HighĆDensity Synchronous Buck
Converter Design Using TPS56xx
Controllers
User’s Guide
June 1999
Mixed-Signal Linear Products
SLVU013
Download from Www.Somanuals.com. All Manuals Search And Download.
Preface
Read This First
About This Manual
This user’s guide describes techniques for designing synchronous buck
converters using TI’sSLVP1111–114evaluationmodules(EVM)andTPS56xx
ripple regulator controllers.
How to Use This Manual
This document contains the following chapters:
- Chapter 1 Introduction
- Chapter 2 Design Procedure
- Chapter 3 Test Results
Information About Cautions and Warnings
This book may contain cautions and warnings.
This is an example of a caution statement.
A caution statement describes a situation that could potentially
damage your software or equipment.
This is an example of a warning statement.
A warning statement describes a situation that could potentially
cause harm to you.
The information in a caution or a warning is provided for your protection.
Please read each caution and warning carefully.
iii
Download from Www.Somanuals.com. All Manuals Search And Download.
Trademarks
Related Documentation From Texas Instruments
- Synchronous Buck Converter Design Using TPS56xx Controllers in
SLVP10x EVMs User’s Guide (literature number SLVU007).
- TPS56xx data sheet (literature number SLVS177A)
- Designer’s Notebook The TPS56xx Family of Power Supply Controllers
(literature number SLVT140A)
- Designing Fast Response Synchronous Buck Regulators Using the
TPS5210 (literaure number SLVA044).
FCC Warning
Thisequipmentisintendedforuseinalaboratorytestenvironmentonly. Itgen-
erates, uses, and can radiate radio frequency energy and has not been tested
for compliance with the limits of computing devices pursuant to subpart J of
part 15 of FCC rules, which are designed to provide reasonable protection
against radio frequency interference. Operation of this equipment in other en-
vironments may cause interference with radio communications, in which case
the user at his own expense will be required to take whatever measures may
be required to correct this interference.
Trademarks
TI is a trademark of Texas Instruments Incorporated.
iv
Download from Www.Somanuals.com. All Manuals Search And Download.
Contents
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
Synchronous Buck Regulator Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Hysteretic Control Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
Design Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
Design Specification Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Board Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
2
Design Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1
2.1 TPS56xx Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-2
2.1.1 Undervoltage Lockout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
V
CC
2.1.2 Inhibit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.1.3 Slowstart Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
2.1.4 Hysteresis Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-5
2.1.5 Noise Suppression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.1.6 Overcurrent Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
2.1.7 Overvoltage Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.1.8 Power Good . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.1.9 Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.1.10 Gate Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
External Component Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2.2.1 Duty Cycle Estimate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2.2.2 Input Capacitance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2.2.3 Output Filter Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2.2.4 Switching Frequency Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-17
2.2.5 Power MOSFET Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-20
2.2
3
Test Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.1
Test Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.1 Static Line and Load Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.2 Output Voltage Ripple . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.3 Efficiency and Power Losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.4 Output Start-Up and Overshoot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
3.1.5 Frequency Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.1.6 Load Current Transient Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.1.7 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
3.1.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4
Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-5
Test Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
3.2
3.3
v
Download from Www.Somanuals.com. All Manuals Search And Download.
Figures
1–1
1–2
1–3
1–4
1–5
1–6
1–7
2–1
2–2
2–3
2–4
2–5
2–6
3–1
3–2
3–3
3–4
3–5
3–6
3–7
3–8
3–9
Simplified Synchronous Buck Converter Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Simplified Hysteretic Controlled Output Voltage Waveform . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
SLVP111–114 EVM Converter Schematic Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7
Top Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Bottom Assembly (Top View) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Top Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-10
Bottom Layer (Top VIew) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-11
TPS56xx Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Block Diagram Showing Noise Suppression Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
V
Sensing Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-8
DS
Gate Driver Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11
I–V Characteristic Curve for Low-Side Gate Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-12
Output Ripple Voltage Detail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18
Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
SLVP111 Measured Load Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
SLVP111Measured Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
SLVP111Measured Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
SLVP111Measured Switching Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8
SLVP111 Measured Switching Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
SLVP111Measured Start-Up (INHIBIT) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9
SLVP111 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
CC
SLVP111Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
IN
3–10 SLVP111 Measured Load Transient Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
3–11 SLVP112 Measured Load Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-11
3–12 SLVP112 Measured Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
3–13 SLVP112 Measured Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
3–14 SLVP112 Measured Switching Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
3–15 SLVP112 Measured Switching Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-13
3–16 SLVP112 Measured Start-Up (INHIBIT) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
3–17 SLVP112 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
CC
3–18 SLVP112 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15
IN
3–19 SLVP112 Measured Load Transient Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15
3–20 SLVP113 Measured Load Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16
3–21 SLVP113 Measured Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16
3–22 SLVP113 Measured Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-17
3–23 SLVP113 Measured Switching Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-17
3–24 SLVP113 Measured Switching Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-18
3–25 SLVP113 Measured Start-Up (INHIBIT) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-18
3–26 SLVP113 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
CC
vi
Download from Www.Somanuals.com. All Manuals Search And Download.
3–27 SLVP113 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-19
IN
3–28 SLVP113 Measured Load Transient Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
3–29 SLVP114 Measured Load Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20
3–30 SLVP114 Measured Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
3–31 SLVP114 Measured Power Dissipation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-21
3–32 SLVP114 Measured Switching Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3–33 SLVP114 Measured Switching Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-22
3–34 SLVP114 Measured Start-Up (INHIBIT) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
3–35 SLVP114 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23
CC
3–36 SLVP114 Measured Start-Up (V ) Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
IN
3–37 SLVP114 Measured Load Transient Waveforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24
Tables
1–1
1–2
1–3
Summary of EVM Converter Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-4
EVM Converter Operating Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
SLVP111–114 EVMs Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-8
Contents
vii
Download from Www.Somanuals.com. All Manuals Search And Download.
viii
Download from Www.Somanuals.com. All Manuals Search And Download.
Chapter 1
Introduction
The SLVP111/112/113/114 evaluation modules (EVMs) have been designed
and tested using the TPS56xx hysteretic controllers. These boards are
synchronous dc-dc buck converters with fixed output voltages of 3.3 V, 2.5 V,
1.8 V and 1.5 V respectively. They use only surface mount components and
are design examples of how to use TI’s TPS56xx controllers in high density,
low loss applications with tight static and dynamic output voltage
requirements. Detailed test results taken from the EVMs are presented.
Design simplicity, low component count, and lower cost make buck converters
popular solutions where low input voltages are available for the converter and
where isolation is not a requirement.
This user’s guide describes techniques for designing synchronous buck
converters using TI’s SLVP111–114 EVMs and TPS56xx ripple regulator
controllers. Synchronous buck converters provide an elegant power supply
solution for rapidly transitioning DSP loads (such as the Texas Instruments
TMS320C62x/67x family), fast memory, and similar processors. An order of
magnitude improvement in dynamic response of this converter over standard
control methods reduces hold-up capacitance needs near the transitioning
loads, thus saving cost and board space.
Topic
Page
1.1 Synchronous Buck Regulator Operation . . . . . . . . . . . . . . . . . . . . . . . . 1–2
1.2 Hysteretic Control Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–3
1.3 Design Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–4
1.4 Design Specification Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–5
1.5 Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
1.6 Bill of Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
1.7 Board Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–10
1-1
Download from Www.Somanuals.com. All Manuals Search And Download.
Synchronous Buck Regulator Operation
1.1 Synchronous Buck Regulator Operation
The synchronous buck converter is a variation of the traditional buck
converter. The main switching device is usually a power MOSFET and is
driven in the same manner as in a traditional buck converter. The freewheeling
rectifier, usually a Schottky device, is replaced by a power MOSFET and is
driven in a complementary or synchronous fashion relative to the main
switching device; when one MOSFET is on, the other is off. The freewheeling
MOSFET is selected so that its ON voltage drop is less than the forward drop
of the original freewheeling rectifier, thus increasing conversion efficiency. A
very important design issue when using a synchronous buck converter is
preventing cross-conduction of the two power MOSFETs, i.e., preventing both
MOSFETs from being on simultaneously. A small amount of deadtime is
necessary.
Figure 1 shows a simplified schematic of a synchronous buck converter. The
TPS56xx senses the output voltage and then drives Q1 and Q2 depending on
thesensedvoltage. TheTPS56xxsensesthevoltageatthejunctionofQ1, Q2,
and L1 and uses it to actively prevent simultaneous conduction of Q1 and Q2.
Figure 1–1. Simplified Synchronous Buck Converter Schematic
V
I
+
Q1
Q2
L1
TPS56xx
V
O
+
C
1-2
Download from Www.Somanuals.com. All Manuals Search And Download.
Hysteretic Control Operation
1.2 Hysteretic Control Operation
Hysteretic control, also called bang-bang control or ripple regulator control,
maintains the output voltage within the hysteresis band centered about the
internal reference voltage. Figure 1–2 shows a simplified example of a
hysteretic controlled output voltage using the TPS5625 with a reference
voltage of 2.500 V and a hysteresis band of 50 mV. If the output voltage is at
or below the level of the reference minus one-half of the hysteresis band
(V = 2.475 V), the TPS5625 turns off the low-side MOSFET (Q2 in Figure
Lo
1–1) and turns on the high-side MOSFET (Q1 in Figure 1–1) of the
synchronous buck converter power stage. This is the power stage on-state,
and it causes the output voltage to increase. When the output voltage reaches
or exceeds the reference plus one-half of the hysteresis band (V = 2.525 V),
Hi
the TPS5625 turns off the high-side MOSFET and turns on the low-side
MOSFET. This is the power stage off-state, and it causes the output voltage
to decrease. This hysteretic method of control keeps the output voltage within
the hysteresis band around the reference voltage. If output-load current steps
or input-voltage transients force the output voltage out of the hysteresis band,
the TPS5625 sets the power-stage MOSFETs in the continuous on or off state
as required to return the output voltage to the hysteresis band. Thus, the
output voltage is corrected as quickly as the output filter allows. There are no
error amplifier sensing and adjusting delays, as is the case with either voltage-
or current-mode controllers. Other advantages of hysteretic control include no
loop compensation design and no input filter interaction problems.
Figure 1–2. Simplified Hysteretic Controlled Output Voltage Waveform
Output
Voltage
V
Hi
2.525 V
2.500 V
2.475 V
Hysteresis
V
Lo
Time
On Time
Off Time
T
s
Introduction
1-3
Download from Www.Somanuals.com. All Manuals Search And Download.
Design Strategy
1.3 Design Strategy
The SLVP111–114 evaluation modules (EVMs) are optimized for 5-V main
input voltage and 6-A output current. The EVMs need an additional low current
12-V (30 mA max) input voltage for the controller. TI’s application report,
Providing a DSP Power Solution from 5 V or 3.3 V Only Systems, TI literature
number SPRA525 describes how one can implement a simple boost circuit for
5-V only input voltage applications. These EVMs are pin to pin compatible with
SLVP104/105/106/115 evaluation boards with 8 A output current, which
combine surface mount and through hole components. This surface mount
version has the same length, 2″, and width, 0.75″, but the height is significantly
lower, 0.375″ versus 0.6″ for through hole version.
The TI SLVP111–114 evaluation modules (EVM) provide synchronous buck
converter circuits for evaluating the capabilities of the TPS56xx family of ripple
regulator controllers. The EVM converters can provide proven, demonstrated
reference designs to aid in the rapid development of application-specific
synchronous buck converters. Output capacities of the EVM converters are
optimized for the Siliconix Si4410 power MOSFET device.
The 6-ampere output current level is a reasonable selection criteria for
powering circuit cards with multiple DSPs, and for providing the regulated
voltage to other hardware on the circuit card. Component size can be reduced
for designs requiring lower power levels.
The TPS56xx controllers each provide one of four popular output voltage
levels. The last two digits of the part number correlate to the set-point voltage
level: TPS5633 is the 3.3-V controller, TPS5625 is the 2.5-V controller,
TPS5618 is the 1.8-V controller, and TPS5615 is the 1.5-V controller. Many
digital devices, memories, and DSP I/O circuits use the 3.3-V level. The core
oftheTMS320C6201requires2.5-V. AlloftheotherDSPsintheTMS320C62x
and the TMS320C67x family need 1.8 V. The GTL bus, as well as various
processors and future DSPs, may require the 1.5-V controller. An external
resistor divider can be used to fine tune the output voltages of these controllers
for other applications including output voltages up to approximately
V
– 0.5 V.
IN
Table 1–1 summarizes the four EVM converter modules.
Table 1–1.Summary of EVM Converter Modules
EVM Part Number
EVM Board
Controller
Output
Voltage
Max. Output
Current
Number
SLVP111
SLVP112
SLVP113
SLVP114
†
TPS5633EVM–111
TPS5625EVM–112
TPS5618EVM–113
TPS5615EVM–114
TPS5633
TPS5625
TPS5618
TPS5615
3.3 V
2.5 V
1.8 V
1.5 V
6 A
†
6 A
†
6 A
†
6 A
†
Output current is limited by the temperature rise of the power MOSFETs chosen. Higher or
lower current designs are possible.
1-4
Download from Www.Somanuals.com. All Manuals Search And Download.
Design Specification Summary
1.4 Design Specification Summary
This section summarizes the design requirements of the EVM converters.
Although every attempt was made to accurately describe the performance of
the EVM converters and the TPS56xx controllers, in case of conflicts, the
TPS56xx data sheet takes precedence over this document.
The TPS56xx family of controllers provides the necessary regulation
functions. In addition to a reference voltage accuracy of ±1% over the full
operating temperature range, the controller has remote sense inputs to
provide a precisely regulated output voltage. The controller also provides
undervoltage lock-out, overload protection, overvoltage protection, and
overtemperature protection. The controller has a logic level INHIBIT input to
control the converter turn-on and turn-off and a power good output to indicate
output voltage status. Undervoltage lock-out prevents operation of the power
supply when the 12 Vdc input voltage is not sufficient for proper operation.
Overload protection protects the power supply from accidental overloads or
short circuits. Overvoltage protection prevents damage to the load in the event
of an internal power supply failure or presence of high voltages on the output
from an external condition. Both overvoltage and overcurrent cause a latched
shutdown. Both power MOSFETs are driven to an OFF state. Recovery from
shutdown requires removal of the 12 V control input supply for reset. Table 1–2
lists the operating specifications of the EVM converters.
Table 1–2.EVM Converter Operating Specifications
Specification
Min
Typ
Max
6
Units
Power input voltage range
Control input voltage range
4.5
V
V
10.8
13.2
†
Static voltage tolerance
SLVP111 (3.3 V)
SLVP112 (2.5 V)
SLVP113 (1.8 V)
SLVP114 (1.5 V)
3.27
2.47
1.78
1.48
3.30
2.50
1.80
1.50
3.33
2.53
1.82
1.52
V
V
V
‡
Line regulation
± 0.05%
± 0.1%
§
Load regulation
± 0.2%
± 0.4%
¶
Transient response
± 100
mV pk
µsec
50
#
Output current range
0
6
A
A
#
Current limit
10
k
Operating frequency
SLVP111 (3.3 V)
SLVP112 (2.5 V)
SLVP113 (1.8 V)
SLVP114 (1.8 V)
135
225
295
360
kHz
kHz
kHz
kHz
Introduction
1-5
Download from Www.Somanuals.com. All Manuals Search And Download.
Design Specification Summary
Table 1–2.EVM Converter Operating Specifications (Continued)
Specification
Min
Typ
Max
Units
||
Output ripple
SLVP111 (3.3 V)
SLVP112 (2.5 V)
SLVP113 (1.8 V)
SLVP114 (1.5 V)
66
50
36
30
mV p–p
mV p–p
mV p–p
mV p–p
Efficiency, 6 A load
SLVP111 (3.3 V)
SLVP112 (2.5 V)
SLVP113 (1.8 V)
SLVP114 (1.5 V)
90%
86.4%
83.2%
79.8%
Efficiency, 4 A load
SLVP111 (3.3 V)
SLVP112 (2.5 V)
SLVP113 (1.8 V)
SLVP114 (1.5 V)
91.6%
88.6%
85.1%
81.9%
†
Vi = 5 V, Io = 6 A
Io = 6 A, Vi = 5 V ±10%
Vi = 5 V
Vi = 5 V, Io stepped repetitively from 0 A to 6.5 A
Output current rating is limited by thermal considerations. Load currents above this rating may
cause damage to the power supply.
‡
§
¶
#
||
k
Unless otherwise specified, all test conditions are T = 25_C, Vi = 5 V, Io = 6 A, Vo = nominal.
Vi = 5 V, Io = 6 A, Vo = nominal
A
1-6
Download from Www.Somanuals.com. All Manuals Search And Download.
Schematic
1.5 Schematic
Figure 1–3 shows the EVM converter schematic diagram. The schematic
diagrams for the other EVM converters are identical except for the controller
IC used.
Figure 1–3. SLVP111–114 EVM Converter Schematic Diagram
+
+
+
V C C
B O O T
H U G H D R V G D N R D
V
D R
1 5
1 6
1 7
1 8
1 9
2 0
2 1
2 2
2 3
2 4
2 5
2 6
2 7
2 8
1 4
L O W D R
1 3
1 2
B O O T L O L O H I B
1 1
H I S E V N S E L O D R
1 0
L O S E N S E
B I A S
9
F µ 0 . 1
I O U T L O
S L O W S T
C 6
8
7
6
5
I N H I B I T
A N A G N D
N C
N C
N C
N C
N C
V S E N S E
V R E F B
V H Y S T
4
O C P
3
A G N D 2
2
P W R G D
I O U T
1
T P S 5 6 3 3 ( S e e N o t e )
U 1
Introduction
1-7
Download from Www.Somanuals.com. All Manuals Search And Download.
Bill of Materials
1.6 Bill of Materials
Table 1–3 lists materials required for the SLVP111–114 EVMs.
Table 1–3.SLVP111–114 EVMs Bill of Materials
Ref Des
C1
Part Number
Description
MFG
10TPA33M
Capacitor, POSCAP, 33 µF, 10 V, 20%
Capacitor, POSCAP, 150 µF, 6.3 V, 20%
Capacitor, POSCAP, 150 µF, 6.3 V, 20%
Capacitor, POSCAP, 150 µF, 6.3 V, 20%
Capacitor, Ceramic, 0.1 µF, 16 V, 10%, X7R
Capacitor, Ceramic, 0.1 µF, 16 V, 10%, X7R
Capacitor, Ceramic, 1.0 µF, 16 V, +80%–20%
Capacitor, Ceramic, 1.0 µF, 16V, +80%–20%
Capacitor, Ceramic, 2.2 µF, 16V, Y5V
Capacitor, POSCAP, 150 µF, 4 V, 20%
Capacitor, POSCAP, 150 µF, 4 V, 20%
Capacitor, POSCAP, 150 µF, 4 V, 20%
Capacitor, POSCAP, 150 µF, 4 V, 20%
Capacitor, Ceramic, 10 µF, 16 V, Y5V
Capacitor, Ceramic, 0.01 µF, 25 V, +80%–20%, Y5V
Capacitor, Ceramic, 1.0 µF, 16 V, +80%–20%
Capacitor, Ceramic, 0.1 µF, 16 V, 10%, X7R
Capacitor, Ceramic, 0.1 µF, 16 V, 10%, X7R
Capacitor, Ceramic, 0.01 µF, 25 V, 10%, X7R
Capacitor, Ceramic, 1000 pF, 50 V, 10%, X7R
Capacitor, Ceramic, 0.1 µF, 16 V, 10%, X7R
Capacitor, Ceramic, 1.0 µF, 16 V, +80%–20%
Header, RA, 18-pin, 0.23 Posts x 0.20 Tails
Inductor, 2.2 µH, 7.4 A
Sanyo
C2
6TPB150M
Sanyo
C3
6TPB150M
Sanyo
C4
6TPB150M
Sanyo
C5
GRM39X7R104K016A
GRM39X7R104K016A
GRM42–6Y5V105Z016A
GRM42–6Y5V105Z016A
GRM42–6Y5V225Z016A
4TPC150M
muRata
muRata
muRata
muRata
muRata
Sanyo
C6
C7
C8
C9
C10
C11
C12
C13
C14
C15
C16
C17
C18
C19
C20
C21
C22
J1
4TPC150M
Sanyo
4TPC150M
Sanyo
4TPC150M
Sanyo
GRM235Y5V106Z016A
GRM42-6Y5V103Z025A
GRM42–6Y5V105Z016A
GRM39X7R104K016A
GRM39X7R104K016A
GRM39X7R103K025A
GRM39X7R102K050A
GRM39X7R104K016A
GRM42–6Y5V105Z016A
S1122–18–ND
DO3316P–222HC
DO3316P–152HC
Si4410DY
muRata
muRata
muRata
muRata
muRata
muRata
muRata
muRata
muRata
Sullins
L1
Coilcraft
Coilcraft
Siliconix
Siliconix
L2
Inductor, 1.5 µH, 9 A
Q1
FET, N-ch, 30-V, 10-A, 13-mΩ
Q2
Si4410DY
FET, N-ch, 30-V, 10-A, 13-mΩ
R1
Std
Resistor, Chip, 1 kΩ, 1/16W, 5%
R2
Std
Resistor, Chip, 10 kΩ, 1/16W, 5%
R3
Std
Resistor, Chip, 10 Ω, 1/10W, 5%
R4
Std
Resistor, Chip, 10 Ω, 1/10W, 5%
R5
Std
Resistor, Chip, 2.7 Ω, 1/4W, 5%
R6
Std
Resistor, Chip, 4.7 Ω, 1/16W, 5%
1-8
Download from Www.Somanuals.com. All Manuals Search And Download.
Bill of Materials
Table 1–3.SLVP111–114 EVMs Bill of Materials (Continued)
Ref Des
R7
Part Number
Description
MFG
Std
Resistor, Chip, 1 kΩ, 1/16W, 5%
Resistor, Chip, 100 Ω, 1/16W, 1%
Resistor, Chip, 11 kΩ, 1/16W, 5%
Resistor, Chip, 100 Ω, 1/16W, 1%
Resistor, Chip, 20 kΩ, 1/16W, 1%
Resistor, Chip, 4.7 Ω, 1/16W, 5%
Resistor, Chip, 750 Ω, 1/16W, 5%
Resistor, Chip, 20 kΩ, 1/16W, 1%
Resistor, Chip, 1 MΩ, 1/16W, 5%
R8
Std
R9
Std
R10
R11
R12
R13
R14
R15
U1a
U1b
U1c
U1d
Std
Std
Std
Std
Std
Std
TPS5633PWP
TPS5625PWP
TPS5618PWP
TPS5615PWP
IC, PWM Ripple Controller, Fixed 3.3-V (SLVP111 only)
TI
IC, PWM Ripple Controller, Fixed 2.5-V (SLVP112 only) TI
IC, PWM Ripple Controller, Fixed 1.8-V (SLVP113 only) TI
IC, PWM Ripple Controller, Fixed 1.5-V (SLVP114 only) TI
Introduction
1-9
Download from Www.Somanuals.com. All Manuals Search And Download.
Board Layout
1.7 Board Layout
Figures 1–4 through 1–7 show the board layouts for the SLVP111–114
evaluation modules.
Figure 1–4. Top Assembly
Top Assembly
Bottom Assembly (Top View)
Top Layer
Figure 1–5. Bottom Assembly (Top View)
Figure 1–6. Top Layer
1-10
Download from Www.Somanuals.com. All Manuals Search And Download.
Board Layout
Figure 1–7. Bottom Layer (Top VIew)
Bottom Layer (Top View)
Introduction
1-11
Download from Www.Somanuals.com. All Manuals Search And Download.
1-12
Download from Www.Somanuals.com. All Manuals Search And Download.
Chapter 2
Design Procedure
The SLVP111–114 are dc-dc synchronous buck converter evaluation modules
(EVMs) that provide a regulated output voltage at up to 6 A with a power input
voltage range of 4.5 V to 6 V. A low power 12-V, 20-mA source is also required
to power the TPS56xx controller. The controller operates at a nominal
frequencyof135kHzfor5Vinputand3.3Voutput. Toprovidethehighestlevel
of performance, the EVM converters use hysteretic, or ripple, control.
Hysteretic-controlled synchronous buck converters have several advantages
over conventional PWM-controlled power supplies:
- Correction of output voltage variations caused by output-load or input-
voltage transients is extremely fast.
- The user controls output ripple by adjusting the operational parameters of
the converter, instead of relying on brute force methods requiring the
choice of an output filter.
- Hysteretic control sets the operational frequency of the converter. For a
given set of external components, the lower the permissible ripple setting,
the higher the operational frequency.
- There is no control loop to design.
Topic
Page
2.1 TPS56xx Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–2
2.2 External Component Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–14
2-1
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
2.1 TPS56xx Functions
The functional block diagram of the TPS56xx family of controllers is given in
Figure 2–1. The controller has the following main features:
-
±1% reference over 0°C to125°C junction temperature range.
- Synchronous-buck gate drivers with adaptive deadtime control
- High-side MOSFET driver voltage rating of 30 V
- MOSFET driver peak current rating of 2 A
- Hysteretic comparator: 250-ns propagation delay to gate driver outputs,
2.5-mV offset voltage, symmetrical hysteresis, hysteresis setting is a
percentage of Vref.
- Lossless output current sensing circuit
- Slowstart circuit; slowstart time independent of VID setting
- Internal 8-V drive regulator for reduced gate charge power losses
- POWERGOOD comparator, 93% of Vref trip
- UVLO, Vcc undervoltage lockout, 10-V start, 2-V hysteresis
- INHIBIT comparator that can also monitor UVLO of the system logic
supply, 2.1-V start, 100-mV hysteresis.
- Latched overcurrent shutdown circuit
- Latched overvoltage shutdown circuit
- LODRV pin that activates the low-side MOSFETs as a crowbar to protect
against a short across the high-side MOSFETs.
2-2
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
Figure 2–1. TPS56xx Functional Block Diagram
15
CC
7
28
20
19
21
ANAGND PWRGD LOSENSE IOUTLO HISENSE
V
V
CC
2 V
+
_
22
1
INHIBIT
IOUT
2X
UVLO
Shutdown
10 V
Q
S
R
V
CC
Fault
Rising
Edge
Delay
3
_
Deglitch
Deglitch
OCP
+
HIGHDR
100mV
HIGHIN
VPGD
0.93 VREF
V
CC
Analog
Bias
VOVP
1.15 VREF
VSENSE
PREREG
Analog
Bias
VREFB
9
BIAS
I
Slowstart
Comparator
8
14
DRV REG
SLOWST
5
DRV
+
_
Shutdown
16
BOOT
17
HIGHDR
_
Shutdown
CM Filters
18
BOOTLO
+
Bandgap
VREF
_
+
_
Hysteresis
Comparator
13
LOWDR
+
12
DRVGND
Hysteresis
Setting
I
VREFB
2
5
4
6
11
LOHIB
10
LODRV
AGND2
VREFB
VHYST VSENSE
This section describes the functions governed by the TPS556xx. A procedure
is given to determine the values of components used in the example design
given in Figure 1–3. Example calculations for the 3.3-V output version accom-
pany the design equations. There are many possible ways to proceed when
designing power supplies and some iteration may be necessary when actual
performance differs from design predictions. Reference designators refer to
the circuit in Figure 1–3.
2.1.1 V
Undervoltage Lockout
CC
TheV undervoltagelockoutcircuitdisablesthecontrollerwhileV isbelow
CC
CC
the 10-V start threshold during power up. While the controller is disabled, the
outputdriverswillbelowandtheslowstartcapacitorwillbeshorted.WhenV
CC
exceeds the start threshold, the short across the slowstart capacitor is
released and normal converter operation begins. There is a 2-V hysteresis in
the undervoltage lockout circuit for noise immunity.
Design Procedure
2-3
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
2.1.2 Inhibit
The inhibit circuit is a comparator with a 2.1-V start voltage and a 100-mV
hysteresis. When inhibit is low, the output drivers are low and the slowstart
capacitor is discharged. When inhibit is above the start threshold, the short
across the slowstart capacitor is released and normal operation begins.
When the system logic supply is connected to the inhibit pin, the inhibit pin also
controls power sequencing by locking out controller operation until the system
logic supply exceeds the input threshold voltage of the inhibit circuit; thus, the
12-V supply and the system logic supply must be above UVLO thresholds
before the controller is allowed to start up.
2.1.3 Slowstart Design
Slowstart or soft-start is added to reduce power-up transients. Without
slowstart, when input power is first applied, the TPS56xx attempts to raise the
output voltage (initially zero) to its normal operating level by turning on the top
MOSFET until the voltage is approximately Vref. This can cause high transient
currents to flow in the output inductor and output capacitor. Although this form
of startup usually does not cause component failures, it does apply stresses
much greater than those typically encountered in normal operation. It is good
design practice to include slowstart circuitry to avoid these unnecessary
stresses.
The slowstart circuit in the TPS56xx controls the rate at which the output
voltagepowersup. Acapacitor, C21, connectedbetweenSLOWST(pin8)and
ANAGND (pin 7), is charged by an internal current source. This current source
is proportional to the reference voltage and is adjustable by an external
resistance selected by the user. The output voltage follows the voltage on the
slowstart capacitor during startup. Since the charging current is proportional
to the reference voltage, the slowstart time is independent of the reference
voltage for a given resistor value.
Choices of the slowstart time and the slowstart capacitor value are largely
arbitrary as long as system start-up time requirements are met. For this
example design, a slowstart time of 10 ms is chosen, and the slowstart
capacitor is chosen to be 0.1 µF. Therefore, to charge 0.1 µF from zero volts
to 3.3 V in 10 ms, the following equation holds:
DV
c
3.3 V
10 ms
I
+ C
+ 0.1 mF
+ 33 mA
SLOWSTART
SLOWSTART
Dt
ss
The slowstart charging current is determined by the following equation:
(
)
I VREFB
I
+
SLOWSTART
5
where I(VREFB) = the current out of VREFB (pin 5).
For an I
(
current equal to 33 µA, I(VREFB) should be set to:
SLOWSTART
)
I VREFB + 5 I
+ 5 33 mA + 165 mA
SLOWSTART
The voltage on VREFB (pin 5) is a buffered reference voltage. The resistance
from the VREFB pin to ANAGND can be calculated as:
2-4
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
3.3 V
165 mA
R
+
+ 20 kW
VREFB
This value is used to determine the values of R10 and R14 that set the
hysteresis level.
The equations above can be used to derive a simplified relationship for the
slowstart time as shown:
t
+ 5 C
R
SLOWSTART
SLOWSTART
VREFB
V
start-up waveforms for different reference voltage settings are given in
O
Figures 3–7, 3–8, 3–9, 3–16, 3–17, 3–18, 3–25, 3–26, 3–27, 3–34, 3–35, and
3–36 in the test results section, showing that slowstart time is independent of
the reference voltage.
2.1.4 Hysteresis Setting
The next step in this design is choosing the desired output voltage ripple. As
a first approximation, the output voltage ripple is simply the differencebetween
the two levels (V
and V ) shown in Figure 1–2. The hysteresis setting of
LO
Hi
the hysteresis comparator of the TPS56xx sets these two levels. The
hysteresis is set by two external resistors and is centered around VREF (pin
5). The hysteretic comparator is designed with low input offset voltage
(±2.5 mV max) low propagation delays (250ns max to gate driver outputs with
10mV overdrive) and accurate hysteresis setting (±3.5 mV max). The
hysteresis is proportional to the reference voltage; setting Vref to a new value
automatically adjusts the hysteresis to be the same percentage of Vref.
The total output ripple is greater than the ripple set by the hysteresis compara-
tor. To accurately choose the output voltage ripple, all of the propagation de-
lays must be considered. The first step is to calculate the amount of ripple ex-
pected due to the delays, in addition to the ripple set by the hysteresis compa-
rator. Using the terminology of Section 2.2.4.1, this additional ripple is equal
to V
– Hyst and will be referred to V for ease of reference. Or:
del
p-p
V
+ Hyst ) V
p-p
del
Where:
V
=the total output ripple,
p-p
Hyst = the hysteresis setting of the hysteretic comparator, and
V
= the output ripple exceeding Hyst and due to all propagation
delays.
del
Assuming the output ripple during the delay time is primarily caused by the
inductor ripple current flowing through the output capacitors’ ESR, the
additional ripple, V , can be estimated as:
del
V –V
V
V t
I
O
O
I
del
+ ƪ ƫ ƪ ƫ
V
t
ESR )
t
ESR +
del
del
del
L
L
L
Where L is the value of the output inductor.
Design Procedure
2-5
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
Note that V is independent of the output voltage.
del
To calculate V for this example design, use the component measurements
del
given in Section 2.2.4.2. They are repeated here for convenience:
L = 1.5 µH
ESR = 10 mΩ
T
= 400 ns
del
Now calculate V
:
del
5
–9
–3
V
+
400 10 10 10 + 13.3 mV
del
–6
1.5 10
Since V does not depend on the output voltage setting, V is a larger por-
del
del
tion of the total output voltage ripple for lower output voltages. So, the hystere-
sis (which is proportional to the output voltage setting) should be set so that
the desired output ripple voltage is obtained for the lowest output voltage. The
total output voltage ripple requirement per Table 1–2 for an output voltage of
1.5 V is 30 mV, the hysteresis is set as:
V
+ 30 mV * V
+ 30 mV * 13.3 mV + 16.7 mV
Hysteresis
del
For convenience, and allowing a little margin, the hysteresis will be designed
for 15 mV.
To set the hysteresis, connect two external resistors to form a resistor divider
fromVREFB(pin5)toANAGND(pin7)withthecenterofthedividerconnected
to VHYST (pin 4). The hysteresis of the comparator is equal to twice the
voltage that is between the VREFB (pin 5) and VHYST (pin 4) pins. Or,
(
)
V
+ 2 VREFB–VHYST
Hysteresis
For this design, 15 mV of hysteresis was chosen for a 1.5 V output voltage.
(
)
V
+ 15 mV + 2 1.5 V–VHYST
Hysteresis
Solving for VHYST:
V
Hysteresis
2
15 mV
2
VHYST + VREFB–
+ 1.5–
+ 1.4925 V
Referring to the schematic, Figure 1–3, the two external resistors are R10 and
R14. From the previous section, the total resistance required is 20 kΩ. Since
R10 is very small compared to R14, for simplicity set R14 = 20 kΩ. To calculate
the value of R13:
R14
R14 ) R10
VHYST + VREFB
Solving for R10:
(
) (
1.4925 V
)
1.5 V 20 kW
VREFB R14
R10 +
–R14 +
–20 kW + 100 W
VHYST
The controller hysteresis is now set to 15 mV. This, in addition to the 13.3 mV
due to propagation delays, results in a total ripple voltage of less than the
design goal of 30 mV for an output voltage of 1.5 V.
2-6
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
2.1.5 Noise Suppression
Hysteretic regulators, by nature, have a fast response time to V transients
O
and are thus inherently noise sensitive due to the very high bandwidth of the
controller. Noise suppression circuits were added to the TPS56xx to improve
thenoiseimmunity, asshowninFigure2–2. Internallow-passfilterswithapole
frequency of 5 MHz were added to the inputs of the hysteretic comparator.
These low-pass filters are referenced to the same analog ground as the
hysteretic comparator. There is a common-mode filter with a 4-MHz pole
between VREFB and VHYST to filter out noise between these pins. A double
pulse suppression circuit prohibits spurious pulses from propagating to the
gate drivers. The double pulse suppression circuit becomes active when the
comparator has toggled or when the LOHIB pin (which is connected to the
power MOSFETs) has transitioned, providing additional noise immunity from
internally and externally generated noise. The suppression circuit is active for
150 ns.
A low-pass filter is recommended between V and the VSENSE pin (R1 and
O
C3inFigure2–2);recommendedvaluesare100ohmsand1nF. Thislow-pass
filter is included in the evaluation design of Figure 1–3 (R8, R11, and C20).
Figure 2–2. Block Diagram Showing Noise Suppression Circuits
V
in
L1
V
ds
Sensing
C1
5 MHz
Filter
HIGHDR
LOHIB
V
ref
Reference
VO
L2
5 MHz
Filter
Adaptive
Deadtime
Control
Double Pulse
Suppression
Circuit
V
phase
LOWDR
C2
–
+
Hysteresis
Setting
4 MHz
Filter
TPS56xx Synchronous-Buck Controller
R1
VSENSE
IOUT
R5
VREFB
R3
VHYST
R2
C3
VH_SET
R6
R4
2.1.6 Overcurrent Protection
Overcurrent protection is provided by measuring the on-state voltage of the
high-side MOSFET, conditioning the measured voltage, and comparing the
result to a reference voltage. If the output current exceeds the current limit
setpoint, a fault latch is set and the output drivers are turned off. Vcc (12 V)
must be reduced to below the undervoltage lockout value to restart the
converter.
A sample-and-hold circuit measures the power supply output current by
sensing the on-state drain-to-source voltage of the high-side MOSFET (Q1 in
Design Procedure
2-7
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
Figure 1–3). This arrangement improves efficiency over solutions having a
separate current sensing resistor. The drain of the high-side MOSFET is
connected to HISENSE (pin 19). The source of the high-side MOSFET is
connected to LOSENSE (pin 20). When the high-side MOSFET is on, a
TPS56xx internal switch is also on and samples the source voltage of the
high-side MOSFET. This sampled voltage is applied to IOUTLO (pin 21) and
is held by the external 0.1-µF capacitor, C6, which is connected from IOUTLO
(pin 21) to HISENSE (pin 19). The TPS56xx amplifies (gain=2) the
sampled-and-held voltage on C6 and sends the output voltage to IOUT (pin
1).
Figure 2–3. V
Sensing Circuit
DS
Cs/h
HISENSE
IOUTLO
V
in
V
Sensing Circuit
+
ds
–
Rising
Edge
Delay
Shutdown
L1
Gain of 2
C1
HIGHDR
LOSENSE
HIGHIN
LOWIN
VO
L2
Adaptive
Deadtime
Control
V
ref
V
phase
LOWDR
C2
TPS56xx Synchronous-Buck Controller
R1
VSENSE
IOUT
R2
C3
Figure 2–3 gives a simple block diagram of the Vds sensing circuit. The Vds
sensing circuit measures the average voltage across the high-side MOSFET
when the high-side MOSFET is on, and holds that value on a sample/hold
capacitor when the high-side MOSFET is off. The voltage on the sample/hold
capacitor is directly proportional to the load current. Sensing across the
high-side MOSFET rather than the low-side MOSFET ensures that shorted
loads can be detected. The RC time constant of the sample/hold network must
be greater than the conduction-time of the high-side MOSFET, otherwise the
sample/hold circuit will function as a peak detector circuit and will not hold the
average Vds voltage. The differential voltage across the sample/hold
capacitor is amplified by 2 and converted to a single-ended signal on the IOUT
pin. TheDCCMRRoftheVdssensingamplifieris69dBminimum. Addedlogic
ensures that sampling begins and ends while the high-side MOSFET is
conducting. The turn-on and turn-off delays of the sample/hold switch are less
than 100 ns. Additional logic and a rising edge delay circuit are included to
guarantee sampling during a short-to-ground fault across the low-side
MOSFET; the rising edge delay time is 500 ns.
Resistors R7 and R13 in Figure 1–3 set the current limit setpoint. This
resistor-divider network applies the IOUT output voltage to OCP (pin 3). The
2-8
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
resistor-divider network is designed so that the voltage applied to OCP is
100 mV for the desired output current limit point. If the voltage on OCP
exceeds 100 mV, a fault latch is set and the output drivers are turned off. The
latch remains set until VCC (pin 15) goes below the undervoltage lockout
value.
The following equations summarize the relationships discussed above.
The on state drain-to-source voltage of the high-side MOSFET, Q1, is:
ǒV
Ǔ
–V
+ I R
HISENSE IOUTLO
O
DS(on)
where R
is the value obtained from the high-side MOSFET and includes
DS(on)
correction for elevated temperature if necessary.
The voltage difference, V – V , is internally amplified by a fixed
HISENSE
IOUTLO
gain of two to produce the IOUT (pin 1) signal.
+ ǒV
Ǔ
V
–V
2
IOUT
HISENSE IOUTLO
The V
signal is scaled for the desired current limit level and applied to the
IOUT
OCP pin:
R13
R13 ) R7
V
+ V
OCP
IOUT
Therefore, to set the power supply output current, first calculate the quantity
I
× R for the value of I desired for current limit. Variations in R
,
O
DS(on)
O
DS(on)
including its temperature dependence, should be considered, since this
parameter can vary a significant amount for typical MOSFETs. Next, multiply
this voltage by two. Finally, set the R7 and R13 voltage divider to produce 100
mV at the desired current limit point.
For this design, the maximum output current is 6 A. In most power supply
designs, the exact current limit set point rarely needs to be very accurate.
Limiting the current to a level that is not destructive is the main consideration.
In this case, the nominal current limit is set for approximately 25% above the
maximum, including an elevated temperature correction factor of 1.4 applied
to a nominal R
of 11 mΩ. The current limit set point, I
, is given by:
DS(on)
OCP
I
+ I
1.25 + 6 (1.25) + 7.5 Amps
OCP
O(Max)
For the above current level, V
is given by:
IOUT(Trip)
(
+ ǒI
Ǔ
)
V
R
2 + 7.5 A 0.011 W 1.4 2 + 0.23 V
IOUT(Trip)
O
DS(on)
Choose R13 = 750 Ω.
Now, calculate R7 using:
V
IOUT(Trip)
R7 +
* 1 R13 + 983 W +u 1 kW
ǒ Ǔ
0.1 V
In the above calculations, the nominal value for R
of 11 mΩ was used.
DS(on)
Worst case analysis should always be performed to insure that current limit
does not interfere with delivering maximum load.
Design Procedure
2-9
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
An alternate current sensing scheme is to insert a current sense resistor in
series with the drain of Q1. Higher accuracy may be obtained at the expense
of lower efficiency.
2.1.7 Overvoltage Protection
If V exceeds Vref by 15%, a fault latch is set and the output gate drivers are
O
turned off. The latch remains set until VCC (pin 15) goes below the
undervoltage lockout value.
In addition to the standard OVP protection, the LODRV circuit protects the
processor against overvoltages due to a short across the high-side power
MOSFET. External components to sense an overvoltage condition are
required to use this feature. When a shorted high-side MOSFET occurs, the
low-side MOSFET is used as a crowbar. LODRV is pulled low and the low-side
MOSFET is turned on, overriding all control signals inside the controller. The
crowbar action shorts the input supply to ground through the faulted high-side
MOSFET. A fuse in series with V must be added to disconnect the
I
short-circuit.
2.1.8 Power Good
The power-good circuit monitors for an undervoltage condition on V . If V
O
O
drops below 93% of VREF, then the PWRGD output is pulled low. PWRGD is
an open-drain output and needs a pullup resistor.
2.1.9 Bias
Analog BIAS (pin 9), the output of the internal analog bias regulator, is
designed to provide a quiet bias supply for the internal TPS56xx circuitry.
External loads should not be driven by the bias regulator. A 1-µF capacitor,
C22, is connected from BIAS to ANAGND.
2.1.10 Gate Drivers
The gate drivers drive large capacitive loads quickly and efficiently. Figure 2–4
is a block diagram of the drivers. The output stage of the drivers consists of
bipolar and MOS transistors in parallel. The bipolar transistors provide the
majority of the 2-A drive current. The driver outputs get pulled to ground
(during sinking) or to the supply rail (during sourcing) by the MOS transistors.
If the MOS transistors were not in the design, the voltage level on the driver
outputs could only be driven to the saturation voltage level of the bipolar
transistors. This could be a serious limitation, especially if logic-level power
MOSFETs are used in the power stage, resulting in shoot-through current
through the power MOSFETs.
2-10
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
Figure 2–4. Gate Driver Block Diagram
TPS56xx Synchronous-Buck Controller
DRV
V
CC
Highside Driver
8 V Drive
Regulator
C3
L1
BOOT
12 V
V
in
M1
45 Ω
C1
HIGHDR
C4
Level
Shifter/
Predriver
V
REF
M2
5 Ω
L2
LOWDR
VO
V
phase
BOOTLO
LOWDR
Lowside Driver
Adaptive
Deadtime
Control
M3
45 Ω
C2
Predriver
M4
5 Ω
DRVGND
R1
C5
R2
Figure 2–5 gives an I–V sweep of the low-side driver during sinking. The
Rds(on) of the MOS transistors for the sink stage is 5 Ω at T = 125°C and is
J
45 Ω for the source stage. The Rds(on) is lower for the sink stage to provide
alowimpedancepathforthedisplacementcurrentthatflowsthroughtheMiller
capacitance of the power MOSFET when the drain switches. This is especially
important for the low-side driver to keep the low-side MOSFET off when the
high-side MOSFET is turned on.
Design Procedure
2-11
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
Figure 2–5. I–V Characteristic Curve for Low-Side Gate Drivers
Driver Output Voltage – 1 V/div
The high-side gate driver is a bootstrap configuration with an internally
integrated Schottky bootstrap diode. The voltage rating of the BOOT pin to
DRVGND is 30 V. The gate drivers are biased from an internal 8-V drive
regulator to minimize the gate drive power losses that are dissipated inside the
TPS55xx.
The gate drivers have also been optimized to reduce the amount of internal
shoot-through current, which will result when either the low-side or high-side
driver is switching states.
The adaptive deadtime control minimizes the deadtime between conduction
intervals of the power MOSFETs.
The low-side gate driver is not allowed to turn on until the Vphase voltage is
below 2 V; the high-side gate driver is not allowed to turn on until the LOWDR
pin falls below 2 V.
Fast switching and short dead times improve efficiency. There is 100-mA
current limiting within the internal 8-V voltage regulator to protect the regulator
and IC against a short fault on one of the driver pins.
2.1.10.1 Low-Side Driver Controls
The TPS56xx contains two control inputs to control the low-side MOSFET
drive for various applications. They are LODRV (pin 10) and LOHIB (pin 11).
LODRV (pin 10) is an enable input for the low-side MOSFET driver. This pin
is connected to the 5-V input supply for normal synchronous operation.
For added overvoltage protection, external sensing circuitry can be included
to drive the LODRV input low in the event of an overvoltage. Applying a logic
low to LODRV causes the driver for the low-side MOSFET to go to a high state
causing the low-side MOSFET to turn on and act as a crowbar for the output.
This input has precedence over any input present at LOHIB (pin 11); i.e., a low
input to LODRV (pin 10) overrides the inhibit function.
2-12
Download from Www.Somanuals.com. All Manuals Search And Download.
TPS56xx Functions
LOHIB (pin 11) is an inhibit input for the low-side MOSFET driver. This input
has to be logic low before the low-side MOSFET is allowed to be turned on,
i.e., a logic high on LOHIB prevents the low-side MOSFET driver from turning
on the low-side MOSFET. For normal synchronous operation, this pin is
connected to the junction of the high and low-side MOSFETs. This prevents
cross-conduction of the two MOSFETs by constraining the low-side MOSFET
to be OFF unless its drain-to-source voltage is at a low level. Shoot-through
current caused by both MOSFETs being ON simultaneously is actively
prevented. However, if LODRV is low, the low-side MOSFET is turned ON
regardless of the LOHIB input.
2.1.10.2 High-Side Driver
The driver for the high-side MOSFET can be configured either as a
ground-referenced driver or as a floating bootstrap driver. When configured as
a floating driver, the bias voltage to the driver is developed from the DRV
regulator. The maximum voltage that can be applied between BOOT and
DRVGND is 30 V. The driver can be referenced to ground by connecting
BOOTLO to DRVGND, and connecting VCC to BOOT. A 1-µF capacitor, C7, is
connected from BOOT (pin 16) to BOOTLO (pin 18) for bypassing.
2.1.10.3 Grounding
Three separate ground connections enable the user to isolate high-current
grounds from low current logic grounds. The low-current logic ground is called
analog ground. ANAGND (pin 7) and AGND2 (pin2) are the connections for
analog ground. The high-current ground is called power ground and must be
connected to DRVGND (pin 12). The maximum voltage difference between
ANAGND and DRVGND must be limited to less than ±0.2 V.
Refer to the Layout Guidelines section of the TPS56xx data sheet for further
information on grounding.
Design Procedure
2-13
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
2.2 External Component Selection
This section shows the procedure used in designing and selecting the power
stage components to meet the performance parameters shown in Table 1–2
for the example circuit shown in Figure 1–3.
2.2.1 Duty Cycle Estimate
An estimate of the duty cycle is used frequently in the following sections. The
duty cycle, D, is the ratio of the high-side power-switch conduction time to the
period ofoneswitchingcycle. Thedutycycleforacontinuousmodestep-down
converter is given by:
V
) V
O
DS(ON)
I
D +
V
Where:
V
= An estimate of the on-voltage of the power MOSFETs.
DS(ON)
For an initial estimate for V
, use 0.2 V. So, for an output voltage of 3.3 V
DS(ON)
and an input voltage of 5 V, the duty cycle calculates to:
3.3 V ) 0.2 V
D +
+ 0.7
5 V
2.2.2 Input Capacitance
The input capacitance provides a low-impedance voltage source for the power
stage. The ESR, ESL, RMS current rating and capacitance value of the input
capacitance are important parameters in the selection process. The most
stringent requirement is often the RMS current that the capacitance must
handle. An equation for the RMS current seen by the input capacitance for a
buck converter is given by:
2
+ ǸD 1–D I
(
)
I
Cin(RMS)
O
The above equation assumes that the output ripple current is small, that there
is an input inductor, and that its ripple current is small. For V = 5 V and
I
I
= 6 A, we get:
O
2
+ Ǹ0.7 (1–0.7) 6 + 2.7 A RMS
I
Cin(RMS)
The input capacitance for this design uses three 150-µF, 6.3-V Sanyo
POSCAP type electrolytic capacitors in parallel. They are C2, C3, and C4 in
Figure 1–3. The ripple current rating for one of these capacitors is 1.9 A RMS
at 45°C. The total ripple current rating for the input capacitance is
3 × 1.9 = 5.7 A RMS.
2.2.3 Output Filter Design
Unlike fixed-frequency PWM-controlled power supplies, the output filter
design is driven primarily by the need to provide satisfactory output voltage
2-14
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
performance in response to fast load transients encountered when supplying
power to current- and next-generation microprocessors. A secondary
consideration is the switching frequency resulting from the output filter
component values. This section discusses important considerations when
selecting/designing the output filter elements. A detailed analysis of the output
voltage ripple characteristics is also presented, resulting in an expression for
predicting the power supply switching frequency.
2.2.3.1 Output Capacitance
Normally, the output capacitor is selected to limit ripple voltage to the level
required by the specification, but in a hysteretic regulator, such as this one, the
TPS55xx essentially determines the output voltage ripple. The output ripple
is previously chosen to be less than 2% of V and is relatively independent of
O
the output capacitor characteristics. Since output voltage ripple is set by the
comparator hysteresis band, the output capacitor is chosen to provide
satisfactory response to fast load transients.
To show the importance of the output capacitor characteristics, consider the
following: This example circuit is designed for a worst case load step of no load
(0 Amps) to full load (6 Amps) with a slew rate of 30 A/µs. For a transient of
thisslewrate, theoutputfilteralonecontrolstheinitialoutputvoltagedeviation.
Further examination shows that the output filter inductor current changes little
during the load transient. Therefore, for fast load transients, the output
capacitor characteristics dominate the output filter performance. In this
design, the output capacitor’s ESR (equivalent series resistance) and ESL
(equivalent series inductance) are the parameters that are most critical.
To calculate the ESR requirement, assume that all the load transient current
is supplied by the output capacitor. Also assume that the output voltage
change due to the capacitor’s capacitance is much smaller than the voltage
change due to the ESR, and that the capacitor’s ESL is negligible. In most
practical applications, these assumptions are reasonable and they greatly
simplify calculations. The ESR required to limit output voltage change to
100 mV due to a 6 amp load step is:
DV
O
100 mV
6 A
ESR v
+
+ 16.7 mW
DI
O
The required level of ESR requires a large amount of capacitance. For this
design, four Sanyo POSCAP type electrolytic capacitors connected in parallel
are selected. These capacitors are a good compromise between
performance, cost, and board area requirements. The particular part used is
an 150-µF, 4-V capacitor with a specified maximum ESR of 45 mΩ, giving a
total maximum ESR of 11.25 mΩ. These capacitors are C10, C11, C12, and
C13 in Figure 1–3. For good design practice, C14, a 10-µF ceramic capacitor,
is placed in parallel with C10–C13. Ceramic capacitors are very effective for
suppressing high frequency switching spikes and reducing the effects of the
ESL of C10–C13.
To summarize, the output capacitance must be selected to provide a
sufficiently low ESR. The capacitor(s) must have an adequate voltage rating
Design Procedure
2-15
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
for the particular application. In addition, the capacitor(s) must have an ample
ripple current rating to handle the applied ripple current. This ripple current is
dependent on the ripple current in the output inductance that is calculated in
the next section. The RMS current in the output capacitance is calculated as
follows for 3.3 V output:
Ǹ
3
6
I
+ DI
+ DI 0.289 + 7.3 A 0.289 + 2.1 A
CRMS
L
L
RMS
Where ∆I is the peak-to-peak ripple current in the output inductor.
L
The ripple current rating for one of these capacitors is 1.7 A RMS at 45°C. The
total ripple current rating for the output capacitance is 4 × 1.7 = 6.8 A RMS as-
suming that each cap shares equally. Poor layout may cause one or more of
the caps to take an unequal share.
2.2.3.2 Output Inductance
The output filter inductance is the next quantity to be determined. Like the out-
put capacitance selection, the primary consideration is providing satisfactory
response to a fast load transient.
The inductance affects the output voltage response to transient loads by
governing the rate at which its current can increase (or decrease). For
example, during a load transient where the output current increases from no
load to full load, the output capacitor supplies all of the output current until the
inductor current has time to increase to full load. So, a relatively small
inductance is desired.
On the other hand, the inductance also plays a part in the power supply
switchingfrequency, because the inductance limits how fast the outputvoltage
traverses through the hysteresis band. As the inductance decreases, the
output voltage changes faster, giving rise to higher switching frequencies.
Therefore, the inductor value is fairly critical and should be stable over the
expected load and temperature range. Care should be exercised in the
inductance selection because design requirements for hysteretic regulators
are different from PWM controlled converters. Recall that for a fixed frequency
PWM controlled buck converter, for a given operating point, the output
inductance governs the peak-to-peak amplitude and the slope (di/dt) of the
inductor current. In a hysteretic regulator, the ripple current is set by the output
voltage ripple setting and the output capacitance impedance.
Other important factors to be considered when designing or selecting the
inductor are current capability, allowable operating frequency, and dc
resistance.
After choosing an initial inductance value, the power supply switching
frequency must be estimated to insure that it is within the desired range. The
switching frequency for the example design given in Figure 1–3 is nominally
135 kHz for 5-V input and a 3.3-V output. A detailed analysis of the switching
frequency and an equation to predict it is given later.
To calculate the maximum inductance allowed for a given response time, load
step, and operating point, the following simple relationship is used:
2-16
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
I
V
TRAN
Dt
L
V + L
å L v
Dt
L
I
TRAN
Where:
V = the voltage applied across the output inductor,
L
I
= the magnitude of the load step, and
TRAN
∆t = the desired response time.
For a load step from light load to heavy load, the voltage applied across the
inductor can be assumed to be V – V . This also assumes that the duty cycle
I
O
is 100% as the output voltage is corrected. Alternatively, for a load step from
heavy load to light load, the voltage across the inductor can be assumed to be
V . This assumes that the duty cycle is 0% as the output voltage is corrected.
O
For the example circuit described here, the condition which gives the smallest
inductor value is a load step from light load to heavy load for a 3.3-V output.
This is because the voltage applied to the output inductor is the lowest, i.e.,
the input voltage minus the output voltage. For this case, allowing 5 µs for the
inductor current to change, an inductance value is calculated as follows:
V
5–3.3
6
L
–6
L v
Dt +
5 10 + 1.4 mH
I
TRAN
For convenience, a 1.5-µH inductor was designed for this circuit.
Figures 3–10, 3–19, 3–28, and 3–37 illustrate satisfactory transient load per-
formance for the output filter values selected in this and the previous sections.
2.2.4 Switching Frequency Analysis
After the elements of the output filter are determined, the power supply
switching frequency must be estimated. If the estimated switching frequency
is too high, the switching losses in the power MOSFETs will be high, resulting
in less than optimum efficiency. If the estimated switching frequency is too low,
the inductor value may be too high, resulting in unsatisfactory transient
response. A switching frequency outside the desired range should be
corrected by changing either the output ripple setting, the output capacitance,
or the output inductance.
To accurately predict the switching frequency of a hysteretic regulator, the
output voltage ripple must be investigated. This should be expected, since the
power supply switching instants are based upon the state of the output
voltage. Asimpleandaccuratemethodofdeterminingtheswitchingfrequency
is described below.
2.2.4.1 Output Ripple
The three elements of the capacitor that contribute to ripple are ESR, ESL, and
capacitance. Assume that all three elements are in series and there are no
other parasitic components to consider. Figure 2–6 shows the voltage
waveforms across each component of the output capacitor and the
corresponding equations.
Design Procedure
2-17
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
Figure 2–6. Output Ripple Voltage Detail
(a)
(b)
Current waveform through output capacitor
I
AC
Voltage waveform across ideal capacitor with initial value at
beginning of high-side MOSFET on-time
∆Ι
t
(a)
(b)
High-side MOSFET ON
2
High-side MOSFET OFF
V
C
2
DI t
DI t
DI t
DI t
V
+
–
V
+
–
C
C
2 Co D Ts 2 Co
2 Co 2 Co (1–D) Ts
t
t
V
ESR
(c)
Voltage waveform across ESR
High-side MOSFET ON
High-side MOSFET OFF
(c)
DI t
DI
2
DI t
DI
V
+ ESR
–
V
+ ESR
–
ESR
D Ts
ESR
V
2
(1–D) Ts
V
ESL
(d)
Voltage waveform across ESL
t
t
(d)
t
High-side MOSFET ON
High-side MOSFET OFF
V
del
ripple
ESL DI
ESL DI
V
+
+ –
ESL
D Ts
ESL
(1–D) Ts
Vp–p
yst
H
(e)
Composite output voltage ripple waveform
t
del
(e)
t
V
+ V ) V
) V
ripple
C
ESR
ESL
t
off
on
T
s
Figures 3–6, 3–15, 3–24, and 3–33 show the phase voltage (voltage at
junction of high-side MOSFET with low-side MOSFET), and output voltage
ripple waveforms for the example circuit of Figure 1–3. The output voltage
waveform is slightly different from the theoretical waveform of Figure 2–6(e)
due to the smoothing effect of the 10-µF ceramic capacitor in parallel with the
electrolytic capacitors.
2.2.4.2 Switching Frequency Equation
Assume that the input and output voltage ripple magnitudes are relatively
negligible compared to the dc component. Also assume that the time constant
L/(R +R ),whereListheoutputinductance,R istheinductorresistance
DS(on)
L
L
and R
is the on-state resistance of the high-side MOSFET(s), is high in
DS(on)
comparisonwiththeswitchingperiod. Assumethebodydiodeconductiontime
and switching transition time are much smaller than the switching period.
These assumptions are reasonable for low voltage ripple and high efficiency
regulators. In such a case the output inductor current can be modeled as
the sum of the dc component, which is equal to the output current Io, and
the ac linear ramp component, which flows through the output capacitor
[Figure 2–6(a)].
The numbered equations in this section are used to derive the switching
frequency equation.
2-18
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
Peak to peak value of the inductor current ∆I is given by the following equation:
V – Io ǒR
) R Ǔ–V
I
DS(on)
L
L
O
(1)
DI +
D Ts
Where:
V =the input voltage
I
V = the output voltage
O
Ts = the switching period
) Io ǒR
) R Ǔ
V
O
DS(on)
L
D +
is the duty cycle which is defined as :
V
I
t
ON
Ts
D +
and t
is the on time of the high-side MOSFET.
ON
Referring to Figure 2–6 (e), the output voltage ripple, V , is higher than the
p-p
hysteresis band, Hyst, because of the delays, t . Assumethatdelaysforboth
del
switching instants are equal for simplicity. The ideal capacitor voltage
component has the same initial value during switching instants t
and
ON
(Ts–t ) (see Figure 2–6 (b)). In this case the voltage V
is:
ON
p-p
ESL
L
(2)
V
+
V ) DI ESR
p-p
I
On the other hand, the hysteresis band is equal to the difference between the
peak-peak values of the V ripple, v
at the moments t
– t
del
and
O
ripple,
ON
t
– t
OFF del
(3)
ǒt
Ǔ* V ǒt
Ǔ
Hyst + V
* t
* t
ripple ON
del
ripple OFF
del
After substituting equation (1) into equations for v , v
and v
ESL
C
ESR
(Figure 2–6) and using equations (2) and (3), the following equation for the
switching frequency, f , can be derived:
S
ǒV – V Ǔ ǒ
ńCoǓ
V
ESR–t
O
I
O
del
(4)
f +
s
V ǒV ESR t ) Hyst L–ESL V Ǔ
I
I
del
I
Equation(4)showsthattheswitchingfrequencystronglydependsonESRand
ESL. It is important that ESL meet the following condition:
ESL t ǒESR t ) Hyst L D ń VoutǓ
del
If it does not, the voltage step across the ESL during switching exceeds the
hysteresis window, and the switching frequency becomes too high and
uncontrollable.
The switching frequency does not depend on the load current in equation (4),
becauseasynchronousregulatorhasthesametwostagesofoperationduring
the switching period over the load current range, including no load condition.
There is no discontinuous mode operation at very light loads. In reality there
is a weak dependence of the switching frequency on the load current because
Design Procedure
2-19
Download from Www.Somanuals.com. All Manuals Search And Download.
External Component Selection
of power losses and additional voltage drops through non-ideal components.
Equation (4) should be sufficiently accurate for the first frequency estimate at
the beginning of a design.
2.2.5 Power MOSFET Selection
The TPS56xx is designed to drive N-channel power MOSFETs in a
synchronousrectifierconfiguration. TheMOSFETchosenforthisdesignisthe
Siliconix Si4410DY. This device is chosen for its low r
drain-to-source breakdown voltage rating of 30 V.
of 13.5 mΩ and
DS(on)
PowerdissipationfortheswitchingMOSFETs, whichincludesbothconduction
and switching losses, is given by:
2
D ) ǒ0.5 V I
Ǔ
+ ǒI
+ ǒI
Ǔ
P
r
t
f
sw
D (Q1)
DS(on)
i
O
r)f
O
2
O
) ǒ0.5 V I
Ǔ
1–D)Ǔ
(
P
r
t
f
sw
D (Q2)
DS(on)
i
O
r)f
An example MOSFET power dissipation calculation for Q1 and Q3 is shown
below with the following assumptions:
The total switching time, t = 100 ns,
r+f
An r
high temperature adjustment factor = 1.4,
DS (on)
A 60°C maximum ambient temperature,
V = 5.0 V, V = 3.3 V, and I = 6 A then :
I
O
O
2
–9
3
3
( )
(
)
P
P
+ 6 0.0135 1.4 0.7 ) 0.5 5 6 100 10 135 10
D (Q1)
D (Q2)
+ 0.48 ) 0.20 + 0.68 W
2
–9
( )
(
)
+ 6 0.0135 1.4 0.3 ) 0.5 5 6 100 10 135 10
+ 0.20 ) 0.20 + 0.40 W
The thermal impedance of these devices, R
copper and a one-inch-square pattern, thus:
= 90°C/W for FR-4 with 2-oz.
θJA
+ T ) ǒR
Ǔ + 60 ) 50 0.90 + 94°C
(
)
T
T
P
J (Q1)
J (Q2)
A
qJA
+ T ) ǒR
DǓ + 60 ) 50 1.40 + 80°C
(
)
P
A
qJA
D
It is good design practice to check power dissipation at the extreme limits of
input voltage to find the worst case.
2-20
Download from Www.Somanuals.com. All Manuals Search And Download.
Chapter 3
Test Results
This chapter shows the test setups used, and the test results obtained, in
designing the SLVP111–114 EVMS.
Topic
Page
3.1 Test Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–2
3.2 Test Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–5
3.2 Test Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
3-1
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Summary
3.1 Test Summary
The detailed test results and waveforms are presented in Figures 3–2 to 3–10
for the SLVP111, Figures 3–11 to 3–19 for the SLVP112, Figures 3–20 to 3–28
for the SLVP113 and Figures 3–29 to 3–37 for the SLVP114. The following are
summarized results.
3.1.1 Static Line and Load Regulation
The precise reference voltage regulator implemented in the TPS56XX
controller using both positive and negative remote sense pins provides
excellent regulation characteristics. The load regulation from no load to
6 amps load current does not exceed 0.26%. The line regulation is less than
0.045% for the input voltage range from 4.5 V to 5.5 V. Line and load regulation
is shown in Figures 3–2, 3–11, 3–20, and 3–29. The set point tolerance is less
than 0.333%. The measured results have a good margin for high volume
production.
3.1.2 Output Voltage Ripple
The output voltage peak to peak ripple is less than ±1%. This is a typical value
but it can be optimized for lower ripple applications. Measured output ripple
waveforms are shown in Figures 3–6, 3–15, 3–24, and 3–33.
The output filter for this EVM design is optimized for fast transient response
due to the high slew-rate load current transitions. Therefore, the output filter
is not optimized for low ripple and has a moderate amount of output ripple.
3.1.3 Efficiency and Power Losses
Efficiency and power losses for 5 V input voltage and maximum output current
6 A are presented in the following table:
Evaluation Board
SLVP111, 3.3V
SLVP112, 2.5V
SLVP113, 1.8V
SLVP114, 1.5V
Efficiency, %
Power Losses, W
90
2.2
86.4
83.2
79.8
2.37
2.18
2.27
Efficiency graphs versus load and at different line voltages are shown in
Figures 3–3, 3–12, 3–21, and 3–30.
Low power loss in each component decreases their temperature rise and
improves long term reliability. The EVMs do not require forced air cooling at
room temperature with good margin.
3.1.4 Output Start-Up and Overshoot
All possible start-up combinations are tested and waveforms are presented.
Output voltage rise time does not depend on the load current and ramps up
3-2
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Summary
in a linear fashion. There is no discernable overshoot in the waveforms. In this
application, output voltage rise time is set to approximately 9 mS with an
external capacitor. Although the EVMs have been tested with a very short Vcc
rise time, (see Figures 3–26 and 3–35) it is recommended to keep the rise time
of Vcc longer than 3mS as shown in Figures 3–8 and 3–17.
3.1.5 Frequency Variation
The switching frequency for a hysteretic controller depends on the input and
output voltages and the output filter characteristics. It has approximately the
same frequency variation as constant OFF time controllers. The precise
equationforswitchingfrequency, confirmedbyexperiment, ispresentedinTI’s
application report Designing Fast Response Synchronous Buck Converters
Using the TPS5210 and in the paper, represented at HFPC–98 A Fast,
Efficient Synchronous-Buck Controller for Microprocessor Power Supplies
The frequency variation over all input voltage and output current combinations
is presented in the following table.
Switching frequency graphs versus load at different line voltages are shown
in Figures 3–5, 3–14, 3–23, and 3–32.
Frequency Variation,
Evaluation Board
(kHz)
SLVP111, 3.3 V
SLVP112, 2.5 V
SLVP113, 1.8 V
SLVP114, 1.5 V
97–175
189–247
262–312
296–378
3.1.6 Load Current Transient Response
The hysteretic controller has excellent dynamic characteristics (see
Figures 3–10, 3–19, 3–28 and 3–37) and does not require any feedback
compensation circuitry. These figures show that the controller responds to a
load transient within the same switching cycle as when the transient occurs.
In addition, it is necessary to optimize the output filter to meet high slew-rate
load current transient requirements thus saving money and board space by
using only the fewest expensive bulk capacitors. Special attention should also
be paid to high frequency decoupling within the output filter to control the initial
transient voltage deviation because it does not depend on controller
characteristics.
3.1.7 Features
The EVMs have all the features, which are described in detail in the datasheet
for TPS56XX controller.
The features include undervoltage lockout for both 12 V and 5 V input, inhibit
signal, power good signal, overvoltage protection, slow start, remote sense,
and overcurrent protection. Overcurrent limit is set to an 8 amp – 10 amp level.
Test Results
3-3
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Summary
3.1.8 Conclusion
The test results of the SLVP111/112/113/114 EVMs demonstrate the
advantages of TPS56xx controllers to meet stringent supply requirements to
power supplies, especially for powering DSPs and microprocessors. The
power system designer has a good solution to optimize system for his
particular application. Detailed information how to design a dc-dc converter by
using TPS56xx controller is represented in TI’s User’s Guide Synchronous
Buck Converter Design Using TPS56xx Controllers in SLVP10x EVMs,
Literature Number SLVU007 or Designing Fast Response Synchronous Buck
Regulators Using the TPS5210, Literature Number SLVA044.
3-4
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Setup
3.2 Test Setup
Follow these steps for initial power up of the SLVP112:
1) Connect an electronic load from Vout to PwrGND (J1-15, -16,
-17, -18 to J1-11, -12, -13, -14) adjusted to draw approximately
1 A at 2.5 V. The exact current is not critical; any nominal current
is sufficient. A fixed resistor can also be used in place of the
electronic load. The output current drawn by the resistor is
2.5 V
I
+
amps where R is the value of the load resistor. The
O
R
2
2.5
R
2 Watts.
resistor power rating, P should be at least
R
Connect the sense lines from the load to VsenseH and VsenseL
(J1-4 and J1-2).
2) Connect a 12-V lab power supply to the 12-V input (J1-7
referenced to PwrGND, J1-8) of the SLVP112. A current limit set
for 20 mA should be adequate for the controller’s power
requirements.
3) Connect another lab power supply to the 5 V dc input of the
SLVP112 (J1-5, -6 referenced to Return, J1-9, -10). Verify that
the current limit is set for at least 2 A and that it is set to 0 V.
4) Turn on the 12-V lab supply. Turn on the 5-V power supply and
ramp the input voltage up to 5 V. Once proper operation is
verified, this order is not important.
5) Verify that the SLVP112 output voltage (measured at the module
output pins) is 2.5 V ± 0.025 V.
6) For subsequent testing, ensure the lab supply output current
capacity and current limit are at least 7 A so that the SLVP112
can be operated at maximum load of 6 A.
7) For initial power up of the other converters, replace any
reference to 2.5 V in the above discussion with a reference to the
appropriate output voltage.
8) Refer to Chapter 3 for selected typical waveforms and operating
conditions for verification of proper module operation.
Figure 3–1 shows the SLVP112 test setup.
Test Results
3-5
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Setup
Figure 3–1. Test Setup
5V Power Supply
+
–
– +
–
+
Load
12-V Power Supply
3-6
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
3.3 Test Results
Figures 3–2 to 3–102 show test results for the SLVP111.
Figure 3–2. SLVP111 Measured Load Regulation
SLVP111 MEASURED LOAD REGULATION
3.305
3.3
3.295
Vin = 5 V
Vin = 5.5 V
3.29
Vin = 4.5 V
3.285
0
1
2
3
4
5
6
I
O
– A
Figure 3–3. SLVP111Measured Efficiency
SLVP111 MEASURED EFFICIENCY
94
92
90
88
86
84
82
Vin = 5 V
Vin = 4.5 V
Vin = 5.5 V
1
2
3
4
5
6
I
O
– A
Test Results
3-7
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–4. SLVP111Measured Power Dissipation
SLVP111 MEASURED POWER DISSIPATION
2.5
2
1.5
1
Vin = 4.5 V
Vin = 5 V
Vin = 5.5 V
0.5
0
0
1
2
3
– A
4
5
6
I
O
Figure 3–5. SLVP111Measured Switching Frequency
SLVP111 MEASURED SWITCHING FREQUENCY
200
175
150
Vin = 5.5 V
Vin = 5 V
125
100
Vin = 4.5 V
75
50
0
1
2
3
4
5
6
I
O
– A
3-8
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–6. SLVP111 Measured Switching Waveforms
C3 Pk–Pk
50.8 mV
C3 Frequency
130.088 kHz
Low Signal
Amplitude
V
O
20 mV/div
V
Q2
DS
2 V/div
C4 Max
5.20 V
C4 + Duty
70.4%
2.5 µs/div
Figure 3–7. SLVP111Measured Start-Up (INHIBIT) Waveforms
V
C3 Pk–Pk
3.36 V
O
2 V/div
C3 Rise
7.500 ms
Low Signal
Amplitude
INHIBIT
1 V/div
C3 + Over
2.5%
2.5 ms/div
Test Results
3-9
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–8. SLVP111 Measured Start-Up (V ) Waveforms
CC
C3 Pk–Pk
3.36 V
V
O
2 V/div
C3 Rise
7.300 ms
Low Signal
Amplitude
V
(12 V)
CC
5 V/div
C3 + Over
2.5%
UVLO
Threshold
2.5 ms/div
Figure 3–9. SLVP111Measured Start-Up (V ) Waveforms
IN
C3 Pk–Pk
3.52 V
V
O
2 V/div
C3 Rise
8.540 ms
Low Signal
Amplitude
V
(5 V)
IN
1 V/div
C3 + Over
2.4%
2.5 ms/div
3-10
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–10. SLVP111 Measured Load Transient Waveforms
V
O
100 mV/div
C3 Pk–Pk
208 mV
C2 High
6.5 V
I
O
5 A/div
6.5 A
2.5 µs/div
Figure 3–11. SLVP112 Measured Load Regulation
SLVP112 MEASURED LOAD REGULATION
2.51
Vin = 5 V
Vin = 5.5 V
Vin = 4.5 V
2.505
2.5
2.495
0
1
2
3
4
5
6
I
O
– A
Test Results
3-11
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–12. SLVP112 Measured Efficiency
SLVP111 MEASURED EFFICIENCY
90
88
86
84
82
80
78
Vin = 4.5 V
Vin = 5 V
Vin = 5.5 V
1
2
3
4
5
6
I
O
– A
Figure 3–13. SLVP112 Measured Power Dissipation
SLVP111 MEASURED POWER DISSIPATION
2.5
2
1.5
1
Vin = 5.5 V
Vin = 5 V
Vin = 4.5 V
0.5
0
0
1
2
3
4
5
6
I
O
– A
3-12
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–14. SLVP112 Measured Switching Frequency
SLVP112 MEASURED SWITCHING FREQUENCY
300
275
250
Vin = 5.5 V
Vin = 5 V
225
Vin = 4.5 V
200
175
150
0
1
2
3
4
5
6
I
O
– A
Figure 3–15. SLVP112 Measured Switching Waveforms
C3 Pk–Pk
43.2 mV
C3 Frequency
218.800 kHz
Low Signal
Amplitude
V
O
V
Q2
DS
2 V/div
20 mV/div
C4 Max
5.72 V
C3 + Duty
54.2%
2.5 µs/div
Test Results
3-13
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–16. SLVP112 Measured Start-Up (INHIBIT) Waveforms
V
O
1 V/div
C3 Pk–Pk
2.64 V
C3 Rise
7.885 ms
INHIBIT
1 V/div
C3 + Over
3.2%
2.5 ms/div
Figure 3–17. SLVP112 Measured Start-Up (V ) Waveforms
CC
C3 Pk–Pk
2.56 V
V
O
2 V/div
C3 Rise
7.995 ms
V
(12 V)
CC
5 V/div
C3 + Over
3.3%
UVLO
Threshold
2.5 ms/div
3-14
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–18. SLVP112 Measured Start-Up (V ) Waveforms
IN
C3 Pk–Pk
2.60 V
V
O
1 V/div
C3 Rise
7.635 ms
V
(5 V)
IN
1 V/div
C3 + Over
3.2%
2.5 ms/div
Figure 3–19. SLVP112 Measured Load Transient Waveforms
V
O
100 mV/div
C3 Pk–Pk
200 mV
C2 High
7.00 V
I
O
2 A/div
5 A
25 µs/div
Test Results
3-15
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–20. SLVP113 Measured Load Regulation
SLVP113 MEASURED LOAD REGULATION
1.805
Vin = 5 V
Vin = 5.5 V
1.8025
Vin = 4.5 V
1.8
1.7975
1.795
0
1
2
3
4
5
6
I
O
– A
Figure 3–21. SLVP113 Measured Efficiency
SLVP113 MEASURED EFFICIENCY
Vin = 4.5 V
88
86
84
82
80
78
Vin = 5 V
Vin = 5.5 V
76
74
72
1
2
3
4
5
6
I
O
– A
3-16
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–22. SLVP113 Measured Power Dissipation
SLVP113 MEASURED POWER DISSIPATION
2.5
2
1.5
1
Vin = 5.5 V
Vin = 4.5 V
Vin = 5 V
0.5
0
0
1
2
3
4
5
6
I
O
– A
Figure 3–23. SLVP113 Measured Switching Frequency
SLVP113 MEASURED SWITCHING FREQUENCY
350
325
300
Vin = 5.5 V
Vin = 5 V
275
Vin = 4.5 V
250
225
200
175
150
0
1
2
3
4
5
6
I
O
– A
Test Results
3-17
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–24. SLVP113 Measured Switching Waveforms
C3 Pk–Pk
34.8 mV
C3 Frequency
285.52 kHz
Low Signal
Amplitude
V
O
20 mV/div
V
Q2
DS
2 V/div
C5 Max
5.80 V
C4 + Duty
40.4%
1 µs/div
Figure 3–25. SLVP113 Measured Start-Up (INHIBIT) Waveforms
C3 Pk–Pk
1.88 V
V
O
1 V/div
C3 Rise
7.360 ms
Low Signal
Amplitude
INHIBIT
1 V/div
C3 + Over
2.3%
2.5 ms/div
3-18
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–26. SLVP113 Measured Start-Up (V ) Waveforms
CC
C3 Pk–Pk
1.84 V
V
O
2 V/div
C3 Rise
7.195 ms
Low Signal
Amplitude
V
(12 V)
CC
5 V/div
C3 + Over
2.3%
2.5 ms/div
Figure 3–27. SLVP113 Measured Start-Up (V ) Waveforms
IN
C3 Pk–Pk
1.88 V
V
O
1 V/div
C3 Rise
8.300 ms
Low Signal
Amplitude
V
(5 V)
IN
1 V/div
C3 + Over
2.2%
2.5 ms/div
Test Results
3-19
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–28. SLVP113 Measured Load Transient Waveforms
V
O
C3 Pk–Pk
112 mV
100 mV/div
C2 High
3.64 V
I
O
5 A/div
3.6 A
25 µs/div
Figure 3–29. SLVP114 Measured Load Regulation
SLVP114 MEASURED LOAD REGULATION
1.5
1.499
1.498
1.497
1.496
1.495
1.494
1.493
1.492
Vin = 5 V
Vin = 5.5 V
Vin = 4.5 V
1.491
1.49
0
1
2
3
4
5
6
I
O
– A
3-20
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–30. SLVP114 Measured Efficiency
SLVP114 MEASURED EFFICIENCY
85
Vin = 4.5 V
Vin = 5 V
83
81
79
77
75
73
71
69
Vin = 5.5 V
67
65
1
2
3
4
5
6
I
O
– A
Figure 3–31. SLVP114 Measured Power Dissipation
SLVP114 MEASURED POWER DISSIPATION
2.5
2
Vin = 5.5 V
1.5
1
Vin = 5 V
Vin = 4.5 V
0.5
0
0
1
2
3
4
5
6
I
O
– A
Test Results
3-21
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–32. SLVP114 Measured Switching Frequency
SLVP114 MEASURED SWITCHING FREQUENCY
400
375
Vin = 5.5 V
350
Vin = 5 V
325
300
Vin = 4.5 V
275
250
225
200
0
1
2
3
4
5
6
I
O
– A
Figure 3–33. SLVP114 Measured Switching Waveforms
C3 Pk–Pk
30.8 mV
C3 Frequency
337.82 kHz
Low Signal
Amplitude
V
O
V
Q2
DS
2 V/div
20 mV/div
C4 Max
7.44 V
C4 + Duty
36.2%
1 µs/div
3-22
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–34. SLVP114 Measured Start-Up (INHIBIT) Waveforms
C3 Pk–Pk
1.56 V
V
O
1 V/div
C3 Rise
6.990 ms
Low Signal
Amplitude
INHIBIT
1 V/div
C3 + Over
2.8%
2.5 ms/div
Figure 3–35. SLVP114 Measured Start-Up (V ) Waveforms
CC
C3 Pk–Pk
1.56 V
V
O
1 V/div
C3 Rise
7.090 ms
Low Signal
Amplitude
V
(12 V)
CC
5 V/div
C3 + Over
2.8%
2.5 ms/div
Test Results
3-23
Download from Www.Somanuals.com. All Manuals Search And Download.
Test Results
Figure 3–36. SLVP114 Measured Start-Up (V ) Waveforms
IN
C3 Pk–Pk
1.56 V
V
O
1 V/div
C3 Rise
7.07 ms
Low Signal
Amplitude
V
(5 V)
IN
1 V/div
C3 + Over
2.7%
2.5 ms/div
Figure 3–37. SLVP114 Measured Load Transient Waveforms
V
O
50 mV/div
C3 Pk–Pk
108 mV
C2 High
3.08 V
I
O
2 A/div
3 A
25 µs/div
3-24
Download from Www.Somanuals.com. All Manuals Search And Download.
|