  
INTRO: INTERRUPT REQUEST  
DIOW-: I/O WRITE  
STOP: STOP DURING ULTRA DMA DATA BURSTS  
DIOR-:I/O READ  
HDMARDY:DMA READY DURING ULTRA DMA DATA IN BURSTS  
HSTROBE:DATA STROBE DURING ULTRA DMA DATA OUT BURST  
PDIAG-: PASSED DIAGNOSTICS  
CBLID-: CABLE TYPE IDENTIFIER  
DASP-: DEVICE ACTIVE/SLAVE PRESENT  
IORDY:I/O READY  
DDMARDY:DMA READY DURING ULTRA DMA DATA OUT BURSTS  
DSTROBE: DATA STROBE DURING ULTRA DMA DATA IN BURSTS  
DA 0-2: DEVICE ADDRESS  
CS0-: CHIP SELECT 0  
CS1-: CHIP SELECT 1  
RESET-: RESET  
CSEL: CABLE SELECT  
MSTR: Master  
ENCSEL: ENABLE CSEL  
+5V DC: +5 volt  
GND: GROUND  
Figure 5.1  
Interface signals  
5-2  
C141-E104-02EN  
5.1 Physical Interface  
5.1.2 Signal assignment on the connector  
Table 5.1 shows the signal assignment on the interface connector.  
Table 5.1 Signal assignment on the interface connector  
Pin No.  
Signal  
Pin No.  
Signal  
A
C
MSTR  
B
D
MSTR/ENCSEL  
ENCSEL  
(KEY)  
unused  
(KEY)  
E
F
1
RESET–  
DATA7  
DATA6  
DATA5  
DATA4  
DATA3  
DATA2  
DATA1  
DATA0  
GND  
2
GND  
3
4
DATA8  
DATA9  
DATA10  
DATA11  
DATA12  
DATA13  
DATA14  
DATA15  
(KEY)  
5
6
7
8
9
10  
12  
14  
16  
18  
20  
22  
24  
26  
11  
13  
15  
17  
19  
21  
23  
25  
DMARQ  
GND  
DIOW-, STOP  
GND  
DIOR-, HDMRDY,  
HSTROBE  
GND  
27  
IORDY, DDMARDY,  
DSTROBE  
28  
CSEL  
29  
31  
33  
35  
37  
39  
41  
43  
DMACK–  
INTRQ  
DA1  
30  
32  
34  
36  
38  
40  
42  
44  
GND  
reserved (IOCS16-)  
PDIAG–, CBLID–  
DA2  
DA0  
CS0–  
CS1–  
DASP–  
+5 VDC  
GND  
GND  
+5 VDC  
unused  
C141-E104-02EN  
5-3  
Interface  
[signal]  
[I/O]  
I
[Description]  
ENCSEL  
This signal is used to set master/slave using the CSEL signal (pin 28).  
Pins B and D  
Open: Sets master/slave using the CSEL signal  
is disabled.  
Short: Sets master/slave using the CSEL signal  
is enabled.  
MSTR-  
I
I
MSTR, I, Master/slave setting  
Pin A, B, C, D open: Master setting  
Pin A, B Short:  
Slave setting  
RESET-  
DATA 0-15  
DIOW-  
Reset signal from the host. This signal is low active and is  
asserted for a minimum of 25 µs during power on.  
I/O Sixteen-bit bi-directional data bus between the host and the  
device. These signals are used for data transfer  
I
Signal asserted by the host to write to the device register or data  
port.  
STOP  
I
DIOW- must be negated by the host before starting the Ultra  
DMA transfer. The STOP signal must be negated by the host  
before data is transferred during the Ultra DMA transfer. During  
data transfer in Ultra DMA mode, the assertion of the STOP  
signal asserted by the host later indicates that the transfer has been  
suspended.  
DIOR-  
I
I
Read strobe signal from the host to read the device register or data  
port  
HDMARDY-  
Flow control signal for Ultra DMA data In transfer (READ DMA  
command). This signal is asserted by the host to inform the  
device that the host is ready to receive the Ultra DMA data In  
transfer. The host can negate the HDMARDY- signal to suspend  
the Ultra DMA data In transfer.  
HSTROBE  
INTRQ  
I
Data Out Strobe signal from the host during Ultra DMA data Out  
transfer (WRITE DMA command). Both the rising and falling  
edges of the HSTROBE signal latch data from Data 15-0 into the  
device. The host can suspend the inversion of the HSTROBE  
signal to suspend the Ultra DMA data Out transfer.  
O
Interrupt signal to the host.  
This signal is negated in the following cases:  
assertion of RESET- signal  
Reset by SRST of the Device Control register  
Write to the command register by the host  
Read of the status register by the host  
Completion of sector data transfer  
(without reading the Status register)  
The signal output line has a high impedance when no devices are  
selected or interruption is disabled.  
5-4  
C141-E104-02EN  
5.1 Physical Interface  
[signal]  
CS0-  
[I/O]  
I
[Description]  
Chip select signal decoded from the host address bus. This signal  
is used by the host to select the command block registers.  
CS1-  
I
I
-
Chip select signal decoded from the host address bus. This signal  
is used by the host to select the control block registers.  
DA 0-2  
Binary decoded address signals asserted by the host to access task  
file registers.  
KEY  
Key pin for prevention of erroneous connector insertion  
PDIAG-  
I/O This signal is an input mode for the master device and an output  
mode for the slave device in a daisy chain configuration. This  
signal indicates that the slave device has been completed self  
diagnostics.  
This signal is pulled up to +5 V through 10 kresistor at each device.  
CBLID-  
DASP-  
IORDY  
I/O This signal is used to detect the type of cable installed in the  
system.  
This signal is pulled up to +5 V through 10 kresistor at each device.  
I/O This is a time-multiplexed signal that indicates that the device is  
active and a slave device is present.  
This signal is pulled up to +5 V through 10 kresistor at each device.  
O
O
This signal requests the host system to delay the transfer cycle  
when the device is not ready to respond to a data transfer request  
from the host system.  
DDMARDY  
-
Flow control signal for Ultra DMA data Out transfer (WRITE  
DMA command). This signal is asserted by the device to inform  
the host that the device is ready to receive the Ultra DMA data  
Out transfer. The device can negate the DDMARDY- signal to  
suspend the Ultra DMA data Out transfer.  
DSTROBE  
CSEL  
O
I
Data In Strobe signal from the device during Ultra DMA data In  
transfer. Both the rising and falling edges of the DSTROBE  
signal latch data from Data 15-0 into the host. The device can  
suspend the inversion of the DSTROBE signal to suspend the  
Ultra DMA data In transfer.  
This signal to configure the device as a master or a slave device.  
When CSEL signal is grounded, the IDD is a master device.  
When CSEL signal is open, the IDD is a slave device.  
This signal is pulled up with 240 kresistor at each device.  
DMACK-  
I
The host system asserts this signal as a response that the host  
system receive data or to indicate that data is valid.  
C141-E104-02EN  
5-5  
Interface  
[signal]  
[I/O]  
O
[Description]  
DMARQ  
This signal is used for DMA transfer between the host system and  
the device. The device asserts this signal when the device  
completes the preparation of DMA data transfer to the host system  
(at reading) or from the host system (at writing).  
The direction of data transfer is controlled by the DIOR and  
DIOW signals. This signal hand shakes with the DMACK-signal.  
In other words, the device negates the DMARQ signal after the  
host system asserts the DMACK signal. When there is other data  
to be transferred, the device asserts the DMARQ signal again.  
When the DMA data transfer is performed, IOCS16-, CS0- and  
CS1- signals are not asserted. The DMA data transfer is a 16-bit  
data transfer.  
+5 VDC  
GND  
I
-
+5 VDC power supplying to the device.  
Grounded signal at each signal wire.  
Note:  
“I” indicates input signal from the host to the device.  
“O” indicates output signal from the device to the host.  
“I/O” indicates common output or bi-directional signal between the host  
and the device.  
5.2 Logical Interface  
The device can operate for command execution in either address-specified mode;  
cylinder-head-sector (CHS) or Logical block address (LBA) mode. The  
IDENTIFY DEVICE information indicates whether the device supports the LBA  
mode. When the host system specifies the LBA mode by setting bit 6 in the  
Device/Head register to 1, HS3 to HS0 bits of the Device/Head register indicates  
the head No. under the LBA mode, and all bits of the Cylinder High, Cylinder  
Low, and Sector Number registers are LBA bits.  
The sector No. under the LBA mode proceeds in the ascending order with the start  
point of LBA0 (defined as follows).  
LBA0 = [Cylinder 0, Head 0, Sector 1]  
Even if the host system changes the assignment of the CHS mode by the  
INITIALIZE DEVICE PARAMETER command, the sector LBA address is not  
changed.  
LBA = [((Cylinder No.) × (Number of head) + (Head No.)) × (Number of  
sector/track)] + (Sector No.) 1  
5-6  
C141-E104-02EN  
5.2 Logical Interface  
5.2.1 I/O registers  
Communication between the host system and the device is done through input-  
output (I/O) registers of the device.  
These I/O registers can be selected by the coded signals, CS0-, CS1-, and DA0 to  
DA2 from the host system. Table 5.2. shows the coding address and the function  
of I/O registers.  
Table 5.2 I/O registers  
I/O registers  
Host I/O  
address  
CS0– CS1–  
DA2  
DA1  
DA0  
Read operation Write operation  
Command block registers  
L
L
L
L
L
L
L
L
L
H
H
H
H
H
H
H
H
L
L
L
L
L
L
H
L
Data  
Data  
X’1F0’  
X’1F1’  
X’1F2’  
X’1F3’  
X’1F4’  
X’1F5’  
X’1F6’  
X’1F7’  
Error Register  
Sector Count  
Sector Number  
Cylinder Low  
Cylinder High  
Device/Head  
Status  
Features  
L
H
H
L
Sector Count  
Sector Number  
Cylinder Low  
Cylinder High  
Device/Head  
Command  
(Invalid)  
L
H
L
H
H
H
H
X
L
H
L
H
H
X
H
X
(Invalid)  
Control block registers  
H
H
L
L
H
H
H
H
L
Alternate Status Device Control  
X’3F6’  
X’3F7’  
H
Notes:  
1.  
The Data register for read or write operation can be accessed by 16 bit data  
bus (DATA0 to DATA15).  
2.  
The registers for read or write operation other than the Data registers can be  
accessed by 8 bit data bus (DATA0 to DATA7).  
3.  
4.  
When reading the Drive Address register, bit 7 is high-impedance state.  
H indicates signal level High and L indicates signal level Low.  
And the LBA mode is specified, the Device/Head, Cylinder High, Cylinder  
Low, and Sector Number registers indicate LBA bits 27 to 24, 23 to 16, 15  
to 8, and 7 to 0.  
C141-E104-02EN  
5-7  
Interface  
5.2.2 Command block registers  
(1) Data register (X’1F0’)  
The Data register is a 16-bit register for data block transfer between the device  
and the host system. Data transfer mode is PIO or DMA mode.  
(2) Error register (X’1F1’)  
The Error register indicates the status of the command executed by the device.  
The contents of this register are valid when the ERR bit of the Status register is 1.  
This register contains a diagnostic code after power is turned on, a reset , or the  
EXECUTIVE DEVICE DIAGNOSTIC command is executed.  
[Status at the completion of command execution other than diagnostic command]  
Bit 7  
Bit 6  
UNC  
Bit 5  
X
Bit 4  
Bit 3  
X
Bit 2  
Bit 1  
Bit 0  
ICRC  
IDNF  
ABRT TK0NF AMNF  
X: Unused  
- Bit 7: Interface CRC Error (ICRC). This bit indicates that a CRC error  
occurred during Ultra DMA transfer.  
- Bit 6: Uncorrectable Data Error (UNC). This bit indicates that an  
uncorrectable data error has been encountered.  
- Bit 5: Unused  
- Bit 4: ID Not Found (IDNF). This bit indicates an error except for bad  
sector, uncorrectable error and SB not found.  
- Bit 3: Unused  
- Bit 2: Aborted Command (ABRT). This bit indicates that the requested  
command was aborted due to a device status error (e.g. Not Ready,  
Write Fault) or the command code was invalid.  
- Bit 1: Track 0 Not Found (TK0NF). This bit indicates that track 0 was not  
found during RECALIBRATE command execution.  
- Bit 0: Address Mark Not Found (AMNF). This bit indicates that the SB Not  
Found error occurred.  
5-8  
C141-E104-02EN  
5.2 Logical Interface  
[Diagnostic code]  
X’01’: No Error Detected.  
X’02’: HDC Register Compare Error  
X’03’: Data Buffer Compare Error.  
X’05’: ROM Sum Check Error.  
X’80’: Device 1 (slave device) Failed.  
Error register of the master device is valid under two devices (master  
and slave) configuration. If the slave device fails, the master device  
posts X’80’ OR (the diagnostic code) with its own status (X’01’ to  
X’05’).  
However, when the host system selects the slave device, the diagnostic  
code of the slave device is posted.  
(3) Features register (X’1F1’)  
The Features register provides specific feature to a command. For instance, it is  
used with SET FEATURES command to enable or disable caching.  
(4) Sector Count register (X’1F2’)  
The Sector Count register indicates the number of sectors of data to be transferred  
in a read or write operation between the host system and the device. When the  
value in this register is X’00’, the sector count is 256.  
When this register indicates X’00’ at the completion of the command execution,  
this indicates that the command is completed successfully. If the command is not  
completed successfully, this register indicates the number of sectors to be  
transferred to complete the request from the host system. That is, this register  
indicates the number of remaining sectors that the data has not been transferred  
due to the error.  
The contents of this register has other definition for the following commands;  
INITIALIZE DEVICE PARAMETERS, SET FEATURES, IDLE, STANDBY and  
SET MULTIPLE MODE.  
(5) Sector Number register (X’1F3’)  
The contents of this register indicates the starting sector number for the  
subsequent command. The sector number should be between X’01’ and [the  
number of sectors per track defined by INITIALIZE DEVICE PARAMETERS  
command.  
Under the LBA mode, this register indicates LBA bits 7 to 0.  
C141-E104-02EN  
5-9  
Interface  
(6) Cylinder Low register (X’1F4’)  
The contents of this register indicates low-order 8 bits of the starting cylinder  
address for any disk-access.  
At the end of a command, the contents of this register are updated to the current  
cylinder number.  
Under the LBA mode, this register indicates LBA bits 15 to 8.  
(7) Cylinder High register (X’1F5’)  
The contents of this register indicates high-order 8 bits of the disk-access start  
cylinder address.  
At the end of a command, the contents of this register are updated to the current  
cylinder number. The high-order 8 bits of the cylinder address are set to the  
Cylinder High register.  
Under the LBA mode, this register indicates LBA bits 23 to 16.  
(8) Device/Head register (X’1F6’)  
The contents of this register indicate the device and the head number.  
When executing INITIALIZE DEVICE PARAMETERS command, the contents  
of this register defines “the number of heads minus 1” (a maximum head No.).  
Bit 7  
X
Bit 6  
L
Bit 5  
X
Bit 4  
DEV  
Bit 3  
HS3  
Bit 2  
HS2  
Bit 1  
HS1  
Bit 0  
HS0  
- Bit 7: Unused  
- Bit 6: L. 0 for CHS mode and 1 for LBA mode.  
- Bit 5: Unused  
- Bit 4: DEV bit. 0 for the master device and 1 for the slave device.  
- Bit 3: HS3 CHS mode head address 3 (23). LBA bit 27.  
- Bit 2: HS2 CHS mode head address 2 (22). LBA bit 26.  
- Bit 1: HS1 CHS mode head address 1 (21). LBA bit 25.  
- Bit 0: HS0 CHS mode head address 0 (20). LBA bit 24.  
5-10  
C141-E104-02EN  
5.2 Logical Interface  
(9) Status register (X’1F7’)  
The contents of this register indicate the status of the device. The contents of this  
register are updated at the completion of each command. When the BSY bit is  
cleared, other bits in this register should be validated within 400 ns. When the  
BSY bit is 1, other bits of this register are invalid. When the host system reads  
this register while an interrupt is pending, it is considered to be the Interrupt  
Acknowledge (the host system acknowledges the interrupt). Any pending  
interrupt is cleared (negating INTRQ signal) whenever this register is read.  
Bit 7  
BSY  
Bit 6  
Bit 5  
DF  
Bit 4  
DSC  
Bit 3  
DRQ  
Bit 2  
0
Bit 1  
0
Bit 0  
ERR  
DRDY  
- Bit 7: Busy (BSY) bit. This bit is set whenever the Command register is  
accessed. Then this bit is cleared when the command is completed.  
However, even if a command is being executed, this bit is 0 while data  
transfer is being requested (DRQ bit = 1).When BSY bit is 1, the host  
system should not write the command block registers. If the host  
system reads any command block register when BSY bit is 1, the  
contents of the Status register are posted. This bit is set by the device  
under following conditions:  
(a) Within 400 ns after RESET- is negated or SRST is set in the  
Device Control register, the BSY bit is set. the BSY bit is cleared,  
when the reset process is completed.  
The BSY bit is set for no longer than 15 seconds after the IDD  
accepts reset.  
(b) Within 400 ns from the host system starts writing to the  
Command register.  
(c) Within 5 µs following transfer of 512 bytes data during execution  
of the READ SECTOR(S), WRITE SECTOR(S), or WRITE  
BUFFER command.  
Within 5 µs following transfer of 512 bytes of data and the  
appropriate number of ECC bytes during execution of READ  
LONG or WRITE LONG command.  
- Bit 6: Device Ready (DRDY) bit. This bit indicates that the device is  
capable to respond to a command.  
The IDD checks its status when it receives a command. If an error is  
detected (not ready state), the IDD clears this bit to 0. This is cleared  
to 0 at power-on and it is cleared until the rotational speed of the  
spindle motor reaches the steady speed.  
C141-E104-02EN  
5-11  
Interface  
- Bit 5: The Device Write Fault (DF) bit. This bit indicates that a device fault  
(write fault) condition has been detected.  
If a write fault is detected during command execution, this bit is  
latched and retained until the device accepts the next command or  
reset.  
- Bit 4: Device Seek Complete (DSC) bit. This bit indicates that the device  
heads are positioned over a track.  
In the IDD, this bit is always set to 1 after the spin-up control is  
completed.  
- Bit 3: Data Request (DRQ) bit. This bit indicates that the device is ready to  
transfer data of word unit or byte unit between the host system and the  
device.  
- Bit 2: Always 0.  
- Bit 1: Always 0.  
- Bit 0: Error (ERR) bit. This bit indicates that an error was detected while the  
previous command was being executed. The Error register indicates  
the additional information of the cause for the error.  
(10) Command register (X’1F7’)  
The Command register contains a command code being sent to the device. After  
this register is written, the command execution starts immediately.  
Table 5.3 lists the executable commands and their command codes. This table  
also lists the necessary parameters for each command which are written to certain  
registers before the Command register is written.  
5-12  
C141-E104-02EN  
5.3 Host Commands  
5.2.3 Control block registers  
(1) Alternate Status register (X’3F6’)  
The Alternate Status register contains the same information as the Status register  
of the command block register.  
The only difference from the Status register is that a read of this register does not  
imply Interrupt Acknowledge and INTRQ signal is not reset.  
Bit 7  
BSY  
Bit 6  
Bit 5  
DF  
Bit 4  
DSC  
Bit 3  
DRQ  
Bit 2  
0
Bit 1  
0
Bit 0  
ERR  
DRDY  
(2) Device Control register (X’3F6’)  
The Device Control register contains device interrupt and software reset.  
Bit 7  
X
Bit 6  
X
Bit 5  
X
Bit 4  
X
Bit 3  
X
Bit 2  
Bit 1  
nIEN  
Bit 0  
0
SRST  
- Bit 2: SRST is the host software reset bit. When this bit is set, the device is  
held reset state. When two device are daisy chained on the interface,  
setting this bit resets both device simultaneously.  
The slave device is not required to execute the DASP- handshake.  
- Bit 1: nIEN bit enables an interrupt (INTRQ signal) from the device to the  
host. When this bit is 0 and the device is selected, an interruption  
(INTRQ signal) can be enabled through a tri-state buffer. When this  
bit is 1 or the device is not selected, the INTRQ signal is in the high-  
impedance state.  
5.3 Host Commands  
The host system issues a command to the device by writing necessary parameters  
in related registers in the command block and writing a command code in the  
Command register.  
The device can accept the command when the BSY bit is 0 (the device is not in  
the busy status).  
The host system can halt the uncompleted command execution only at execution  
of hardware or software reset.  
C141-E104-02EN  
5-13  
Interface  
When the BSY bit is 1 or the DRQ bit is 1 (the device is requesting the data  
transfer) and the host system writes to the command register, the correct device  
operation is not guaranteed.  
5.3.1 Command code and parameters  
Table 5.3 lists the supported commands, command code and the registers that  
needed parameters are written.  
Table 5.3 Command code and parameters (1 of 2)  
Command code (Bit)  
Parameters used  
Command name  
7
6
5
4
3
2
1
0
FR SC SN CY DH  
READ SECTOR(S)  
0
1
1
0
1
1
0
0
0
0
1
1
1
1
1
1
1
1
0
0
1
1
0
1
1
1
1
1
0
0
0
1
0
1
1
1
1
1
1
0
0
0
1
1
1
0
0
0
0
0
1
1
0
1
0
1
1
1
0
1
1
0
1
1
1
1
0
0
0
0
0
0
1
1
1
1
1
0
0
0
0
1
1
1
0
1
0
0
0
0
1
0
0
1
1
0
X
X
0
1
1
1
0
1
1
0
0
0
0
1
0
1
0
0
1
0
1
0
X
X
0
1
1
1
1
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
X
X
0
0
0
1
1
0
0
0
1
1
0
0
R
0
N
N
N
N
N
N
N
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
N
N
Y
N
N
N*  
Y
Y
N
N
Y
Y
N
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
N
N
N
N
N
Y
N
N
Y
Y
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
N
N
N
N
N
Y
N
N
Y
Y
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
D
Y
Y
D
D
D
D
Y
D
D*  
Y
Y
D
D
D
READ MULTIPLE  
READ DMA  
R
R
1
READ VERIFY SECTOR(S)  
WRITE MULTIPLE  
WRITE DMA  
R
0
WRITE VERIFY  
WRITE SECTOR(S)  
RECALIBRATE  
R
X
X
1
SEEK  
INITIALIZE DEVICE PARAMETERS  
IDENTIFY DEVICE  
IDENTIFY DEVICE DMA  
SET FEATURES  
0
0
1
SET MULTIPLE MODE  
SET MAX  
0
1
READ NATIVE MAX ADDRESS  
EXECUTE DEVICE DIAGNOSTIC  
READ LONG  
0
0
R
R
0
WRITE LONG  
READ BUFFER  
WRITE BUFFER  
0
IDLE  
1
1
0
1
0
1
1
0
0
0
1
0
1
1
1
1
5-14  
C141-E104-02EN  
5.3 Host Commands  
Table 5.3 Command code and parameters (2 of 2)  
Command code (Bit)  
Parameters used  
Command name  
7
6
5
4
3
2
1
0
FR SC SN CY DH  
IDLE IMMEDIATE  
1
1
0
1
0
1
1
0
0
0
1
0
0
0
1
1
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
D
D
D
D
D
STANDBY  
1
1
0
1
0
1
1
0
0
0
1
0
1
1
0
0
STANDBY IMMEDIATE  
SLEEP  
1
1
0
1
0
1
1
0
0
0
1
0
0
0
0
0
1
1
0
1
0
1
1
0
1
0
0
1
0
1
1
0
CHECK POWER MODE  
1
1
0
1
0
1
1
0
1
0
0
1
0
0
0
1
SMART  
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
1
0
1
1
0
0
1
0
1
1
0
0
0
1
1
0
0
1
0
1
1
0
1
Y
N
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
Y
N
N
N
N
N
N
N
D
D
D
D
D
D
D
D
SECURITY DISABLE PASSWORD  
SECURITY ERASE PREPARE  
SECURITY ERASE UNIT  
SECURITY FREEZE LOCK  
SECURITY SET PASSWORD  
SECURITY UNLOCK  
FLUSH CACHE  
Notes:  
FR: Features Register  
CY: Cylinder Registers  
SC: Sector Count Register  
DH: Drive/Head Register  
SN: Sector Number Register  
R:  
Retry at error  
1 = Without retry  
0 = With retry  
Y:  
Necessary to set parameters  
C141-E104-02EN  
5-15  
Interface  
Y*: Necessary to set parameters under the LBA mode.  
N: Not necessary to set parameters (The parameter is ignored if it is set.)  
N*: May set parameters  
D: The device parameter is valid, and the head parameter is ignored.  
D*: The command is addressed to the master device, but both the master device  
and the slave device execute it.  
X:  
Do not care  
5.3.2 Command descriptions  
The contents of the I/O registers to be necessary for issuing a command and the  
example indication of the I/O registers at command completion are shown as  
following in this subsection.  
Example: READ SECTOR(S)  
At command issuance (I/O registers setting contents)  
Bit  
7
6
5
4
0
3
0
2
0
1
0
0
0
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
0
x
0
1
x
L
DV Head No. / LBA [MSB]  
Start cylinder address [MSB] / LBA  
Start cylinder address [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
Bit  
7
6
5
4
3
2
1
0
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV Head No. / LBA [MSB]  
End cylinder address [MSB] / LBA  
End cylinder address [LSB] / LBA  
End sector No. / LBA [LSB]  
X’00’  
Error information  
5-16  
C141-E104-02EN  
5.3 Host Commands  
CM: Command register  
DH: Device/Head register  
FR: Features register  
ST: Status register  
CH: Cylinder High register ER: Error register  
CL: Cylinder Low register  
L: LBA (logical block address) setting bit  
SN: Sector Number register DV: Device address. bit  
SC: Sector Count register  
Note:  
x, xx: Do not care (no necessary to set)  
1.  
When the L bit is specified to 1, the lower 4 bits of the DH register and all  
bits of the CH, CL and SN registers indicate the LBA bits (bits of the DH  
register are the MSB (most significant bit) and bits of the SN register are  
the LSB (least significant bit).  
2.  
3.  
At error occurrence, the SC register indicates the remaining sector count of data  
transfer.  
In the table indicating I/O registers contents in this subsection, bit indication is  
omitted.  
(1) READ SECTOR(S) (X’20’ or X’21’)  
This command reads data of sectors specified in the Sector Count register from  
the address specified in the Device/Head, Cylinder High, Cylinder Low and  
Sector Number registers. Number of sectors can be specified to 256 sectors in  
maximum. To specify 256 sectors reading, ‘00’ is specified. For the DRQ,  
INTRQ, and BSY protocols related to data transfer, see Subsection 5.4.1.  
If the head is not on the track specified by the host, the device performs an  
implied seek. After the head reaches to the specified track, the device reads the  
target sector.  
If an error occurs, retry reads are attempted to read the target sector before  
reporting an error, irrespective of the R bit setting.  
The DRQ bit of the Status register is always set prior to the data transfer  
regardless of an error condition.  
Upon the completion of the command execution, command block registers contain  
the cylinder, head, and sector addresses (in the CHS mode) or logical block  
address (in the LBA mode) of the last sector read.  
If an unrecoverable error occurs in a sector, the read operation is terminated at the  
sector where the error occurred.  
C141-E104-02EN  
5-17  
Interface  
Command block registers contain the cylinder, the head, and the sector addresses  
of the sector (in the CHS mode) or the logical block address (in the LBA mode)  
where the error occurred, and remaining number of sectors of which data was not  
transferred.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
0
x
0
1
x
0
0
0
0
R
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
(R: Retry)  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
(2) READ MULTIPLE (X’C4’)  
This command operates similarly to the READ SECTOR(S) command. The  
device does not generate an interrupt (assertion of the INTRQ signal) on each  
every sector. An interrupt is generated after the transfer of a block of sectors for  
which the number is specified by the SET MULTIPLE MODE command.  
5-18  
C141-E104-02EN  
5.3 Host Commands  
The implementation of the READ MULTIPLE command is identical to that of the  
READ SECTOR(S) command except that the number of sectors is specified by  
the SET MULTIPLE MODE command are transferred without intervening  
interrupts. In the READ MULTIPLE command operation, the DRQ bit of the  
Status register is set only at the start of the data block, and is not set on each  
sector.  
The number of sectors (block count) to be transferred without interruption is  
specified by the SET MULTIPLE MODE command. The SET MULTIPLE  
MODE command should be executed prior to the READ MULTIPLE command.  
When the READ MULTIPLE command is issued, the Sector Count register  
contains the number of sectors requested (not a number of the block count or a  
number of sectors in a block).  
Upon receipt of this command, the device executes this command even if the  
value of the Sector Count register is less than the defined block count (the value  
of the Sector Count should not be 0).  
If the number of requested sectors is not divided evenly (having the same number  
of sectors [block count]), as many full blocks as possible are transferred, then a  
final partial block is transferred. The number of sectors in the partial block to be  
transferred is n where n = remainder of (“number of sectors”/”block count”).  
If the READ MULTIPLE command is issued before the SET MULTIPLE MODE  
command is executed or when the READ MULTIPLE command is disabled, the  
device rejects the READ MULTIPLE command with an ABORTED COMMAND  
error.  
If an unrecoverable error occurs, reading sector is stopped at the sector where the  
error occurred. Command block registers contain the cylinder, the head, the  
sector addresses (in the CHS mode) or the logical block address (in the LBA  
mode) of the sector where the error occurred, and remaining number of sectors  
that had not transferred after the sector where the error occurred.  
An interrupt is generated when the DRQ bit is set at the beginning of each block or a  
partial block.  
Figure 5.2 shows an example of the execution of the READ MULTIPLE  
command.  
Block count specified by SET MULTIPLE MODE command = 4 (number of  
sectors in a block)  
READ MULTIPLE command specifies;  
Number of requested sectors = 9 (Sector Count register = 9)  
Number of sectors in incomplete block = remainder of 9/4 =1  
C141-E104-02EN  
5-19  
Interface  
Figure 5.2 Execution example of READ MULTIPLE command  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1
x
1
0
x
0
0
1
0
0
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00(*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors for which data was not transferred is set in this register.  
5-20  
C141-E104-02EN  
5.3 Host Commands  
(3) READ DMA (X’C8’ or X’C9’)  
This command operates similarly to the READ SECTOR(S) command except for  
following events.  
The data transfer starts at the timing of DMARQ signal assertion.  
The device controls the assertion or negation timing of the DMARQ signal.  
The device posts a status as the result of command execution only once at  
completion of the data transfer.  
When an error, such as an unrecoverable medium error, that the command  
execution cannot be continued is detected, the data transfer is stopped without  
transferring data of sectors after the erred sector. The device generates an  
interrupt using the INTRQ signal and posts a status to the host system. The  
format of the error information is the same as the READ SECTOR(S) command.  
In LBA mode  
The logical block address is specified using the start head No., start cylinder No.,  
and first sector No. fields. At command completion, the logical block address of  
the last sector and remaining number of sectors of which data was not transferred,  
like in the CHS mode, are set.  
The host system can select the DMA transfer mode by using the SET FEATURES  
command.  
Multiword DMA transfer mode 0 to 2  
Ultra DMA transfer mode 0 to 4  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1
x
1
0
x
0
1
0
0
R
1F6H(DH)  
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
C141-E104-02EN  
5-21  
Interface  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
(4) READ VERIFY SECTOR(S) (X’40’ or X’41’)  
This command operates similarly to the READ SECTOR(S) command except that  
the data is not transferred to the host system.  
After all requested sectors are verified, the device clears the BSY bit of the Status  
register and generates an interrupt. Upon the completion of the command  
execution, the command block registers contain the cylinder, head, and sector  
number of the last sector verified.  
If an unrecoverable error occurs, the verify operation is terminated at the sector  
where the error occurred. The command block registers contain the cylinder, the  
head, and the sector addresses (in the CHS mode) or the logical block address (in  
the LBA mode) of the sector where the error occurred. The Sector Count register  
indicates the number of sectors that have not been verified.  
5-22  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
0
x
1
0
x
0
0
0
0
R
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
(5) WRITE SECTOR(S) (X’30’ or X’31’)  
This command writes data of sectors from the address specified in the  
Device/Head, Cylinder High, Cylinder Low, and Sector Number registers to the  
address specified in the Sector Count register. Number of sectors can be specified  
to 256 sectors in maximum. Data transfer begins at the sector specified in the  
Sector Number register. For the DRQ, INTRQ, and BSY protocols related to data  
transfer, see Subsection 5.4.2.  
If the head is not on the track specified by the host, the device performs an  
implied seek. After the head reaches to the specified track, the device writes the  
target sector.  
If an error occurs when writing to the target sector, retries are attempted  
irrespectively of the R bit setting.  
The data stored in the buffer, and CRC code and ECC bytes are written to the data  
field of the corresponding sector(s). Upon the completion of the command  
execution, the command block registers contain the cylinder, head, and sector  
addresses of the last sector written.  
C141-E104-02EN  
5-23  
Interface  
If an error occurs during multiple sector write operation, the write operation is  
terminated at the sector where the error occurred. Command block registers  
contain the cylinder, the head, the sector addresses (in the CHS mode) or the  
logical block address (in the LBA mode) of the sector where the error occurred.  
Then the host can read the command block registers to determine what error has  
occurred and on which sector the error has occurred.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
0
x
0
1
x
1
0
0
0
R
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
Status information  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
5-24  
C141-E104-02EN  
5.3 Host Commands  
(6) WRITE MULTIPLE (X’C5’)  
This command is similar to the WRITE SECTOR(S) command. The device does  
not generate interrupts (assertion of the INTRQ) signal) on each sector but on the  
transfer of a block which contains the number of sectors for which the number is  
defined by the SET MULTIPLE MODE command.  
The implementation of the WRITE MULTIPLE command is identical to that of  
the WRITE SECTOR(S) command except that the number of sectors is specified  
by the SET MULTIPLE MODE command are transferred without intervening  
interrupts. In the WRITE MULTIPLE command operation, the DRQ bit of the  
Status register is required to set only at the start of the data block, not on each  
sector.  
The number of sectors (block count) to be transferred without interruption is  
specified by the SET MULTIPLE MODE command. The SET MULTIPLE  
MODE command should be executed prior to the WRITE MULTIPLE command.  
When the WRITE MULTIPLE command is issued, the Sector Count register contains  
the number of sectors requested (not a number of the block count or a number of  
sectors in a block).  
Upon receipt of this command, the device executes this command even if the  
value of the Sector Count register is less than the defined block count the value of  
the Sector Count should not be 0).  
If the number of requested sectors is not divided evenly (having the same number  
of sectors [block count]), as many full blocks as possible are transferred, then a  
final partial block is transferred. The number of sectors in the partial block to be  
transferred is n where n = remainder of (“number of sectors”/”block count”).  
If the WRITE MULTIPLE command is issued before the SET MULTIPLE  
MODE command is executed or when WRITE MULTIPLE command is disabled,  
the device rejects the WRITE MULTIPLE command with an ABORTED  
COMMAND error.  
Disk errors encountered during execution of the WRITE MULTIPLE command are  
posted after attempting to write the block or the partial block that was transferred.  
Write operation ends at the sector where the error was encountered even if the sector is  
in the middle of a block. If an error occurs, the subsequent block shall not be  
transferred. Interrupts are generated when the DRQ bit of the Status register is set at  
the beginning of each block or partial block.  
The contents of the command block registers related to addresses after the transfer  
of a data block containing an erred sector are undefined. To obtain a valid error  
information, the host should retry data transfer as an individual request.  
C141-E104-02EN  
5-25  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1
x
1
0
x
0
0
1
0
1
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00  
Error information  
(7) WRITE DMA (X’CA’ or X’CB’)  
This command operates similarly to the WRITE SECTOR(S) command except for  
following events.  
The data transfer starts at the timing of DMARQ signal assertion.  
The device controls the assertion or negation timing of the DMARQ signal.  
The device posts a status as the result of command execution only once at  
completion of the data transfer or completion of processing in the device.  
The device posts a status as the result of command execution only once at  
completion of the data transfer.  
When an error, such as an unrecoverable medium error, that the command  
execution cannot be continued is detected, the data transfer is stopped without  
transferring data of sectors after the erred sector. The device generates an  
interrupt using the INTRQ signal and posts a status to the host system. The  
format of the error information is the same as the WRITE SECTOR(S) command.  
5-26  
C141-E104-02EN  
5.3 Host Commands  
A host system can select the following transfer mode using the SET FEATURES  
command.  
Multiword DMA transfer mode 0 to 2  
Ultra DMA transfer mode 0 to 4  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1
x
1
0
x
0
1
0
1
R
1F6H(DH)  
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
(8) WRITE VERIFY (X’3C’)  
This command operates similarly to the WRITE SECTOR(S) command except  
that the device verifies each sector immediately after being written. The verify  
operation is a read and check for data errors without data transfer. Any error that  
is detected during the verify operation is posted.  
After all sectors are verified, the last interruption (INTRQ for command  
termination) is generated.  
C141-E104-02EN  
5-27  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
0
x
0
1
x
1
1
1
0
0
L
DV Start head No. /LBA  
[MSB]  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
Start cylinder No. [MSB] / LBA  
Start cylinder No. [LSB] / LBA  
Start sector No. / LBA [LSB]  
Transfer sector count  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV End head No. /LBA [MSB]  
End cylinder No. [MSB] / LBA  
End cylinder No. [LSB] / LBA  
End sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, the remaining number of  
sectors of which data was not transferred is set in this register.  
(9) RECALIBRATE (X’1x’, x: X’0’ to X’F’)  
This command performs the calibration. Upon receipt of this command, the  
device sets BSY bit of the Status register and performs a calibration. When the  
device completes the calibration, the device updates the Status register, clears the  
BSY bit, and generates an interrupt.  
This command can be issued in the LBA mode.  
5-28  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
0
x
0
x
0
x
1
x
x
x
x
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
Note:  
Also executable in LBA mode.  
(10) SEEK (X’7x’, x : X’0’ to X’F’)  
This command performs a seek operation to the track and selects the head  
specified in the command block registers. After completing the seek operation,  
the device clears the BSY bit in the Status register and generates an interrupt.  
The IDD always sets the DSC bit (Drive Seek Complete status) of the Status  
register to 1.  
In the LBA mode, this command performs the seek operation to the cylinder and  
head position in which the sector is specified with the logical block address.  
C141-E104-02EN  
5-29  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
0
x
1
1
x
1
x
x
x
x
L
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
xx  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
xx  
Error information  
(11) INITIALIZE DEVICE PARAMETERS (X’91’)  
The host system can set the number of sectors per track and the maximum head  
number (maximum head number is “number of heads minus 1”) per cylinder with  
this command. Upon receipt of this command, the device sets the BSY bit of  
Status register and saves the parameters. Then the device clears the BSY bit and  
generates an interrupt.  
When the SC register is specified to X’00’, an ABORTED COMMAND error is  
posted. Other than X’00’ is specified, this command terminates normally.  
The parameters set by this command are retained even after reset or power save  
operation regardless of the setting of disabling the reverting to default setting.  
In LBA mode  
The device ignores the L bit specification and operates with the CHS mode  
specification. An accessible area of this command within head moving in the LBA  
mode is always within a default area. It is recommended that the host system refers the  
addressable user sectors (total number of sectors) in word 60 to 61 of the parameter  
information by the IDENTIFY DEVICE command.  
5-30  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
0
x
0
x
1
0
0
0
1
DV Max. head No.  
xx  
xx  
xx  
Number of sectors/track  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
x
x
DV Max. head No.  
Number of sectors/track  
Error information  
(12) IDENTIFY DEVICE (X’EC’)  
The host system issues the IDENTIFY DEVICE command to read parameter  
information (512 bytes) from the device. Upon receipt of this command, the drive  
sets the BSY bit of Status register and sets required parameter information in the  
sector buffer. The device then sets the DRQ bit of the Status register, and  
generates an interrupt. After that, the host system reads the information out of the  
sector buffer. Table 5.4 shows the arrangements and values of the parameter  
words and the meaning in the buffer.  
C141-E104-02EN  
5-31  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
1
x
0
1
1
0
0
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
Table 5.4 Information to be read by IDENTIFY DEVICE command (1 of 8)  
Word  
0
Value  
X’045A’  
*2  
Description  
General Configuration *1  
1
Number of cylinders *2  
Detailed Configuration  
Number of Heads *2  
Undefined  
2
X’C837’  
*2  
3
4-5  
6
X’0000’  
*2  
Number of sectors per track *2  
Undefined  
7-9  
10-19  
20  
21  
22  
X’0000’  
Set by a device  
X’0000’  
X’0400’  
X’0004’  
Serial number (ASCII code, 20 characters, right)  
Undefined  
Buffer Size (1 LSB: 512 Byte)  
Number of ECC bytes transferred at READ LONG or  
WRITE LONG command  
5-32  
C141-E104-02EN  
5.3 Host Commands  
Table 5.4 Information to be read by IDENTIFY DEVICE command (2 of 8)  
Word  
23-26  
27-46  
47  
Value  
Description  
Firmware revision (ASCII code, 8 characters, left)  
Model name (ASCII code, 40 characters, left)  
Set by a device  
X’8010’  
Maximum number of sectors per interrupt on  
READ/WRITE MULTIPLE command  
48  
49  
X’0000’  
X’0B00’  
X’0000’  
X’0200’  
X’0000’  
X’0007’  
(Variable)  
(Variable)  
(Variable)  
(Variable)  
*6  
Reserved  
Capabilities *3  
50  
Reserved  
51  
PIO data transfer mode *4  
Reserved  
52  
53  
Enable/disable setting of words 54-58 and 64-70, 88 *5  
Number of current Cylinders  
Number of current Head  
Number of current sectors per track  
Total number of current sectors  
54  
55  
56  
57-58  
59  
Transfer sector count currently set by READ/WRITE  
MULTIPLE command *6  
60-61  
*2  
Total number of user addressable sectors (LBA mode only)  
*2  
62  
63  
64  
65  
X’0000’  
X’xx07’  
X’0003’  
X’0078’  
Reserved  
Multiword DMA transfer mode *7  
Advance PIO transfer mode support status *8  
Minimum multiword DMA transfer cycle time per word :  
120 [ns]  
66  
67  
68  
X’0078’  
X’00F0’  
X’0078’  
Manufacturer’s recommended DMA transfer cycle time :  
120 [ns]  
Minimum PIO transfer cycle time without IORDY flow  
control : 240 [ns]  
Minimum PIO transfer cycle time with IORDY flow control  
: 120 [ns]  
69-79  
80  
X’0000’  
X’003C’  
X’0000’  
X’346B’  
X’4008’  
X’4000’  
Reserved  
Major version number *9  
Minor version number (not reported)  
Support of command sets *10  
Support of command sets *11  
Support of command sets/function  
81  
82  
83  
84  
C141-E104-02EN  
5-33  
Interface  
Table 5.4 Information to be read by IDENTIFY DEVICE command (3 of 8)  
Word  
85  
Value  
*12  
Description  
Valid of command sets/function *12  
Valid of command sets/function *13  
Default of command sets/function  
Ultra DMA transfer mode *14  
Security Erase Unit execution time (Unit: 2 min.)  
Enhanced Security Erase Unit execution time (Unit: 2 min.)  
Advance power management level  
Master password revision  
86  
*13  
87  
X’4000’  
X’xx1F’  
Set by a device  
X’0000’  
(Variable)  
(Variable)  
*15  
88  
89  
90  
91  
92  
93  
Hardware configuration  
94-127  
128  
129-159  
160-254  
255  
X’0000’  
(Variable)  
X’0000’  
X’0000’  
X’xxA5’  
Reserved  
Security status *16  
Undefined  
Reserved  
Check sum (The 2 complement of the lower order byte  
resulting from summing bits 7 to 0 of word 0 to 254 and  
word 255, in byte units.)  
*1 Word 0: General configuration  
Bit 15:  
ATA device = 0, ATAPI device = 1  
Bit 14-8: Undefined  
Bit 7:  
Removable disk drive = 1  
Bit 6:  
Fixed drive = 1  
Bit 5-3:  
Bit 2:  
Undefined  
IDENTIFY DEVICE Valid = 0  
Reserved  
Bit 1-0:  
*2 Word 1, 3, 6, 60-61  
MHL2300AT  
MHM2200AT  
X’3FFF’  
X’10’  
MHM2150AT  
X’3FFF’  
X’10’  
MHM2100AT  
X’3FFF’  
X’10’  
Word 01  
Word 03  
Word 06  
Word 60-61  
X’3FFF’  
X’10’  
X’3F’  
X’3F’  
X’3F’  
X’3F’  
X’37E3E40’  
X’2542980’  
X’1C21B00’  
X’12BB230’  
*3 Word 49: Capabilities  
Bit 15-14: Reserved  
5-34  
C141-E104-02EN  
5.3 Host Commands  
Table 5.4 Information to be read by IDENTIFY DEVICE command (4 of 8)  
Bit 13:  
Bit 12:  
Bit 11:  
Standby timer value. Factory default is 0.  
Reserved  
IORDY support  
1=Supported  
Bit 10:  
IORDY inhibition  
Undefined  
0=Disable inhibition  
Bit 9-0:  
Bit 9, 8: Always 1  
*4 Word 51: PIO data transfer mode  
Bit 15-8: PIO data transfer mode  
X’02’=PIO mode 2  
Bit 7-0:  
Undefined  
*5 Word 53: Enable/disable setting of word 54-58 and 64-70  
Bit 15-3: Reserved  
Bit 2:  
Bit 1:  
Bit 0:  
Enable/disable setting of word 88  
Enable/disable setting of word 64-70  
Enable/disable setting of word 54-58  
1=Enable  
1=Enable  
1=Enable  
*6 Word 59: Transfer sector count currently set by READ/WRITE  
MULTIPLE  
command  
Bit 15-9: Reserved  
Bit 8:  
Multiple sector transfer 1=Enable  
Bit 7-0:  
Transfer sector count currently set by READ/WRITE  
MULTIPLE command without interrupt supports 2, 4, 8 and 16  
sectors.  
*7 Word 63: Multiword DMA transfer mode  
Bit 15-8: Currently used multiword DMA transfer mode  
Bit 7-0:  
Supportable multiword DMA transfer mode  
Bit 2=1 Mode 2  
Bit 1=1 Mode 1  
Bit 0=1 Mode 0  
*8 Word 64: Advance PIO transfer mode support status  
Bit 15-8: Reserved  
Bit 7-0:  
Advance PIO transfer mode  
C141-E104-02EN  
5-35  
Interface  
Table 5.4 Information to be read by IDENTIFY DEVICE command (5 of 8)  
Bit 1 = 1 Mode 4  
Bit 0 = 1 Mode 3  
*9 WORD 80  
Bit 15-6: Reserved  
Bit 5:  
Bit 4:  
ATA/ATAPI-5 supported = 1  
ATA/ATAPI-4 supported = 1  
ATA-3 supported = 1  
ATA-2 supported = 1  
Undefined  
Bit 3:  
Bit 2:  
Bit 1-0:  
*10 WORD 82  
Bit 15:  
Bit 14:  
Bit 13:  
Bit 12:  
Bit 11:  
Bit 10:  
Bit 9:  
Undefined  
'1' = Supports the NOP command.  
'1' = Supports the READ BUFFER command.  
'1' = Supports the WRITE BUFFER command.  
Undefined  
'1' = Supports the Host Protected Area feature set.  
'1' = Supports the DEVICE RESET command.  
'1' = Supports the SERVICE interrupt.  
'1' = Supports the release interrupt.  
Bit 8:  
Bit 7:  
Bit 6:  
'1' = Supports the read cache function.  
'1' = Supports the write cache function.  
'1' = Supports the PACKET command feature set.  
'1' = Supports the power management feature set.  
'1' = Supports the Removable Media feature set.  
'1' = Supports the Security Mode feature set.  
'1' = Supports the SMART feature set.  
Bit 5:  
Bit 4:  
Bit 3:  
Bit 2:  
Bit 1:  
Bit 0:  
*11 WORD 83  
Bits 15-7: Undefined  
Bit 6:  
'1' = When the power is turned on, spin is started by the SET  
FEATURES sub-command.  
5-36  
C141-E104-02EN  
5.3 Host Commands  
Table 5.4 Information to be read by IDENTIFY DEVICE command (6 of 8)  
Bit 5:  
Bit 4:  
'1' = Supports the Power-Up In Standby set.  
'1' = Supports the Removable Media Status Notification feature  
set.  
Bit 3:  
Bit 2:  
'1' = Supports the Advanced Power Management feature set.  
'1' = Supports the CFA (Compact Flash Association) feature set.  
'1' = Supports the READ/WRITE DMA QUEUED command.  
'1' = Supports the DOWNLOAD MICROCODE command.  
Bit 1:  
Bit 0:  
*12 WORD 85  
Bit 15:  
Bit 14:  
Bit 13:  
Bit 12:  
Bit 11:  
Bit 10:  
Bit 9:  
Undefined.  
'1' = Enables the NOP command.  
'1' = Enables the READ BUFFER command.  
'1' = Enables the WRITE BUFFER command.  
Undefined.  
'1' = Enables the Host Protected Area function.  
'1' = Enables the DEVICE RESET command.  
'1' = Enables the SERVICE interrupt.  
'1' = Enables the release interrupt.  
Bit 8:  
Bit 7:  
Bit 6:  
'1' = Enables the read cache function.  
'1' = Enables the write cache function.  
'1' = Enables the P PACKET command set.  
'1' = Enables the Power Management function.  
'1' = Enables the Removable Media function.  
'1' = Enables the Security Mode function.  
'1' = Enables the SMART function.  
Bit 5:  
Bit 4:  
Bit 3:  
Bit 2:  
Bit 1:  
Bit 0:  
*13 WORD 86  
Bits 15-7: '1' = Reserved  
Bit 6:  
Bit 5  
Same definition as WORD 83.  
Enables the Power-Up In Standby function.  
Bit 4:  
Bit 3:  
'1' = Enables the Removable Media Status Notification function.  
'1' = Enables the Advanced Power Management function.  
C141-E104-02EN  
5-37  
Interface  
Table 5.4 Information to be read by IDENTIFY DEVICE command (7 of 8)  
Bits 2-0: Same definition as WORD 83.  
*14 WORD 88  
Bit 15-8: Currently used Ultra DMA transfer mode  
Bit 7-0:  
Supportable Ultra DMA transfer mode  
Bit 4 = '1': Mode 4  
Bit 3 = '1': Mode 3  
Bit 2 = '1': Mode 2  
Bit 1 = '1': Mode 1  
Bit 0 = '1': Mode 0  
*15 WORD 93  
Bits 15-14: Reserved  
Bit 13:  
'1' = CBLID- is a level higher than VIH.  
'0' = CBLID- is a level lower than VIL.  
Bits 12-8: In the case of Device 1 (slave drive), a valid value is set.  
Bit 12:  
Bit 11:  
Reserved  
'1' = Device asserts PDIAG-.  
Bit 10, 9: Method for deciding the device No. of Device 1.  
'00' = Reserved  
'01' = Using a jumper.  
'10' = Using the CSEL signal.  
'11' = Other method.  
Bit 8:  
Reserved  
Bits 7-0: In the case of Device 0 (master drive), a valid value is set.  
Bit 7:  
Bit 6:  
Bit 5:  
Bit 4:  
Bit 3:  
Reserved  
'1' = Device 1 is selected, Device 0 responds.  
'1' = Device 0, assertion of DASP- was detected.  
'1' = Device 0, assertion of PDIAG- was detected.  
'1' = Device 0, an error was not detected in the self-  
diagnosis.  
Bit 2, 1: Method for deciding the device No. of Device 0.  
5-38  
C141-E104-02EN  
5.3 Host Commands  
Table 5.4 Information to be read by IDENTIFY DEVICE command (8 of 8)  
'00' = Reserved  
'01' = Using a jumper.  
'10' = Using the CSEL signal.  
'11' = Other method.  
Reserved  
Bit 0:  
*16 WORD 128  
Bit 15-9: Reserved  
Bit 8:  
Bit 7-6:  
Bit 5:  
Bit 4:  
Bit 3:  
Bit 2:  
Bit 1:  
Bit 0:  
Security level. 0: High, 1: Maximum  
Reserved  
1: Enhanced security erase supported  
1: Security counter expired  
1: Security frozen  
1: Security locked  
1: Security enabled  
1: Security supported  
(13) IDENTIFY DEVICE DMA (X’EE’)  
When this command is not used to transfer data to the host in DMA mode, this  
command functions in the same way as the Identify Device command.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
1
x
0
1
1
1
0
DV xx  
xx  
xx  
xx  
xx  
xx  
C141-E104-02EN  
5-39  
Interface  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(14) SET FEATURES (X’EF’)  
The host system issues the SET FEATURES command to set parameters in the  
Features register for the purpose of changing the device features to be executed.  
For the transfer mode (Feature register = 03), detail setting can be done using the  
Sector Count register.  
Upon receipt of this command, the device sets the BSY bit of the Status register  
and saves the parameters in the Features register. Then, the device clears the BSY  
bit, and generates an interrupt.  
If the value in the Features register is not supported or it is invalid, the device  
posts an ABORTED COMMAND error.  
Table 5.5 lists the available values and operational modes that may be set in the  
Features register.  
5-40  
C141-E104-02EN  
5.3 Host Commands  
Table 5.5 Features register values and settable modes  
Drive operation mode  
Features  
Register  
X’02’  
X’03’  
Enables the write cache function.  
Transfer mode depends on the contents of the Sector Count register.  
(Details are given later.)  
X’05’  
X’55’  
X’66’  
X’82’  
X’85’  
X’AA’  
X’BB’  
Enables the advanced power management function.  
Disables read cache function.  
Disables the reverting to power-on default settings after software reset.  
Disables the write cache function.  
Disables the advanced power management function.  
Enables the read cache function.  
Specifies the transfer of 4-byte ECC for READ LONG and WRITE  
LONG commands.  
X’CC’  
Enables the reverting to power-on default settings after software reset.  
At power-on or after hardware reset, the default mode is the same as that is set  
with a value greater than X’85’ (except for write cache). If X’66’ is specified, it  
allows the setting value greater than X’AA’ which may have been modified to a  
new value since power-on, to remain the same even after software reset.  
C141-E104-02EN  
5-41  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
1
x
0
1
1
1
1
DV xx  
xx  
xx  
xx  
xx or transfer mode  
[See Table 5.5]  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
The host sets X’03’ to the Features register. By issuing this command with  
setting a value to the Sector Count register, the transfer mode can be selected.  
Upper 5 bits of the Sector Count register defines the transfer type and lower 3 bits  
specifies the binary mode value.  
The IDD supports following values in the Sector Count register value. If other  
value than below is specified, an ABORTED COMMAND error is posted.  
PIO default transfer mode  
00000 000 (X’00’)  
PIO flow control transfer mode X  
00001 000 (X’08’: Mode 0)  
00001 001 (X’09’: Mode 1)  
00001 010 (X’0A’: Mode 2)  
00001 011 (X’0B’: Mode 3)  
00001 100 (X’0C’: Mode 4)  
5-42  
C141-E104-02EN  
5.3 Host Commands  
Multiword DMA transfer mode X  
Ultra DMA transfer mode X  
00100 000 (X’20’: Mode 0)  
00100 001 (X’21’: Mode 1)  
00100 010 (X’22’: Mode 2)  
01000 000 (X’40’: Mode 0)  
01000 001 (X’41’: Mode 1)  
01000 010 (X’42’: Mode 2)  
01000 011 (X’43’: Mode 3)  
01000 100 (X’44’: Mode 4)  
The host writes the Sector Count register with the desired power management  
level and executes this command with the Features register X’05’, and then  
Advanced Power Management is enabled.  
Level  
Sector Count register  
Power management without standby  
Power management with standby  
Reserved  
80h-FEh  
01h-7Fh  
FFh, 00h  
(15) SET MULTIPLE MODE (X’C6’)  
This command enables the device to perform the READ MULTIPLE and  
WRITE MULTIPLE commands. The block count (number of sectors in a  
block) for these commands are also specified by the SET MULTIPLE MODE  
command.  
The number of sectors per block is written into the Sector Count register. The  
IDD supports 2, 4, 8, 16 and 32 (sectors) as the block counts.  
Upon receipt of this command, the device sets the BSY bit of the Status register  
and checks the contents of the Sector Count register. If the contents of the Sector  
Count register is valid and is a supported block count, the value is stored for all  
subsequent READ MULTIPLE and WRITE MULTIPLE commands. Execution  
of these commands is then enabled. If the value of the Sector Count register is not  
a supported block count, an ABORTED COMMAND error is posted and the  
READ MULTIPLE and WRITE MULTIPLE commands are disabled.  
If the contents of the Sector Count register is 0, 1 when the SET MULTIPLE  
MODE command is issued, the READ MULTIPLE and WRITE MULTIPLE  
commands are disabled.  
When the SET MULTIPLE MODE command operation is completed, the device  
clears the BSY bit and generates an interrupt.  
C141-E104-02EN  
5-43  
Interface  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
0
x
0
0
1
1
0
DV xx  
xx  
xx  
xx  
Sector count/block  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
x
x
DV xx  
Sector count/block  
Error information  
After power-on or after hardware reset, the READ MULTIPLE and WRITE  
MULTIPLE command operation are disabled as the default mode.  
The mode established before software reset is retained if disable default (Features  
Reg. = 66h setting) has been defined by the SET FEATURES command. If  
disable default has not been defined after the software is the READ MULTIPLE  
and WRITE MULTIPLE commands are disabled.  
The parameters for the multiple commands which are posted to the host system  
when the IDENTIFY DEVICE command is issued are listed below. See  
Subsection 5.32 for the IDENTIFY DEVICE command.  
5-44  
C141-E104-02EN  
5.3 Host Commands  
Word 47  
Bit 7-0 = 10:  
Maximum number of sectors that can be transferred per interrupt  
by the READ MULTIPLE and WRITE MULTIPLE commands.  
The READ MULTIPLE and WRITE MULTIPLE commands are  
Word 59 = 0000: disabled.  
The READ MULTIPLE and WRITE MULTIPLE commands are  
= 01xx: enabled. “xx” indicates the current setting for number of sectors  
that can be transferred per interrupt by the READ MULTIPLE and  
WRITE MULTIPLE commands.  
e.g. 0110 = Block count of 16 has been set by the SET  
MULTIPLE MODE command.  
(16) SET MAX (F9)  
Each SET MAX command is determined by the Features Register value.  
SET MAX Features Register Values  
Value  
Command  
00h  
01h  
02h  
03h  
04h  
Obsolete  
SET MAX SET PASSWORD  
SET MAX LOCK  
SET MAX UNLOCK  
SET MAX FREEZE LOCK  
Reserved  
05h - FFh  
SET MAX ADDRESS  
This command allows the maximum address accessible by the user to be set in  
LBA or CHS mode. Upon receipt of the command, the device sets the BSY bit  
and saves the maximum address specified in the DH, CH, CL and SN registers.  
Then, it clears BSY and generates an interrupt.  
The new address information set by this command is reflected in Words 1, 54, 57,  
58, 60 and 61 of IDENTIFY DEVICE information. If an attempt is made to  
perform a read or write operation for an address beyond the new address space, an  
ID Not Found error will result.  
When SC register bit 0, VV (Value Volatile), is 1, the value set by this command  
is held even after power on and the occurrence of a hard reset. When the VV bit is  
0, the value set by this command becomes invalid when the power is turned on or  
a hard reset occurs, and the maximum address returns to the value (default value if  
not set) most lately set when VV bit = 1.  
C141-E104-02EN  
5-45  
Interface  
After power on and the occurrence of a hard reset, the host can issue this  
command only once when VV bit = 1. If this command with VV bit = 1 is issued  
twice or more, any command following the first time will result in an Aborted  
Command error.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
x
1
1
0
0
1
L
DV Max head/LBA [MSB]  
Max. cylinder [MSB]/Max. LBA  
Max. cylinder [LSB]/Max. LBA  
Max. sector/Max. LBA [LSB]  
xx  
xx  
VV  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
x
x
DV Max head/LBA [MSB]  
Max. cylinder [MSB]/Max. LBA  
Max. cylinder [LSB]/Max. LBA  
Max. sector/Max. LBA [LSB]  
xx  
Error information  
SET MAX SET PASSWORD (FR=01h)  
This command requests a transfer of 1 sector of data from the host, and defines  
the contents of SET MAX password.  
The READ NATIVE MAX ADDRESS command is not executed just before this  
command. The command is the SET MAX ADDRESS command if it is the  
command just after the READ NATIVE MAX ADDRESS command is executed.  
If the device is in the Set Max Locked or Set Max Freeze Locked state:  
51h, 04h: ABORTED command  
5-46  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
x
1
1
0
0
1
L
DV xx  
xx  
xx  
xx  
xx  
01  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
xx  
xx  
xx  
xx  
xx  
Error information  
Password information  
Words  
Contents  
0
Reserved  
1 to 16  
Password (32 bytes)  
Reserved  
17 to 255  
SET MAX LOCK (FR=02h)  
After this command is completed, any other SET MAX commands except SET  
MAX UNLOCK and SET MAX FREEZE LOCK commands are rejected. And  
the device returns command aborted.  
The device remains in the SET MAX LOCK state until a power cycle or the  
acceptance of SET MAX UNLOCK or SET MAX FREEZE LOCK command.  
The READ NATIVE MAX ADDRESS command is not executed just before this  
command. The command is the SET MAX ADDRESS command if it is the  
command just after the READ NATIVE MAX ADDRESS command is executed.  
C141-E104-02EN  
5-47  
Interface  
If the device is in the Set Max Locked or Set Max Freeze Locked state:  
51h, 04h: ABORTED command  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
x
1
1
0
0
1
L
DV xx  
xx  
xx  
xx  
xx  
02  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
xx  
xx  
xx  
xx  
xx  
Error information  
SET MAX UNLOCK (FR=03h)  
This command requests a transfer of single sector of data from the host, and  
defines the contents of SET MAX ADDRESS password. The data transferred  
controls the function of this command.  
The password supplied in the sector of data transferred shall be compared with the  
stored password.  
If the password compare fails, the device returns command aborted and  
decrements the Unlock counter, and remains in the Set Max Lock state. On the  
acceptance of the SET MAX LOCK command, the Unlock counter is set to a  
value of five. When this counter reaches zero, then SET MAX UNLOCK  
command shall return command aborted until a power cycle.  
If the password compare matches, then the device shall make a transition to the  
Set Max Unlocked state and all SET MAX commands will be accepted.  
5-48  
C141-E104-02EN  
5.3 Host Commands  
If this command is accepted in the Set Max Unlocked state, the device terminates  
normally.  
The READ NATIVE MAX ADDRESS command is not executed just before this  
command. The command is the SET MAX ADDRESS command if it is the  
command just after the READ NATIVE MAX ADDRESS command is executed.  
If the device is in the Set Max Locked or Set Max Freeze Locked state:  
51h, 04h: ABORTED command  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
x
1
1
0
0
1
L
DV xx  
xx  
xx  
xx  
xx  
03  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
xx  
xx  
xx  
xx  
xx  
Error information  
SET MAX FREEZE LOCK (FR=04h)  
After the device made a transition to the Set Max Freeze Lock state, the following  
commands are rejected, then the device returns command aborted:  
SET MAX ADDRESS  
SET MAX SET PASSWORD  
SET MAX LOCK  
SET MAX UNLOCK  
C141-E104-02EN  
5-49  
Interface  
The READ NATIVE MAX ADDRESS command is not executed just before this  
command. The command is the SET MAX ADDRESS command if it is the  
command just after the READ NATIVE MAX ADDRESS command is executed.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
1
x
1
1
0
0
1
DV xx  
xx  
xx  
xx  
xx  
04  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
xx  
xx  
xx  
xx  
xx  
Error information  
(17) READ NATIVE MAX ADDRESS (F8)  
This command posts the maximum address intrinsic to the device, which can be  
set by the SET MAX ADDRESS command. Upon receipt of this command, the  
device sets the BSY bit and indicates the maximum address in the DH, CH, CL  
and SN registers. Then, it clears BSY and generates an interrupt.  
5-50  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
x
1
1
0
0
0
L
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
x
x
DV Max head/LBA [MSB]  
Max. cylinder [MSB]/Max. LBA  
Max. cylinder [LSB]/Max. LBA  
Max. sector/Max. LBA [LSB]  
xx  
Error information  
(18) EXECUTE DEVICE DIAGNOSTIC (X’90’)  
This command performs an internal diagnostic test (self-diagnosis) of the device.  
This command usually sets the DRV bit of the Drive/Head register is to 0  
(however, the DV bit is not checked). If two devices are present, both devices  
execute self-diagnosis.  
If device 1 is present:  
Both devices shall execute self-diagnosis.  
The device 0 waits for up to 5 seconds until device 1 asserts the PDIAG-  
signal.  
If the device 1 does not assert the PDIAG- signal but indicates an error, the  
device 0 shall append X’80’ to its own diagnostic status.  
The device 0 clears the BSY bit of the Status register and generates an  
interrupt. (The device 1 does not generate an interrupt.)  
A diagnostic status of the device 0 is read by the host system. When a  
diagnostic failure of the device 1 is detected, the host system can read a status  
of the device 1 by setting the DV bit (selecting the device 1).  
C141-E104-02EN  
5-51  
Interface  
When device 1 is not present:  
The device 0 posts only the results of its own self-diagnosis.  
The device 0 clears the BSY bit of the Status register, and generates an  
interrupt.  
Table 5.6 lists the diagnostic code written in the Error register which is 8-bit code.  
If the device 1 fails the self-diagnosis, the device 0 “ORs” X’80’ with its own  
status and sets that code to the Error register.  
Table 5.6 Diagnostic code  
Code  
Result of diagnostic  
No error detected.  
X’01’  
X’03’  
X’05’  
X’8x’  
Data buffer compare error  
ROM sum check error  
Failure of device 1  
attention: The device responds normally to this command without executing  
internal diagnostic test.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
0
x
0
x
1
0
0
0
0
DV  
Head No. /LBA [MSB]  
xx  
xx  
xx  
xx  
xx  
5-52  
C141-E104-02EN  
5.3 Host Commands  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
x
x
DV  
Head No. /LBA [MSB]  
01H (*1)  
01H  
Diagnostic code  
*1  
This register indicates X’00’ in the LBA mode.  
(19) READ LONG (X’22’ or X’23’)  
This command operates similarly to the READ SECTOR(S) command except that  
the device transfers the data in the requested sector and the ECC bytes to the host  
system. The ECC error correction is not performed for this command. This  
command is used for checking ECC function by combining with the WRITE  
LONG command.  
Number of ECC bytes to be transferred is fixed to 4 bytes and cannot be changed  
by the SET FEATURES command.  
The READ LONG command supports only single sector operation.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
0
x
0
1
x
0
0
0
1
R
L
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
01  
xx  
(R: Retry)  
C141-E104-02EN  
5-53  
Interface  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, this register indicates 01.  
(20) WRITE LONG (X’32’ or X’33’)  
This command operates similarly to the READ SECTOR(S) command except that  
the device writes the data and the ECC bytes transferred from the host system to  
the disk medium. The device does not generate ECC bytes by itself. The WRITE  
LONG command supports only single sector operation.  
The number of ECC bytes to be transferred is fixed to 4 bytes and can not be  
changed by the SET FEATURES command.  
This command is operated under the following conditions:  
READ LONG issued WRITE LONG (Same address) issues sequence  
(After READ LONG is issued, WRITE LONG can be issued consecutively.)  
If above condition is not satisfied, the command operation is not guaranteed.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
0
x
0
1
x
1
0
0
1
R
L
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
01  
xx  
5-54  
C141-E104-02EN  
5.3 Host Commands  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
L
x
DV Head No. /LBA [MSB]  
Cylinder No. [MSB] / LBA  
Cylinder No. [LSB] / LBA  
Sector No. / LBA [LSB]  
00 (*1)  
Error information  
*1  
If the command is terminated due to an error, this register indicates 01.  
(21) READ BUFFER (X’E4’)  
The host system can read the current contents of the sector buffer of the device by  
issuing this command. Upon receipt of this command, the device sets the BSY bit  
of Status register and sets up the sector buffer for a read operation. Then the  
device sets the DRQ bit of Status register, clears the BSY bit, and generates an  
interrupt. After that, the host system can read up to 512 bytes of data from the  
buffer.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
1
1
0
1
0
0
x
x
DV xx  
xx  
xx  
xx  
xx  
xx  
C141-E104-02EN  
5-55  
Interface  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(22) WRITE BUFFER (X’E8’)  
The host system can overwrite the contents of the sector buffer of the device with  
a desired data pattern by issuing this command. Upon receipt of this command,  
the device sets the BSY bit of the Status register. Then the device sets the DRQ  
bit of Status register and clears the BSY bit when the device is ready to receive  
the data. After that, 512 bytes of data is transferred from the host and the device  
writes the data to the sector buffer, then generates an interrupt.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
1
x
1
x
1
1
0
0
0
DV xx  
xx  
xx  
xx  
xx  
xx  
5-56  
C141-E104-02EN  
5.3 Host Commands  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(23) IDLE (X’97’ or X’E3’)  
Upon receipt of this command, the device sets the BSY bit of the Status register,  
and enters the idle mode. Then, the device clears the BSY bit, and generates an  
interrupt. The device generates an interrupt even if the device has not fully  
entered the idle mode. If the spindle of the device is already rotating, the spin-up  
sequence shall not be implemented.  
By using this command, the automatic power-down function is enabled and the  
timer immediately starts the countdown. When the timer reaches the specified  
value, the device enters standby mode.  
Enabling the automatic power-down function means that the device automatically  
enters the standby mode after a certain period of time. When the device enters the  
idle mode, the timer starts countdown. If any command is not issued while the  
timer is counting down, the device automatically enters the standby mode. If any  
command is issued while the timer is counting down, the timer is initialized and  
the command is executed. The timer restarts countdown after completion of the  
command execution.  
The period of timer count is set depending on the value of the Sector Count  
register as shown below.  
Sector Count register value  
[X’00’]  
Point of timer  
30 minutes  
0
1 to 3  
[X’01’ to X’03’]  
[X’04’ to X’F0’]  
15 seconds  
4 to 240  
(Value ×5) seconds  
30 minutes  
241 to 251 [X’F1’ to X’FB’]  
252  
253  
[X’FC’]  
[X’FD’]  
21 minutes  
30 minutes  
254 to 255 [X’FE’ to X’FF’]  
21 minutes 15 seconds  
C141-E104-02EN  
5-57  
Interface  
attention: The automatic power-down is executed if no command is coming for  
30 min.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
X’97’ or X’E3’  
x
xx  
xx  
xx  
x
x
DV xx  
Period of timer  
xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(24) IDLE IMMEDIATE (X’95’ or X’E1’)  
Upon receipt of this command, the device sets the BSY bit of the Status register,  
and enters the idle mode. Then, the device clears the BSY bit, and generates an  
interrupt. This command does not support the automatic power-down function.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
X’95’ or X’E1’  
x
xx  
xx  
xx  
xx  
xx  
x
x
DV xx  
5-58  
C141-E104-02EN  
5.3 Host Commands  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(25) STANDBY (X’96’ or X’E2’)  
Upon receipt of this command, the device sets the BSY bit of the Status register  
and enters the standby mode. The device then clears the BSY bit and generates an  
interrupt. The device generates an interrupt even if the device has not fully  
entered the standby mode. If the device has already spun down, the spin-down  
sequence is not implemented.  
By using this command, the automatic power-down function is enabled and the  
timer starts the countdown when the device returns to idle mode.  
When the timer value reaches 0 (a specified time has padded), the device enters  
standby mode.  
Under the standby mode, the spindle motor is stopped. Thus, when the command  
involving a seek such as the READ SECTOR(s) command is received, the device  
processes the command after driving the spindle motor.  
attention: The automatic power-down is executed if no command is coming for  
30 min.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
X’96’ or X’E2’  
x
xx  
xx  
xx  
x
x
DV xx  
Period of timer  
xx  
C141-E104-02EN  
5-59  
Interface  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(26) STANDBY IMMEDIATE (X’94’ or X’E0’)  
Upon receipt of this command, the device sets the BSY bit of the Status register  
and enters the standby mode. The device then clears the BSY bit and generates an  
interrupt. This command does not support the automatic power-down sequence.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
X’94’ or X’E0’  
x
xx  
xx  
xx  
xx  
xx  
x
x
DV xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
5-60  
C141-E104-02EN  
5.3 Host Commands  
(27) SLEEP (X’99’ or X’E6’)  
This command is the only way to make the device enter the sleep mode.  
Upon receipt of this command, the device sets the BSY bit of the Status register  
and enters the sleep mode. The device then clears the BSY bit and generates an  
interrupt. The device generates an interrupt even if the device has not fully  
entered the sleep mode.  
In the sleep mode, the spindle motor is stopped and the ATA interface section is  
inactive. All I/O register outputs are in high-impedance state.  
The only way to release the device from sleep mode is to execute a software or  
hardware reset.  
At command issuance (I/O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
X’99’ or X’E6’  
x
xx  
xx  
xx  
xx  
xx  
x
x
DV xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
C141-E104-02EN  
5-61  
Interface  
(28) CHECK POWER MODE (X’98’ or X’E5’)  
The host checks the power mode of the device with this command.  
The host system can confirm the power save mode of the device by analyzing the  
contents of the Sector Count and Sector registers.  
The device sets the BSY bit and sets the following register value. After that, the  
device clears the BSY bit and generates an interrupt.  
Power save mode  
Sector Count register  
X’00’  
• During moving to standby mode  
• Standby mode  
• During returning from the standby mode  
• Idle mode  
X’FF’  
X’FF’  
• Active mode  
At command issuance (I/O registers setting contents)  
1F7H(CM) X’98’ or X’E5’  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
x
xx  
xx  
xx  
xx  
xx  
x
x
DV xx  
At command completion (I/O registers contents to be read)  
1F7H(ST)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
Status information  
x
xx  
xx  
xx  
x
x
DV xx  
X’00’ ,X’80’ or X’FF’  
Error information  
5-62  
C141-E104-02EN  
5.3 Host Commands  
(29) SMART (X’B0)  
This command performs operations for device failure predictions according to a  
subcommand specified in the FR register. If the value specified in the FR register  
is supported, the Aborted Command error is posted.  
It is necessary for the host to set the keys (CL = 4Fh and CH = C2h) in the CL and  
CH registers prior to issuing this command. If the keys are set incorrectly, the  
Aborted Command error is posted.  
In the default setting, the failure prediction feature is enabled.  
The device collects or updates several items to forecast failures. In the following  
sections, the values of items collected or updated by the device to forecast failures  
are referred to as attribute values.  
C141-E104-02EN  
5-63  
Interface  
Table 5.7 Features Register values (subcommands) and functions (1 of 3)  
Features Resister  
X’D0’  
Function  
SMART Read Attribute Values:  
A device that received this subcommand asserts the BSY bit and saves all  
the updated attribute values. The device then clears the BSY bit and  
transfers 512-byte attribute value information to the host.  
* For information about the format of the attribute value information, see  
Table 5.8.  
X’D1’  
X’D2’  
SMART Read Attribute Thresholds:  
This subcommand is used to transfer 512-byte insurance failure threshold  
value data to the host.  
* For information about the format of the insurance failure threshold value  
data, see Table 5.9.  
SMART Enable-Disable Attribute AutoSave:  
This subcommand is used to enable (SC register 00h) or disable (SC  
register = 00h) the setting of the automatic saving feature for the device  
attribute data. The setting is maintained every time the device is turned off  
and then on. When the automatic saving feature is enabled, the attribute  
values are saved before the device enters the power saving mode. However,  
if the failure prediction feature is disabled, the attribute values are not  
automatically saved.  
When the device receives this subcommand, it asserts the BSY bit, enables  
or disables the automatic saving feature, then clears the BSY bit.  
X’D3’  
X’D4’  
SMART Save Attribute Values:  
When the device receives this subcommand, it asserts the BSY bit, saves  
device attribute value data, then clears the BSY bit.  
SMART Executive Off-line Immediate:  
A device which receives this command asserts the BSY bit, then starts  
collecting the off-line data specified in the SN register, or stops.  
In the off-line mode, after BSY is cleared, off-line data are collected. In the  
captive mode, it collects off-line data with the BSY assertion as is, then  
clears the BSY when collection of data is completed.  
SN Off-line data collection mode  
00h: Off-line diagnosis (off-line mode)  
01h: Simple self test (off-line mode)  
02h: Comprehensive self test (off-line mode)  
7Fh: Self test stop  
81h: Simple self test (captive mode)  
82h: Comprehensive self test (captive mode)  
5-64  
C141-E104-02EN  
5.3 Host Commands  
Table 5.7 Features Register values (subcommands) and functions (2 of 3)  
Features Resister  
X’D5’  
Function  
SMART Read Log Sector:  
A device which receives this sub-command asserts the BSY bit, then reads  
the log sector specified in the SN register. Next, it clears the BSY bit and  
transmits the log sector to the host computer.  
SN:  
01h:  
06h:  
Log sector  
SMART error log  
SMART self test log  
80h-9Fh: Host vendor log  
* See Table 5.10 concerning the SMART error log data format.  
See Table 5.11 concerning the SMART self test log data format.  
X’D6’  
SMART Write Log Sector:  
A device which receives this sub-command asserts the BSY bit and when it  
has prepared to receive data from the host computer, it sets DRQ and clears  
the BSY bit. Next, it receives data from the host computer and writes the  
specified log sector in the SN register.  
SN:  
Log sector  
80h-9Fh: Host vendor log  
* The host can write any desired data in the host vendor log.  
SMART Enable Operations:  
This subcommand enables the failure prediction feature. The setting is  
maintained even when the device is turned off and then on.  
When the device receives this subcommand, it asserts the BSY bit, enables  
the failure prediction feature, then clears the BSY bit.  
X’D8’  
X’D9’  
SMART Disable Operations:  
This subcommand disables the failure prediction feature. The setting is  
maintained even when the device is turned off and then on.  
When the device receives this subcommand, it asserts the BSY bit, disables  
the failure prediction feature, then clears the BSY bit.  
C141-E104-02EN  
5-65  
Interface  
Table 5.7 Features Register values (subcommands) and functions (3 of 3)  
Features Resister  
X’DA’  
Function  
SMART Return Status:  
When the device receives this subcommand, it asserts the BSY bit and saves the  
current device attribute values. Then the device compares the device attribute  
values with insurance failure threshold values. If there is an attribute value  
exceeding the threshold, F4h and 2Ch are loaded into the CL and CH registers.  
If there are no attribute values exceeding the thresholds, 4Fh and C2h are loaded  
into the CL and CH registers. After the settings for the CL and CH registers  
have been determined, the device clears the BSY bit  
X’DB’  
SMART Enable/Disable Auto Off-line:  
This sets automatic off-line data collection in the enabled (when the SC  
register specification 00h) or disabled (when the SC register specification  
= 00) state. This setting is preserved whether the drive’s power is switched  
on or off.  
If four hours have passed since the power was switched on, or since the last  
time that off-line data were collected, off-line data collection is performed  
without relation to any command from the host computer.  
The host must regularly issue the SMART Read Attribute Values subcommand  
(FR register = D0h), SMART Save Attribute Values subcommand (FR register =  
D3h), or SMART Return Status subcommand (FR register = DAh) to save the  
device attribute value data on a medium.  
Alternative, the device must issue the SMART Enable-Disable Attribute  
AutoSave subcommand (FR register = D2h) to use a feature which regularly save  
the device attribute value data to a medium.  
The host can predict failures in the device by periodically issuing the SMART Return  
Status subcommand (FR register = DAh) to reference the CL and CH registers.  
If an attribute value is below the insurance failure threshold value, the device is  
about to fail or the device is nearing the end of its life . In this case, the host  
recommends that the user quickly backs up the data.  
At command issuance (I-O registers setting contents)  
1F7H(CM)  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(FR)  
1
x
0
x
1
x
1
0
0
0
0
DV xx  
Key (C2h)  
Key (4Fh)  
xx  
xx  
Subcommand  
5-66  
C141-E104-02EN  
5.3 Host Commands  
At command completion (I-O registers setting contents)  
1F7H(ST) Status information  
1F6H(DH)  
1F5H(CH)  
1F4H(CL)  
1F3H(SN)  
1F2H(SC)  
1F1H(ER)  
x
x
x
DV xx  
Key-failure prediction status (C2h-2Ch)  
Key-failure prediction status (4Fh-F4h)  
xx  
xx  
Error information  
The attribute value information is 512-byte data; the format of this data is shown  
below. The host can access this data using the SMART Read Attribute Values  
subcommand (FR register = D0h). The insurance failure threshold value data is  
512-byte data; the format of this data is shown below. The host can access this  
data using the SMART Read Attribute Thresholds subcommand (FR register =  
D1h).  
C141-E104-02EN  
5-67  
Interface  
Table 5.8 Format of device attribute value data  
Byte  
00  
Item  
Data format version number  
01  
02  
Attribute 1  
Attribute ID  
Status flag  
03  
04  
05  
Current attribute value  
Attribute value for worst case so far  
Raw attribute value  
06  
07 to 0C  
0D  
Reserved  
0E to 169 Attribute 2 to  
attribute 30  
(The format of each attribute value is the same  
as that of bytes 02 to 0D.)  
16A  
16B  
Off-line data collection status  
Self test execution status  
16C, 16D Off-line data collection execution time [sec.]  
16E  
16F  
Reserved  
Off-line data collection capability  
Trouble prediction capability flag  
Error logging capability  
170, 171  
172  
173  
Vendor unique  
174  
Simple self test execution time [min.]  
Comprehensive self test execution time [min.]  
175  
176 to 181 Reserved  
182 to 1FE Vendor unique  
1FF  
Check sum  
Table 5.9 Format of insurance failure threshold value data  
Byte Item  
00  
01  
Data format version number  
02  
Threshold 1 Attribute ID  
03  
Insurance failure threshold  
Reserved  
04 to 0D  
0E to 169 Threshold 2 to (The format of each threshold value is the same as  
Threshold 30 that of bytes 02 to 0D.)  
16A to 17B Reserved  
17C to 1FE Vendor unique  
1FF Check sum  
5-68  
C141-E104-02EN  
5.3 Host Commands  
Data format version number  
The data format version number indicates the version number of the data  
format of the device attribute values or insurance failure thresholds. The data  
format version numbers of the device attribute values and insurance failure  
thresholds are the same. When a data format is changed, the data format  
version numbers are updated.  
Attribute ID  
The attribute ID is defined as follows:  
Attribute ID  
Attribute name  
(Indicates unused attribute data.)  
Read error rate  
0
1
2
Throughput performance  
Spin up time  
3
4
Start/stop count  
5
Re-allocated sector count  
Seek error rate  
7
8
Seek time performance  
9
Power-on time  
10  
12  
Number of retries made to activate the spindle motor  
Number of power-on-power-off times  
13 to 198 (Reserved)  
199  
200  
Ultra ATA CRC error rate  
Write error rate  
201 to 255 (Unique to vendor)  
Status Flag  
Bit  
0
Meaning  
If this bit 1, it indicates that if the attribute exceeds the threshold,  
it is the attribute covered by the drive warranty.  
1
If this bit is 1 (0), it indicates the attribute only updated by an on-  
line test (off-line test).  
2
3
4
If this bit 1, it indicates the attribute that represents performance.  
If this bit 1, it indicates the attribute that represents an error rate.  
If this bit 1, it indicates the attribute that represents the  
number of occurrences.  
5
If this bit 1, it indicates the attribute that can be collected/saved  
even if the drive fault prediction function is disabled.  
6 to 15  
Reserve bit  
C141-E104-02EN  
5-69  
Interface  
Current attribute value  
The current attribute value is the normalized raw attribute data. The value  
varies between 01h and 64h. The closer the value gets to 01h, the higher the  
possibility of a failure. The device compares the attribute values with  
thresholds. When the attribute values are larger than the thresholds, the  
device is operating normally.  
Attribute value for the worst case so far  
This is the worst attribute value among the attribute values collected to date.  
This value indicates the state nearest to a failure so far.  
Raw attribute value  
Raw attributes data is retained.  
Off-line data collection status  
Bits 0 to 6: Indicates the situation of off-line data collection according to the  
table below.  
Bit 7:  
If this bit is 1, it indicates that the automatic off-line data  
collection function is enabled.  
Status Byte  
Meaning  
0
2
4
Off-line data collection is not started.  
Off-line data collection has been completed normally.  
Off-line data collection has been suspended by a command  
interrupt.  
5
6
Off-line data collection has been aborted by a command  
interrupt.  
Off-line data collection has been aborted by a fatal error.  
Self test execution status  
Bits 0 to 3: Indicates the rest of self-test in 0 to 9 (corresponding 0 to 90%).  
Bits 4 to 7: Indicates the self-test execution status at the following table.  
5-70  
C141-E104-02EN  
5.3 Host Commands  
Self-test  
Meaning  
execution status  
0
Self-test has been completed normally or has not been  
executed.  
1
2
3
4
Self-test has been stopped by the host computer.  
Self-test has been suspended by hard or soft reset.  
Self-test has been aborted by a fatal error.  
Self-test has been completed abnormally by an unknown  
meaning.  
5
Self-test has been completed abnormally by write test.  
Self-test has been completed abnormally by serbo test.  
Self-test has been completed abnormally by read test.  
Reserved  
6
7
8 to 14  
15  
Self-test is in progress.  
Off-line data collection capability  
Indicates the method of off-line data collection carried out by the drive. If the  
off-line data collection capability is 0, it indicates that off-line data collection  
is not supported.  
Bit  
Meaning  
Indicates that Execute Off-Line Immediate is supported.  
Vendor unique  
0
1
2
Indicates that off-line data collection being executed is  
aborted when a new command is received.  
3
4
Indicates that supports off-line read scan function.  
Indicates that supports self-test function.  
Failure prediction capability flag  
Bit 0: The attribute value data is saved to a media before the device enters  
power saving mode.  
Bit 1: The device automatically saves the attribute value data to a media  
after the previously set operation.  
Bits 2 to 15: Reserved bits  
Error logging capability  
Bit 0: Indicates that error logging function.  
Bits 1 to 7: Reserved bits  
C141-E104-02EN  
5-71  
Interface  
Check sum  
Two’s complement of the lower byte, obtained by adding 511-byte data one  
byte at a time from the beginning.  
Insurance failure threshold  
The limit of a varying attribute value. The host compares the attribute values  
with the thresholds to identify a failure.  
If an unrecoverable error is detected during execution of a command received by  
the device from the host computer, the device saves the SMART error log on the  
disk medium.  
The host computer can issue the SMART Read Log Sector sub-command (FR  
register = D5h, SN register = 01h) and read the SMART error log.  
Table 5.10 SMART error log data format (1/2)  
Byte  
00  
Item  
Error log version number  
Error log index  
01  
02  
Error log 1  
Command Data 1  
Device Control register  
Features register  
03  
04  
Sector Count register  
Sector Number register  
Cylinder Low register  
Cylinder High register  
Device/Head register  
Command register  
05  
06  
07  
08  
09  
0A to 0D  
Elapsed time [ms] from the point  
when the power is turned on  
until command reception  
0E to 3D  
Command Data 2 to 5 (The format of each type of  
command data is the same as  
that of byte 02 to 0D.)  
3E  
3F  
40  
41  
42  
43  
Error data  
Reserved  
Error register  
Sector Count register  
Sector Number register  
Cylinder Low register  
Cylinder High register  
5-72  
C141-E104-02EN  
5.3 Host Commands  
Table 5.10 SMART error log data format (2/2)  
Byte  
44  
Item  
Error log 1  
Error data  
Device/Head register  
Status register  
Vendor unique  
Status  
45  
46 to 58  
59  
5A, 5B  
5C to 1C3  
Total power on time [hour]  
Error log 2 (The format of each error log is the same as Byte 02 to  
to  
5B.)  
Error log 5  
1C4, 1C5 Number of unrecoverable errors that have occurred.  
1C6 to 1FE Reserved  
1FF  
Check sum  
Error log index  
Indicates the latest error log number. If an error has not occurred, 00 is  
displayed.  
Error log 1 to 5  
When an error occurs, the error log index value is incremented and  
information at the time the error occurred is recorded in the error log area  
specified by this value. When the error log index exceeds 05, it returns to 01.  
Command data 1 to 5  
Indicates five commands data in order received by the device until the error  
occurs. Commands for which an error occurred are included in Command  
Data 5.  
Error data  
Indicates the I/O register values when the error is reported.  
Status  
Bits 0 to 3: Indicates the drive status when received error commands  
according to the following table.  
Bits 4 to 7: Vendor unique  
C141-E104-02EN  
5-73  
Interface  
Status  
Meaning  
0
Unclear status  
Sleep status  
1
2
3
Standby status  
Active status or idle status (BSY bit = 0)  
Off-line data collection being executed  
Reserved  
4
5 to F  
The host computer can issue the SMART Execute Off-line Immediate sub-  
command (FR Register = D4h) and cause the device to execute a self test. When  
the self test is completed, the device saves the SMART self test log to the disk  
medium.  
The host computer can issue the SMART Read Log Sector sub-command (FR  
Register = D5h, SN Register = 06h) and can read the SMART self test log.  
Table 5.11 SMART self test log data format  
Byte  
Item  
00, 01  
02  
Self test log data format version number  
Self test log 1 Self test mode (SN Register Value)  
03  
Self test execution status  
04, 05  
Total power on time until the self test is  
completed. [hours]  
06  
Self test error No.  
Error LBA  
07 to 0A  
0B to 19  
Vendor unique  
1A to 1F9 Self test log 2 to 21  
(Each log data format is the same as that in  
byte 02 to 19.)  
1FA, 1FB Vendor unique  
1FC  
Self test index  
1FD, 1FE Reserved  
1FF  
Check sum  
Self test log 1 to 21  
When executes self test, the self test index value is incremented and the self  
test execution result is recorded in the self log test area specified by this  
value. When the self test index exceeds 21, it returns to 01.  
Self test index  
Indicates the latest self test log number. If the self test has not been executed,  
00h is displayed.  
5-74  
C141-E104-02EN  
5.3 Host Commands  
(30) SECURITY DISABLE PASSWORD (F6h)  
This command invalidates the user password already set and releases the lock  
function.  
The host transfers the 512-byte data shown in Table 5.12 to the device. The  
device compares the user password or master password in the transferred data  
with the user password or master password already set, and releases the lock  
function if the passwords are the same.  
Although this command invalidates the user password, the master password is  
retained. To recover the master password, issue the SECURITY SET  
PASSWORD command and reset the user password.  
If the user password or master password transferred from the host does not match,  
the Aborted Command error is returned.  
Issuing this command while in LOCKED MODE or FROZEN MODE returns the  
Aborted Command error.  
(The section about the SECURITY FREEZE LOCK command describes  
LOCKED MODE and FROZEN MODE.)  
C141-E104-02EN  
5-75  
Interface  
Table 5.12 Contents of security password  
Word Contents  
0
Control word  
Bit 0: Identifier  
0 = Compares the user passwords.  
1 = Compares the master passwords.  
Bits 1 to 15: Reserved  
1 to 16  
Password (32 bytes)  
17 to 255 Reserved  
At command issuance (I-O register contents))  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
1
1
0
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(31) SECURITY ERASE PREPARE (F3h)  
The SECURITY ERASE UNIT command feature is enabled by issuing the  
SECURITY ERASE PREPARE command and then the SECURITY ERASE  
UNIT command. The SECURITY ERASE PREPARE command prevents data  
from being erased unnecessarily by the SECURITY ERASE UNIT command.  
Issuing this command during FROZEN MODE returns the Aborted Command  
error.  
5-76  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
0
1
1
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(32) SECURITY ERASE UNIT (F4h)  
This command erases all user data. This command also invalidates the user  
password and releases the lock function.  
The host transfers the 512-byte data shown in Table 5.10 to the device. The  
device compares the user password or master password in the transferred data  
with the user password or master password already set. The device erases user  
data, invalidates the user password, and releases the lock function if the  
passwords are the same.  
Although this command invalidates the user password, the master password is  
retained. To recover the master password, issue the SECURITY SET  
PASSWORD command and reset the user password.  
If the SECURITY ERASE PREPARE command is not issued immediately before  
this command is issued, the Aborted Command error is returned.  
Issuing this command while in FROZEN MODE returns the Aborted Command  
error.  
C141-E104-02EN  
5-77  
Interface  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
1
0
0
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(33) SECURITY FREEZE LOCK (F5h)  
This command puts the device into FROZEN MODE. The following commands  
used to change the lock function return the Aborted Command error if the device  
is in FROZEN MODE.  
SECURITY SET PASSWORD  
SECURITY UNLOCK  
SECURITY DISABLE PASSWORD  
SECURITY ERASE UNIT  
FROZEN MODE is canceled when the power is turned off, or when hardware is  
reseted. If this command is reissued in FROZEN MODE, the command is  
completed and FROZEN MODE remains unchanged.  
Issuing this command during LOCKED MODE returns the Aborted Command  
error.  
The following medium access commands return the Aborted Command error  
when the device is in LOCKED MODE:  
5-78  
C141-E104-02EN  
5.3 Host Commands  
READ DMA  
WRITE DMA  
SECURITY DISABLE PASSWORD  
READ LONG  
WRITE LONG  
SECURITY FREEZE LOCK  
READ MULTIPLE WRITE MULTIPLE SECURITY SET PASSWORD  
READ SECTORS  
WRITE SECTORS  
WRITE VERIFY  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
1
0
1
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
C141-E104-02EN  
5-79  
Interface  
(34) SECURITY SET PASSWORD (F1h)  
This command enables a user password or master password to be set.  
The host transfers the 512-byte data shown in Table 5.13 to the device. The  
device determines the operation of the lock function according to the  
specifications of the Identifier bit and Security level bit in the transferred data.  
(Table 5.14)  
Issuing this command in LOCKED MODE or FROZEN MODE returns the  
Aborted Command error.  
Table 5.13 Contents of SECURITY SET PASSWORD data  
Word  
0
Contents  
Control word  
Bit 0 Identifier  
0 = Sets a user password.  
1 = Sets a master password.  
Bits 1 to 7 Reserved  
Bit 8 Security level  
0 = High  
1 = Maximum  
Bits 9 to 15 Reserved  
1 to 16  
17  
Password (32 bytes)  
Master password version number  
18 to 255 Reserved  
Table 5.14 Relationship between combination of Identifier and Security level, and  
operation of the lock function  
Identifier  
User  
Level  
High  
Description  
The specified password is saved as a new user password.  
The lock function is enabled after the device is turned off  
and then on. LOCKED MODE can be canceled using the  
user password or the master password already set.  
Master  
User  
High  
The specified password is saved as a new master password.  
The lock function is not enabled.  
Maximum The specified password is saved as a new user password.  
The lock function is enabled after the device is turned off  
and then on. LOCKED MODE can be canceled using the  
user password only. The master password already set  
cannot cancel LOCKED MODE.  
Master  
Maximum The specified password is saved as a new master password.  
The lock function is not enabled.  
5-80  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
0
0
1
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(35) SECURITY UNLOCK  
This command cancels LOCKED MODE.  
The host transfers the 512-byte data shown in Table 5.12 to the device. Operation  
of the device varies as follows depending on whether the host specifies the master  
password.  
When the master password is selected  
When the security level is LOCKED MODE is high, the password is  
compared with the master password already set. If the passwords are the  
same, LOCKED MODE is canceled. Otherwise, the Aborted Command error  
is returned. If the security level in LOCKED MODE is set to the highest  
level, the Aborted Command error is always returned.  
When the user password is selected  
The password is compared with the user password already set. If the  
passwords are the same, LOCKED MODE is canceled. Otherwise, the  
Aborted Command error is returned.  
If the password comparison fails, the device decrements the UNLOCK counter.  
The UNLOCK counter initially has a value of five. When the value of the  
UNLOCK counter reaches zero, this command or the SECURITY ERASE UNIT  
command causes the Aborted Command error until the device is turned off and  
then on, or until a hardware reset is executed. Issuing this command with  
C141-E104-02EN  
5-81  
Interface  
LOCKED MODE canceled (in UNLOCK MODE) has no affect on the UNLOCK  
counter.  
Issuing this command in FROZEN MODE returns the Aborted Command error.  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
1
0
0
1
0
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
(36) FLUSH CACHE (E7)  
This command is used to order to write every write cache data stored by the  
device into the medium. BSY bit is held at "1" until every data has been written  
normally or an error has occurred. The device performs every error recovery so  
that the data are read correctly.  
When executing this command, the reading of the data may take several seconds  
if much data are to be read.  
In case a non-recoverable error has occurred while the data is being read, the error  
generation address is put into the command block register before ending the  
command. This error sector is deleted from the write cache data, and the  
remaining cache data is written into the medium by the execution of the next  
Flush Cache command.  
5-82  
C141-E104-02EN  
5.3 Host Commands  
At command issuance (I-O register contents)  
1F7h(CM)  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(FR)  
1
x
1
x
1
x
0
0
1
1
1
DV xx  
xx  
xx  
xx  
xx  
xx  
At command completion (I-O register contents to be read)  
1F7h(ST) Status information  
1F6h(DH)  
1F5h(CH)  
1F4h(CL)  
1F3h(SN)  
1F2h(SC)  
1F1h(ER)  
x
xx  
xx  
xx  
xx  
x
x
DV xx  
Error information  
5.3.3 Error posting  
Table 5.15 lists the defined errors that are valid for each command.  
Table 5.15 Command code and parameters (1 of 2)  
Command name  
Error register (X’1F1’)  
Status register (X’1F7’)  
ICRC  
UNC  
V
INDF  
V
ABRT  
V
TK0NF  
DRDY  
DWF  
V
ERR  
V
READ SECTOR(S)  
WRITE SECTOR(S)  
READ MULTIPLE  
WRITE MULTIPLE  
READ DMA  
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
WRITE DMA  
V
V
V
V
WRITE VERIFY  
V
V
V
V
V
V
READ VERIFY SECTOR(S)  
V
V
V
V
V:  
*:  
Valid on this command  
See the command descriptions.  
C141-E104-02EN  
5-83  
Interface  
Table 5.15 Command code and parameters (2 of 2)  
Command name  
Error register (X’1F1’)  
Status register (X’1F7’)  
ICRC  
UNC  
INDF  
ABRT  
V
V
V
V
V
V
V
V
V
*
TK0NF  
V
DRDY  
DWF  
V
ERR  
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
RECALIBRATE  
V
V
V
V
V
V
V
V
V
SEEK  
V
V
INITIALIZE DEVICE PARAMETERS  
IDENTIFY DEVICE  
IDENTIFY DEVICE DMA  
SET FEATURES  
V
V
V
V
SET MULTIPLE MODE  
SET MAX ADDRESS  
READ NATIVE MAX ADDRESS  
EXECUTE DEVICE DIAGNOSTIC  
READ LONG  
V
V
V
V
*
*
*
V
V
*
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
WRITE LONG  
READ BUFFER  
WRITE BUFFER  
IDLE  
IDLE IMMEDIATE  
STANDBY  
STANDBY IMMEDIATE  
SLEEP  
CHECK POWER MODE  
SMART  
V
SECURITY DISABLE PASSWORD  
SECURITY ERASE PREPARE  
SECURITY ERASE UNIT  
SECURITY FREEZE LOCK  
SECURITY SET PASSWORD  
SECURITY UNLOCK  
FLUSH CACHE  
V
Invalid command  
V:  
*:  
Valid on this command  
See the command descriptions.  
5-84  
C141-E104-02EN  
5.4 Command Protocol  
5.4 Command Protocol  
The host should confirm that the BSY bit of the Status register of the device is 0  
prior to issue a command. If BSY bit is 1, the host should wait for issuing a  
command until BSY bit is cleared to 0.  
Commands can be executed only when the DRDY bit of the Status register is 1.  
However, the following commands can be executed even if DRDY bit is 0.  
EXECUTE DEVICE DIAGNOSTIC  
INITIALIZE DEVICE PARAMETERS  
5.4.1 Data transferring commands from device to host  
The execution of the following commands involves data transfer from the device  
to the host.  
IDENTIFY DEVICE.  
IDENTIFY DEVICE DMA  
READ SECTOR(S)  
READ LONG  
READ BUFFER  
SMART  
The execution of these commands includes the transfer one or more sectors of  
data from the device to the host. In the READ LONG command, 516 bytes are  
transferred. Following shows the protocol outline.  
a) The host writes any required parameters to the Features, Sector Count, Sector  
Number, Cylinder, and Device/Head registers.  
b) The host writes a command code to the Command register.  
c) The device sets the BSY bit of the Status register and prepares for data  
transfer.  
d) When one sector of data is available for transfer to the host, the device sets  
DRQ bit and clears BSY bit. The drive then asserts INTRQ signal.  
e) After detecting the INTRQ signal assertion, the host reads the Status register.  
The host reads one sector of data via the Data register. In response to the  
Status register being read, the device negates the INTRQ signal.  
f) The drive clears DRQ bit to 0. If transfer of another sector is requested, the  
device sets the BSY bit and steps d) and after are repeated.  
Even if an error is encountered, the device prepares for data transfer by setting the  
DRQ bit. Whether or not to transfer the data is determined for each host. In other  
C141-E104-02EN  
5-85  
Interface  
words, the host should receive the relevant sector of data (512 bytes of uninsured  
dummy data) or release the DRQ status by resetting.  
Figure 5.3 shows an example of READ SECTOR(S) command protocol, and  
Figure 5.4 shows an example protocol for command abort.  
Figure 5.3 Read Sector(s) command protocol  
IMPORTANT  
For transfer of a sector of data, the host needs to read Status register  
(X’1F7’) in order to clear INTRQ (interrupt) signal. The Status  
register should be read within a period from the DRQ setting by the  
device to 50 ms after the completion of the sector data transfer.  
5-86  
C141-E104-02EN  
5.4 Command Protocol  
Note that the host does not need to read the Status register for the  
reading of a single sector or the last sector in multiple-sector  
reading. If the timing to read the Status register does not meet  
above condition, normal data transfer operation is not guaranteed.  
When the host new command even if the device requests the data  
transfer (setting in DRQ bit), the correct device operation is not  
guaranteed.  
Figure 5.4 Protocol for command abort  
5.4.2 Data transferring commands from host to device  
The execution of the following commands involves Data transfer from the host to  
the drive.  
WRITE SECTOR(S)  
WRITE LONG  
WRITE BUFFER  
WRITE VERIFY  
SECURITY DISABLE PASSWORD  
SECURITY ERASE UNIT  
SECURITY SET PASSWORD  
SECURITY UNCLOK  
The execution of these commands includes the transfer one or more sectors of  
data from the host to the device. In the WRITE LONG command, 516 bytes are  
transferred. Following shows the protocol outline.  
C141-E104-02EN  
5-87  
Interface  
a) The host writes any required parameters to the Features, Sector Count, Sector  
Number, Cylinder, and Device/Head registers.  
b) The host writes a command code in the Command register. The drive sets the  
BSY bit of the Status register.  
c) When the device is ready to receive the data of the first sector, the device sets  
DRQ bit and clears BSY bit.  
d) The host writes one sector of data through the Data register.  
e) The device clears the DRQ bit and sets the BSY bit.  
f) When the drive completes transferring the data of the sector, the device clears  
BSY bit and asserts INTRQ signal. If transfer of another sector is requested,  
the drive sets the DRQ bit.  
g) After detecting the INTRQ signal assertion, the host reads the Status register.  
h) The device resets INTRQ (the interrupt signal).  
i) If transfer of another sector is requested, steps d) and after are repeated.  
Figure 5.5 shows an example of WRITE SECTOR(S) command protocol.  
Figure 5.5 WRITE SECTOR(S) command protocol  
5-88  
C141-E104-02EN  
5.4 Command Protocol  
IMPORTANT  
For transfer of a sector of data, the host needs to read Status register  
(X’1F7’) in order to clear INTRQ (interrupt) signal. The Status  
register should be read within a period from the DRQ setting by the  
device to 50 µs after the completion of the sector data transfer.  
Note that the host does not need to read the Status register for the  
first and the last sector to be transferred. If the timing to read the  
Status register does not meet above condition, normal data transfer  
operation is not assured guaranteed.  
When the host issues the command even if the drive requests the  
data transfer (DRQ bit is set), or when the host executes resetting,  
the device correct operation is not guaranteed.  
5.4.3 Commands without data transfer  
Execution of the following commands does not involve data transfer between the  
host and the device.  
RECABLIBRATE  
SEEK  
READY VERIFY SECTOR(S)  
EXECUTE DEVICE DIAGNOSTIC  
INITIALIZE DEVICE PARAMETERS  
SET FEATURES  
SET MULTIPLE MODE  
SET MAX ADDRESS  
READ NATIVE MAX ADDRESS  
IDLE  
IDLE IMMEDIATE  
STANDBY  
STANDBY IMMEDIATE  
CHECK POWER MODE  
SECURITY ERASE PREPARE  
SECURITY FREEZE LOCK  
FLUSH CACHE  
Figure 5.6 shows the protocol for the command execution without data transfer.  
C141-E104-02EN  
5-89  
Interface  
Figure 5.6 Protocol for the command execution without data transfer  
5.4.4 Other commands  
READ MULTIPLE  
SLEEP  
WRITE MULTIPLE  
See the description of each command.  
5.4.5 DMA data transfer commands  
READ DMA  
WRITE DMA  
Starting the DMA transfer command is the same as the READ SECTOR(S) or  
WRITE SECTOR(S) command except the point that the host initializes the DMA  
channel preceding the command issuance.  
Interruption processing for DMA transfer does not issue interruptions in any  
intermediate sector when a multisector command is executed.  
The following outlines the protocol:  
The interrupt processing for the DMA transfer differs the following point.  
The interrupt processing for the DMA transfer differs the following point.  
a) The host writes any parameters to the Features, Sector Count, Sector Number,  
Cylinder, and Device/Head register.  
b) The host initializes the DMA channel  
c) The host writes a command code in the Command register.  
d) The device sets the BSY bit of the Status register.  
e) The device asserts the DMARQ signal after completing the preparation of  
data transfer. The device asserts either the BSY bit or DRQ bit during DMA  
data transfer.  
5-90  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
f) When the command execution is completed, the device clears both BSY and  
DRQ bits and asserts the INTRQ signal. Then, the host reads the Status  
register.  
g) The host resets the DMA channel.  
Figure 5.7 shows the correct DMA data transfer protocol.  
f
g
d
d
f
e
d
f
e
Figure 5.7 Normal DMA data transfer  
C141-E104-02EN  
5-91  
Interface  
5.5 Ultra DMA Feature Set  
5.5.1 Overview  
Ultra DMA is a data transfer protocol used with the READ DMA and WRITE  
DMA commands. When this protocol is enabled it shall be used instead of the  
Multiword DMA protocol when these commands are issued by the host. This  
protocol applies to the Ultra DMA data burst only. When this protocol is used  
there are no changes to other elements of the ATA protocol (e.g.: Command  
Block Register access).  
Several signal lines are redefined to provide new functions during an Ultra DMA  
burst. These lines assume these definitions when 1) an Ultra DMA Mode is  
selected, and 2) a host issues a READ DMA or a WRITE DMA, command  
requiring data transfer, and 3) the host asserts DMACK-. These signal lines  
revert back to the definitions used for non-Ultra DMA transfers upon the negation  
of DMACK- by the host at the termination of an Ultra DMA burst. All of the  
control signals are unidirectional. DMARQ and DMACK- retain their standard  
definitions.  
With the Ultra DMA protocol, the control signal (STROBE) that latches data  
from DD (15:0) is generated by the same agent (either host or device) that drives  
the data onto the bus. Ownership of DD (15:0) and this data strobe signal are  
given either to the device during an Ultra DMA data in burst or to the host for an  
Ultra DMA data out burst.  
During an Ultra DMA burst a sender shall always drive data onto the bus, and  
after a sufficient time to allow for propagation delay, cable settling, and setup  
time, the sender shall generate a STROBE edge to latch the data. Both edges of  
STROBE are used for data transfers so that the frequency of STROBE is limited  
to the same frequency as the data. The highest fundamental frequency on the  
cable shall be 16.67 million transitions per second or 8.33 MHz (the same as the  
maximum frequency for PIO Mode 4 and DMA Mode 2).  
Words in the IDENTIFY DEVICE data indicate support of the Ultra DMA feature  
and the Ultra DMA Modes the device is capable of supporting. The Set transfer  
mode subcommand in the SET FEATURES command shall be used by a host to  
select the Ultra DMA Mode at which the system operates. The Ultra DMA Mode  
selected by a host shall be less than or equal to the fastest mode of which the  
device is capable. Only the Ultra DMA Mode shall be selected at any given time.  
All timing requirements for a selected Ultra DMA Mode shall be satisfied.  
Devices supporting Ultra DMA Mode 2 shall also support Ultra DMA Modes 0  
and 1. Devices supporting Ultra DMA Mode 1 shall also support Ultra DMA  
Mode 0.  
An Ultra DMA capable device shall retain its previously selected Ultra DMA  
Mode after executing a Software reset sequence. An Ultra DMA capable device  
shall clear any previously selected Ultra DMA Mode and revert to its default non-  
Ultra DMA Modes after executing a Power on or hardware reset.  
5-92  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
Both the host and device perform a CRC function during an Ultra DMA burst. At  
the end of an Ultra DMA burst the host sends the its CRC data to the device. The  
device compares its CRC data to the data sent from the host. If the two values do  
not match the device reports an error in the error register at the end of the  
command. If an error occurs during one or more Ultra DMA bursts for any one  
command, at the end of the command, the device shall report the first error that  
occurred.  
5.5.2 Phases of operation  
An Ultra DMA data transfer is accomplished through a series of Ultra DMA data  
in or data out bursts. Each Ultra DMA burst has three mandatory phases of  
operation: the initiation phase, the data transfer phase, and the Ultra DMA burst  
termination phase. In addition, an Ultra DMA burst may be paused during the  
data transfer phase (see 5.5.3 and 5.5.4 for the detailed protocol descriptions for  
each of these phases, 5.6.4 defines the specific timing requirements). In the  
following rules DMARDY- is used in cases that could apply to either  
DDMARDY- or HDMARDY-, and STROBE is used in cases that could apply to  
either DSTROBE or HSTROBE. The following are general Ultra DMA rules.  
a) An Ultra DMA burst is defined as the period from an assertion of DMACK-  
by the host to the subsequent negation of DMACK-.  
b) A recipient shall be prepared to receive at least two data words whenever it  
enters or resumes an Ultra DMA burst.  
5.5.2.1 Ultra DMA burst initiation phase  
a) The Ultra DMA burst initiation phase is started by the assertion of DMARQ  
signal by the device, and is ended when the transmitting side has inverted  
STROBE signal for transmitting the first data.  
b) The Ultra DMA burst requires the assertion of DMARQ signal by the device.  
c) The host asserts DMACK-signal when it is able to start the requested burst.  
d) The host always asserts DMACK signal after detecting the first assertion of  
DMARQ signal.  
e) Ultra DMA data in burst  
The device starts transmission of the data to DD (15 : 0) when;  
DMACK-signal assertion has been detected,  
STOP signal negation has been detected, or  
HDMARDY-signal assertion has been detected.  
f) Ultra DMA data out burst  
The device should not invert the state of this signal in the period from the  
moment of DMARQ signal assertion or DDMARDY-signal assertion to the  
moment of inversion of the first STROBE signal.  
C141-E104-02EN  
5-93  
Interface  
g) Ultra DMA data in burst  
The device should not invert the state of this signal in the period from the  
moment of STOP signal negation or HDMARDY-signal assertion to the  
moment of inversion of the first STROBE signal.  
5.5.2.2 Data transfer phase  
a) The Data transfer phase is defined as the period from The Ultra DMA burst  
initiation phase to Ultra DMA burst termination phase.  
b) The receiving side stops the Ultra DMA burst temporarily by negating  
DMARDY-signal, and then restarts the Ultra DMA burst by asserting again.  
c) The transmitting side stops the Ultra DMA burst temporarily by not-  
performing inversion of STROBE signal, and then restarts the Ultra DMA  
burst by restarting the inversion.  
d) When the transmitting side has stopped the inversion of STROBE signal, the  
receiving side should not output termination request signal immediately.  
The receiving side should negate DMARDY signal when no termination  
request signal has been received from the transmission side, and then should  
output the termination request signal when a certain wait time has elapsed.  
e) The transmitting side is allowed to send STROBE signal at a transfer speed  
that is lower than the one in the transferable fastest Ultra DMA mode, but is  
not allowed to send the STROBE signal at a higher speed than this.  
The receiving side should be able to receive the data in the transferable  
fastest Ultra DMA mode.  
5.5.2.3 Ultra DMA burst termination phase  
a) The transmitting side or receiving side is allowed to end the Ultra DMA  
burst.  
b) The Ultra DMA burst termination is not the end of the command execution.  
When the Ultra DMA burst termination has occurred before the ending of the  
command, the command should be ended by starting a new Ultra DMA burst,  
or the host should issue command abort by outputting hard reset or soft reset  
to the device.  
c) The Ultra DMA burst should be stopped temporarily before the receiving side  
outputs the ending request.  
d) The host outputs the ending request by asserting STOP signal, and then the  
device negates DMARQ signal to confirm it.  
e) The device outputs the ending request by negating DMARQ signal, and then  
the host asserts STOP signal to confirm it.  
5-94  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
f) Once the transmitting side has outputted the ending request, the output state  
of STROBE signal should not be changed unless the receiving side has  
confirmed it. Then, if the STROBE signal is not in asserted state, The  
transmitting side should assert the STROBE signal. However, the assertion of  
the STROBE signal should not cause the data transfer to occur.  
g) The transmitting side should return the STROBE signal to its asserted state  
immediately after receiving the ending request from the receiving side.  
However, the returning of the STROBE signal to its asserted state should not  
cause the data transfer to occur and CRC to be perform.  
h) Once the receiving side has outputted the ending request, the negated state of  
the DMARDY signal should not be changed for the remaining Ultra DMA  
burst to be performed.  
i) The receiving side should neglect the inversion of the STROBE signal if  
DMARQ signal has been negated or STOP signal has been asserted.  
5.5.3 Ultra DMA data in commands  
5.5.3.1 Initiating an Ultra DMA data in burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.1 and 5.6.4.2 for specific timing requirements):  
1) The host shall keep DMACK- in the negated state before an Ultra DMA burst  
is initiated.  
2) The device shall assert DMARQ to initiate an Ultra DMA burst. After  
assertion of DMARQ the device shall not negate DMARQ until after the first  
negation of DSTROBE.  
3) Steps (3), (4) and (5) may occur in any order or at the same time. The host  
shall assert STOP.  
4) The host shall negate HDMARDY-.  
5) The host shall negate CS0-, CS1-, DA2, DA1, and DA0. The host shall keep  
CS0-, CS1-, DA2, DA1, and DA0 negated until after negating DMACK- at  
the end of the burst.  
6) Steps (3), (4) and (5) shall have occurred at least tACK before the host asserts  
DMACK-. The host shall keep DMACK- asserted until the end of an Ultra  
DMA burst.  
7) The host shall release DD (15:0) within tAZ after asserting DMACK-.  
8) The device may assert DSTROBE tZIORDY after the host has asserted  
DMACK-. Once the device has driven DSTROBE the device shall not  
release DSTROBE until after the host has negated DMACK- at the end of an  
Ultra DMA burst.  
9) The host shall negate STOP and assert HDMARDY- within tENV after  
asserting DMACK-. After negating STOP and asserting HDMARDY-, the  
C141-E104-02EN  
5-95  
Interface  
host shall not change the state of either signal until after receiving the first  
transition of DSTROBE from the device (i.e., after the first data word has  
been received).  
10) The device shall drive DD (15:0) no sooner than tZAD after the host has  
asserted DMACK-, negated STOP, and asserted HDMARDY-.  
11) The device shall drive the first word of the data transfer onto DD (15:0).  
This step may occur when the device first drives DD (15:0) in step (10).  
12) To transfer the first word of data the device shall negate DSTROBE within tFS  
after the host has negated STOP and asserted HDMARDY-. The device shall  
negate DSTROBE no sooner than tDVS after driving the first word of data onto  
DD (15:0).  
5.5.3.2 The data in transfer  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.3 and 5.6.4.2):  
1) The device shall drive a data word onto DD (15:0).  
2) The device shall generate a DSTROBE edge to latch the new word no sooner  
than tDVS after changing the state of DD (15:0). The device shall generate a  
DSTROBE edge no more frequently than tCYC for the selected Ultra DMA  
Mode. The device shall not generate two rising or two falling DSTROBE  
edges more frequently than 2tCYC for the selected Ultra DMA mode.  
3) The device shall not change the state of DD (15:0) until at least tDVH after  
generating a DSTROBE edge to latch the data.  
4) The device shall repeat steps (1), (2) and (3) until the data transfer is  
complete or an Ultra DMA burst is paused, whichever occurs first.  
5.5.3.3 Pausing an Ultra DMA data in burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.4 and 5.6.4.2 for specific timing requirements).  
a) Device pausing an Ultra DMA data in burst  
1) The device shall not pause an Ultra DMA burst until at least one data  
word of an Ultra DMA burst has been transferred.  
2) The device shall pause an Ultra DMA burst by not generating DSTROBE  
edges.  
NOTE - The host shall not immediately assert STOP to initiate Ultra  
DMA burst termination when the device stops generating STROBE  
edges. If the device does not negate DMARQ, in order to initiate  
ULTRA DMA burst termination, the host shall negate HDMARDY- and  
wait tRP before asserting STOP.  
3) The device shall resume an Ultra DMA burst by generating a DSTROBE  
edge.  
5-96  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
b) Host pausing an Ultra DMA data in burst  
1) The host shall not pause an Ultra DMA burst until at least one data word  
of an Ultra DMA burst has been transferred.  
2) The host shall pause an Ultra DMA burst by negating HDMARDY-.  
3) The device shall stop generating DSTROBE edges within tRFS of the host  
negating HDMARDY-.  
4) If the host negates HDMARDY- within tSR after the device has generated  
a DSTROBE edge, then the host shall be prepared to receive zero or one  
additional data words. If the host negates HDMARDY- greater than tSR  
after the device has generated a DSTROBE edge, then the host shall be  
prepared to receive zero, one or two additional data words. The  
additional data words are a result of cable round trip delay and tRFS  
timing for the device.  
5) The host shall resume an Ultra DMA burst by asserting HDMARDY-.  
5.5.3.4 Terminating an Ultra DMA data in burst  
a) Device terminating an Ultra DMA data in burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.5 and 5.6.4.2 for specific timing  
requirements):  
1) The device shall initiate termination of an Ultra DMA burst by not  
generating DSTROBE edges.  
2) The device shall negate DMARQ no sooner than tSS after generating the  
last DSTROBE edge. The device shall not assert DMARQ again until  
after the Ultra DMA burst is terminated.  
3) The device shall release DD (15:0) no later than tAZ after negating  
DMARQ.  
4) The host shall assert STOP within tLI after the device has negated  
DMARQ. The host shall not negate STOP again until after the Ultra  
DMA burst is terminated.  
5) The host shall negate HDMARDY- within tLI after the device has  
negated DMARQ. The host shall continue to negate HDMARDY- until  
the Ultra DMA burst is terminated. Steps (4) and (5) may occur at the  
same time.  
6) The host shall drive DD (15:0) no sooner than tZAH after the device has  
negated DMARQ. For this step, the host may first drive DD (15:0) with  
the result of its CRC calculation (see 5.5.5):  
C141-E104-02EN  
5-97  
Interface  
7) If DSTROBE is negated, the device shall assert DSTROBE within tLI  
after the host has asserted STOP. No data shall be transferred during this  
assertion. The host shall ignore this transition on DSTROBE.  
DSTROBE shall remain asserted until the Ultra DMA burst is  
terminated.  
8) If the host has not placed the result of its CRC calculation on DD (15:0)  
since first driving DD (15:0) during (6), the host shall place the result of  
its CRC calculation on DD (15:0) (see 5.5.5).  
9) The host shall negate DMACK- no sooner than tMLI after the device has  
asserted DSTROBE and negated DMARQ and the host has asserted  
STOP and negated HDMARDY-, and no sooner than tDVS after the host  
places the result of its CRC calculation on DD (15:0).  
10) The device shall latch the host's CRC data from DD (15:0) on the  
negating edge of DMACK-.  
11) The device shall compare the CRC data received from the host with the  
results of its own CRC calculation. If a miscompare error occurs during  
one or more Ultra DMA bursts for any one command, at the end of the  
command the device shall report the first error that occurred (see 5.5.5).  
12) The device shall release DSTROBE within tIORDYZ after the host negates  
DMACK-.  
13) The host shall not negate STOP no assert HDMARDY- until at least tACK  
after negating DMACK-.  
14) The host shall not assert DIOR-, CS0-, CS1-, DA2, DA1, or DA0 until at  
least tACK after negating DMACK.  
b) Host terminating an Ultra DMA data in burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.6 and 5.6.4.2 for specific timing  
requirements):  
1) The host shall not initiate Ultra DMA burst termination until at least one  
data word of an Ultra DMA burst has been transferred.  
2) The host shall initiate Ultra DMA burst termination by negating  
HDMARDY-. The host shall continue to negate HDMARDY- until the  
Ultra DMA burst is terminated.  
3) The device shall stop generating DSTROBE edges within tRFS of the host  
negating HDMARDY-.  
4) If the host negates HDMARDY- within tSR after the device has generated  
a DSTROBE edge, then the host shall be prepared to receive zero or one  
additional data words. If the host negates HDMARDY- greater than tSR  
after the device has generated a DSTROBE edge, then the host shall be  
prepared to receive zero, one or two additional data words. The  
additional data words are a result of cable round trip delay and tRFS  
timing for the device.  
5-98  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
5) The host shall assert STOP no sooner than tRP after negating  
HDMARDY-. The host shall not negate STOP again until after the Ultra  
DMA burst is terminated.  
6) The device shall negate DMARQ within tLI after the host has asserted  
STOP. The device shall not assert DMARQ again until after the Ultra  
DMA burst is terminated.  
7) If DSTROBE is negated, the device shall assert DSTROBE within tLI  
after the host has asserted STOP. No data shall be transferred during this  
assertion. The host shall ignore this transition on DSTROBE.  
DSTROBE shall remain asserted until the Ultra DMA burst is  
terminated.  
8) The device shall release DD (15:0) no later than tAZ after negating  
DMARQ.  
9) The host shall drive DD (15:0) no sooner than tZAH after the device has  
negated DMARQ. For this step, the host may first drive DD (15:0) with  
the result of its CRC calculation (see 5.5.5).  
10) If the host has not placed the result of its CRC calculation on DD (15:0)  
since first driving DD (15:0) during (9), the host shall place the result of  
its CRC calculation on DD (15:0) (see 5.5.5).  
11) The host shall negate DMACK- no sooner than tMLI after the device has  
asserted DSTROBE and negated DMARQ and the host has asserted  
STOP and negated HDMARDY-, and no sooner than tDVS after the host  
places the result of its CRC calculation on DD (15:0).  
12) The device shall latch the host's CRC data from DD (15:0) on the  
negating edge of DMACK-.  
13) The device shall compare the CRC data received from the host with the  
results of its own CRC calculation. If a miscompare error occurs during  
one or more Ultra DMA burst for any one command, at the end of the  
command, the device shall report the first error that occurred (see 5.5.5).  
14) The device shall release DSTROBE within tIORDYZ after the host negates  
DMACK-.  
15) The host shall neither negate STOP nor assert HDMARDY- until at least  
t
ACK after the host has negated DMACK-.  
16) The host shall not assert DIOR-, CS0-, CS1-, DA2, DA1, or DA0 until at  
least tACK after negating DMACK.  
C141-E104-02EN  
5-99  
Interface  
5.5.4 Ultra DMA data out commands  
5.5.4.1 Initiating an Ultra DMA data out burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.7 and 5.6.4.2 for specific timing requirements):  
1) The host shall keep DMACK- in the negated state before an Ultra DMA burst  
is initiated.  
2) The device shall assert DMARQ to initiate an Ultra DMA burst.  
3) Steps (3), (4), and (5) may occur in any order or at the same time. The host  
shall assert STOP.  
4) The host shall assert HSTROBE.  
5) The host shall negate CS0-, CS1-, DA2, DA1, and DA0. The host shall keep  
CS0-, CS1-, DA2, DA1, and DA0 negated until after negating DMACK- at  
the end of the burst.  
6) Steps (3), (4), and (5) shall have occurred at least tACK before the host asserts  
DMACK-. The host shall keep DMACK- asserted until the end of an Ultra  
DMA burst.  
7) The device may negate DDMARDY- tZIORDY after the host has asserted  
DMACK-. Once the device has negated DDMARDY-, the device shall not  
release DDMARDY- until after the host has negated DMACK- at the end of  
an Ultra DMA burst.  
8) The host shall negate STOP within tENV after asserting DMACK-. The host  
shall not assert STOP until after the first negation of HSTROBE.  
9) The device shall assert DDMARDY- within tLI after the host has negated  
STOP. After asserting DMARQ and DDMARDY- the device shall not  
negate either signal until after the first negation of HSTROBE by the host.  
10) The host shall drive the first word of the data transfer onto DD (15:0). This  
step may occur any time during Ultra DMA burst initiation.  
11) To transfer the first word of data: the host shall negate HSTROBE no sooner  
than tLI after the device has asserted DDMARDY-. The host shall negate  
HSTROBE no sooner than tDVS after the driving the first word of data onto  
DD (15:0).  
5.5.4.2 The data out transfer  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.8 and 5.6.4.2 for specific timing requirements):  
1) The host shall drive a data word onto DD (15:0).  
2) The host shall generate an HSTROBE edge to latch the new word no sooner  
than tDVS after changing the state of DD (15:0). The host shall generate an  
5-100  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
HSTROBE edge no more frequently than tCYC for the selected Ultra DMA  
Mode. The host shall not generate two rising or falling HSTROBE edges  
more frequently than 2 tCYC for the selected Ultra DMA mode.  
3) The host shall not change the state of DD (15:0) until at least tDVH after  
generating an HSTROBE edge to latch the data.  
4) The host shall repeat steps (1), (2) and (3) until the data transfer is complete  
or an Ultra DMA burst is paused, whichever occurs first.  
5.5.4.3 Pausing an Ultra DMA data out burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.9 and 5.6.4.2 for specific timing requirements).  
a) Host pausing an Ultra DMA data out burst  
1) The host shall not pause an Ultra DMA burst until at least one data word  
of an Ultra DMA burst has been transferred.  
2) The host shall pause an Ultra DMA burst by not generating an  
HSTROBE edge.  
Note: The device shall not immediately negate DMARQ to initiate Ultra  
DMA burst termination when the host stops generating HSTROBE  
edges. If the host does not assert STOP, in order to initiate Ultra DMA  
burst termination, the device shall negate DDMARDY- and wait tRP  
before negating DMARQ.  
3) The host shall resume an Ultra DMA burst by generating an HSTROBE  
edge.  
b) Device pausing an Ultra DMA data out burst  
1) The device shall not pause an Ultra DMA burst until at least one data  
word of an Ultra DMA burst has been transferred.  
2) The device shall pause an Ultra DMA burst by negating DDMARDY-.  
3) The host shall stop generating HSTROBE edges within tRFS of the device  
negating DDMARDY-.  
4) If the device negates DDMARDY- within tSR after the host has generated  
an HSTROBE edge, then the device shall be prepared to receive zero or  
one additional data words. If the device negates DDMARDY- greater  
than tSR after the host has generated an HSTROBE edge, then the device  
shall be prepared to receive zero, one or two additional data words. The  
additional data words are a result of cable round trip delay and tRFS  
timing for the host.  
5) The device shall resume an Ultra DMA burst by asserting DDMARDY-.  
C141-E104-02EN  
5-101  
Interface  
5.5.4.4 Terminating an Ultra DMA data out burst  
a) Host terminating an Ultra DMA data out burst  
The following stops shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.10 and 5.6.4.2 for specific timing  
requirements):  
1) The host shall initiate termination of an Ultra DMA burst by not  
generating HSTROBE edges.  
2) The host shall assert STOP no sooner than tSS after it last generated an  
HSTROBE edge. The host shall not negate STOP again until after the  
Ultra DMA burst is terminated.  
3) The device shall negate DMARQ within tLI after the host asserts STOP.  
The device shall not assert DMARQ again until after the Ultra DMA  
burst is terminated.  
4) The device shall negate DDMARDY- with tLI after the host has negated  
STOP. The device shall not assert DDMARDY- again until after the  
Ultra DMA burst termination is complete.  
5) If HSTROBE is negated, the host shall assert HSTROBE with tLI after  
the device has negated DMARQ. No data shall be transferred during this  
assertion. The device shall ignore this transition on HSTROBE.  
HSTROBE shall remain asserted until the Ultra DMA burst is  
terminated.  
6) The host shall place the result of its CRC calculation on DD (15:0) (see  
5.5.5)  
7) The host shall negate DMACK- no sooner than tMLI after the host has  
asserted HSTROBE and STOP and the device has negated DMARQ and  
DDMARDY-, and no sooner than tDVS after placing the result of its CRC  
calculation on DD (15:0).  
8) The device shall latch the host's CRC data from DD (15:0) on the  
negating edge of DMACK-.  
9) The device shall compare the CRC data received from the host with the  
results of its own CRC calculation. If a miscompare error occurs during  
one or more Ultra DMA bursts for any one command, at the end of the  
command, the device shall report the first error that occurred (see 5.5.5).  
10) The device shall release DDMARDY- within tIORDYZ after the host has  
negated DMACK-.  
11) The host shall neither negate STOP nor negate HSTROBE until at least  
t
ACK after negating DMACK-.  
12) The host shall not assert DIOW-, CS0-, CS1-, DA2, DA1, or DA0 until  
at least tACK after negating DMACK.  
5-102  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
b) Device terminating an Ultra DMA data out burst  
The following steps shall occur in the order they are listed unless otherwise  
specifically allowed (see 5.6.4.11 and 5.6.4.2 for specific timing  
requirements):  
1) The device shall not initiate Ultra DMA burst termination until at least  
one data word of an Ultra DMA burst has been transferred.  
2) The device shall initiate Ultra DMA burst termination by negating  
DDMARDY-.  
3) The host shall stop generating an HSTROBE edges within tRFS of the  
device negating DDMARDY-.  
4) If the device negates DDMARDY- within tSR after the host has generated  
an HSTROBE edge, then the device shall be prepared to receive zero or  
one additional data words. If the device negates DDMARDY- greater  
than tSR after the host has generated an HSTROBE edge, then the device  
shall be prepared to receive zero, one or two additional data words. The  
additional data words are a result of cable round trip delay and tRFS  
timing for the host.  
5) The device shall negate DMARQ no sooner than tRP after negating  
DDMARDY-. The device shall not assert DMARQ again until after the  
Ultra DMA burst is terminated.  
6) The host shall assert STOP with tLI after the device has negated  
DMARQ. The host shall not negate STOP again until after the Ultra  
DMA burst is terminated.  
7) If HSTROBE is negated, the host shall assert HSTROBE with tLI after  
the device has negated DMARQ. No data shall be transferred during this  
assertion. The device shall ignore this transition of HSTROBE.  
HSTROBE shall remain asserted until the Ultra DMA burst is  
terminated.  
8) The host shall place the result of its CRC calculation on DD (15:0) (see  
5.5.5).  
9) The host shall negate DMACK- no sooner than tMLI after the host has  
asserted HSTROBE and STOP and the device has negated DMARQ and  
DDMARDY-, and no sooner than tDVS after placing the result of its CRC  
calculation on DD (15:0).  
10) The device shall latch the host's CRC data from DD (15:0) on the  
negating edge of DMACK-.  
11) The device shall compare the CRC data received from the host with the  
results of its own CRC calculation. If a miscompare error occurs during  
one or more Ultra DMA bursts for any one command, at the end of the  
command, the device shall report the first error that occurred (see 5.5.5).  
12) The device shall release DDMARDY- within tIORDYZ after the host has  
negated DMACK-.  
C141-E104-02EN  
5-103  
Interface  
13) The host shall neither negate STOP nor HSTROBE until at least tACK after  
negating DMACK-.  
14) The host shall not assert DIOW-, CS0-, CS1-, DA2, DA1, or DA0 until  
at least tACK after negating DMACK.  
5.5.5 Ultra DMA CRC rules  
The following is a list of rules for calculating CRC, determining if a CRC error  
has occurred during an Ultra DMA burst, and reporting any error that occurs at  
the end of a command.  
a) Both the host and the device shall have a 16-bit CRC calculation function.  
b) Both the host and the device shall calculate a CRC value for each Ultra DMA  
burst.  
c) The CRC function in the host and the device shall be initialized with a seed  
of 4ABAh at the beginning of an Ultra DMA burst before any data is  
transferred.  
d) For each STROBE transition used for data transfer, both the host and the  
device shall calculate a new CRC value by applying the CRC polynomial to  
the current value of their individual CRC functions and the word being  
transferred. CRC is not calculated for the return of STROBE to the asserted  
state after the Ultra DMA burst termination request has been acknowledged.  
e) At the end of any Ultra DMA burst the host shall send the results of its CRC  
calculation function to the device on DD (15:0) with the negation of  
DMACK-.  
f) The device shall then compare the CRC data from the host with the calculated  
value in its own CRC calculation function. If the two values do not match,  
the device shall save the error and report it at the end of the command. A  
subsequent Ultra DMA burst for the same command that does not have a  
CRC error shall not clear an error saved from a previous Ultra DMa burst in  
the same command. If a miscompare error occurs during one or more Ultra  
DMA bursts for any one command, at the end of the command, the device  
shall report the first error that occurred.  
g) For READ DMA or WRITE DMA commands: When a CRC error is  
detected, it shall be reported by setting both ICRC and ABRT (bit 7 and bit 2  
in the Error register) to one. ICRC is defined as the "Interface CRC Error"  
bit. The host shall respond to this error by re-issuing the command.  
h) A host may send extra data words on the last Ultra DMA burst of a data out  
command. If a device determines that all data has been transferred for a  
command, the device shall terminate the burst. A device may have already  
received more data words than were required for the command. These extra  
words are used by both the host and the device to calculate the CRC, but, on  
an Ultra DMA data out burst, the extra words shall be discarded by the  
device.  
i) The CRC generator polynomial is : G (X) = X16 + X12 + X5 + 1.  
5-104  
C141-E104-02EN  
5.5 Ultra DMA Feature Set  
Note: Since no bit clock is available, the recommended approach for  
calculating CRC is to use a word clock derived from the bus strobe. The  
combinational logic shall then be equivalent to shifting sixteen bits serially  
through the generator polynomial where DD0 is shifted in first and DD15 is  
shifted in last.  
CRCOUT  
16  
DD0–DD15  
16  
CRCIN  
F1–f16  
Latch  
16  
Combination  
logic  
Word  
clock  
16  
Figure 5.8 An example of generation of parallel CRC  
Table 5.16 Parallel generation equation of CRC polynomial  
CRCINO=f16  
CRCIN8 = f8 XOR f13  
CRCIN1=f15  
CRCIN9 = f7 XOR f12  
CRCIN2=f14  
CRCIN10 = f6 XOR f11  
CRCIN3=f13  
CRCIN11 = f5 XOR f10  
CRCIN4=f12  
CRCIN12 = f4 XOR f9 XOR f16  
CRCIN13 = f3 XOR f8 XOR f15  
CRCIN14 = f2 XOR f7 XOR f14  
CRCIN15 = f1 XOR f6 XOR f13  
CRCIN5=f11 XOR f  
CRCIN6=f10 XOR f15  
CRCIN7=f9 XOR f14  
f1 = DD0 XOR CRCOUT15  
f2 = DD1 XOR CRCOUT14  
f3 = DD2 XOR CRCOUT13  
f4 = DD3 XOR CRCOUT12  
f9 = DD8 XOR CRCOUT7 XOR f5  
f10 = DD9 XOR CRCOUT6 XOR f6  
f11 = DD10 XOR CRCOUT5 XOR f7  
f12 = DD11 XOR CRCOUT4 XOR f1 XOR f8  
f5 = DD4 XOR CRCOUT11 XOR f1 f13 = DD12 XOR CRCOUT3 XOR f2 XOR f9  
f6 = DD5 XOR CRCOUT10 XOR f2 f14 = DD13 XOR CRCOUT2 XOR f3 XOR f10  
f7 = DD6 XOR CRCOUT9 XOR f3 f15 = DD14 XOR CRCOUT1 XOR f4 XOR f11  
f8 = DD7 XOR CRCOUT8 XOR f4 f16 = DD15 XOR CRCOUT0 XOR f5 XOR f12  
DD :  
Data from bust  
f : Feedback  
CRCIN : Output of combination logic (the next CRC)  
CROUT : Result of 16 bit latch (current CRC)  
C141-E104-02EN  
5-105  
Interface  
5.5.6 Series termination required for Ultra DMA  
Series termination resistors are required at both the host and the device for  
operation in any of the Ultra DMA Modes. The following table describes  
recommended values for series termination at the host and the device.  
Table 5.17 Recommended series termination for Ultra DMA  
Signal  
Host Termination  
Device Termination  
DIOR-:HDMARDY-:HSTROBE  
DIOW-:STOP  
22 Ω  
22 Ω  
33 Ω  
33 Ω  
22 Ω  
33 Ω  
82 Ω  
82 Ω  
82 Ω  
82 Ω  
82 Ω  
CS0-, CS1-  
82 Ω  
DA0, DA1, DA2  
DMACK-  
82 Ω  
82 Ω  
DD15 through DD0  
DMARQ  
120 (100 MHz)  
22 Ω  
INTRQ  
22 Ω  
IORDY:DDMARDY-:DSTROBE  
22 Ω  
Note: Only those signals requiring termination are listed in this table. If a signal is  
not listed, series termination is not required for operation in an Ultra DMA Mode.  
For signals also requiring a pull-up or pull-down resistor at the host see Figure 5.9.  
Vcc  
DMARQ  
IORDY  
Figure 5.9 Ultra DMA termination with pull-up or pull-down  
Configuration of cable  
For the configuration of the cable (common use of primary port and secondary  
port), DMACK signal should not be used in common. It is not recommended to  
use DIOR-, DIOW- and IORDY signal in common.  
5-106  
C141-E104-02EN  
5.6 Timing  
5.6 Timing  
5.6.1 PIO data transfer  
Figure 5.10 shows of the data transfer timing between the device and the host  
system.  
C141-E104-02EN  
5-107  
Interface  
Figure 5.10 Data transfer timing  
5-108  
C141-E104-02EN  
5.6 Timing  
5.6.2 Multiword DMA data transfer  
Figure 5.11 shows the multiword DMA data transfer timing between the device  
and the host system.  
Delay time from DIOR-/DIOW- assertion to DMARQ negation  
Figure 5.11 Multiword DMA data transfer timing (mode 2)  
C141-E104-02EN  
5-109  
Interface  
5.6.3 Transfer of Ultra DMA data  
Figures 5.12 to 5.21 define the timings concerning every phase for the Ultra DMA  
Burst.  
Table 5.18 includes the timing for each Ultra DMA mode.  
5.6.3.1 Starting of Ultra DMA data In Burst  
The timing for each Ultra DMA mode is included in 5.6.3.2.  
DMARQ  
(device)  
tUI  
DMACK-  
(host)  
tFS  
tACK  
tENV  
tZAD  
STOP  
(host)  
tACK  
tENV  
tFS  
HDMARDY-  
(host)  
tZAD  
tZIORDY  
DSTROBE  
(device)  
tAZ  
tVDS  
tDVH  
DD (15:0)  
tACK  
DA0,DA1,DA2,  
CS0-,CS1-  
Note :  
The definitions of STOP, HDMARDY- and DSTROBE signals are  
valid before the assertion of DMACK signal.  
Figure 5.12 Starting of Ultra DMA data In Burst transfer  
5-110  
C141-E104-02EN  
5.6 Timing  
5.6.3.2 Ultra DMA data burst timing requirements  
Table 5.18 Ultra DMA data burst timing requirements (1 of 2)  
NAME  
MODE 0  
(in ns)  
MODE 1  
(in ns)  
MODE 2  
(in ns)  
MODE 3  
(in ns)  
MODE 4  
(in ns)  
COMMENT  
MIN MAX MIN MAX MIN MAX MIN MAX MIN MAX (see Notes 1 and 2)  
t2CYCTYP  
tCYC  
240  
160  
120  
90  
60  
Typical sustained average two cycle  
time  
112  
73  
54  
39  
25  
Cycle time allowing for asymmetry  
and clock variations (from  
STROBE edge to STROBE edge)  
t2CYC  
230  
154  
115  
86  
57  
Two cycle time allowing for clock  
variations (from rising edge to next  
rising edge or from falling edge to  
next falling edge of STROBE)  
tDS  
15  
5
10  
5
7
5
7
5
5
5
6
Data setup time (at recipient)  
(see Note 4)  
tDH  
tDVS  
Data hold time (at recipient)  
(see Note 4)  
70  
48  
30  
20  
Data valid setup time at sender (from  
data valid until STROBE edge)  
(see Note 5)  
tDVH  
6
0
6
0
6
0
6
0
6
0
Data valid hold time at sender (from  
STROBE edge until data may  
become invalid) (see Note 5)  
tFS  
230  
150  
200  
150  
170  
150  
130  
100  
120 First STROBE time (for device to  
first negate DSTROBE from STOP  
during a data in burst)  
tLI  
0
0
0
0
0
100 Limited interlock time (see Note 3)  
tMLI  
20  
20  
20  
20  
20  
Interlock time with minimum  
(see Note 3)  
tUI  
0
0
0
0
0
Unlimited interlock time (see Note 3)  
tAZ  
10  
10  
10  
10  
10  
Maximum time allowed for output  
drivers to release (from asserted or  
negated)  
tZAH  
tZAD  
tENV  
20  
0
20  
0
20  
0
20  
0
20  
0
Minimum delay time required  
for output  
Drivers to assert or negate (from  
released)  
20  
70  
50  
75  
20  
70  
30  
70  
20  
70  
20  
60  
20  
55  
NA  
60  
20  
55  
Envelope time (from DMACK- to  
STOP and HDMARDY- during  
data in burst initiation and from  
DMACK to STOP during data out  
burst initiation)  
tSR  
NA STROBE-to-DMARDY-time (if  
DMARDY- is negated before this long  
after STROBE edge, the recipient shall  
receive no more than one additional  
data word)  
tRFS  
60  
Ready-to-Final-STROBE time  
(After negating DMARDY-,  
STROBE edges should not be sent  
beyond this period)  
tRP  
160  
125  
100  
100  
100  
Ready-to-pause time (that recipient  
shall wait to pause after negating  
DMARDY-)  
tIORDYZ  
20  
20  
20  
20  
20  
MaximumtimebeforereleasingIORDY  
C141-E104-02EN  
5-111  
Interface  
Table 5.18 Ultra DMA data burst timing requirements (2 of 2)  
NAME  
MODE 0  
(in ns)  
MODE 1  
(in ns)  
MODE 2  
(in ns)  
MODE 3  
(in ns)  
MODE 4  
(in ns)  
COMMENT  
MIN MAX MIN MAX MIN MAX MIN MAX MIN MAX (see Notes 1 and 2)  
tZIORDY  
tACK  
tSS  
0
0
0
0
0
Minimum time before driving  
IORDY  
20  
50  
20  
50  
20  
50  
20  
50  
20  
50  
Setup and hold times for DMACK-  
(before assertion or negation)  
Time from STROBE edge to  
negation of DMARQ or assertion of  
STOP (when sender terminates a  
burst)  
Notes:  
1) Unless otherwise specified, timing parameters shall be measured at the connector of the sender or receiver to which the parameter  
applies (see Note 5 for exceptions). For example, the sender shall stop generating STROBE edges tRFS after the negation of  
DMARDY-. Both STROBE and DMARDY- timing measurements are taken at the connector of the sender.  
2) All timing measurement switching points (low to high and high to low) shall be taken at 1.5 V.  
3)  
tUI, tMLI and tLI indicate sender-to-recipient or recipient-to-sender interlocks, i.e., one agent (either sender or recipient) is waiting for  
the other agent to respond with a signal before proceeding. tUI is an unlimited interlock that has no maximum time value. tMLI is a  
limited time-out that has a defined minimum. tLI is a limited time-out that has a defined maximum.  
4) Special cabling shall be required in order to meet data setup (tDS) and data hold (tDH) times in modes 3 and 4.  
5) Timing for tDVS and tDVH shall be met for all capacitive loads from 15 to 40 pf where all signals have the same capacitive load value.  
5-112  
C141-E104-02EN  
5.6 Timing  
5.6.3.3 Sustained Ultra DMA data in burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
t2CYC  
tCYC  
tCYC  
t2CYC  
DSTROBE  
at device  
tDVH  
tDVH  
tDVH  
tDVS  
tDVS  
DD(15:0)  
at device  
DSTROBE  
at host  
tDH  
tDS  
tDS  
tDH  
tDH  
DD(15:0)  
at host  
Note:  
DD (15:0) and DSTROBE are shown at both the host and the device  
to emphasize that cable setting time as well as cable propagation  
delay shall not allow the data signals to be considered stable at the  
host until some time after they are driven by the device.  
Figure 5.13 Sustained Ultra DMA data in burst  
C141-E104-02EN  
5-113  
Interface  
5.6.3.4 Host pausing an Ultra DMA data in burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
DMARQ  
(device)  
DMACK-  
(host)  
tRP  
STOP  
(host)  
tSR  
HDMARDY-  
(host)  
tRFS  
DSTROBE  
(device)  
DD(15:0)  
(device)  
Notes:  
1) The host may assert STOP to request termination of the Ultra DMA  
burst no sooner than tRP after HDMARDY- is negated.  
2) If the tSR timing is not satisfied, the host may receive zero, one or  
two more data words from the device.  
Figure 5.14 Host pausing an Ultra DMA data in burst  
5-114  
C141-E104-02EN  
5.6 Timing  
5.6.3.5 Device terminating an Ultra DMA data in burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
DMARQ  
(device)  
tMLI  
DMACK-  
(host)  
tACK  
tLI  
tLI  
STOP  
(host)  
tACK  
tLI  
HDMARDY-  
(host)  
tSS  
tIORDYZ  
DSTROBE  
(device)  
tZAH  
tAZ  
tDVS  
tDVH  
DD(15:0)  
CRC  
tACK  
DA0, DA1, DA2,  
CS0-, CS1-  
Note:  
The definitions for the STOP, HDMARDY- and DSTROBE signal  
lines are no longer in effect after DMARQ and DMACK are  
negated.  
Figure 5.15 Device terminating an Ultra DMA data in burst  
C141-E104-02EN  
5-115  
Interface  
5.6.3.6 Host terminating an Ultra DMA data in burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
DMARQ  
(device)  
tLI  
tMLI  
DMACK-  
(host)  
tZAH  
tACK  
tAZ  
tRP  
STOP  
(host)  
tACK  
HDMARDY-  
(host)  
tRFS  
tMLI  
tLI  
tIORDYZ  
DSTROBE  
(device)  
tDVS tDVH  
CRC  
DD(15:0)  
tACK  
DA0, DA1, DA2,  
CS0, CS1  
Note:  
The definitions for the STOP, HDMARDY- and DSTROBE signal  
lines are no longer in effect after DMARQ and DMACK are  
negated.  
Figure 5.16 Host terminating an Ultra DMA data in burst  
5-116  
C141-E104-02EN  
5.6 Timing  
5.6.3.7 Initiating an Ultra DMA data out burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
DMARQ  
(device)  
tUI  
DMACK-  
(host)  
tENV  
tACK  
STOP  
(host)  
tLI  
tUI  
tZIORDY  
DDMARDY-  
(device)  
tACK  
HSTROBE  
(host)  
tDVS  
tDVH  
DD(15:0)  
(host)  
tACK  
DA0, DA1, DA2  
CS0-, CS1-  
Note:  
The definitions for the STOP, DDMARDY- and HSTROBE signal  
lines are not in effect until DMARQ and DMACK are asserted.  
Figure 5.17 Initiating an Ultra DMA data out burst  
C141-E104-02EN  
5-117  
Interface  
5.6.3.8 Sustained Ultra DMA data out burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
t2CYC  
tCYC  
tCYC  
t2CYC  
HSTROBE  
at host  
tDVH  
tDVS  
tDVH  
tDVS  
tDVH  
DD(15:0)  
at host  
HSTROBE  
at device  
tDS  
tDH  
tDS  
tDH  
tDH  
DD(15:0)  
at device  
Note:  
DD (15:0) and HSTROBE signals are shown at both the device and  
the host to emphasize that cable setting time as well as cable  
propagation delay shall not allow the data signals to be considered  
stable at the device until some time after they are driven by the host.  
Figure 5.18 Sustained Ultra DMA data out burst  
5-118  
C141-E104-02EN  
5.6 Timing  
5.6.3.9 Device pausing an Ultra DMA data out burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
tRP  
DMARQ  
(device)  
DMACK-  
(host)  
STOP  
(host)  
tSR  
DDMARDY-  
(device)  
tRFS  
HSTROBE  
(host)  
DD(15:0)  
(host)  
Notes:  
1) The device may negate DMARQ to request termination of the Ultra  
DMA burst no sooner than tRP after DDMARDY- is negated.  
2) If the tSR timing is not satisfied, the device may receive zero, one or  
two more data words from the host.  
Figure 5.19 Device pausing an Ultra DMA data out burst  
C141-E104-02EN  
5-119  
Interface  
5.6.3.10 Host terminating an Ultra DMA data out burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
tLI  
DMARQ  
(device)  
tMLI  
DMACK-  
(host)  
tLI  
tACK  
tSS  
STOP  
(host)  
tLI  
tIORDYZ  
DDMARDY-  
(device)  
tACK  
HSTROBE  
(host)  
tDVS  
tDVH  
DD(15:0)  
(host)  
CRC  
tACK  
DA0, DA1, DA2  
CS0-, CS1-  
Note:  
The definitions for the STOP, DDMARDY- and HSTROBE signal  
lines are no longer in effect after DMARQ and DMACK are  
negated.  
Figure 5.20 Host terminating an Ultra DMA data out burst  
5-120  
C141-E104-02EN  
5.6 Timing  
5.6.3.11 Device terminating an Ultra DMA data in burst  
5.6.3.2 contains the values for the timings for each of the Ultra DMA Modes.  
DMARQ  
(device)  
DMACK-  
(host)  
tACK  
tLI  
tMLI  
STOP  
(host)  
tRP  
tIORDYZ  
DDMARDY-  
(device)  
tRFS  
tACK  
tLI  
tMLI  
HSTROBE  
(host)  
tDVS  
tDVS  
DD(15:0)  
(host)  
CRC  
tACK  
DA0, DA1, DA2,  
CS0-, CS1-  
Note:  
The definitions for the STOP, DDMARDY- and HSTROBE signal  
lines are no longer in effect after DMARQ and DMACK are  
negated.  
Figure 5.21 Device terminating an Ultra DMA data out burst  
C141-E104-02EN  
5-121  
Interface  
5.6.4 Power-on and reset  
Figure 5.22 shows power-on and reset (hardware and software reset) timing.  
(1) Only master device is present  
Power-on Reset  
RESET–  
(2) Master and slave devices are present (2-drives configuration)  
PDIAG- negation  
31  
Figure 5.22 Power on Reset Timing  
5-122  
C141-E104-02EN  
CHAPTER 6 Operations  
6.1 Device Response to the Reset  
6.2 Address Translation  
6.3 Power Save  
6.4 Defect Management  
6.5 Read-Ahead Cache  
6.6 Write Cache  
C141-E104-02EN  
6-1  
Operations  
6.1 Device Response to the Reset  
This section describes how the PDIAG- and DASP- signals responds when the  
power of the IDD is turned on or the IDD receives a reset or diagnostic command.  
6.1.1 Response to power-on  
After the master device (device 0) releases its own power-on reset state, the  
master device shall check a DASP- signal for up to 450 ms to confirm presence of  
a slave device (device 1). The master device recognizes presence of the slave  
device when it confirms assertion of the DASP- signal. Then, the master device  
checks a PDIAG- signal to see if the slave device has successfully completed the  
power-on diagnostics.  
If the master device cannot confirm assertion of the DASP- signal within 450 ms,  
the master device recognizes that no slave device is connected.  
After the slave device (device 1) releases its own power-on reset state, the slave  
device shall report its presence and the result of power-on diagnostics to the  
master device as described below:  
DASP- signal: Asserted within 400 ms.  
PDIAG- signal: Negated within 1 ms and asserted within 30 seconds.  
6-2  
C141-E104-02EN  
6.1 Device Response to the Reset  
Power on  
Master device  
Power On Reset-  
Status Reg.  
BSY bit  
Max. 31 sec.  
Checks DASP- for up to  
450 ms.  
If presence of a slave device is  
confirmed, PDIAG- is checked for  
up to 31 seconds.  
Slave device  
Power On Reset-  
BSY bit  
Max. 1 ms.  
PDIAG-  
DASP-  
Max. 30 sec.  
Max. 400 ms.  
Figure 6.1 Response to power-on  
C141-E104-02EN  
6-3  
Operations  
6.1.2 Response to hardware reset  
Response to RESET- (hardware reset through the interface) is similar to the  
power-on reset.  
Upon receipt of hardware reset, the master device checks a DASP- signal for up to  
450 ms to confirm presence of a slave device. The master device recognizes the  
presence of the slave device when it confirms assertion of the DASP- signal.  
Then the master device checks a PDIAG- signal to see if the slave device has  
successfully completed the self-diagnostics.  
If the master device cannot confirm assertion of the DASP- signal within 450 ms,  
the master device recognizes that no slave device is connected.  
After the slave device receives the hardware reset, the slave device shall report its  
presence and the result of the self-diagnostics to the master device as described  
below:  
DASP- signal: Asserted within 400 ms.  
PDIAG- signal: Negated within 1 ms and asserted within 30 seconds.  
Reset-  
Master device  
Status Reg.  
BSY bit  
Max. 31 sec.  
If presence of a slave device is  
Checks DASP- for up to  
450 ms.  
confirmed, PDIAG- is checked for  
up to 31 seconds.  
Slave device  
BSY bit  
Max. 1 ms.  
PDIAG-  
DASP-  
Max. 30 sec.  
Max. 400 ms.  
.
Figure 6.2 Response to hardware reset  
6-4  
C141-E104-02EN  
6.1 Device Response to the Reset  
6.1.3 Response to software reset  
The master device does not check the DASP- signal for a software reset. If a  
slave device is present, the master device checks the PDIAG- signal for up to 15  
seconds to see if the slave device has completed the self-diagnosis successfully.  
After the slave device receives the software reset, the slave device shall report its  
presence and the result of the self-diagnostics to the master device as described  
below:  
PDIAG- signal: negated within 1 ms and asserted within 30 seconds  
When the IDD is set to a slave device, the IDD asserts the DASP- signal when  
negating the PDIAG- signal, and negates the DASP- signal when asserting the  
PDIAG- signal.  
X'3F6' Reg.  
X"0C"  
X"00"  
or X"04"  
Master device  
Status Reg.  
BSY bit  
Max. 31 sec.  
If the slave device is preset, PDIAG- is checked for  
up to 31 seconds.  
Slave device  
BSY bit  
Max. 1 ms.  
PDIAG-  
DASP-  
Max. 30 sec.  
Figure 6.3 Response to software reset  
C141-E104-02EN  
6-5  
Operations  
6.1.4 Response to diagnostic command  
When the master device receives an EXECUTE DEVICE DIAGNOSTIC  
command and the slave device is present, the master device checks the PDIAG-  
signal for up to 6 seconds to see if the slave device has completed the self-  
diagnosis successfully.  
The master device does not check the DASP- signal.  
After the slave device receives the EXECUTE DEVICE DIAGNOSTIC  
command, it shall report the result of the self-diagnostics to the master device as  
described below:  
PDIAG- signal: negated within 1 ms and asserted within 5 seconds  
When the IDD is set to a slave device, the IDD asserts the DASP- signal when  
negating the PDIAG- signal, and negates the DASP- signal when asserting the  
PDIAG- signal.  
X'1F7' Reg.  
Write  
Master device  
Status Reg.  
BSY bit  
Max. 6 sec.  
If the slave device is preset, PDIAG- signal is checked for  
up to6 seconds.  
Slave device  
BSY bit  
Max. 1 ms.  
PDIAG-  
DASP-  
Max. 5 sec.  
Figure 6.4 Response to diagnostic command  
6-6  
C141-E104-02EN  
6.2 Address Translation  
6.2 Address Translation  
When the IDD receives any command which involves access to the disk medium,  
the IDD always implements the address translation from the logical address (a  
host-specified address) to the physical address (logical to physical address  
translation).  
Following subsections explains the CHS translation mode.  
6.2.1 Default parameters  
In the logical to physical address translation, the logical cylinder, head, and sector  
addresses are translated to the physical cylinder, head, and sector addresses based  
on the number of heads and the number of sectors per track which are specified  
with an INITIALIZE DEVICE PARAMETERS command. This is called as the  
current translation mode.  
If the number of heads and the number of sectors are not specified with an  
INITIALIZE DEVICE PARAMETERS command, the default values listed in  
Table 6.1 are used. This is called the default translation mode. The parameters in  
Table 6.1 are called BIOS specification.  
Table 6.1 Default parameters  
MHL2300AT  
16,383  
16  
MHM2200AT MHM2150AT MHM2100AT  
Number of cylinders  
Number of heads  
16,383  
16  
16,383  
16  
16,383  
16  
Parameters  
(logical)  
Number of sectors/track  
63  
63  
63  
63  
Formatted capacity (MB)  
30,005.82  
20,003.88  
15,103.03  
10,056.13  
As long as the formatted capacity of the IDD does not exceed the value shown on  
Table 6.1, the host can freely specify the number of cylinders, heads, and sectors  
per track.  
Generally, the device recognizes the number of heads and sectors per track with  
the INITIALIZE DEVICE PARAMETER command. However, it cannot  
recognize the number of cylinders. In other words, there is no way for the device  
to recognize a host access area on logical cylinders. Thus the host should manage  
cylinder access to the device.  
The host can specify a logical address freely within an area where an address can  
be specified (within the specified number of cylinders, heads, and sectors per  
track) in the current translation mode.  
The host can read an addressable parameter information from the device by the  
IDENTIFY DEVICE command (Words 54 to 56).  
C141-E104-02EN  
6-7  
Operations  
6.2.2 Logical address  
(1) CHS mode  
Logical address assignment starts from physical cylinder (PC) 0, physical head  
(PH) 0, and physical sector (PS) 1 and is assigned by calculating the number of  
sectors per track that is specified by the INITIALIZE DEVICE PARAMETERS  
command. If the last sector of a physical track is used, the track is switched and  
the next logical sector is placed in the initial sector of the subsequent physical  
track.  
Figure 6.5 shows an example of 6 heads configuration. (assuming there is no track  
skew).  
Physical sector  
1
2
3
62 63 64  
126 127  
504 505 506  
Physical cylinder 0  
Physical head 0  
LS  
63  
LS  
LS1  
63  
LS  
LS1  
LS1 LS2  
63  
LH0  
LH8  
LH1  
LH9  
LH8  
Physical sector  
1
2
61 62  
LS  
Physical cylinder 1  
Physical head 0  
LS3LS4  
LS1LS2  
63  
ex: Zone 0 in 6-head device  
Physical parameter  
– Physical head: 0 to 5  
– Physical sector: 1 to 506  
Specification of INITIALIZE DEVICE PARAMETERS command  
– Logical head: 0 to 15  
– Logical sector: 1 to 63  
Figure 6.5 Address translation (example in CHS mode)  
6-8  
C141-E104-02EN  
6.3 Power Save  
(2) LBA mode  
Logical address assignment in the LBA mode starts from physical cylinder 0,  
physical head 0, and physical sector 1. If the last sector of a physical track is  
used, the track is switched and the next LBA is assigned to the initial sector of the  
subsequent physical track.  
Figure 6.6 shows an example of 4 heads configuration (assuming there is no track  
skew).  
626  
627  
625  
626  
626  
627  
627  
628  
629  
1252 1253  
6
5
627  
Figure 6.6 Address translation (example in LBA mode)  
6.3 Power Save  
The host can change the power consumption state of the device by issuing a  
power command to the device.  
6.3.1 Power save mode  
There are four types of power consumption state of the device including active  
mode where all circuits are active.  
In the power save mode, power supplying to the part of the circuit is turned off.  
There are three types of power save modes:  
Idle mode  
Standby mode  
Sleep mode  
C141-E104-02EN  
6-9  
Operations  
The drive moves from the Active mode to the idle mode by itself.  
Regardless of whether the power down is enabled, the device enters the idle  
mode. The device also enters the idle mode in the same way after power-on  
sequence is completed.  
And, the automatic power-down is executed if no command is coming for 30 min.  
(default)  
(1) Active mode  
In this mode, all the electric circuit in the device are active or the device is under  
seek, read or write operation.  
A device enters the active mode under the following conditions:  
A command other than power commands is issued.  
A reset command is received.  
(2) Idle mode  
In this mode, circuits on the device is set to power save mode.  
The device enters the Idle mode under the following conditions:  
After completion of power-on sequence.  
After completion of the command execution other than SLEEP and STANDBY  
commands.  
After completion of the reset sequence  
(3) Standby mode  
In this mode, the VCM circuit is turned off and the spindle motor is stopped.  
The device can receive commands through the interface. However if a command  
with disk access is issued, response time to the command under the standby mode  
takes longer than the active or Idle mode because the access to the disk medium  
cannot be made immediately.  
The drive enters the standby mode under the following conditions:  
A STANDBY or STANDBY IMMEDIATE command is issued in the active  
or idle mode.  
When automatic power down sequence is enabled, the timer has elapsed.  
A reset is issued in the sleep mode.  
6-10  
C141-E104-02EN  
6.4 Defect Management  
When one of following commands is issued, the command is executed normally  
and the device is still stayed in the standby mode.  
Reset (hardware or software)  
STANDBY command  
STANDBY IMMEDIATE command  
INITIALIZE DEVICE PARAMETERS command  
CHECK POWER MODE command  
(4) Sleep mode  
The power consumption of the drive is minimal in this mode. The drive enters  
only the standby mode from the sleep mode. The only method to return from the  
standby mode is to execute a software or hardware reset.  
The drive enters the sleep mode under the following condition:  
A SLEEP command is issued.  
Issued commands are invalid (ignored) in this mode.  
6.3.2 Power commands  
The following commands are available as power commands.  
IDLE  
IDLE IMMEDIATE  
STANDBY  
STANDBY IMMEDIATE  
SLEEP  
CHECK POWER MODE  
6.4 Defect Management  
Defective sectors of which the medium defect location is registered in the system  
space are replaced with spare sectors in the formatting at the factory shipment.  
All the user space area are formatted at shipment from the factory based on the  
default parameters listed in Table 6.1.  
C141-E104-02EN  
6-11  
Operations  
6.4.1 Spare area  
Following two types of spare area are provided for every physical head.  
1) Spare cylinder for sector slip:  
used for alternating defective sectors at formatting in shipment (4 cylinders)  
2) Spare cylinder for alternative assignment:  
used for automatic alternative assignment at read error occurrence.  
(4 cylinders)  
6.4.2 Alternating defective sectors  
The two alternating methods described below are available:  
(1) Sector slip processing  
A defective sector is not used and is skipped and a logical sector address is  
assigned to the subsequent normal sector (physically adjacent sector to the  
defective sector).  
When defective sector is present, the sector slip processing is performed in the  
formatting.  
Figure 6.7 shows an example where (physical) sector 5 is defective on head 0 in  
cylinder 0.  
504  
505  
506  
1
1
2
3
3
4
4
5
6
5
7
6
8
7
2
(unused)  
503504505
Note:  
If an access request to physical sector 5 is specified, the device  
accesses physical sector 6 instead of sector 5.  
Figure 6.7 Sector slip processing  
6-12  
C141-E104-02EN  
6.4 Defect Management  
(2) Alternate cylinder assignment  
A defective sector is assigned to the spare sector in the alternate cylinder.  
This processing is performed when the alternate assignment is specified in the  
FORMAT TRACK command or when the automatic alternate processing is  
performed at read error occurrence.  
Figure 6.8 shows an example where (physical) sector 5 is detective on head 0 in  
cylinder 0.  
Index  
Sector (Physical)  
1
2
3
4
4
5
6
6
7
7
505  
505  
506  
506  
Cylinder 0  
Head 0  
Defective  
sector  
1
2
3
(unused)  
Sector (Logical)  
Alternate  
cylinder  
Already  
assigned  
Head 0  
Defective sector is assigned to unassigned sector.  
Notes:  
1) 4 alternate cylinders are provided for each head in zone 14 (inner  
side).  
2) When an access request to physical sector 5 is specified, the device  
accesses the alternated sector in the alternate cylinder instead of  
sector 5. When an access request to sectors next to sector 5 is  
specified, the device seeks to cylinder 0, head 0, and continues the  
processing.  
Figure 6.8 Alternate cylinder assignment  
(3) Automatic alternate assignment  
The device performs the automatic alternate assignment when ECC correction  
performance is increased during read error retry, a read error is recovered. Before  
automatic alternate assignment, the device performs rewriting the corrected data  
to the erred sector and rereading. If no error occurs at rereading, the automatic  
alternate assignment is not performed.  
C141-E104-02EN  
6-13  
Operations  
An unrecoverable write error occurs during write error retry, automatic alternate  
assignment is performed.  
6.5 Read-Ahead Cache  
After read command which involves read data from the disk medium is  
completed, the read-ahead cache function reads the subsequent data blocks  
automatically and stores the data to the data buffer.  
When the next command requests to read the read-ahead data, the data can be  
transferred from the data buffer without accessing the disk medium. The host can  
thus access data at higher speed.  
6.5.1 Data buffer configuration  
The drive has a 2 MB data buffer. The buffer is used by divided into three parts;  
for read commands, for write commands, and for MPU work (see Figure 6.9).  
2 MB (2,097,152 bytes)  
for read commands  
for write commands  
for MPU work  
65,536 bytes  
1,048,576 bytes (2048 sector)  
983,040 bytes (1920 sector)  
Figure 6.9 Data buffer configuration  
The read-ahead operation is performed at execution of the READ SECTOR(S),  
READ MULTIPLE, or READ DMA command, and read-ahead data is stored in  
the buffer for read commands.  
6.5.2 Caching operation  
Caching operation is performed only at issuance of the following commands. The  
device transfers data from the data buffer to the host system at issuance of  
following command if following data exist in the data buffer.  
All sectors to be processed by the command  
A part of data including load sector to be processed by the command  
When a part of data to be processed exist in the data buffer, remaining data are  
read from the medium and are transferred to the host system.  
(1) Commands that are object of caching operation  
Follow commands are object of caching operation.  
6-14  
C141-E104-02EN  
6.5 Read-Ahead Cache  
READ SECTOR (S)  
READ MULTIPLE  
READ DMA  
When caching operation is disabled by the SET FEATURES command, no  
caching operation is performed.  
(2) Data that are object of caching operation  
Follow data are object of caching operation.  
1) Read-ahead data read from the medium to the data buffer after completion of  
the command that are object of caching operation.  
2) Data transferred to the host system once by requesting with the command that  
are object of caching operation (except for the cache invalid data by some  
reasons).  
3) Remaining data in the data buffer (for write command) transferred from the  
host system by the command that writes data onto the disk medium, such as  
the WRITE SECTOR (S), WRITE DMA, WRITE MULTIPLE.  
Followings are definition of in case that the write data is treated as a cache data.  
However, since the hit check at issuance of read command is performed to the  
data buffer for read command preferentially, caching write data is limited to the  
case that the hit check is missed at the data buffer for read command.  
When all data requested by the read command are stored in the data buffer for  
write command (hit all), the device transfers data from the data buffer for  
write command. At this time, the read-ahead operation to the data subsequent  
to the requested data is not performed.  
Even if a part of data requested by the read command are stored in the data  
buffer for write command (hit partially), all data are read from the disk  
medium without transferring from the data buffer for write command.  
(3) Invalidating caching data  
Caching data in the data buffer is invalidated in the following case.  
1) Following command is issued to the same data block as caching data.  
WRITE SECTOR(S)  
WRITE DMA  
WRITE MULTIPLE  
2) Command other than following commands is issued (all caching data are  
invalidated)  
READ SECTOR (S)  
READ DMA  
C141-E104-02EN  
6-15  
Operations  
READ MULTIPLE  
WRITE SECTOR(S)  
WRITE MULTIPLE  
WRITE VERIFY SECTOR(S)  
3) Caching operation is inhibited by the SET FEATURES command.  
4) Issued command is terminated with an error.  
5) Soft reset or hard reset occurs, or power is turned off.  
6) The device enters the sleep mode.  
7) Under the state that the write data is kept in the data buffer for write  
command as a caching data, new write command is issued. (write data kept  
until now are invalidated)  
6.5.3 Usage of read segment  
This subsection explains the usage of the read segment buffer at following cases.  
6.5.3.1 Mis-hit (no hit)  
A lead block of the read-requested data is not stored in the data buffer. The  
requested data is read from the disk media.  
The read-ahead operation is performed only when the last sector address of the  
previous read command and the lead sector address of this read command is  
sequential (see item (2)).  
1) Sets the host address pointer (HAP) and the disk address pointer (DAP) to the  
lead of segment.  
HAP  
DAP  
Segment only for read  
6-16  
C141-E104-02EN  
6.5 Read-Ahead Cache  
2) Transfers the requested data that already read to the host system with reading  
the requested data from the disk media.  
Stores the read-requested  
data upto this point  
HAP  
Empty area  
Read-requested data  
DAP  
3) After reading the requested data and transferring the requested data to the  
host system had been completed, the disk drive stops command execution  
without performing the read-ahead operation.  
HAP  
(stopped)  
Empty area  
Read-requested data  
(stopped)  
DAP  
4) Following shows the cache enabled data for next read command.  
Empty area  
Cache enabled data  
Start LBA  
6.5.3.2 Sequential read  
Last LBA  
When the disk drive receives the read command that targets the sequential address  
to the previous read command, the disk drive starts the read-ahead operation.  
a. Sequential command just after non-sequential command  
When the previously executed read command is an non-sequential  
command and the last sector address of the previous read command is  
sequential to the lead sector address of the received read command, the  
disk drive assumes the received command is a sequential command and  
performs the read-ahead operation after reading the requested data.  
C141-E104-02EN  
6-17  
Operations  
1) At receiving the sequential read command, the disk drive sets the DAP and  
HAP to the start address of the segment and reads the requested data from the  
load of the segment.  
HAP  
DAP  
Mis-hit data  
Empty area  
2) The disk drive transfers the requested data that is already read to the host  
system with reading the requested data.  
HAP  
Requested data  
Mis-hit data  
Empty area  
DAP  
3) After completion of the reading and transferring the requested data to the host  
system, the disk drive performs the read-ahead operation continuously.  
HAP (Completion of transferring requested data)  
Requested data  
Empty area  
Read-ahead data  
DAP  
4) The disk drive performs the read-ahead operation for all area of segment with  
overwriting the requested data. Finally, the cache data in the buffer is as  
follows.  
HAP  
Read-ahead data  
DAP  
Last LBA Start LBA  
6-18  
C141-E104-02EN  
6.5 Read-Ahead Cache  
b. Sequential hit  
When the previously executed read command is the sequential command  
and the last sector address of the previous read command is sequential to  
the lead sector address of the received read command, the disk drive  
transfers the hit data in the buffer to the host system.  
The disk drive performs the read-ahead operation of the new continuous  
data to the empty area that becomes vacant by data transfer at the same  
time as the disk drive starts transferring data to the host system.  
1) In the case that the contents of buffer is as follows at receiving a read  
command;  
HAP (Continued from the previous read request data)  
Read-ahead data  
Hit data  
DAP  
Last LBA Start LBA  
2) The disk drive starts the read-ahead operation to the empty area that becomes  
vacant by data transfer at the same time as the disk drive starts transferring hit  
data.  
HAP  
Read-ahead data  
New read-ahead data  
Hit data  
DAP  
3) After completion of data transfer of hit data, the disk drive performs the read-  
ahead operation for the data area of which the disk drive transferred hit data.  
HAP  
DAP  
Read-ahead data  
C141-E104-02EN  
6-19  
Operations  
4) Finally, the cache data in the buffer is as follows.  
Read-ahead data  
Start LBA  
Last LBA  
c. Non-sequential command immediately after sequential command  
When a sequential read command (first read) has been executed, the first  
read operation should be stopped if a non-sequential read command has  
been received and then, ten or more of the non-sequential read  
commands have been received. (Refer to 6.5.3.1.)  
The figure that describes the first read operation is the same as that  
shown in the sub-section a.  
6.5.3.3 Full hit (hit all)  
All requested data are stored in the data buffer. The disk drive starts transferring  
the requested data from the address of which the requested data is stored. After  
completion of command, a previously existed cache data before the full hit  
reading are still kept in the buffer, and the disk drive does not perform the read-  
ahead operation.  
1) In the case that the contents of the data buffer is as follows for example and  
the previous command is a sequential read command, the disk drive sets the  
HAP to the address of which the hit data is stored.  
Last position at previous read command  
HAP (set to hit position for data transfer)  
HAP  
Cache data  
Full hit data  
Cache data  
DAP  
Last position at previous read command  
0
2) The disk drive transfers the requested data but does not perform the read-  
ahead operation.  
HAP  
(stopped)  
Cache data  
Full hit data  
Cache data  
6-20  
C141-E104-02EN  
6.5 Read-Ahead Cache  
3) The cache data for next read command is as follows.  
Cache data  
Start LBA  
Last LBA  
6.5.3.4 Partially hit  
A part of requested data including a lead sector are stored in the data buffer. The  
disk drive starts the data transfer from the address of the hit data corresponding to  
the lead sector of the requested data, and reads remaining requested data from the  
disk media directly. The disk drive does not perform the read-ahead operation  
after data transfer.  
Following is an example of partially hit to the cache data.  
Cache data  
Last LBA  
Start LBA  
1) The disk drive sets the HAP to the address where the partially hit data is  
stored, and sets the DAP to the address just after the partially hit data.  
HAP  
Partially hit data  
Lack data  
DAP  
2) The disk drive starts transferring partially hit data and reads lack data from  
the disk media at the same time. However, the disk drive does not perform  
the read-ahead operation newly.  
HAP  
(stopped)  
Requested data to be transferred  
Partially hit data  
Lack data  
DAP  
(stopped)  
C141-E104-02EN  
6-21  
Operations  
3) The cache data for next read command is as follows.  
Cache data  
Start LBA  
Last LBA  
6.6 Write Cache  
The write cache function of the drive makes a high speed processing in the case  
that data to be written by a write command is physically sequent the data of  
previous command and random write operation is performed.  
When the drive receives a write command, the drive starts transferring data of  
sectors requested by the host system and writing on the disk medium. After  
transferring data of sectors requested by the host system, the drive generates the  
interrupt of command complete. Also, the drive sets the normal end status in the  
Status register.  
The drive continues writing data on the disk medium. When all data requested by  
the host are written on the disk medium, actual write operation is completed.  
The drive receives the next command continuously. If the received command is a  
“sequential write” (data to be written by a command is physically sequent to data  
of previous command), the drive starts data transfer and receives data of sectors  
requested by the host system. At this time, if the write operation of the previous  
command is still been executed, the drive continuously executes the write  
operation of the next command from the sector next to the last sector of the  
previous write operation. Thus, the latency time for detecting a target sector of  
the next command is eliminated. This shortens the access time.  
The drive generates an interrupt of command complete after completion of data  
transfer requested by the host system as same as at previous command.  
When the write operation of the previous command had been completed, the  
latency time occurs to search the target sector.  
If the received command is not a “sequential write”, the drive receives data of sectors  
requested by the host system as same as “sequential write”. The drive generates the  
interrupt of command complete after completion of data transfer requested by the host  
system. Received data is processed after completion of the write operation to the disk  
medium of the previous command.  
Even if a hard reset or soft reset is received or the write cache function is disabled  
by the SET FEATURES command during unwritten data is kept, the instruction is  
not enabled until remaining unwritten data is written onto the disk medium.  
6-22  
C141-E104-02EN  
6.6 Write Cache  
The drive uses a cache data of the last write command as a read cache data. When  
a read command is issued to the same address after the write command (cache  
hit), the read operation to the disk medium is not performed.  
If an error occurs during the write operation, the device retries the processing. If the  
error cannot be recovered by retry, automatic alternate assignment is performed. For  
details about automate alternate assignment, see item (3) of Section 6.4.2.  
The write cache function is operated with the following command.  
WRITE SECTOR(S)  
WRITE MULTIPLE  
WRITE DMA  
IMPORTANT  
When Write Cache is permitted, the writing of the data transferred  
from the host by the above mentioned Write Cache permit  
command into the disk medium may not be completed at the  
moment a normal ending interrupt has occurred.  
In case a non-recoverable error has occurred during receiving more  
than one write command, it is difficult for the host to identify a  
command that caused the error.  
(However, the error is not reported to the hose if an error at writing  
has been processed normally.)  
Therefore, note that it is difficult for the host to retry an operation  
that caused a non-recoverable error.  
C141-E104-02EN  
6-23  
This page is intentionally left blank.  
Glossary  
Actuator  
AT bus  
Head positioning assembly. The actuator consists of a voice coil motor and head  
arm. If positions the read-write (R-W) head.  
A bus between the host CPU and adapter board  
ATA (AT Attachment) standard  
The ATA standard is for a PC AT interface regulated to establish compatibility  
between products manufactured by different vendors. Interfaces based on this  
standard are called ATA interfaces.  
BIOS standard for drives  
The BIOS standard collectively refers to the parameters defined by the host,  
which, for example, include the number of cylinders, the number of heads, and the  
number of sectors per track in the drive. The physical specifications of the drive  
do not always correspond to these parameters.  
The BIOS of a PC AT cannot make full use of the physical specifications of these  
drivers. To make the best use of these drives, a BIOS that can handle the standard  
parameters of these drives is required.  
Command  
Data block  
DE  
Commands are instructions to input data to and output data from a drive.  
Commands are written in command registers.  
A data block is the unit used to transfer data. A data block normally indicates a  
single sector.  
Disk enclosure. The DE includes the disks, built-in spindle motor, actuator,  
heads, and air filter. The DE is sealed to protect these components from dust.  
Master (Device 0)  
The master is the first drive that can operate on the AT bus. The master is daisy-  
chained with the second drive which can operate in conformity with the ATA  
standard.  
C141-E104-02EN  
GL-1  
Glossary  
MTBF  
Mean time between failures. The MTBF is calculated by dividing the total  
operation time (total power-on time) by the number of failures in the disk drive  
during operation.  
MTTR  
Mean time to repair. The MTTR is the average time required for a service person  
to diagnose and repair a faulty drive.  
PIO (Programmed input-output)  
Mode to transfer data under control of the host CPU  
Positioning  
Sum of the seek time and mean rotational delay  
Power save mode  
The power save modes are idle mode, standby mode, and sleep mode.  
In idle mode, the drive is neither reading, writing, nor seeking data. In standby  
mode, the spindle motor is stopped and circuits other than the interface control  
circuit are sleeping. The drive enters sleep mode when the host issues the SLEEP  
command.  
Reserved  
Reserved bits, bytes, and fields are set to zero and unusable because they are  
reserved for future standards.  
Rotational delay  
Time delay due to disk rotation. The mean delay is the time required for half a  
disk rotation. The mean delay is the average time required for a head to reach a  
sector after the head is positioned on a track.  
Seek time  
The seek time is the time required for a head to move from the current track to  
another track. The seek time does not include the mean rotational delay.  
Slave (Device 1)  
The slave is a second drive that can operate on the AT bus. The slave is daisy-  
chained with the first drive operating in conformity with the ATA standard.  
GL-2  
C141-E104-02EN  
Glossary  
Status  
VCM  
The status is a piece of one-byte information posted from the drive to the host  
when command execution is ended. The status indicates the command  
termination state.  
Voice coil motor. The voice coil motor is excited by one or more magnets. In  
this drive, the VCM is used to position the heads accurately and quickly.  
C141-E104-02EN  
GL-3  
This page is intentionally left blank.  
Acronyms and Abbreviations  
HDD  
Hard disk drive  
A
I
ABRT Aborted command  
AIC  
AMNF Address mark not found  
ATA AT attachment  
Automatic idle control  
IDNF  
ID not found  
IRQ14 Interrupt request 14  
L
AWG American wire gage  
LED  
MB  
Light emitting diode  
B
M
BBK  
BIOS  
Bad block detected  
Basic input-output system  
Mega-byte  
MB/S Mega-byte per seconds  
C
MPU  
Micro processor unit  
CORR Corrected data  
P
CH  
Cylinder high register  
CL  
Cylinder low register  
Command register  
Current sense register  
Current start/stop  
Cylinder register  
PCA  
PIO  
Printed circuit assembly  
Programmed input-output  
CM  
CSR  
CSS  
CY  
R
RLL  
Run-length-limited  
D
S
dBA  
DE  
DH  
dB A-scale weighting  
Disk enclosure  
Device/head register  
SA  
SC  
SG  
SN  
ST  
System area  
Sector count register  
Signal ground  
Sector number register  
Status register  
DRDY Drive ready  
DRQ  
DSC  
DWF  
Ddata request bit  
Drive seek complete  
Drive write fault  
T
E
TPI  
Track per inches  
TRONF Track 0 not found  
ECC  
ER  
Error checking and correction  
Error register  
Typ  
Typical  
ERR  
Error  
U
F
UNC  
VCM  
Uncorrectable ECC error  
FR  
Feature register  
V
H
Voice coil motor  
HA  
Host adapter  
C141-E104-02EN  
AB-1  
This page is intentionally left blank.  
Comment Form  
We would appreciate your comments and suggestions regarding this manual.  
Manual code  
Manual name  
C141-E104-02EN  
MHL2300AT, MHM2200AT, MHM2150AT, MHM2100AT DISK  
DRIVES PRODUCT MANUAL  
Please mark each item: E(Excellent), G(Good), F(Fair), P(Poor).  
General appearance  
Technical level  
Organization  
Clarity  
(
(
(
(
(
)
)
)
)
)
Illustration  
Glossary  
Acronyms & Abbreviations  
Index  
(
(
(
(
)
)
)
)
Accuracy  
Comments & Suggestions  
List any errors or suggestions for improvement.  
Page  
Line  
Contents  
Please send this form to the address below. We will use your comments in planning future editions.  
Address: Fujitsu Learning Media Limited  
22-7 Minami-Ooi 6-Chome  
Shinagawa-Ku  
Tokyo 140-0013  
JAPAN  
Fax: 81-3-5762-8073  
Organization:  
Name:  
C141-E104-02EN  

Motorola V65p User Manual
Morphy Richards Coffeemaker 47160 User Manual
Kenwood CD Player KDC 448U User Manual
JVC RA P31B User Manual
Jensen JCR 540 User Manual
IBM 71P7285 User Manual
Hypertec HYFLUSB024GB User Manual
Hitachi Computer Drive HTS541680J9SA00 User Manual
DeLonghi Coffeemaker EC220 CD User Manual
Belkin F4D162 2 User Manual