TPS548B28 [TI]

具有遥感功能和断续限流功能的 2.7V 至 16V 输入 20A 同步降压转换器;
TPS548B28
型号: TPS548B28
厂家: TEXAS INSTRUMENTS    TEXAS INSTRUMENTS
描述:

具有遥感功能和断续限流功能的 2.7V 至 16V 输入 20A 同步降压转换器

转换器
文件: 总49页 (文件大小:2551K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
TPS548B28  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
具有遥感功能、3V LDO 和断续电流限制功能的  
TPS548B28 2.7V 16V 输入、20A 同步降压转换器  
1 特性  
2 应用  
• 输入范围4V 16V电流高20A无外部偏  
• 输入范围2.7V 16V电流高20A外部偏  
压范围3.13V 3.6V  
机架式服务器和刀片式服务器  
硬件加速卡和插件卡  
数据中心交换机  
PC  
• 输出电压范围0.6V 5.5V  
• 集7.7mΩ2.4mΩMOSFET20A 持续  
输出电流  
3 说明  
TPS548B28 器件是一款具有自适应导通时间 D-CAP3  
控制模式的高效率、小尺寸同步降压转换器。该器件不  
需要外部补偿因此易于使用并且仅需要很少的外部元  
件。该器件非常适合空间受限的数据中心应用。  
D-CAP3控制模式下可提供超快负载阶跃响应  
• 支持所有陶瓷输出电容器  
40°C +125°C 结温下实现差分遥感VREF  
0.6V ±1%  
• 可FCCM 或自动跳Eco-mode可实现较高的  
轻负载效率  
• 通RTRIP 实现可编程电流限制  
• 引脚可选开关频率600kHz800kHz1MHz  
• 可实现高输出精度的差分遥感功能  
• 可编程软启动时间  
• 外部基准输入用于跟踪  
• 预偏置启动功能  
TPS548B28 件具有差分遥感功能和高性能集成  
MOSFET整个工作结温范围具有高精度 (±1%)  
0.6V 电压基准。该器件具有快速负载瞬态响应、精确  
负载调节和线路调节、跳跃模式或 FCCM 运行模式以  
及可编程软启动功能。  
TPS548B28 是一款无铅器件完全符合 RoHS 标准,  
无需豁免。  
封装信息  
封装(1)  
• 开漏电源正常状态输出  
• 在发OC UV 故障时进入断续模式在发生  
OV 故障时进入闭锁模式  
封装尺寸标称值)  
器件型号  
RWWVQFN-HR,  
21)  
TPS548B28  
4.00mm × 3.00mm  
4mm × 3mm 21 QFN 封装  
• 引脚15A TPS548A28 兼容  
• 完全符RoHS 标准无需豁免  
(1) 如需了解所有可用封装请参阅数据表末尾的可订购产品附  
录。  
100  
95  
90  
85  
80  
75  
VIN  
10  
21  
1
VIN  
VIN  
BOOT  
VOUT  
8
20  
EN  
SW  
TPS548B28  
19  
9
VCC  
Vosns+  
VOUT = 0.9 V  
VOUT = 1.0 V  
7
FB  
70  
PGOOD  
MODE  
TRIP  
65  
60  
55  
50  
VOUT = 1.2 V  
VOUT = 1.8 V  
VOUT = 2.5 V  
VOUT = 3.3 V  
VOUT = 5.0 V  
4
Vosns-  
6
5
VSNS-  
3
SS/  
2
REFIN  
AGND  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
PGND  
Net-tie  
效率图12VIN800 kHz,  
DCMVCC  
简化原理图  
本文档旨在为方便起见提供有TI 产品中文版本的信息以确认产品的概要。有关适用的官方英文版本的最新信息请访问  
www.ti.com其内容始终优先。TI 不保证翻译的准确性和有效性。在实际设计之前请务必参考最新版本的英文版本。  
English Data Sheet: SNVSBC5  
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
Table of Contents  
7.4 Device Functional Modes..........................................23  
8 Application and Implementation..................................26  
8.1 Application Information............................................. 26  
8.2 Typical Application.................................................... 26  
8.3 Power Supply Recommendations.............................37  
8.4 Layout....................................................................... 37  
9 Device and Documentation Support............................41  
9.1 Documentation Support............................................ 41  
9.2 接收文档更新通知..................................................... 41  
9.3 支持资源....................................................................41  
9.4 Trademarks...............................................................41  
9.5 Electrostatic Discharge Caution................................41  
9.6 术语表....................................................................... 41  
10 Mechanical, Packaging, and Orderable  
1 特性................................................................................... 1  
2 应用................................................................................... 1  
3 说明................................................................................... 1  
4 Revision History.............................................................. 2  
5 Pin Configuration and Functions...................................3  
6 Specifications.................................................................. 5  
6.1 Absolute Maximum Ratings........................................ 5  
6.2 ESD Ratings............................................................... 5  
6.3 Recommended Operating Conditions.........................5  
6.4 Thermal Information....................................................6  
6.5 Electrical Characteristics.............................................6  
6.6 Typical Characteristics..............................................10  
7 Detailed Description......................................................13  
7.1 Overview...................................................................13  
7.2 Functional Block Diagram.........................................13  
7.3 Feature Description...................................................14  
Information.................................................................... 42  
4 Revision History  
以前版本的页码可能与当前版本的页码不同  
Changes from Revision * (December 2020) to Revision A (December 2022)  
Page  
Changed transient time 10 ns to 20 ns in Absolute Maximum Ratings.............................................................. 5  
Updated ESD Ratings (CDM) to current standards............................................................................................5  
Corrected Overvoltage and Undervoltage Protection procedure......................................................................22  
Copyright © 2022 Texas Instruments Incorporated  
2
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
5 Pin Configuration and Functions  
VIN  
21  
SW  
20  
SW  
20  
VIN  
21  
BOOT  
1
19 VCC  
VCC  
19  
1
BOOT  
AGND  
TRIP  
2
3
4
5
6
7
8
18 PGND  
17 PGND  
16 PGND  
15 PGND  
14 PGND  
13 PGND  
12 PGND  
PGND  
PGND  
PGND  
PGND  
PGND  
PGND  
PGND  
18  
17  
16  
15  
14  
13  
12  
2
3
4
5
6
7
8
AGND  
TRIP  
MODE  
SS/REFIN  
VSNS-  
FB  
MODE  
SS/REFIN  
VSNS-  
FB  
EN  
EN  
PGOOD  
9
11 PGND  
PGND  
11  
9
PGOOD  
10  
10  
VIN  
VIN  
5-1. RWW Package 21-Pin VQFN-HR Top View  
5-2. RWW Package 21-Pin VQFN-HR Bottom  
View  
5-1. Pin Functions  
NO.  
NAME  
BOOT  
AGND  
TRIP  
I/O(1)  
I/O  
G
DESCRIPTION  
Supply rail for the high-side gate driver (boost terminal). Connect the bootstrap capacitor from  
this pin to SW node.  
1
2
3
Ground pin. Reference point for the internal control circuits  
Current limit setting pin. Connect a resistor to AGND to set the current limit trip point. ±1%  
tolerance resistor is highly recommended. See 7.3.9 for details on OCL setting.  
I/O  
The MODE pin sets the forced continuous-conduction mode (FCCM) or skip-mode operation. It  
also selects the operating frequency by connecting a resistor from MODE pin to AGND pin.  
±1% tolerance resistor is recommended. See 7-1 for details.  
4
5
MODE  
I
Dual-function pin. Soft-start function: Connecting a capacitor to VSNSpin programs soft-start  
time. Minimum soft-start time (1.5 ms) is fixed internally. A minimum 1-nF capacitor is required  
for this pin to avoid overshoot during the charge of soft-start capacitor. REFIN function: The  
device always looks at the voltage on this SS/REFIN pin as the reference for the control loop.  
The internal reference voltage can be overridden by an external DC voltage source on this pin  
for tracking application.  
SS/REFIN  
I/O  
The return connection for a remote voltage sensing configuration. It is also used as ground for  
the internal reference. Short to AGND for single-end sense configuration  
6
7
I
I
VSNS–  
Output voltage feedback input. A resistor divider from the VOUT to VSNS(tapped to FB pin)  
sets the output voltage.  
FB  
Enable pin. The enable pin turns the DC/DC switching converter on or off. Floating EN pin  
before start-up disables the converter. The recommended operating condition for EN pin is  
maximum 5.5-V. Do not connect EN pin to VIN pin directly.  
8
9
EN  
I
Open-drain power-good status signal. When the FB voltage moves outside the specified limits,  
PGOOD goes low after 2-µs delay.  
PGOOD  
VIN  
O
P
Power-supply input pins for both integrated power MOSFET pair and the internal LDO. Place  
the decoupling input capacitors from VIN pins to PGND pins as close as possible.  
10, 21  
11, 12,  
13, 14,  
15, 16,  
17, 18  
Power ground of internal low-side MOSFET. At least six PGND vias are required to be placed  
as close as possible to the PGND pins. This minimizes parasitic impedance and also lowers  
thermal resistance.  
PGND  
G
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
3
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
5-1. Pin Functions (continued)  
NO.  
NAME  
VCC  
SW  
I/O(1)  
I/O  
O
DESCRIPTION  
Internal 3-V LDO output. An external bias with 3.3-V or higher voltage can be connected to this  
pin to save the power losses on the internal LDO. The voltage source on this pin powers both  
the internal circuitry and gate driver. Requires a 2.2-µF, at least 6.3-V, rating ceramic capacitor  
from VCC pin to PGND pins as the decoupling capacitor and the placement is required to be  
as close as possible.  
19  
20  
Output switching terminal of the power converter. Connect this pin to the output inductor.  
(1) I = Input, O = Output, P = Supply, G = Ground  
Copyright © 2022 Texas Instruments Incorporated  
4
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6 Specifications  
6.1 Absolute Maximum Ratings  
Over operating junction temperature range (unless otherwise noted) (1)  
MIN  
0.3  
0.3  
1.5  
0.3  
5  
MAX  
18  
18  
25  
18  
21.5  
22  
4
UNIT  
V
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
Pin voltage  
VIN  
V
VIN SW, DC  
VIN SW, < 20 ns transient  
SW PGND, DC  
SW PGND, < 20 ns transient  
BOOT PGND  
BOOT SW  
V
V
V
V
0.3  
0.3  
0.3  
0.3  
0.3  
0.3  
0.3  
V
VCC  
4
V
EN, PGOOD  
6
V
MODE  
4
V
TRIP, SS/REFIN, FB  
VSNS–  
3
V
0.3  
10  
150  
150  
V
Sinking current Power Good sinking current capability  
Operating junction temperature, TJ  
mA  
°C  
°C  
40  
55  
Storage temperature, Tstg  
(1) Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings  
only, which do not imply functional operation of the device at these or any other conditions beyond those indicated under  
Recommended Operating Conditions. Exposure to absolute-maximum-rated conditions for extended periods may affect device  
reliability.  
6.2 ESD Ratings  
VALUE  
UNIT  
Human-body model (HBM), per ANSI/ESDA/  
JEDEC JS-001 (1)  
±2000  
V(ESD)  
Electrostatic discharge  
V
Charged-device model (CDM), per ANSI/ESDA/  
JEDEC JS-002(2)  
±500  
(1) JEDEC document JEP155 states that 500-V HBM allows safe manufacturing with a standard ESD control process.  
(2) JEDEC document JEP157 states that 250-V CDM allows safe manufacturing with a standard ESD control process.  
6.3 Recommended Operating Conditions  
Over operating junction temperature range (unless otherwise noted)  
MIN  
2.7  
NOM  
MAX  
16  
UNIT  
V
VIN  
VIN  
Input voltage range when VCC pin is powered by a valid external bias  
Input voltage range when using the internal VCC LDO  
4.0  
16  
V
Minimum VIN before enabling the converter when using the internal VCC  
LDO  
VIN  
3.3  
V
VOUT  
Output voltage range  
0.6  
3.13  
0.1  
0.1  
0.1  
0.1  
50  
0
5.5  
3.6  
3.6  
5.5  
VCC  
1.5  
50  
V
V
Pin voltage External VCC bias  
Pin voltage BOOT to SW  
Pin voltage EN, PGOOD  
Pin voltage MODE  
V
V
V
Pin voltage TRIP, SS/REFIN, FB  
V
Pin voltage  
IPG  
mV  
mA  
VSNS(refer to AGND)  
Power Good input current capability  
10  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
5
Product Folder Links: TPS548B28  
 
 
 
 
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
Over operating junction temperature range (unless otherwise noted)  
MIN  
NOM  
MAX  
UNIT  
ILPEAK  
Maximum peak inductor current  
Minimum RTRIP  
35  
A
0
TJ  
Operating junction temperature  
125  
°C  
40  
6.4 Thermal Information  
TPS548B28  
THERMAL METRIC(1)  
RWW (QFN, JEDEC)  
RWW (QFN, TI EVM)  
UNIT  
21 PINS  
49.0  
23.0  
9.2  
21 PINS  
RθJA  
RθJC(top)  
RθJB  
ψJT  
Junction-to-ambient thermal resistance  
Junction-to-case (top) thermal resistance  
Junction-to-board thermal resistance  
24.1  
°C/W  
°C/W  
°C/W  
°C/W  
°C/W  
Not applicable (2)  
Not applicable (2)  
0.7  
Junction-to-top characterization parameter  
Junction-to-board characterization parameter  
0.5  
9.0  
8.7  
ψJB  
(1) For more information about traditional and new thermal metrics, see the Semiconductor and IC Package Thermal Metrics application  
report.  
(2) Not applicable to an EVM layout.  
6.5 Electrical Characteristics  
TJ = 40°C to +125°C, VCC = 3 V (unless otherwise noted)  
PARAMETER  
TEST CONDITIONS  
MIN  
TYP  
MAX UNIT  
SUPPLY  
VIN = 12 V, VEN = 2 V, VFB = VINTREF  
50mV (non-switching), no external  
bias on VCC pin  
+
IQ(VIN)  
VIN quiescent current  
910  
9.5  
680  
40  
1007  
20  
µA  
µA  
µA  
µA  
VIN = 12 V, VEN = 0 V, no external bias  
on VCC pin  
ISD(VIN)  
IQ(VCC)  
ISD(VCC)  
VIN shutdown supply current  
VCC quiescent current  
VCC shutdown current  
TJ = 25°C, VIN = 12 V, VEN = 2 V, VFB  
= VINTREF + 50mV (non-switching), 3.3  
V external bias on VCC pin  
820  
60  
VEN = 0 V, VIN = 0 V, 3.3 V external  
bias on VCC pin  
UVLO  
VINUVLO(rise)  
VINUVLO(fall)  
ENABLE  
VEN(rise)  
VIN UVLO rising threshold  
VIN UVLO falling threshold  
VIN rising, VCC = 3.3 V external bias  
VIN falling, VCC = 3.3 V external bias  
2.1  
2.4  
2.7  
V
V
1.55  
1.85  
2.15  
EN voltage rising threshold  
EN voltage falling threshold  
EN voltage hysteresis  
EN rising, enable switching  
EN falling, disable switching  
1.17  
0.97  
1.22  
1.02  
0.2  
1.27  
1.07  
V
V
VEN(fall)  
VEN(hyst)  
VEN(LKG)  
V
Input leakage current into EN pin  
VEN = 3.3 V  
0.5  
5
µA  
EN pin to AGND. EN floating disables  
the converter.  
EN internal pull-down resistance  
6500  
kΩ  
INTERNAL LDO (VCC PIN)  
Internal LDO output voltage  
VIN = 12 V, IVCC(Load) = 2 mA  
VCC rising  
2.90  
2.80  
2.62  
3.02  
2.87  
2.70  
0.17  
3.12  
2.94  
2.77  
V
V
VCCUVLO(rise)  
VCCUVLO(fall)  
VCCUVLO(hys)  
VCC UVLO rising threshold  
VCC UVLO falling threshold  
VCC UVLO hysteresis  
VCC falling  
V
V
TJ = 25°C, VIN = 4.0 V, IVCC(Load) = 20  
mA, non-switching  
VCC LDO dropout voltage, 20mA load  
VCC LDO short-circuit current limit  
1.037  
158  
VIN = 12 V, all temperature  
52  
105  
mA  
Copyright © 2022 Texas Instruments Incorporated  
6
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.5 Electrical Characteristics (continued)  
TJ = 40°C to +125°C, VCC = 3 V (unless otherwise noted)  
PARAMETER  
TEST CONDITIONS  
MIN  
TYP  
MAX UNIT  
VCC LDO turn-off is controlled by FB  
voltage during EN shutdown event  
FB Threshold to turn off VCC LDO  
REFERENCE VOLTAGE  
90  
146  
mV  
VINTREF  
Internal voltage reference  
TJ = 25°C  
600  
1
mV  
mV  
mV  
nA  
Internal voltage reference range  
Internal voltage reference range  
Input leakage current into FB pin  
TJ = 0°C to 85°C  
TJ = 40°C to 125°C  
VFB = VINTREF  
597  
594  
603  
606  
40  
IFB(LKG)  
TJ = -40°C to 125°C, VSS/REFIN = 0.6 V,  
VSNS- = AGND, refer to VINTREF  
SS/REFIN-to-FB Accuracy  
0.6%  
0.6%  
3.6  
PSEUDO REMOTE SENSE  
SS/REFIN-to-FB Accuracy  
SWITCHING FREQUENCY  
TJ = -40°C to 125°C, VSS/REFIN = 0.6 V,  
VSNS- = AGND, refer to VINTREF  
3.6  
mV  
TJ = 25°C, VIN = 12 V, VOUT=1.25V,  
RMODE = 0 Ωto AGND  
0.5  
0.6  
0.6  
0.7  
0.7  
TJ = 25°C, VIN = 12 V, VOUT=1.25V,  
RMODE = 30.1 kΩto AGND  
SW switching frequency, FCCM  
operation  
fSW  
0.8 MHz  
1.0  
TJ = 25°C, VIN = 12 V, VOUT=1.25V,  
RMODE = 60.4 kΩto AGND  
0.70  
0.85(2)  
STARTUP  
The delay from EN goes high to the  
first SW rising edge with internal LDO  
configuration. CVCC = 2.2 µF. CSS/REFIN  
= 220 nF.  
EN to first switching delay, internal  
LDO  
0.93  
2
ms  
The delay from EN goes high to the  
first SW rising edge with external VCC  
bias configuration. VCC bias should  
reach regulation before EN ramp up.  
CSS/REFIN = 220 nF.  
EN to first switching delay, external  
VCC bias  
0.55  
1.5  
0.9  
ms  
ms  
VO rising from 0 V to 95% of final  
setpoint, CSS/REFIN = 1nF  
tSS  
Internal fixed Soft-start time  
1
SS/REFIN sourcing current  
SS/REFIN sinking current  
VSS/REFIN = 0 V  
VSS/REFIN = 1 V  
36  
12  
µA  
µA  
POWER STAGE  
RDSON(HS)  
RDSON(LS)  
High-side MOSFET on-resistance  
Low-side MOSFET on-resistance  
Minimum on-time  
7.7  
2.4  
70  
TJ = 25°C, BOOTSW = 3 V  
TJ = 25°C, VCC = 3 V  
mΩ  
mΩ  
ns  
tON(min)  
TJ = 25°C, VCC = Internal LDO  
85  
TJ = 25°C, VCC = Internal LDO, HS  
FET Gate falling to rising  
tOFF(min)  
Minimum off-time  
220  
ns  
BOOT CIRCUIT  
IBOOT(LKG)  
BOOT leakage current  
TJ = 25°C, VBOOT-SW = 3.3 V  
35  
50  
µA  
V
VBOOT-SW(UV_F)  
BOOT-SW UVLO falling threshold  
TJ = 25°C, VIN = 12 V, VBOOT-SW falling  
2.0  
OVERCURRENT PROTECTION  
RTRIP  
TRIP pin resistance range  
0
20  
25  
kΩ  
Valley current on LS FET, 0-≤  
Current limit clamp  
19.2  
22.9  
120000  
22.9  
A
RTRIP 5.24-kΩ  
KOCL  
Constant for RTRIP equation  
Current limit threshold  
Ω  
Valley current on LS FET, RTRIP  
AGND  
=
IOCL (valley)  
19.2  
25  
A
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
7
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.5 Electrical Characteristics (continued)  
TJ = 40°C to +125°C, VCC = 3 V (unless otherwise noted)  
PARAMETER  
TEST CONDITIONS  
MIN  
TYP  
MAX UNIT  
Valley current on LS FET, RTRIP = 5.23  
kΩ  
IOCL (valley)  
IOCL (valley)  
IOCL (valley)  
IOCL (valley)  
IOCL (valley)  
IOCL (valley)  
Current limit threshold  
Current limit threshold  
Current limit threshold  
Current limit threshold  
Current limit threshold  
Current limit threshold  
22.9  
A
Valley current on LS FET, RTRIP = 6.04  
kΩ  
19.9  
16  
12  
8.2  
6
A
A
A
A
A
Valley current on LS FET, RTRIP = 7.5  
kΩ  
Valley current on LS FET, RTRIP = 10  
kΩ  
Valley current on LS FET, RTRIP = 14.7  
kΩ  
Valley current on LS FET, RTRIP = 20  
kΩ  
KOCL  
KOCL  
KOCL  
KOCL  
INOCL  
Constant KOCL tolerance  
Constant KOCL tolerance  
Constant KOCL tolerance  
Constant KOCL tolerance  
Negative current limit threshold  
-16.4%  
-12%  
-18%  
-21%  
12  
9%  
12%  
18%  
21%  
RTRIP = 5.23 kΩ  
6.04 kRTRIP 10 kΩ  
RTRIP = 14.7 kΩ  
RTRIP = 20 kΩ  
All VINs  
A
10  
8  
Zero-cross detection current threshold,  
open loop  
IZC  
VIN = 12 V, VCC = Internal LDO  
400  
mA  
OUTPUT OVP AND UVP  
Output Overvoltage-protection (OVP)  
threshold voltage  
VOVP  
113%  
77%  
116%  
400  
80%  
68  
119%  
83%  
tOVP(delay)  
VUVP  
Output OVP response delay  
With 100-mV overdrive  
ns  
µs  
Output Undervoltage-protection (UVP)  
threshold voltage  
tUVP(delay)  
Output UVP filter delay  
PGOOD threshold  
POWER GOOD  
PGOOD high, FB rising  
PGOOD low, FB rising  
PGOOD low, FB falling  
PGOOD high, FB rising  
89%  
113%  
77%  
92.5%  
116%  
80%  
95%  
119%  
83%  
VPGTH  
OOB (Out-Of-Bounds) threshold  
PGOOD sink current  
103% 105.5%  
108%  
VPGOOD = 0.4 V, VIN = 12 V, VCC =  
Internal LDO  
IPG  
17  
mA  
mV  
IPGOOD = 5.5 mA, VIN = 12 V, VCC =  
Internal LDO  
VPG(low)  
PGOOD low-level output voltage  
400  
tPGDLY(rise)  
tPGDLY(fall)  
Delay for PGOOD from low to high  
Delay for PGOOD from high to low  
1.06  
0.5  
1.33  
5
ms  
µs  
PGOOD leakage current when pulled TJ = 25°C, VPGOOD = 3.3 V, VFB  
=
IPG(LKG)  
5
µA  
high  
VINTREF  
VIN = 0 V, VCC = 0 V, VEN = 0 V,  
PGOOD pulled up to 3.3 V through a  
100-kΩresistor  
710  
850  
850  
mV  
PGOOD clamp low-level output  
voltage  
VIN = 0 V, VCC = 0 V, VEN = 0 V,  
PGOOD pulled up to 3.3 V through a  
10-kΩresistor  
1000  
1.5  
mV  
V
Min VCC for valid PGOOD output  
V
PGOOD 0.4 V  
OUTPUT DISCHARGE  
RDischg Output discharge resistance  
VIN = 12 V, VCC = Internal LDO, VSW  
= 0.5 V, power conversion disabled  
70  
Ω
Copyright © 2022 Texas Instruments Incorporated  
8
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.5 Electrical Characteristics (continued)  
TJ = 40°C to +125°C, VCC = 3 V (unless otherwise noted)  
PARAMETER  
TEST CONDITIONS  
MIN  
TYP  
MAX UNIT  
THERMAL SHUTDOWN  
TSDN  
Thermal shutdown threshold(1)  
Thermal shutdown hysteresis(1)  
Temperature rising  
150  
165  
30  
°C  
°C  
THYST  
(1) Specified by design. Not production tested.  
(2) Fsw variates with Vout due to D-CAP3 control mode.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
9
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.6 Typical Characteristics  
100  
95  
90  
85  
80  
75  
100  
95  
90  
85  
80  
75  
70  
65  
60  
55  
50  
VOUT = 0.9 V  
VOUT = 0.9 V  
VOUT = 1.0 V  
VOUT = 1.2 V  
VOUT = 1.8 V  
VOUT = 2.5 V  
VOUT = 3.3 V  
VOUT = 5.0 V  
70  
65  
60  
55  
50  
VOUT = 1.0 V  
VOUT = 1.2 V  
VOUT = 1.8 V  
VOUT = 2.5 V  
VOUT = 3.3 V  
VOUT = 5.0 V  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
6-1. Efficiency vs Output Current, 12 VIN, 800 kHz, DCM, VCC 6-2. Efficiency vs Output Current, 12 VIN, 800 kHz, DCM, VCC  
= Int  
= EXT 3.3 V  
100  
95  
90  
85  
80  
75  
70  
65  
60  
55  
50  
100  
95  
90  
85  
80  
75  
70  
65  
60  
55  
50  
VOUT = 0.9 V  
VOUT = 1.0 V  
VOUT = 1.2 V  
VOUT = 1.8 V  
VOUT = 2.5 V  
VOUT = 3.3 V  
VOUT = 5.0 V  
VOUT = 0.9 V  
VOUT = 1.0 V  
VOUT = 1.2 V  
VOUT = 1.8 V  
VOUT = 2.5 V  
VOUT = 3.3 V  
VOUT = 5.0 V  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
6-3. Efficiency vs Output Current, 12 VIN, 800 kHz, FCCM,  
6-4. Efficiency vs Output Current, 12 VIN, 800 kHz, FCCM,  
VCC = Int  
VCC = EXT 3.3 V  
115  
105  
95  
115  
105  
95  
85  
85  
75  
75  
65  
65  
55  
55  
Nat Conv  
100 LFM  
Nat Conv  
100 LFM  
45  
45  
200LFM  
400LFM  
200LFM  
400LFM  
35  
35  
25  
25  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
D002  
D001  
Vin = 12 V  
800 nH  
Vout = 5 V  
600 kHz  
Int Vcc  
Vin = 12 V  
300 nH  
Vout = 1 V  
800 kHz  
Int Vcc  
6-6. Safe Operating Area, VOUT = 5 V  
6-5. Safe Operating Area, VOUT = 1.0 V  
Copyright © 2022 Texas Instruments Incorporated  
10  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.6 Typical Characteristics (continued)  
115  
105  
95  
115  
105  
95  
85  
75  
65  
55  
45  
35  
25  
85  
75  
65  
55  
Nat Conv  
Nat Conv  
100 LFM  
200LFM  
400LFM  
45  
35  
25  
100 LFM  
200LFM  
400LFM  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
0
2
4
6
8
Output Current (A)  
10  
12  
14  
16  
18  
20  
D003  
D004  
Vin = 12 V  
300 nH  
Vout = 1 V  
800 kHz  
Ext Vcc 3.3 V  
Vin = 12 V  
800 nH  
Vout = 5 V  
600 kHz  
Ext Vcc 3.3 V  
6-7. Safe Operating Area, VOUT = 1 V  
6-8. Safe Operating Area, VOUT = 5 V  
3.355  
3.35  
1100  
1050  
1000  
950  
900  
850  
800  
750  
700  
650  
600  
550  
500  
450  
400  
350  
300  
Vin = 8V  
Vin = 12V  
Vin = 16V  
800 nH  
600 kHz  
VCC = Int  
FCCM  
3.345  
3.34  
3.335  
3.33  
3.325  
3.32  
3.315  
3.31  
600 kHz  
800 kHz  
1000 kHz  
3.305  
3.3  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
1
1.4 1.8 2.2 2.6 3  
Output Voltage (V)  
3.4 3.8 4.2 4.6  
5
D010  
6-10. Output Voltage vs Output Current  
6-9. Switching Frequency vs Output Voltage  
20  
15  
10  
5
3.04  
3.03  
3.02  
3.01  
3
0
2.99  
-50  
-25  
0
25  
50  
75  
100  
125  
150  
-50  
-25  
0
25  
50  
75  
100  
125  
150  
Junction Temperature (èC)  
Junction Temperature (èC)  
D014  
D015  
VIN = 12 V  
VEN = 0 V  
Internal VCC LDO  
VIN = 12 V IVCC = 2 mA  
6-12. VCC LDO vs Junction Temperature  
6-11. ISD(VIN) vs Junction Temperature  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
11  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
6.6 Typical Characteristics (continued)  
0.904  
40  
39  
38  
37  
36  
35  
34  
33  
32  
0.902  
0.9  
0.898  
0.896  
-50  
-25  
0
25  
50  
75  
100  
125  
150  
-50  
-25  
0
25  
50  
75  
100  
125  
150  
Junction Temperature (èC)  
Junction Temperature (èC)  
D016  
D020  
VIN = 12 V  
VIN = 12 V  
6-13. VINTREF vs Junction Temperature  
6-14. ISS(source) vs Junction Temperature  
Copyright © 2022 Texas Instruments Incorporated  
12  
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
7 Detailed Description  
7.1 Overview  
The TPS548B28 device is a high-efficiency, single-channel, small-sized, synchronous-buck converter. The  
device suits low output voltage point-of-load applications with 20-A or lower output current in server, storage, and  
similar computing applications. The TPS548B28 features proprietary D-CAP3 control mode combined with  
adaptive on-time architecture. This combination builds modern low-duty-ratio and ultra-fast load-step-response  
DC/DC converters in an ideal fashion. The output voltage ranges from 0.6 V to 5.5 V. The conversion input  
voltage ranges from 2.7 V to 16 V and the VCC input voltage ranges from 3.13 V to 3.6 V. The D-CAP3 control  
mode uses emulated current information to control the modulation. An advantage of this control scheme is that it  
does not require a phase-compensation network outside which makes the device easy-to-use and also allows  
low external component count. Another advantage of this control scheme is that it supports stable operation with  
all low-ESR output capacitors (such as ceramic capacitor and low ESR polymer capacitor). Adaptive on-time  
control tracks the preset switching frequency over a wide range of input and output voltages while increasing  
switching frequency as needed during load-step transient.  
7.2 Functional Block Diagram  
PGOOD  
SS/  
REFIN  
Soft-start  
generator  
PG Falling  
Threshold  
+
UV  
Internal  
Soft-start  
VIN  
EN  
PGOOD Driver  
LDO  
+
OV  
VCC  
Reference  
generator  
VCC  
PG Rising  
Threshold  
FB  
BOOT  
REG  
PGOOD  
+
+
+
VCCOK  
Control Logic  
VINOK  
BOOT  
VIN  
VCC UVLO  
VSNS-  
VIN  
PWM  
+
VIN UVLO  
ñ
ñ
ñ
ñ
ñ
ñ
ñ
ñ
tON generator  
Minimum On/Off  
Light Load  
FCCM/Skip  
Internal  
Ramp  
VCC UVLO  
VIN UVLO  
SW  
XCON  
EN  
Enable  
Output OVP/UVP  
Thermal Shutdown  
EN  
+
1.22V / 1.02V  
SW  
Valley Current  
Limit & ZCD  
OC  
Limit  
TRIP  
MODE  
AGND  
PGND  
MODE  
Selection  
Fsw &  
Mode  
+
ThermalOK  
165°C /  
135°C  
Output Soft  
Discharge  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
13  
Product Folder Links: TPS548B28  
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
7.3 Feature Description  
7.3.1 Internal VCC LDO And Using External Bias On VCC Pin  
The TPS548B28 has an internal 3.0-V LDO, featuring input from VIN and output to VCC. When the EN voltage  
rises above the enable threshold (typically 1.22 V), the internal LDO is enabled and starts regulating output  
voltage on the VCC pin. The VCC voltage provides the bias voltage for the internal analog circuitry and also  
provides the supply voltage for the gate drives.  
The VCC pin must be bypassed with a 2.2-µF, at least 6.3-V rating ceramic capacitor. An external bias that is  
above the output voltage of the internal LDO can override the internal LDO. This enhances the efficiency of the  
converter because the VCC current now runs off this external bias instead of the internal linear regulator.  
The VCC UVLO circuit monitors the VCC pin voltage and disables the whole converter when VCC falls below the  
VCC UVLO falling threshold. Maintaining a stable and clean VCC voltage is required for a smooth operation of  
the device.  
Considerations when using an external bias on the VCC pin are as follows:  
When the external bias is applied on the VCC pin early enough (for example, before EN signal comes in), the  
internal LDO will be always forced off and the internal analog circuits will have a stable power supply rail at  
their power enable.  
(Not recommended) When the external bias is applied on the VCC pin late (for example, after EN signal  
comes in), any power-up and power-down sequencing can be applied as long as there is no excess current  
pulled out of the VCC pin. Understand that an external discharge path on the VCC pin, which can pull a  
current higher than the current limit of the internal LDO from the VCC pin, can potentially turn off VCC LDO,  
thereby shutting down the converter output.  
A good power-up sequence is when at least one of VIN UVLO rising threshold or EN rising threshold is  
satisfied later than VCC UVLO rising threshold. For example, a practical power-up sequence is: VIN applied  
first, then the external bias applied, and then EN signal goes high.  
7.3.2 Enable  
When the EN pin voltage rises above the enable threshold voltage (typically 1.22 V) and VIN rises above the VIN  
UVLO rising threshold, the device enters its internal power-up sequence. The EN to first switching delay is  
specified in Start-up section in 6.5.  
When using the internal VCC LDO, the internal power-up sequence includes three sequential steps. During the  
first period, the VCC voltage is charged up on a VCC bypass capacitor by an 11-mA current source. The length  
of this VCC LDO start-up time varies with the capacitance on VCC pin. However, if the VIN voltage ramps up  
very slowly, the VCC LDO output voltage will be limited by the VIN voltage level, thus the VCC LDO start-up time  
can be extended longer. Because the VCC LDO start-up time is relatively long, the internal VINTREF build-up  
happens and finishes during this period. After the VCC voltage crosses above the VCC UVLO rising threshold  
(typically 2.87 V), the device moves to the second step, power-on delay. The MODE pin setting detection, SS/  
REFIN pin detection, and control loop initialization are finished within this 285-μs delay. Soft-start ramp starts  
when the 285-μs power-on delay finishes. During the soft-start ramp power stage, switching does not happen  
until the SS/REFIN pin voltage reaches 50 mV. This introduces a SS delay that varies with the external  
capacitance on SS/REFIN pin.  
7-1 shows an example where the VIN UVLO rising threshold is satisfied earlier than the EN rising threshold.  
In this scenario, the VCC UVLO rising threshold becomes the gating signal to start the internal power-up  
sequence, and the sequence between VIN and EN does not matter.  
When using an external bias on the VCC pin, the internal power-up sequence still includes three sequential  
steps. The first period is much shorter because VCC voltage is built up already. A 100-µs period allows the  
internal references to start up and reach regulation points. This 100-µs period includes not only the 0.6-V  
VINTREF, but also all of the other reference voltages for various functions. The device then moves to the second  
step, power-on delay. The MODE pin setting detection, SS/REFIN pin detection, and control loop initialization  
are finished within this 285-μs delay. Soft-start ramp starts when the 285-μs power-on delay finishes. During  
Copyright © 2022 Texas Instruments Incorporated  
14  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
the soft-start ramp power stage, switching does not happen until the SS/REFIN pin voltage reaches 50 mV. This  
introduces a SS delay that varies with the external capacitance on SS/REFIN pin.  
7-2 shows an example where the VIN UVLO rising threshold and EN rising threshold are satisfied later than  
the VCC UVLO rising threshold. In this scenario, the VIN UVLO rising threshold or EN rising threshold, whoever  
is satisfied later, becomes the gating signal to start the internal power-up sequence.  
2.4V  
VIN  
1.22V  
EN  
2.87V  
VCC LDO  
Power-on  
delay  
VCC LDO  
Startup  
50mV  
SS/REFIN  
SS delay  
FB  
SW pulses are omitted to  
simplify the illustration  
……  
SW  
7-1. Internal Power-up Sequence Using Internal LDO  
2.87V  
VCC  
External  
3.3V Bias  
2.4V  
VIN  
EN  
1.22V  
Power-on  
delay  
VREF  
Build-up  
SS/REFIN  
50mV  
SS delay  
FB  
SW pulses are omitted to  
simplify the illustration  
……  
SW  
7-2. Internal Power-up Sequence Using External Bias  
The EN pin has an internal filter to avoid unexpected ON or OFF due to small glitches. The time constant of this  
RC filter is 5 µs. For example, when applying 3.3-V voltage source on the EN pin that jumps from 0 V to 3.3 V  
with an ideal rising edge, the internal EN signal will reach 2.086 V after 5 µs, which is 63.2% of applied 3.3-V  
voltage level.  
A internal pulldown resistor is implemented between the EN pin and AGND pin. To avoid impact to the EN rising/  
falling threshold, this internal pulldown resistor is set to 6.5 MΩ. With this pulldown resistor, floating the EN pin  
before start-up keeps the device under disabled state. During nominal operation when the power stage switches,  
this large internal pulldown resistor can not have enough noise immunity to hold EN pin low.  
The recommended operating condition for EN pin is maximum 5.5 V. Do not connect the EN pin to the VIN pin  
directly.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
15  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
7.3.3 Output Voltage Setting  
The output voltage is programmed by the voltage-divider resistors, RFB_HS and RFB_LS. Connect RFB_HS between  
the FB pin and the positive node of the load, and connect RFB_LS between the FB pin and VSNSpin. The  
recommended RFB_LS value is 10 kΩ, ranging from 1 kΩto 20 kΩ. Determine RFB_HS by using 方程1.  
VO - V  
INTREF  
RFB _HS  
=
ìRFB _LS  
V
INTREF  
(1)  
The FB accuracy is determined by two elements. The first element is the accuracy of the internal 600-mV  
reference, which will be applied to the SS/REFIN pin unless an external VREF is applied. The TPS548B28 device  
offers ±0.5% VINTREF accuracy from a 0°C to 85°C temperature range, and ±1.0% VINTREF accuracy from 40°C  
to 125°C temperature range. The second element is the SS/REFIN-to-FB accuracy which tells how accurately  
the control loop regulates FB node to SS/REFIN pin. The TPS548B28 device offers ±0.6% SS/REFIN-to-FB  
accuracy from 40°C to 125°C temperature range. For example, when operating from a 0°C to 85°C  
temperature range, the total FB accuracy is ±1.1% which includes the impact from chip junction temperature and  
also the variation from part to part.  
To improve the overall VOUT accuracy, using a ±1% accuracy or better resistor for the FB voltage divider is highly  
recommended.  
Regardless of remote sensing or single-end sensing connection, the FB voltage divider, RFB_HS and RFB_LS  
,
must be always placed as close as possible to the device.  
7.3.3.1 Remote Sense  
The TPS548B28 device offers remote sense function through FB and VSNSpins. Remote sense function  
compensates a potential voltage drop on the PCB traces thus helps maintain VOUT tolerance under steady state  
operation and load transient event. Connecting the FB voltage divider resistors to the remote location allows  
sensing to the output voltage at a remote location. The connections from the FB voltage divider resistors to the  
remote location must be a pair of PCB traces with at least 12-mil trace width, and must implement Kelvin sensing  
across a high bypass capacitor of 0.1 μF or higher. The ground connection of the remote sensing signal must  
be connected to the VSNSpin. The VOUT connection of the remote sensing signal must be connected to the  
feedback resistor divider with the lower feedback resistor, RFB_LS, terminated at VSNSpin. To maintain stable  
output voltage and minimize the ripple, the pair of remote sensing lines must stay away from any noise sources  
such as inductor and SW nodes, or high frequency clock lines. TI recommends to shield the pair of remote  
sensing lines with ground planes above and below.  
Single-ended Vo sensing is often used for local sensing. For this configuration, connect the higher FB resistor,  
RFB_HS, to a high-frequency local bypass capacitor of 0.1 μF or higher, and short VSNSto AGND.  
The recommended VSNSoperating range (refer to AGND pin) is 50 mV to +50 mV.  
7.3.4 Internal Fixed Soft Start and External Adjustable Soft Start  
The TPS548B28 implements a circuit to allow both internal fixed soft start and external adjustable soft start. The  
internal soft-start time is typically 1.5 ms. The soft-start time can be increased by adding a soft-start (SS)  
capacitor between the SS/REFIN and VSNS- pins. The total SS capacitor value can be determined by Equation  
2. The device follows the longer SS ramp among the internal SS time and the SS time determined by the  
external SS capacitors. The recommended maximum SS capacitor is 1 µF. A minimum 1-nF SS capacitor is  
required.  
The device does not require a capacitor from SS/REFIN pin to AGND, thus TI does not recommend to place a  
capacitor from the SS/REFIN pin to AGND. If both CSS/REFIN-to-VSNSand CSS/REFIN-to-AGND capacitors  
exist, place CSS/REFIN-to-VSNSmore closely with shortest trace back to VSNSpin.  
t
SS(ms)ì36(mA )  
CSS(nF)=  
VINTREF( V)  
(2)  
Copyright © 2022 Texas Instruments Incorporated  
16  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
The TPS548B28 provides an analog input pin (SS/REFIN) to accept an external reference. When an external  
voltage signal is applied between SS/REFIN pin and VSNS- pin, it acts as the reference voltage thus FB voltage  
follows this external voltage signal exactly. Apply this external reference to the SS/REFIN pin before the EN high  
signal is recommended. The external reference must be equal to or higher than the internal reference level to  
ensure correct Power Good thresholds during soft start.  
With an external reference applied, the internal fixed soft start controls output voltage ramp during start-up. After  
soft start finishes, the external voltage signal can be in a range of 0.5 V to 1.2 V.  
When driving the SS/REFIN pin with an external resistor divider, the resistance must be low enough so that the  
external voltage source can overdrive the internal current source.  
7.3.5 External REFIN For Output Voltage Tracking  
The TPS548B28 provides an analog input pin (SS/REFIN) to accept an external reference (that is a DC voltage  
source). The device always looks at the voltage on this SS/REFIN pin as the reference for the control loop.  
When an external voltage reference is applied between the SS/REFIN pin and VSNSpin, it acts as the  
reference voltage thus FB voltage follows this external voltage reference exactly. The same ±0.6% SS/REFIN-to-  
FB accuracy from -40°C to 125°C temperature range applies here too.  
In the middle of internal power-on delay, a detection circuit senses the voltage on the SS/REFIN pin to tell  
whether an active DC voltage source is applied. Before the detection happens, the SS/REFIN pin tries to  
discharge any energy on SS/REFIN capacitors through an internal 120-Ωresistor to AGND. This discharge lasts  
for 125 µs. Then, within a 32-µs window, the detection circuit compares the SS/REFIN pin voltage with an  
internal reference equal to 89% of VINTREF. This discharge operation ensure a SS capacitor with left-over energy  
will not be wrongly detected as a voltage reference. If the external voltage reference failed to supply sufficient  
current and hold voltage level higher than 89% of VINTREF, the SS/REFIN detection circuit will provide wrong  
detection result.  
If the detection result is that SS/REFIN pin voltage falls below 89% of VINTREF which tells no external reference is  
connected, the device first uses the internal fixed VINTREF as the reference for PGOOD, VOUT OVP, and VOUT  
UVP threshold. On this configuration, given SS/REFIN pin sees a soft-start ramp on this pin, the slower ramp  
among the internal fixed soft start and the external soft start determines the start-up of FB. After both the internal  
and external soft-start ramp finishes, the power-good signal becomes high after a 1.06-ms internal delay. The  
whole internal soft-start ramp takes 2 ms to finish. The external soft-start done signal goes high when FB  
reaches a threshold equal to VINTREF 50 mV. The device waits for the PGOOD status transition from low to  
high, then starts using the SS/REFIN pin voltage, instead of the internal VINTREF as the reference for PGOOD,  
VOUT OVP, and VOUT UVP threshold.  
If the detection result is that SS/REFIN pin voltage holds higher than 89% of VINTREF which tells an active DC  
voltage source is used as an external reference, the device always uses the SS/REFIN pin voltage instead of the  
internal VINTREF as the reference for PGOOD, VOUT OVP, and VOUT UVP threshold. On this configuration,  
because the SS/REFIN pin sees a DC voltage and no soft-start ramp on this pin, the internal fixed soft start is  
used for start-up. After the internal soft-start ramp finishes, the power-good signal becomes high after a 1.06-ms  
internal delay. The whole internal soft-start ramp takes 2 ms to finish because the soft-start ramp goes beyond  
VINTREF  
.
On this external REFIN configuration, applying a stabilized DC external reference to the SS/REFIN pin before  
EN high signal is recommended. During the internal power-on delay, the external reference must be capable of  
holding the SS/REFIN pin equal to or higher than 89% of VINTREF, so that the device can correctly detect the  
external reference and choose the right thresholds for power good, VOUT OVP, and VOUT UVP. After the power  
good status transits from low to high, the external reference can be set in a range of 0.5 V to 1.2 V. To overdrive  
the SS/REFIN pin during nominal operation, the external reference has to be able to sink more than 36-µA  
current if the external reference is lower than the internal VINTREF, or source more than 12-µA current if the  
external reference is higher than the internal VINTREF. When driving the SS/REFIN pin by an external reference  
through a resistor divider, the resistance of the divider must be low enough to provide the sinking, or sourcing  
current capability.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
17  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
The configuration of applying EN high signal first, then applying an external ramp on the SS/REFIN pin as a  
tracking reference can be achieved, as long as design considerations for power good, VOUT OVP, and VOUT UVP  
have been taken. Please contact Texas Instruments for detailed information about this configuration.  
If the external voltage source must transition up and down between any two voltage levels, the slew rate must be  
no more than 1 mV/μs.  
7.3.6 Frequency and Operation Mode Selection  
The TPS548B28 provides forced CCM operation for tight output ripple application and auto-skip Eco-mode for  
high light-load efficiency. The TPS548B28 allows users to select the switching frequency and operation mode by  
connecting a resistor from the MODE pin to AGND pin. 7-1 lists the resistor values for the switching frequency  
and operation mode selection. TI recommends ±1% tolerance resistors with a typical temperature coefficient of  
±100 ppm/°C.  
The MODE state will be set and latched during the internal power-on delay period. Changing the MODE pin  
resistance after the power-on delay will not change the status of the device. The internal circuit will set the  
MODE pin status to 600 kHz / skip-mode if MODE pin is left open during the power-on delay period.  
To make sure the internal circuit detects the desired option correctly, do not place any capacitor on the MODE  
pin.  
7-1. MODE Pin Selection  
MODE PIN  
CONNECTIONS  
OPERATION MODE UNDER LIGHT  
LOAD  
SWITCHING FREQUENCY  
(fSW) (kHz)(1)  
Short to VCC  
Skip-mode  
Skip-mode  
600  
800  
243-kΩ± 10% to AGND  
121-kΩ± 10% to AGND  
60.4-kΩ±10% to AGND  
30.1-kΩ±10% to AGND  
Short to AGND  
Skip-mode  
1000  
1000  
800  
Forced CCM  
Forced CCM  
Forced CCM  
600  
(1) Switch frequency is based on 3.3 Vout. Frequency varies with Vout.  
7.3.7 D-CAP3Control Mode  
RR  
SW  
To comparator  
CR  
VOUT  
7-3. Internal RAMP Generation Circuit  
The TPS548B28 uses D-CAP3 control mode to achieve the fast load transient while maintaining the ease-of-use  
feature. The D-CAP3 control mode architecture includes an internal ripple generation network enabling the use  
of very low-ESR output capacitors such as multi-layered ceramic capacitors (MLCC) and low-ESR polymer  
capacitors. No external current sensing network or voltage compensators are required with D-CAP3 control  
mode architecture. The role of the internal ripple generation network is to emulate the ripple component of the  
inductor current information and then combine it with the voltage feedback signal to regulate the loop operation.  
The amplitude of the ramp is determined by VIN, VOUT, operating frequency, and the R-C time-constant of the  
internal ramp circuit. At different switching frequency settings (see 7-1), the R-C time-constant varies to  
maintain relatively constant ramp amplitude. Also, the device uses internal circuitry to cancel the dc offset  
caused by injected ramp, and significantly reduces the dc offset caused by the output ripple voltage, especially  
under light load condition.  
Copyright © 2022 Texas Instruments Incorporated  
18  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
For any control topologies supporting no external compensation design, there is a minimum range, maximum  
range, or both, of the output filter it can support. The output filter used with the TPS548B28 is a low-pass L-C  
circuit. This L-C filter has double pole that is described in 方程3.  
1
f
=
P
2´ p´ L  
´ C  
OUT  
OUT  
(3)  
At low frequencies, the overall loop gain is set by the output set-point resistor divider network and the internal  
gain of the TPS548B28. The low frequency L-C double pole has a 180-degree drop in phase. At the output filter  
frequency, the gain rolls off at a 40 dB per decade rate and the phase drops rapidly. The internal ripple  
generation network introduces a high-frequency zero that reduces the gain roll off from 40 dB to 20 dB per  
decade and increases the phase by 90 degrees per decade above the zero frequency.  
After identifying the application requirements, the output inductance must be designed so that the inductor peak-  
to-peak ripple current is approximately between 15% and 40% of the maximum output current.  
The inductor and capacitor selected for the output filter must be such that the double pole of 方程式 3 is located  
no higher than 1/30th of operating frequency. Choose very small output capacitance leads to relatively high  
frequency L-C double pole, which allows that overall loop gain stays high until the L-C double frequency. Given  
the zero from the internal ripple generation network is relatively high frequency as well, the loop with very small  
output capacitance can have too high crossover frequency which is not desired. Use 7-2 to help locate the  
internal zero based on the selected switching frequency.  
7-2. Locating the Zero  
SWITCHING FREQUENCIES  
ZERO (fZ) LOCATION (kHz)  
(fSW) (kHz)  
600  
800  
84.5  
84.5  
106  
1000  
In general, where reasonable (or smaller) output capacitance is desired, output ripple requirement and load  
transient requirements can be used to determine the necessary output capacitance for stable operation.  
For the maximum output capacitance recommendation, select the inductor and capacitor values so that the L-C  
double pole frequency is no less than 1/100th of operating frequency. With this starting point, verify the small  
signal response on the board using the phase margin at the loop crossover is greater than 50 degrees.  
The actual maximum output capacitance can go higher as long as phase margin is greater than 50 degrees.  
However, small signal measurement (bode plot) must be done to confirm the design.  
If MLCC is used, consider the derating characteristics to determine the final output capacitance for the design.  
For example, when using an MLCC with specifications of 10 µF, X5R and 6.3 V, the derating by DC bias and AC  
bias are 80% and 50%, respectively. The effective derating is the product of these two factors, which in this case  
is 40% and 4 µF. Consult with capacitor manufacturers for specific characteristics of the capacitors to be used in  
the system/applications.  
For higher output voltage at or above 2 V, additional phase boost can be required to secure sufficient phase  
margin due to phase delay/loss for higher output voltage (large on-time (tON)) setting in a fixed-on-time topology  
based operation. A feedforward capacitor placed in parallel with RFB_HS is found to be very effective to boost the  
phase margin at loop crossover. Refer to the Optimizing Transient Response of Internally Compensated dc-dc  
Converters With Feedforward Capacitor application report for details.  
Besides boosting the phase, a feedforward capacitor feeds more VOUT node information into FB node by the AC  
coupling. This feedforward during load transient event enables the control loop a faster response to VOUT  
deviation. However, this feedforward during steady state operation also feeds more VOUT ripple and noise into  
FB. High ripple and noise on FB usually leads to more jitter, or even double pulse behavior. To determine the  
final feedforward capacitor value, impacts to phase margin, load transient performance, and ripple and nosie on  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
19  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
FB must be all considered. Using Frequency Analysis equipment to measure the crossover frequency and the  
phase margin is recommended.  
7.3.8 Low-side FET Zero-Crossing  
The TPS548B28 uses a zero-crossing circuit to perform the zero inductor-current detection during skip-mode  
operation. The function compensates the inherent offset voltage of the Z-C comparator and delay time of the Z-C  
detection circuit. The zero-crossing threshold is set to a positive value to avoid negative inductor current. As a  
result, the device delivers better light-load efficiency.  
7.3.9 Current Sense and Positive Overcurrent Protection  
For a buck converter, during the on-time of the high-side FET, the switch current increases at a linear rate  
determined by input voltage, output voltage, the on-time, and the output inductor value. During the on-time of the  
low-side FET, this current decreases linearly. The average value of the switch current equals to the load current.  
The output overcurrent limit (OCL) in the TPS548B28 device is implemented using a cycle-by-cycle valley  
current detect control circuit. The inductor current is monitored during the on-time of the low-side FET by  
measuring the low-side FET drain-to-source current. If the measured drain-to-source current of the low-side FET  
is above the current limit threshold, the low-side FET stays ON until the current level becomes lower than the  
current limit threshold. This type of behavior reduces the average output current sourced by the device. During  
an overcurrent condition, the current to the load exceeds the current to the output capacitors, thus the output  
voltage tends to decrease. Eventually, when the output voltage falls below the undervoltage-protection threshold  
(80%), the UVP comparator detects it and shuts down the device after a wait time of 68 µs. The device then  
enters a hiccup sleep period for approximately 14 ms. After this waiting period, the device attempts to start up  
again. 7-4 shows the cycle-by-cycle valley current limit behavior as well as the wait time before the device  
shuts down.  
If an OCL condition happens during start-up, the device still has cycle-by-cycle current limit based on low-side  
valley current. After soft start is finished, the UV event which is caused by the OC event shuts down the device  
and enters hiccup mode mode with a wait time of 68 µs.  
The resistor, RTRIP connected from the TRIP pin to AGND sets current limit threshold. A ±1% tolerance resistor  
is highly recommended because a worse tolerance resistor provides less accurate OCL threshold. Equation 4  
calculates the RTRIP for a given overcurrent limit threshold on the device. To simplify the calculation, use a  
constant, KOCL, to replace the value of 12x104. Equation 4 calculates the overcurrent limit threshold for a given  
RTRIP value. The tolerance of KOCL is listed in 6.5 to help you analyze the tolerance of the overcurrent limit  
threshold.  
To protect the device from unexpected connection on TRIP pin, an internal fixed OCL clamp is implemented.  
This internal OCL clamp limits the maximum valley current on LS FET when TRIP pin has too small resistance to  
AGND, or is accidentally shorted to ground.  
(4)  
where  
IOCLIM is overcurrent limit threshold for load current in A  
RTRIP is TRIP resistor value in Ω  
KOCL is a constant for the calculation  
VIN is input voltage value in V  
VO is output voltage value in V  
L is output inductor value in µH  
fSW is switching frequency in MHz  
Copyright © 2022 Texas Instruments Incorporated  
20  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
V -V ì V  
KOCL  
(
)
1
2
1
IN  
O
O
IOCLIM  
=
+
ì
ì
RTRIP  
V
L ì fSW  
IN  
(5)  
where  
IOCLIM is overcurrent limit threshold for load current in A  
RTRIP is TRIP resistor value in Ω  
KOCL is a constant for the calculation  
VIN is input voltage value in V  
VO is output voltage value in V  
L is output inductor value in µH  
fSW is switching frequency in MHz  
7-4. Overcurrent Protection  
7.3.10 Low-side FET Negative Current Limit  
The device has a fixed, cycle-by-cycle negative current limit. Similar with the positive overcurrent limit, the  
inductor current is monitored during the on-time of the low-side FET. To prevent too large negative current  
flowing through low-side FET, when the low-side FET detects a 10-A current (typical threshold), the device  
turns off low-side FET and then turns on the high-side FET for a proper on-time (determined by VIN/VO/fSW).  
After the high-side FET on-time expires, the low-side FET turns on again.  
The device must not trigger the 10-A negative current limit threshold during nominal operation, unless a small  
inductor value that is too small is chosen or the inductor becomes saturated. This negative current limit is utilized  
to discharge output capacitors during an output OVP or an OOB event. See 7.3.12 and 7.3.13 for details.  
7.3.11 Power Good  
The device has power-good output that indicates high when the converter output is within the target. The power-  
good output is an open-drain output and must be pulled up to the VCC pin or an external voltage source (<5.5 V)  
through a pullup resistor (typically 30.1 kΩ). The recommended power-good pullup resistor value is 1 kΩto 100  
kΩ.  
After both the internal and external soft-start ramp finishes, the power-good signal becomes high after a 1.06-ms  
internal delay. The whole internal soft-start ramp takes 2 ms to finish. The external soft-start done signal goes  
high when FB reaches threshold equal to VINTREF 50 mV. If the FB voltage drops to 80% of the VINTREF  
voltage or exceeds 116% of the VINTREF voltage, the power-good signal latches low after a 2-µs internal delay.  
The power-good signal can only be pulled high again after re-toggling EN or a reset of VIN.  
If the input supply fails to power up the device, for example VIN and VCC both stay at zero volt, the power-good  
pin clamps low by itself when this pin is pulled up through an external resistor.  
After VCC voltage level rises above the minimum VCC threshold for valid PGOOD output (maximum 1.5 V),  
internal power-good circuit is enabled to hold the PGOOD pin to the default status. By default, PGOOD is pulled  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
21  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
low and this low-level output voltage is no more than 400 mV with 5.5-mA sinking current. The power-good  
function is fully activated after the soft-start operation is completed.  
7.3.12 Overvoltage and Undervoltage Protection  
The device monitors a resistor-divided feedback voltage to detect overvoltage and undervoltage events. When  
the FB voltage becomes lower than 80% of the VINTREF voltage, the UVP comparator detects and an internal  
UVP delay counter begins counting. After the 68-µs UVP delay time, the device enters hiccup mode and re-  
starts with a sleep time of 14 ms. The UVP function enables after the soft-start period is complete.  
When the FB voltage becomes higher than 116% of the VINTREF voltage, the OVP comparator detects and the  
circuit latches OFF the high-side MOSFET driver and turns on the low-side MOSFET until reaching a negative  
current limit INOCL. Upon reaching the negative current limit, the low-side FET is turned off, and the high-side  
FET is turned on again for a proper on-time (determined by VIN/VO/fSW). The device operates in this cycle until  
the output voltage is pulled down under the UVP threshold voltage for 68 µs. After the 68-µs UVP delay time,  
both the high-side FET and the low-side FET are latched OFF. The fault is cleared with a reset of VIN or by  
retoggling the EN pin.  
During the 68-μs UVP delay time, if the output voltage becomes higher than the UV threshold, thus is not  
qualified for UV event, the timer will be reset to zero. When the output voltage triggers UV threshold again, the  
timer of the 68 μs re-starts.  
7.3.13 Out-Of-Bounds (OOB) Operation  
The device has an out-of-bounds (OOB) overvoltage protection that protects the output load at a much lower  
overvoltage threshold of 5% above the VINTREF voltage. OOB protection does not trigger an overvoltage fault, so  
the device is on non-latch mode after an OOB event. OOB protection operates as an early no-fault overvoltage-  
protection mechanism. During the OOB operation, the controller operates in forced CCM mode. Turning on the  
low-side FET beyond the zero inductor current quickly discharges the output capacitor thus helping the output  
voltage to fall quickly towards the setpoint. During the operation, the cycle-by-cycle negative current limit is also  
activated to ensure the safe operation of the internal FETs.  
7.3.14 Output Voltage Discharge  
When the device is disabled through EN, it enables the output voltage discharge mode. This mode forces both  
high-side and low-side FETs to latch off, but turns on the discharge FET, which is connected from SW to PGND,  
to discharge the output voltage. After the FB voltage drops below 90 mV, the discharge FET is turned off.  
The output voltage discharge mode is activated by any of the following fault events:  
1. EN pin goes low to disable the converter.  
2. Thermal shutdown (OTP) is triggered.  
3. VCC UVLO (falling) is triggered.  
4. VIN UVLO (falling) is triggered.  
7.3.15 UVLO Protection  
The device monitors the voltage on both the VIN and the VCC pins. If the VCC pin voltage is lower than the  
VCCUVLO falling threshold voltage, the device shuts off. If the VCC voltage increases beyond the VCCUVLO rising  
threshold voltage, the device turns back on. VCC UVLO is a non-latch protection.  
When the VIN pin voltage is lower than the VINUVLO falling threshold voltage but the VCC pin voltage is still  
higher than VCCUVLO rising threshold voltage, the device stops switching and discharges SS/REFIN pin. After  
the VIN voltage increases beyond the VINUVLO rising threshold voltage, the device re-initiates the soft start and  
switches again. VIN UVLO is a non-latch protection.  
7.3.16 Thermal Shutdown  
The device monitors internal junction temperature. If the temperature exceeds the threshold value (typically  
165°C), the device stops switching and discharges SS/REFIN pin. When the temperature falls approximately  
30°C below the threshold value, the device turns back on with a re-initiated soft start. Thermal shutdown is a  
non-latch protection.  
Copyright © 2022 Texas Instruments Incorporated  
22  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
7.4 Device Functional Modes  
7.4.1 Auto-Skip Eco-modeLight Load Operation  
While the MODE pin is pulled to VCC directly or connected to AGND pin through a resistor larger than 121 kΩ,  
the device automatically reduces the switching frequency at light-load conditions to maintain high efficiency. This  
section describes the operation in detail.  
As the output current decreases from heavy load condition, the inductor current also decreases until the rippled  
valley of the inductor current touches zero level. Zero level is the boundary between the continuous-conduction  
and discontinuous-conduction modes. The synchronous MOSFET turns off when this zero inductor current is  
detected. As the load current decreases further, the converter runs into discontinuous-conduction mode (DCM).  
The on-time is maintained to a level approximately the same as during continuous-conduction mode operation  
so that discharging the output capacitor with a smaller load current to the level of the reference voltage requires  
more time. The transition point to the light-load operation IO(LL) (for example: the threshold between continuous-  
and discontinuous-conduction mode) is calculated as shown in Equation 5.  
V
- V  
´ V  
(
)
OUT OUT  
V
IN  
1
IN  
I
=
´
OUT LL  
( )  
2´L ´ f  
SW  
(6)  
where  
fSW is the switching frequency  
Using only ceramic capacitors is recommended for skip-mode.  
7.4.2 Forced Continuous-Conduction Mode  
When the MODE pin is tied to the AGND pin through a resistor less than 60.4 kΩ, the controller operates in  
continuous conduction mode (CCM) during light-load conditions. During CCM, the switching frequency is  
maintained to an almost constant level over the entire load range which is suitable for applications requiring tight  
control of the switching frequency at the cost of lower efficiency.  
7.4.3 Powering the Device From a 12-V Bus  
The device works well when powering from a 12-V bus with single VIN configuration. As a single VIN  
configuration, the internal LDO is powered by a 12-V bus and generates a 3.0-V output to bias the internal  
analog circuitry and also powers up the gate drives. The VIN input range under this configuration is 4 V to 16 V  
for up to 20-A load current. 7-5 shows an example for this single VIN configuration.  
VIN and EN are the two signals to enable the part. For start-up sequence, any sequence between the VIN and  
EN signals can power the device up correctly.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
23  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
VIN: 4V œ 16V  
CBOOT  
10  
21  
1
VIN  
VIN  
BOOT  
SW  
CIN  
LOUT  
VOUT  
20  
EN  
8
9
EN  
CFF, Optional  
PGOOD  
PGOOD  
Vosns+  
COUT  
RPG_pullup  
7
FB  
19  
4
VCC  
RFB_HS  
RFB_LS  
CVCC  
RMODE  
MODE  
TRIP  
Vosns-  
RTRIP  
6
5
VSNS-  
3
CSS  
SS/  
REFIN  
2
AGND  
PGND  
7-5. Single VIN Configuration For 12-V Bus  
7.4.4 Powering the Device From a 3.3-V Bus  
The device can also work for up to a 20-A load current when powering from a 3.3-V bus with a single VIN  
configuration. To ensure the internal analog circuitry and the gate drives are powered up properly, VCC pin must  
be shorted to VIN pins with low impedance trace. A trace with at least 24-mil width is recommended. A 2.2-µF, at  
least 6.3-V rating, VCC-to-PGND decoupling capacitor is still recommended to be placed as close as possible to  
VCC pin. Due to the maximum rating limit on VCC pin, the VIN input range under this configuration is 3 V to 3.6  
V. The input voltage must stay higher than both VIN UVLO and VCC UVLO, otherwise the device will shut down  
immediately. 7-6 shows an example for this single VIN configuration.  
VIN and EN are the two signals to enable the part. For start-up sequence, any sequence between the VIN and  
EN signals can power the device up correctly.  
VIN: 3.3V bus  
CBOOT  
10  
21  
1
VIN  
VIN  
BOOT  
SW  
CIN  
LOUT  
VOUT  
20  
EN  
8
9
EN  
CFF, Optional  
PGOOD  
PGOOD  
Vosns+  
COUT  
RPG_pullup  
7
FB  
19  
4
VCC  
RFB_HS  
RFB_LS  
CVCC  
RMODE  
MODE  
TRIP  
Vosns-  
RTRIP  
6
5
VSNS-  
3
CSS  
SS/  
REFIN  
2
AGND  
PGND  
7-6. Single VIN Configuration For 3.3-V Bus  
Copyright © 2022 Texas Instruments Incorporated  
24  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
7.4.5 Powering the Device From a Split-rail Configuration  
When an external bias, which is at a different level from main VIN bus, is applied onto the VCC pin, the device  
can be configured to split-rail by utilizing both the main VIN bus and VCC bias. Connecting a valid VCC bias to  
VCC pin overrides the internal LDO, thus saves power loss on that linear regulator. This configuration helps to  
improve overall system level efficiency but requires a valid VCC bias. A 3.3-V rail is the common choice as VCC  
bias. With a stable VCC bias, the VIN input range under this configuration can be as low as 2.7 V and up to 16 V.  
The noise of the external bias affects the internal analog circuitry. To ensure a proper operation, a clean, low-  
noise external bias and good local decoupling capacitor from VCC pin to PGND pin are required. 7-7 shows  
an example for this split rail configuration.  
The VCC external bias current during nominal operation varies with the bias voltage level and also the operating  
frequency. For example, by setting the device to skip-mode, the VCC pins draw less and less current from the  
external bias when the frequency decreases under light load condition. The typical VCC external bias current  
under FCCM operation is listed in 6.5 to help you prepare the capacity of the external bias.  
Under split rail configuration, VIN, VCC bias, and EN are the signals to enable the part. For start-up sequence,  
TI recommends that at least one of VIN UVLO rising threshold and EN rising threshold is satisfied later than  
VCC UVLO rising threshold. A practical start-up sequence example is: VIN applied first, the external bias  
applied, and then EN signal goes high.  
VIN: 2.7V œ 16V  
CBOOT  
10  
21  
1
VIN  
VIN  
BOOT  
SW  
CIN  
LOUT  
VOUT  
20  
EN  
8
9
EN  
CFF, Optional  
PGOOD  
PGOOD  
Vosns+  
COUT  
RPG_pullup  
VCC bias  
7
FB  
19  
4
VCC  
RFB_HS  
RFB_LS  
RMODE  
CVCC  
MODE  
TRIP  
Vosns-  
RTRIP  
6
5
VSNS-  
3
CSS  
SS/  
REFIN  
2
AGND  
PGND  
7-7. Split Rail Configuration With External VCC Bias  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
25  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8 Application and Implementation  
备注  
以下应用部分中的信息不属TI 器件规格的范围TI 不担保其准确性和完整性。TI 的客 户应负责确定  
器件是否适用于其应用。客户应验证并测试其设计以确保系统功能。  
8.1 Application Information  
The TPS548B28 device is a high-efficiency, single-channel, small-sized, synchronous-buck converter. The  
device suits low output voltage point-of-load applications with 20-A or lower output current in server, storage, and  
similar computing applications. The TPS548B28 features proprietary D-CAP3 control mode combined with  
adaptive on-time architecture. This combination builds modern low-duty-ratio and ultra-fast load-step-response  
DC/DC converters in an ideal fashion. The output voltage ranges from 0.6 V to 5.5 V. The conversion input  
voltage ranges from 2.7 V to 16 V, and the VCC input voltage ranges from 3.13 V to 3.6 V. The D-CAP3 control  
mode uses emulated current information to control the modulation. An advantage of this control scheme is that it  
does not require an external phase-compensation network, which makes the device easy-to-use and also allows  
for a low external component count. Another advantage of this control scheme is that it supports stable operation  
with all low-ESR output capacitors (such as ceramic capacitor and low-ESR polymer capacitor). Adaptive on-  
time control tracks the preset switching frequency over a wide range of input and output voltages while  
increasing switching frequency as needed during a load-step transient.  
8.2 Typical Application  
The schematic shows a typical application for TPS548B28. This example describes the design procedure of  
converting an input voltage range of 8 V to 14 V down to 1 V with a maximum output current of 20 A.  
8-1. TPS548B28 Application Circuit Diagram  
Copyright © 2022 Texas Instruments Incorporated  
26  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8.2.1 Design Requirements  
This design uses the parameters listed in 8-1.  
8-1. Design Example Specifications  
DESIGN PARAMETER  
Voltage range  
CONDITION  
MIN  
TYP  
12  
MAX  
UNIT  
V
VIN  
8
14  
VOUT  
ILOAD  
VRIPPLE  
Output voltage  
1.0  
V
Output load current  
Output voltage ripple  
20  
A
VIN = 12 V, IOUT = 20 A  
IOUT = 25% to 75% step, 2 A/µs slew rate  
10  
mVPP  
Output voltage undershoot  
and overshoot after load  
step  
V TRANS  
+/- 50  
mV  
IOVER  
tSS  
fSW  
Output overcurrent  
Soft-start time  
24  
A
3.7  
ms  
Switching frequency  
800  
kHz  
Operating mode  
FCCM  
25  
TA  
Operating temperature  
°C  
8.2.2 Detailed Design Procedure  
The external component selection is a simple process using D-CAP3 control mode. Select the external  
components using the following steps.  
8.2.2.1 Output Voltage Setting Point  
The output voltage is programmed by the voltage-divider resistors, R1 and R2, shown in 方程式 7. Connect R1  
between the FB pin and the output, and connect R2 between the FB pin and VSNS. The recommended R2  
value is 10 kΩ, but it can also be set to another value between the range of 1 kΩ to 20 kΩ. Determine R1 for  
TPS548B28 by using 方程7.  
«
÷
VOUT - VREF  
VREF  
1.0 V - 0.6 V  
0.6 V  
R = R ì  
= 10 kWì  
= 6.67 kW  
1
2
«
÷
(7)  
8.2.2.2 Choose the Switching Frequency and the Operation Mode  
The switching frequency and operation mode are configured by the resistor on the MODE pin. Select one of  
three switching frequencies: 600 kHz, 800 kHz, or 1 MHz. Refer to 7-1 for the relationship between the  
switching frequency, operation mode, and RMODE  
.
Switching frequency selection is a tradeoff between higher efficiency and smaller system solution size. Lower  
switching frequency yields higher overall efficiency but relatively larger external components. Higher switching  
frequencies cause additional switching losses which impact efficiency and thermal performance. For this design,  
connect the MODE pin to AGND through a 30.1-kΩ resistor to set the switching frequency to 800 kHz and set  
operation mode as FCCM.  
When selecting the switching frequency of a buck converter, the minimum on-time and minimum off-time must  
be considered. 方程式 8 calculates the maximum fSW before being limited by the minimum on-time. When hitting  
the minimum on-time limits of a converter with D-CAP3 control mode, the effective switching frequency will  
change to keep the output voltage regulated. This calculation ignores resistive drops in the converter to give a  
worst case estimation.  
VOUT  
max tON_MIN max  
1
1.0 V  
1
fSW max =  
ì
=
ì
= 840 kHz  
(
)
V
14 V 85 ns  
(
)
(
)
IN  
(8)  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
27  
Product Folder Links: TPS548B28  
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
方程式 8 calculates the maximum fSW before being limited by the minimum off-time. When hitting the minimum  
off-time limits of a converter with D-CAP3 control mode, the operating duty cycle will max out and the output  
voltage will begin to drop with the input voltage. This equation requires the DC resistance of the inductor, RDCR  
,
selected in the following step so this preliminary calculation assumes a resistance of 2.2 mΩ. If operating near  
the maximum fSW limited by the minimum off-time, the variation in resistance across temperature must be  
considered when using 方程9. The selected fSW of 800 kHz is below the two calculated maximum values.  
V
min - V  
-IOUT max ì R  
+ R  
DCR  
DS ON _HS  
(
)
(
)
(
)
(
IN  
OUT  
(
)
fSW max =  
(
)
tOFF_MIN max ì V min -I  
max ì R  
-RDS ON _LS  
(
)
(
)
(
)
(
)
IN  
OUT  
)
(
DS ON _HS  
(
)
)
8 V -1.0 V - 20 A ì 2.2 mW + 7.2 mW  
(
(
)
fSW max =  
= 3918 kHz  
(
)
220 nsì 8 V - 20 A ì 7.2 mW - 2.3 mW  
)
)
(
(9)  
8.2.2.3 Choose the Inductor  
To calculate the value of the output inductor (LOUT), use 方程10. The output capacitor filters the inductor-ripple  
current (IIND(ripple)). Therefore, selecting a high inductor-ripple current impacts the selection of the output  
capacitor because the output capacitor must have a ripple-current rating equal to or greater than the inductor-  
ripple current. On the other hand, larger ripple current increases output ripple voltage, but improves signal-to-  
noise ratio and helps to stabilize operation. Generally speaking, the inductance value must set the ripple current  
at approximately 15% to 40% of the maximum output current for a balanced performance.  
For this design, the inductor-ripple current is set to 30% of 20-A output current. With a 800-kHz switching  
frequency, 14 V as maximum VIN, and 1.0 V as the output voltage, the calculated inductance is 0.290 µH.  
V
max - V  
ì V  
(
)
14 V -1.0 V ì1.0 V  
(
)
(
)
IN  
OUT  
OUT  
L =  
=
= 0.290 H  
IRIPPLE ì V max ì f  
0.2ì 20A ì14 V ì800 kHz  
(
)
IN  
SW  
(10)  
The inductor requires a low DCR to achieve good efficiency. The inductor also requires enough room above  
peak inductor current before saturation. The peak inductor current is estimated using 方程式 11. For this design,  
by selecting 6.04 kΩ as the RTRIP, IOC(valley) is set to 20 A, thus peak inductor current under maximum VIN is  
calculated as 3.869 A.  
V
max - V  
ì V  
(
)
14 V -1.0 V ì1.0 V  
(
)
(
)
IN  
OUT  
OUT  
IRIPPLE  
=
=
= 3.869 A  
Lì V max ì f  
0.3Hì14 V ì800 kHz  
(
)
IN  
SW  
(11)  
(12)  
IRIPPLE  
3.869 A  
2
IL PEAK = IOUT  
+
= 20 A +  
= 21.93 A  
(
)
2
2
3.869 A2  
IRIPPLE  
2
IL RMS = IOUT  
+
=
20 A2 +  
= 20.03 A  
(
)
12  
12  
(13)  
The selected inductance is a Coilcraft XAL7070-301MEB. This has a saturation current rating of 55.6 A, RMS  
current rating of 26.1 A, and a DCR of 1.17-mΩmax. This inductor was selected for its low DCR to get high  
efficiency.  
8.2.2.4 Set the Current Limit (TRIP)  
The RTRIP resistor sets the valley current limit. 方程式 14 calculates the recommended current limit target as  
18.18 A. The value was rounded up to 20 A. Use 方程式 15 to calculate the RTRIP resistor to set the current limit.  
The typical valley current limit target is 20 A and the closest standard value for RTRIP is 6.04 kΩ.  
Copyright © 2022 Texas Instruments Incorporated  
28  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
V
min - V  
ì V  
(
)
(
)
OUT OUT  
1
2
IN  
÷
÷
ILIM_ VALLEY = IOUT  
-
ì
ì
Lì V min ì f  
(
)
IN  
SW  
«
8 V -1.0 V ì1.0 V  
(
)
1
2
ILIM_ VALLEY = 20A -  
= 18.18 A  
÷
÷
«
0.3 µHì8 V ì800 kHz  
(14)  
(15)  
120000  
120000  
20 A  
RTRIP  
=
=
= 6.0 kW  
ILIM_ VALLEY  
With the current limit set, 方程式 16 calculates the typical maximum output current at current limit. 方程式 17  
calculates the typical peak current at current limit. As mentioned in 8.2.2.3, the saturation behavior of the  
inductor at the peak current during current limit must be considered. For worst case calculations, the tolerance of  
the inductance and the current limit must be included.  
V
min - V  
ì V  
(
)
8 V -1.0 V ì1.0 V  
(
)
OUT OUT  
(
)
1
2
1
IN  
IOUT_LIM min = I  
+
ì
= 20 A +  
ì
= 21.82 A  
(
)
LIM_ VALLEY  
Lì V min ìf  
2 0.3 µHì8 Vì800 kHz  
(
)
IN  
SW  
(16)  
V
(
max - V  
ì V  
OUT  
(
)
14 V -1.0 V ì1.0 V  
)
(
)
IN  
OUT  
IL PEAK = ILIM_ VALLEY  
+
= 20 A +  
= 21.935 A  
(
)
Lì V max ì f  
0.3 µHì14 V ì800 kHz  
(
)
IN  
SW  
(17)  
8.2.2.5 Choose the Output Capacitor  
There are three considerations for selecting the value of the output capacitor:  
1. Stability  
2. Steady state output voltage ripple  
3. Regulator transient response to a change load current  
First, the minimum output capacitance must be calculated based on these three requirements. 程式 18  
calculates the minimum capacitance to keep the LC double pole below 1/30th the fSW to meet stability  
requirements. This requirement helps keep the LC double pole close to the internal zero. 方程式 19 calculates  
the minimum capacitance to meet the steady state output voltage ripple requirement of 10 mV. This calculation is  
for CCM operation and does not include the portion of the output voltage ripple caused by the ESR or ESL of the  
output capacitors.  
2
2
«
÷
30  
1
L
30  
1
COUT _STABILITY  
>
ì
=
ì
= 118.7 µF  
«
÷
2pì fSW ◊  
2pì800 kHz  
0.3 µH  
(18)  
(19)  
IRIPPLE  
3.869 A  
COUT _RIPPLE  
>
=
= 60.5 µF  
8ì VRIPPLE ì fSW 8ì10 mV ì800 kHz  
方程式 20 and 方程式 21 calculate the minimum capacitance to meet the transient response requirement of 50  
mV with a 10-A step. These equations calculate the necessary output capacitance to hold the output voltage  
steady while the inductor current ramps up or ramps down after a load step.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
29  
Product Folder Links: TPS548B28  
 
 
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
VOUT  
LìISTEP2 ì  
+ tOFF_MIN max  
(
)
÷
÷
V
min ì f  
(
)
IN  
SW  
«
COUT _UNDERSHOOT  
>
>
V
min - V  
(
IN  
)
IN  
OUT  
2ì VTRANS ì VOUT  
0.3 µHì10 A2 ì  
ì
- tOFF_MIN max  
(
)
÷
÷
V
min ì f  
(
1.0 V  
)
SW  
«
+ 220 ns  
÷
8 V ì800 kHz  
8 V -1.0 V  
8 V ì800 kHz  
«
COUT _UNDERSHOOT  
= 129.2 µF  
2ì50 mV ì1.0 V ì  
- 220 ns  
«
÷
(20)  
(21)  
2
0.3 µHì10A2  
LìISTEP  
COUT _OVERSHOOT  
>
=
= 300 µF  
2ì VTRANS ì VOUT 2ì50 mV ì1.0 V  
The output capacitance needed to meet the overshoot requirement is the highest value so this sets the required  
minimum output capacitance for this example. Stability requirements can also limit the maximum output  
capacitance and 方程式 22 calculates the recommended maximum output capacitance. This calculation keeps  
the LC double pole above 1/100th the fSW. It can be possible to use more output capacitance but the stability  
must be checked through a bode plot or transient response measurement. The selected output capacitance is 8  
x 47-µF 6.3-V ceramic capacitors. When using ceramic capacitors, the capacitance must be derated due to DC  
and AC bias effects. The selected capacitors derate to 85% their nominal value giving an effective total  
capacitance of 320 μF. This effective capacitance meets the minimum and maximum requirements.  
2
2
«
÷
50  
1
L
50  
1
COUT _STABILITY  
<
ì
=
ì
= 1319.3 µF  
«
÷
fSW ◊  
800 kHz  
0.3 µH  
(22)  
This application uses all ceramic capacitors so the effects of ESR on the ripple and transient were ignored. If  
using non-ceramic capacitors, as a starting point, the ESR must be below the values calculated in 方程式 23 to  
meet the ripple requirement and 方程式 24 to meet the transient requirement. For more accurate calculations or  
if using mixed output capacitors, the impedance of the output capacitors must be used to determine if the ripple  
and transient requirements can be met.  
VRIPPLE  
10 mV  
RESR_RIPPLE  
<
=
= 2.58 mW  
IRIPPLE  
3.869 A  
(23)  
(24)  
VTRANS  
50 mV  
RESR _ TRANS  
<
=
= 5 mW  
ISTEP  
10 A  
8.2.2.6 Choose the Input Capacitors (CIN)  
The device requires input bypass capacitors between the VIN and PGND pins to bypass the power-stage. The  
bypass capacitors must be placed as close as possible to the pins of the IC as the layout will allow. At least 10  
µF of ceramic capacitance and 1-µF high frequency ceramic bypass capacitors are required. A 1-μF, 16-V X6S  
size 0402 ceramic capacitor on VIN pin 10 and pin 21 is required. A 1-μF 16-V X6S ceramic capacitor on the  
bottom layer is recommended for high current applications. The high frequency bypass capacitor minimizes high  
frequency voltage overshoot across the power-stage. The ceramic capacitors must be a high-quality dielectric of  
X6S or better for their high capacitance-to-volume ratio and stable characteristics across temperature. In  
addition to this, more bulk capacitance can be needed on the input depending on the application to minimize  
variations on the input voltage during transient conditions.  
The input capacitance required to meet a specific input ripple target can be calculated with 方程式 25. A  
recommended target input voltage ripple is 5% the minimum input voltage, which is 400 mV in this example. The  
Copyright © 2022 Texas Instruments Incorporated  
30  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
calculated input capacitance needed is 6.84 μF and four 22-µF ceramic capacitors are recommended for this  
example.  
VOUT  
min  
1.0V  
8
«
VOUT ìIOUT ì 1-  
÷
÷
1.0 V ì 20 A ì 1-  
÷
V
(
)
IN  
«
CIN  
>
=
= 6.84 F  
fSW ì V min ì V  
800 kHzì8 V ì 400 mV  
(
)
IN  
IN_RIPPLE  
(25)  
The capacitor must also have an RMS current rating greater than the maximum input RMS current in the  
application. The input RMS current the input capacitors must support is calculated by 方程式 26 and is 6.625 A  
in this example. The ceramic input capacitors have a current rating greater than this.  
Iripple2  
12  
V
min - V  
OUT  
(
)
min  
(
)
ìIOUT  
VOUT  
min  
IN  
2
«
÷
÷
ICIN RMS  
=
=
ì
+
=
(
)
)
V
V
(
)
(
)
IN  
IN  
3.6462  
12  
8 V -1.0 V  
(
)
ì 202 +  
1.0 V  
8 V  
ICIN RMS  
ì
= 6.625 A  
÷
÷
(
8 V  
«
(26)  
For applications requiring bulk capacitance on the input, such as ones with low input voltage and high current,  
the selection process in this article is recommended.  
8.2.2.7 Soft Start Capacitor (SS/REFIN Pin)  
The capacitor placed on the SS/REFIN pin can be used to extend the soft-start time past the internal 1.5-ms soft  
start. This example uses a 3.7-ms soft-start time and the required external capacitance can be calculated with 方  
27. In this example, a 220-nF capacitor is used.  
ISS ì tSS  
VREF  
36 µA ì3.7 ms  
0.6 V  
CSS  
=
=
= 200 nF  
(27)  
A minimum capacitor value of 1 nF is required at the SS/REFIN pin. The SS/REFIN capacitor must use the  
VSNSpin for its ground.  
8.2.2.8 EN Pin Resistor Divider  
A resistor divider on the EN pin can be used to increase the input voltage the converter begins its start-up  
sequence. To set the start voltage, first select the bottom resistor (REN_B). The recommended value is between 1  
kΩand 100 kΩ. There is an internal pulldown resistance with a nominal value of 6 MΩ, which must be included  
for the most accurate calculations. This is especially important when the bottom resistor is a higher value, near  
100 kΩ. This example uses a 10-kΩ resistor and combined with the internal resistance in parallel results in an  
equivalent bottom resistance of 9.98 kΩ. The top resistor value for the target start voltage is calculated with 方程  
28. In this example, the nearest standard value of 20 kΩis selected for REN_T. When selecting a start voltage  
in a wide input range application, be cautious that the EN pin absolute maximum voltage of 6 V is not exceeded.  
REN_B ì VSTART  
10 kWì3.7 V  
1.22 V  
REN_ T  
=
-REN_B  
=
-10 kW = 20 kW  
VENH  
(28)  
The start and stop voltages with the selected EN resistor divider can be calculated with 方程式 29 and 方程式  
30.  
REN_B + REN_ T  
10 kW + 20 kW  
10 kW  
VSTART = VENH  
ì
= 1.22 V ì  
= 3.66V  
REN_B  
(29)  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
31  
Product Folder Links: TPS548B28  
 
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
REN_B + REN_ T  
10 kW + 20 kW  
10 kW  
VSTOP = VENL  
ì
= 1.02 V ì  
= 3.06 V  
REN_B  
(30)  
8.2.2.9 VCC Bypass Capacitor  
At a minimum, a 2.2-µF, at least 6.3-V rating, X5R ceramic bypass capacitor is needed on VCC pin located as  
close to the pin as the layout will allow.  
8.2.2.10 BOOT Capacitor  
At a minimum, a 0.1-µF 10-V X5R ceramic bypass capacitor is needed between the BOOT and SW pins located  
as close to the pin as the layout will allow. It is good practice to use a 0-Ω resistor in series with the BOOT  
capacitor.  
8.2.2.11 PGOOD Pullup Resistor  
The PGOOD pin is open-drain so a pullup resistor is required when using this pin. The recommended value is  
between 1 kΩand 100 kΩ.  
Copyright © 2022 Texas Instruments Incorporated  
32  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8.2.3 Application Curves  
90  
88  
86  
84  
82  
80  
100  
96  
92  
88  
84  
80  
76  
72  
68  
64  
60  
78  
Vin = 12 V  
800 kHz  
300 nH  
VCC = INT  
Skip  
Vin = 12 V  
800 kHz  
300 nH  
VCC = INT  
FCCM  
76  
74  
72  
70  
Vout = 1.0V  
Vout = 1.0 V  
16 18 20  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
0
2
4
6
8
10  
Output Current (A)  
12  
14  
D002  
D001  
8-3. Efficiency vs Output Current, Internal VCC  
8-2. Efficiency vs Output Current, Internal VCC  
Bias  
Bias  
100  
96  
92  
88  
84  
80  
1000  
920  
840  
760  
680  
600  
520  
76  
VIN = 12 V  
Vout = 1.0 V  
800 nH  
VCC = Int  
CCM  
Vin = 12 V  
800 kHz  
300 nH  
VCC = 3.3 V  
FCCM  
440  
360  
280  
200  
72  
68  
64  
60  
600 kHz  
800 kHz  
1000 kHz  
Vout = 1.0 V  
16 18 20  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
0
2
4
6
8
10  
Output Current (A)  
12  
14  
D005  
D003  
8-5. Switching Frequency vs Output Current  
8-4. Efficiency vs Output Current, VCC = 3.3V  
External VCC Bias  
1.005  
1100  
1000  
900  
Vin = 8V  
Vin = 12V  
Vin = 16V  
300 nH  
1.00375  
1.0025  
1.00125  
1
800 kHz  
VCC = Int  
Skip  
800  
700  
600  
0.99875  
0.9975  
0.99625  
0.995  
500  
VIN = 12 V  
Vout = 1.0 V  
800 nH  
VCC = Int  
CCM  
400  
300  
200  
100  
600 kHz  
800 kHz  
1000 kHz  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
5
6
7
8
9
10 11 12 13 14 15 16  
D008  
Input Voltage (V)  
D006  
8-7. Output Voltage vs Output Current  
8-6. Switching Frequency vs Input Voltage  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
33  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
1.005  
1.005  
1.00375  
1.0025  
1.00125  
1
Vin = 8V  
Vin = 12V  
Vin = 16V  
Vin = 8V  
Vin = 12V  
Vin = 16V  
300 nH  
300 nH  
800 kHz  
VCC = Int  
FCCM  
800 kHz  
VCC = 3.3 V  
Skip  
1.00375  
1.0025  
1.00125  
1
0.99875  
0.9975  
0.99625  
0.995  
0.99875  
0.9975  
0.99625  
0.995  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
D010  
D011  
8-8. Output Voltage vs Output Current  
8-9. Output Voltage vs Output Current, VCC =  
3.3-V External Bias  
1.005  
1.01  
Vin = 8V  
Vin = 12V  
Vin = 16V  
Skip 0 A  
FCCM 0 A  
FCCM 5 A  
FCCM 10 A  
FCCM 20A  
300 nH  
800 kHz  
VCC = 3.3 V  
FCCM  
1.00375  
1.0025  
1.00125  
1
1.006  
1.002  
0.998  
0.994  
0.99  
0.99875  
0.9975  
0.99625  
0.995  
0
2
4
6
8
10  
12  
Output Current (A)  
14  
16  
18  
20  
5
7
9 11  
Input Voltage (V)  
13  
15 16  
D012  
D013  
8-10. Output Voltage vs Output Current, VCC =  
8-11. Output Voltage vs Input Voltage VCC = Int  
3.3-V External Bias  
1.01  
0.018  
0.016  
0.014  
0.012  
0.01  
Skip 0 A  
FCCM 0 A  
FCCM 5 A  
FCCM 10 A  
FCCM 20A  
1.006  
1.002  
0.998  
0.994  
0.99  
0.008  
0.006  
VIN = 12 V  
Vout = 3.3 V  
Iout = 5 A  
800 nH  
0.004  
0.002  
0
600 kHz  
800 kHz  
1000 kHz  
5
7
9 11  
Input Voltage (V)  
13  
15 16  
3.2 3.24 3.28 3.32 3.36 3.4 3.44 3.48 3.52 3.56 3.6  
VCC Voltage (V)  
D014  
D015  
8-12. Output Voltage vs Input Voltage VCC = 3.3-  
8-13. ICC Current vs External VCC Voltage  
V External Bias  
Copyright © 2022 Texas Instruments Incorporated  
34  
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8-14. Enable Start-Up, Prebias  
8-15. Prebias Power-Down  
8-16. Enable Power Up, Skip  
8-17. Enable Power Down, Skip  
8-18. Enable Power Up into Pre-bias, Skip  
8-19. Enable Power Down with Pre-bias, Skip  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
35  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8-20. Load Transient  
8-22. Output Voltage Ripple  
8-24. Output Voltage Ripple  
8-26. Output Voltage Ripple, Skip  
8-21. Unload Transient  
8-23. Output Voltage Ripple  
8-25. Output Voltage Ripple, Skip  
8-27. Over Temperature Protection in FCCM  
Copyright © 2022 Texas Instruments Incorporated  
36  
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8-29. Enabled into Overcurrent  
8-28. Exiting Overtemperature Protection in  
FCCM  
8-30. Hiccup Overcurrent  
8.3 Power Supply Recommendations  
The device is designed to operate from a wide input voltage supply range between 2.7 V and 16 V when VCC  
pin is powered by external bias ranging from 3.13 V to 3.6 V. Both input supplies (VIN and VCC bias) must be  
well regulated. Proper bypassing of input supplies (VIN and VCC bias) is also critical for noise performance, as  
are PCB layout and grounding scheme. See the recommendations in 8.4.  
8.4 Layout  
8.4.1 Layout Guidelines  
Before beginning a design using the device, consider the following:  
Place the power components (including input and output capacitors, the inductor, and the IC) on the top side  
of the PCB. To shield and isolate the small signal traces from noisy power lines, insert at least one solid  
ground inner plane.  
VIN decoupling capacitors are important for FET robustness. A 1-μF/25-V/X6S/0402 ceramic capacitor on  
VIN pin 21 is required. The PGND vias for this decoupling capacitor must be placed so that the decoupling  
capacitor is closer to IC than the PGND vias. To lower ESL from via connection, two 8-mil vias are  
recommended for the PGND connection to inner PGND plane.  
A 1-μF/25-V/X6S/0402 ceramic capacitor on VIN pin 10 is highly recommended. If this 0402 size capacitor is  
not used, the bigger size VIN decoupling capacitors (0603 or 0805 size) are required to be placed as close as  
possible to IC pin 10 and pin 11.  
Two 1-μF/25-V/X6S/0402 ceramic capacitors on the bottom layer are recommended for high current  
applications (Iout > 13 A). One of these two capacitors must be centered between VIN pin 10 and pin 21. To  
have good connection for this capacitor, a VIN copper on bottom layer and two VIN vias are needed. The  
other one can be placed close to IC package just like a mirrored copy to the 0402 capacitor on top layer.  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
37  
Product Folder Links: TPS548B28  
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
At least six PGND vias are required to be placed as close as possible to the PGND pins (pin 11 to pin 15).  
This minimizes parasitic impedance and also lowers thermal resistance.  
Place the VCC decoupling capacitor (2.2-μF/6.3-V/X6S/0402 or 2.2-μF/6.3-V/X7R/0603) as close as  
possible to the device. Ensure the VCC decoupling loop is smallest.  
Place BOOT capacitor as close as possible to the BOOT and SW pins. Use traces with a width of 12 mil or  
wider to route the connection. TI recommends using a 0.1-µF to 1-µF bootstrap capacitor with 10-V rating.  
The PCB trace, which connects the SW pin and high-voltage side of the inductor, is defined as switch node.  
The switch node must be as short and wide as possible.  
Always place the feedback resistors near the device to minimize the FB trace distance, no matter single-end  
sensing or remote sensing.  
For remote sensing, the connections from the FB voltage divider resistors to the remote location must be a  
pair of PCB traces with at least 12-mil trace width, and must implement Kelvin sensing across a high  
bypass capacitor of 0.1 μF or higher. The ground connection of the remote sensing signal must be  
connected to VSNSpin. The VOUT connection of the remote sensing signal must be connected to the  
feedback resistor divider with the lower feedback resistor terminated at VSNSpin. To maintain stable  
output voltage and minimize the ripple, the pair of remote sensing lines must stay away from any noise  
sources such as inductor and SW nodes, or high frequency clock lines. TI recommends to shield the pair  
of remote sensing lines with ground planes above and below.  
For single-end sensing, connect the higher FB resistor to a high-frequency local bypass capacitor of 0.1  
μF or higher, and short VSNSto AGND with shortest trace.  
This device does not require a capacitor from SS/REFIN pin to AGND, thus TI does not recommend to place  
a capacitor from SS/REFIN pin to AGND. If both CSS/REFIN-to-VSNSand CSS/REFIN-to-AGND capacitors  
exist, place CSS/REFIN-to-VSNSmore closely with shortest trace to VSNSpin.  
Pin 2 (AGND pin) must be connected to a solid PGND plane on inner layer. Use the common AGND via to  
connect the resistors to the inner ground plane if applicable.  
See 8.4.2 for the layout recommendation.  
Copyright © 2022 Texas Instruments Incorporated  
38  
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8.4.2 Layout Example  
Vosns+  
Vosns-  
Via down to connect to solid  
PGND plane on inner layer  
8/20  
8/20  
0402  
0402  
8/20  
8/20  
2x VIN-to-PGND decoupling  
capacitors on bottom layer  
8/20  
PGND  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
SW  
VIN  
VIN  
SW  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
0603  
LOUT  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
0402  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
8/20  
PGND  
0805  
0805  
VOUT  
8-31. Layout Recommendation  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
39  
Product Folder Links: TPS548B28  
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
8.4.2.1 Thermal Performance On TI EVM  
Test conditions:  
fSW = 800-kHz, VIN = 12-V, VCC = Int LDO, VOUT = 1-V, IOUT = 20-A, Inductor LOUT = 0.3-µH (1.06-mΩ typ),  
COUT= 8 × 47-µF (1206/6.3V/X6S), no RBOOT, no RC Snubber  
SP1 (IC): 88.0°C, SP2 (Inductor): 58.7°C  
8-32. Thermal Image At 25°C Ambient  
Copyright © 2022 Texas Instruments Incorporated  
40  
Submit Document Feedback  
Product Folder Links: TPS548B28  
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
9 Device and Documentation Support  
9.1 Documentation Support  
9.1.1 Related Documentation  
Texas Instruments, Optimizing Transient Response of Internally Compensated DC-DC Converters with  
Feedforward Capacitor application note  
Texas Instruments, Non-isolated Point-of-load Solutions for VR13.HC in Rack Server and Datacenter  
Applications application note  
9.2 接收文档更新通知  
要接收文档更新通知请导航至 ti.com 上的器件产品文件夹。点击订阅更新 进行注册即可每周接收产品信息更  
改摘要。有关更改的详细信息请查看任何已修订文档中包含的修订历史记录。  
9.3 支持资源  
TI E2E支持论坛是工程师的重要参考资料可直接从专家获得快速、经过验证的解答和设计帮助。搜索现有解  
答或提出自己的问题可获得所需的快速设计帮助。  
链接的内容由各个贡献者“按原样”提供。这些内容并不构成 TI 技术规范并且不一定反映 TI 的观点请参阅  
TI 《使用条款》。  
9.4 Trademarks  
D-CAP3, Eco-mode, and TI E2Eare trademarks of Texas Instruments.  
所有商标均为其各自所有者的财产。  
9.5 Electrostatic Discharge Caution  
This integrated circuit can be damaged by ESD. Texas Instruments recommends that all integrated circuits be handled  
with appropriate precautions. Failure to observe proper handling and installation procedures can cause damage.  
ESD damage can range from subtle performance degradation to complete device failure. Precision integrated circuits may  
be more susceptible to damage because very small parametric changes could cause the device not to meet its published  
specifications.  
9.6 术语表  
TI 术语表  
本术语表列出并解释了术语、首字母缩略词和定义。  
Copyright © 2022 Texas Instruments Incorporated  
Submit Document Feedback  
41  
Product Folder Links: TPS548B28  
 
 
 
 
 
 
 
TPS548B28  
www.ti.com.cn  
ZHCSMV4A DECEMBER 2020 REVISED DECEMBER 2022  
10 Mechanical, Packaging, and Orderable Information  
The following pages include mechanical, packaging, and orderable information. This information is the most  
current data available for the designated devices. This data is subject to change without notice and revision of  
this document. For browser-based versions of this data sheet, refer to the left-hand navigation.  
Copyright © 2022 Texas Instruments Incorporated  
42  
Submit Document Feedback  
Product Folder Links: TPS548B28  
 
PACKAGE OPTION ADDENDUM  
www.ti.com  
1-Feb-2023  
PACKAGING INFORMATION  
Orderable Device  
Status Package Type Package Pins Package  
Eco Plan  
Lead finish/  
Ball material  
MSL Peak Temp  
Op Temp (°C)  
Device Marking  
Samples  
Drawing  
Qty  
(1)  
(2)  
(3)  
(4/5)  
(6)  
TPS548B28RWWR  
ACTIVE  
VQFN-HR  
RWW  
21  
3000 RoHS & Green  
NIPDAU | SN  
Level-2-260C-1 YEAR  
-40 to 125  
T548B8A  
Samples  
(1) The marketing status values are defined as follows:  
ACTIVE: Product device recommended for new designs.  
LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.  
NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in a new design.  
PREVIEW: Device has been announced but is not in production. Samples may or may not be available.  
OBSOLETE: TI has discontinued the production of the device.  
(2) RoHS: TI defines "RoHS" to mean semiconductor products that are compliant with the current EU RoHS requirements for all 10 RoHS substances, including the requirement that RoHS substance  
do not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered at high temperatures, "RoHS" products are suitable for use in specified lead-free processes. TI may  
reference these types of products as "Pb-Free".  
RoHS Exempt: TI defines "RoHS Exempt" to mean products that contain lead but are compliant with EU RoHS pursuant to a specific EU RoHS exemption.  
Green: TI defines "Green" to mean the content of Chlorine (Cl) and Bromine (Br) based flame retardants meet JS709B low halogen requirements of <=1000ppm threshold. Antimony trioxide based  
flame retardants must also meet the <=1000ppm threshold requirement.  
(3) MSL, Peak Temp. - The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder temperature.  
(4) There may be additional marking, which relates to the logo, the lot trace code information, or the environmental category on the device.  
(5) Multiple Device Markings will be inside parentheses. Only one Device Marking contained in parentheses and separated by a "~" will appear on a device. If a line is indented then it is a continuation  
of the previous line and the two combined represent the entire Device Marking for that device.  
(6)  
Lead finish/Ball material - Orderable Devices may have multiple material finish options. Finish options are separated by a vertical ruled line. Lead finish/Ball material values may wrap to two  
lines if the finish value exceeds the maximum column width.  
Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is provided. TI bases its knowledge and belief on information  
provided by third parties, and makes no representation or warranty as to the accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and  
continues to take reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on incoming materials and chemicals.  
TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited information may not be available for release.  
In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI to Customer on an annual basis.  
Addendum-Page 1  
PACKAGE MATERIALS INFORMATION  
www.ti.com  
20-Apr-2023  
TAPE AND REEL INFORMATION  
REEL DIMENSIONS  
TAPE DIMENSIONS  
K0  
P1  
W
B0  
Reel  
Diameter  
Cavity  
A0  
A0 Dimension designed to accommodate the component width  
B0 Dimension designed to accommodate the component length  
K0 Dimension designed to accommodate the component thickness  
Overall width of the carrier tape  
W
P1 Pitch between successive cavity centers  
Reel Width (W1)  
QUADRANT ASSIGNMENTS FOR PIN 1 ORIENTATION IN TAPE  
Sprocket Holes  
Q1 Q2  
Q3 Q4  
Q1 Q2  
Q3 Q4  
User Direction of Feed  
Pocket Quadrants  
*All dimensions are nominal  
Device  
Package Package Pins  
Type Drawing  
SPQ  
Reel  
Reel  
A0  
B0  
K0  
P1  
W
Pin1  
Diameter Width (mm) (mm) (mm) (mm) (mm) Quadrant  
(mm) W1 (mm)  
TPS548B28RWWR  
TPS548B28RWWR  
VQFN-  
HR  
RWW  
RWW  
21  
21  
3000  
3000  
330.0  
12.4  
3.3  
4.3  
1.1  
8.0  
12.0  
Q1  
VQFN-  
HR  
330.0  
12.4  
3.3  
4.3  
1.1  
8.0  
12.0  
Q1  
Pack Materials-Page 1  
PACKAGE MATERIALS INFORMATION  
www.ti.com  
20-Apr-2023  
TAPE AND REEL BOX DIMENSIONS  
Width (mm)  
H
W
L
*All dimensions are nominal  
Device  
Package Type Package Drawing Pins  
SPQ  
Length (mm) Width (mm) Height (mm)  
TPS548B28RWWR  
TPS548B28RWWR  
VQFN-HR  
VQFN-HR  
RWW  
RWW  
21  
21  
3000  
3000  
367.0  
346.0  
367.0  
346.0  
35.0  
33.0  
Pack Materials-Page 2  
PACKAGE OUTLINE  
VQFN-HR - 1 mm max height  
PLASTIC QUAD FLAT-NO LEAD  
RWW0021A  
3.1  
2.9  
A
B
4.1  
3.9  
PIN 1  
INDEX AREA  
C0.15  
PIN 1 IDENTIFICATION  
(OPTIONAL)  
1 MAX  
C
SEATING PLANE  
0.08 C  
0.05  
0.00  
PKG  
0.4  
0.3  
0.35  
0.25  
REF 0.15  
3X  
10  
(0.2) TYP  
9
8
11  
12X 0.4  
1.8  
3X  
1.6  
3.25  
3.05  
2X  
3.4  
0.25  
PKG  
14X  
2X  
2.4  
0.15  
0.1  
C A B  
C
0.05  
2
1
18  
19  
4X 0.5  
21  
0.35  
0.25  
20  
4X  
PIN 1 ID  
(OPTIONAL)  
3X 0.3  
0.1  
C A B  
C
0.6  
0.4  
18X  
0.05  
4223950/C 04/2019  
NOTES:  
1. All linear dimensions are in millimeters. Any dimensions in parenthesis are for reference only. Dimensioning and tolerancing  
per ASME Y14.5M.  
2. This drawing is subject to change without notice.  
www.ti.com  
EXAMPLE BOARD LAYOUT  
VQFN-HR - 1 mm max height  
PLASTIC QUAD FLAT-NO LEAD  
RWW0021A  
(2.7)  
3X (0.3)  
18X (0.7)  
4X (0.3)  
20  
21  
1
2
19  
14X (0.2)  
4X (0.5)  
18  
EXPOSED METAL  
PKG  
2X  
2X  
(3.4) (2.4)  
3X  
(1.9)  
12  
8
9
12X (0.4)  
11  
(0.35)  
(R0.05) TYP  
10  
3X (0.3)  
PKG  
LAND PATTERN EXAMPLE  
SOLDER MASK DEFINED  
SCALE: 20X  
0.05 MAX  
ALL AROUND  
0.05 MAX  
ALL AROUND  
EXPOSED METAL  
EXPOSED METAL  
SOLDER MASK  
OPENING  
SOLDER MASK  
OPENING  
METAL UNDER  
SOLDER MASK  
METAL EDGE  
NON SOLDER MASK  
DEFINED  
SOLDER MASK  
DEFINED  
(PREFERRED)  
SOLDER MASK DETAILS  
4223950/C 04/2019  
NOTES: (continued)  
3. For more information, see Texas Instruments literature number SLUA271 (www.ti.com/lit/slua271).  
4. Solder mask tolerances between and around signal pads can vary based on board fabrication site.  
www.ti.com  
EXAMPLE STENCIL DESIGN  
VQFN-HR - 1 mm max height  
PLASTIC QUAD FLAT-NO LEAD  
RWW0021A  
(2.7)  
PKG  
18X (0.7)  
4X (0.3)  
20  
21  
1
2
19  
4X (0.5)  
18  
14X (0.2)  
PKG  
2X  
2X  
(3.4) (2.4)  
3X  
(0.73)  
3X  
(1.05)  
12  
11  
8
9
12X (0.4)  
(0.35)  
(R0.05) TYP  
6X  
(0.85)  
10  
6X (0.3)  
(0.3)  
SOLDER PASTE EXAMPLE  
BASED ON 0.1mm THICK STENCIL  
EXPOSED PAD  
89% PRINTED COVERAGE BY AREA  
SCALE: 20X  
4223950/C 04/2019  
NOTES: (continued)  
5.  
Laser cutting apertures with trapezoidal walls and rounded corners may offer better paste release. IPC-7525 may have alternate  
design recommendations.  
www.ti.com  
重要声明和免责声明  
TI“按原样提供技术和可靠性数据(包括数据表)、设计资源(包括参考设计)、应用或其他设计建议、网络工具、安全信息和其他资源,  
不保证没有瑕疵且不做出任何明示或暗示的担保,包括但不限于对适销性、某特定用途方面的适用性或不侵犯任何第三方知识产权的暗示担  
保。  
这些资源可供使用 TI 产品进行设计的熟练开发人员使用。您将自行承担以下全部责任:(1) 针对您的应用选择合适的 TI 产品,(2) 设计、验  
证并测试您的应用,(3) 确保您的应用满足相应标准以及任何其他功能安全、信息安全、监管或其他要求。  
这些资源如有变更,恕不另行通知。TI 授权您仅可将这些资源用于研发本资源所述的 TI 产品的应用。严禁对这些资源进行其他复制或展示。  
您无权使用任何其他 TI 知识产权或任何第三方知识产权。您应全额赔偿因在这些资源的使用中对 TI 及其代表造成的任何索赔、损害、成  
本、损失和债务,TI 对此概不负责。  
TI 提供的产品受 TI 的销售条款ti.com 上其他适用条款/TI 产品随附的其他适用条款的约束。TI 提供这些资源并不会扩展或以其他方式更改  
TI 针对 TI 产品发布的适用的担保或担保免责声明。  
TI 反对并拒绝您可能提出的任何其他或不同的条款。IMPORTANT NOTICE  
邮寄地址:Texas Instruments, Post Office Box 655303, Dallas, Texas 75265  
Copyright © 2023,德州仪器 (TI) 公司  

相关型号:

TPS548B28RWWR

具有遥感功能和断续限流功能的 2.7V 至 16V 输入 20A 同步降压转换器 | RWW | 21 | -40 to 125
TI

TPS548C26

采用 5mm x 6mm QFN 封装的 4V 至 16V、35A 同步 D-CAP+™ 降压转换器
TI

TPS548C26RXXR

采用 5mm x 6mm QFN 封装的 4V 至 16V、35A 同步 D-CAP+™ 降压转换器 | RXX | 37 | -40 to 125
TI

TPS548D21

具有 AVSO 和差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器
TI

TPS548D21RVFR

具有 AVSO 和差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器 | RVF | 40 | -40 to 125
TI

TPS548D21RVFT

具有 AVSO 和差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器 | RVF | 40 | -40 to 125
TI

TPS548D22

具有差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器
TI

TPS548D22RVFR

具有差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器 | RVF | 40 | -40 to 125
TI

TPS548D22RVFT

具有差分遥感功能的 1.5V 至 16V、40A 同步 SWIFT™ 降压转换器 | RVF | 40 | -40 to 125
TI

TPS54900

PROGRAMMABLE FOUR-CHANNEL STEP-DOWN DC/DC CONVERTER
TI

TPS54900PW

PROGRAMMABLE FOUR-CHANNEL STEP-DOWN DC/DC CONVERTER
TI

TPS54900PWR

PROGRAMMABLE FOUR-CHANNEL STEP-DOWN DC/DC CONVERTER
TI