XMG3522R030QRQSTQ1 [TI]
具有集成驱动器、保护和温度报告功能的汽车类 650V 30mΩ GaN FET | RQS | 52 | -40 to 150;型号: | XMG3522R030QRQSTQ1 |
厂家: | TEXAS INSTRUMENTS |
描述: | 具有集成驱动器、保护和温度报告功能的汽车类 650V 30mΩ GaN FET | RQS | 52 | -40 to 150 驱动 驱动器 |
文件: | 总41页 (文件大小:2549K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
LMG3522R030-Q1 650V 30mΩ 具有集成驱动器、保护和温度报告功能的GaN
FET
1 特性
3 说明
• 符合面向汽车应用的AEC-Q100 标准
LMG3522R030-Q1 GaN FET 具有集成式驱动器和保
护功能,适用于开关模式电源转换器,可让设计人员实
现更高水平的功率密度和效率。
– 温度等级1:–40°C 至+125°C,TA
– 结温:–40°C 至+150°C,TJ
• 具有集成栅极驱动器的650V GaN-on-Si FET
LMG3522R030-Q1 集成了一个硅驱动器,可实现高达
150V/ns 的开关速度。与分立式硅栅极驱动器相比,TI
的集成式精密栅极偏置可实现更高的开关 SOA。这种
集成特性与 TI 的低电感封装技术相结合,可在硬开关
电源拓扑中提供干净的开关和超小的振铃。可调栅极驱
动强度允许将压摆率控制在20V/ns 至150V/ns 之间,
这可用于主动控制EMI 并优化开关性能。
– 集成高精度栅极偏置电压
– 200V/ns FET 释抑
– 23.6MHz 开关频率
– 20V/ns 至150V/ns 压摆率,用于优化开关性能
和缓解EMI
– 在7.5V 至18V 电源下工作
• 强大的保护
高级电源管理功能包括数字温度报告和故障检测。
GaN FET 的温度通过可变占空比 PWM 输出进行报
告,这可简化器件加载管理。报告的故障包括过热、过
流和UVLO 监控。
– 响应时间少于100ns 的逐周期过流和锁存短路
保护
– 硬开关时可承受720V 浪涌
– 针对内部过热和UVLO 监控的自我保护
• 高级电源管理
封装信息
封装(1)
封装尺寸(标称值)
器件型号
– 数字温度PWM 输出
• 顶部冷却12mm × 12mm VQFN 封装将电气路径和
散热路径分开,可实现更低的电源环路电感
LMG3522R030-Q1
VQFN (52)
12.00mm × 12.00mm
(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。
2 应用
• 开关模式电源转换器
• 商用网络和服务器PSU
• 商用通信电源整流器
• 车载充电器(OBC) 和无线充电器
• 直流/直流转换器
DRAIN
Direct-Drive
Slew
GaN
Rate
SOURCE
RDRV
IN
VDD
VNEG
LDO,
BB
OCP, SCP,
OTP, UVLO
Current
LDO5V
TEMP
FAULT
OC
SOURCE
高于100V/ns 时的开关性能
简化版方框图
本文档旨在为方便起见,提供有关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: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
Table of Contents
8.4 Start-Up Sequence................................................... 25
8.5 Safe Operation Area (SOA)...................................... 26
8.6 Device Functional Modes..........................................26
9 Application and Implementation..................................27
9.1 Application Information............................................. 27
9.2 Typical Application.................................................... 28
9.3 Do's and Don'ts.........................................................32
9.4 Power Supply Recommendations.............................32
9.5 Layout....................................................................... 34
10 Device and Documentation Support..........................38
10.1 Documentation Support.......................................... 38
10.2 接收文档更新通知................................................... 38
10.3 支持资源..................................................................38
10.4 Trademarks.............................................................38
10.5 静电放电警告.......................................................... 38
10.6 Export Control Notice..............................................38
10.7 术语表..................................................................... 38
11 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 Switching Characteristics............................................8
6.7 Typical Characteristics..............................................10
7 Parameter Measurement Information..........................12
7.1 Switching Parameters...............................................12
8 Detailed Description......................................................14
8.1 Overview...................................................................14
8.2 Functional Block Diagram.........................................15
8.3 Feature Description...................................................16
Information.................................................................... 38
4 Revision History
注:以前版本的页码可能与当前版本的页码不同
Changes from Revision B (June 2021) to Revision C (June 2023)
Page
• 将数据表状态从“预告信息”更改为“量产数据”............................................................................................ 1
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
2
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
5 Pin Configuration and Functions
DRAIN
16
15 14
13 12 11 10
9
8
7
6
5
4
3
2
1
NC2
17
52
NC2
18
19
20
21
22
23
24
25
26
51
50
49
48
47
46
45
44
43
LDO5V
RDRV
TEMP
OC
THERMAL PAD
NC2
FAULT
IN
VDD
NC2
27
28 29
30 31 32 33 34 35 36 37
38 39 40 41
42
SOURCE
VNEG
图5-1. RQS Package, 52-Pin VQFN (Top View)
表5-1. Pin Functions
PIN
TYPE(1)
DESCRIPTION
NAME
NC1
NO.
1, 16
2–15
Used to anchor QFN package to PCB. Pins must be soldered to PCB landing pads. The PCB landing pads are
non-solder mask defined pads and must not be physically connected to any other metal on the PCB. Internally
connected to DRAIN.
—
DRAIN
NC2
P
GaN FET drain. Internally connected to NC1.
Used to anchor QFN package to PCB. Pins must be soldered to PCB landing pads. The PCB landing pads are
non-solder mask defined pads and must not be physically connected to any other metal on the PCB. Internally
connected to SOURCE and THERMAL PAD.
17, 27, 43, 47,
52
—
SOURCE
VNEG
P
P
GaN FET source. Internally connected to NC2 and THERMAL PAD.
18–26, 28–39
Internal buck-boost converter negative output. Used as the negative supply to turn off the depletion mode GaN
FET. Bypass to SOURCE with a 2.2-µF capacitor.
40, 41
BBSW
VDD
IN
42
44
45
46
P
P
I
Internal buck-boost converter switch pin. Connect an inductor from this point to SOURCE.
Device input supply.
CMOS-compatible non-inverting input used to turn the FET on and off.
Push-pull digital output that asserts low during a fault condition. Refer to Fault Detection for details.
FAULT
O
Push-pull digital output that asserts low during overcurrent and short-circuit fault conditions. Refer to Fault
Detection for details.
OC
48
49
50
51
O
O
I
Push-pull digital output that gives information about the GaN FET temperature. Outputs a fixed 9-kHz pulsed
waveform. The device temperature is encoded as the duty cycle of the waveform.
TEMP
RDRV
LDO5V
Drive-strength selection pin. Connect a resistor from this pin to SOURCE to set the turn-on drive strength to
control slew rate. Tie the pin to SOURCE to enable 150 V/ns and tie the pin to LDO5V to enable 100 V/ns.
5-V LDO output for external digital isolator. If using this externally, connect a 0.1-µF or greater capacitor to
SOURCE.
P
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
3
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
表5-1. Pin Functions (continued)
PIN
TYPE(1)
DESCRIPTION
NAME
NO.
THERMAL
PAD
Thermal pad. Internally connected to SOURCE and NC2.
—
—
(1) I = input, O = output, P = power
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
4
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6 Specifications
6.1 Absolute Maximum Ratings
Unless otherwise noted: voltages are in respect to SOURCE connected to reference ground(1)
MIN
MAX
650
720
800
20
UNIT
V
VDS
Drain-source voltage, FET off
VDS(surge)
VDS,tr
Drain-source voltage, FET switching, surge condition(2)
V
Drain-source transient ringing peak voltage, FET off, surge condition(2) (3)
V
VDD
V
–0.3
–0.3
–16
LDO5V
VNEG
5.5
V
0.5
V
VVNEG
–
1
BBSW
VVDD+0.5
20
V
V
V
Pin voltage
IN
–0.3
–0.3
–0.3
VLDO5V+0
.3
FAULT, OC, TEMP
RDRV
5.5
55
V
A
ID(RMS)
ID(pulse)
Drain RMS current, FET on
Internally
limited
Drain pulsed current, FET on, tp < 10 µs(4)
A
–125
IS(pulse)
TJ
Source pulsed current, FET off, tp < 1 µs
Operating junction temperature(5)
Storage temperature
80
150
150
A
°C
°C
–40
–55
TSTG
(1) Operation outside the Absolute Maximum Ratings may cause permanent device damage. Absolute Maximum Ratings do not imply
functional operation of the device at these or any other conditions beyond those listed under Recommended Operating Conditions. If
used outside the Recommended Operating Conditions but within the Absolute Maximum Ratings, the device may not be fully
functional, and this may affect device reliability, functionality, performance, and shorten the device lifetime.
(2) See 节8.3.3 for an explanation of the switching cycle drain-source voltage ratings.
(3) t1 < 200 ns in 图8-1.
(4) The positive pulsed current must remain below the overcurrent threshold to avoid the FET being automatically shut off. The FET drain
intrinsic positive pulsed current rating for tp < 10 µs is 120 A.
(5) Refer to the Electrical and Switching Characteristics Tables for junction temperature test conditions.
6.2 ESD Ratings
PARAMETER
VALUE
±2000
±500
UNIT
Human-body model (HBM), per AEC Q100–002(3)
Electrostatic
discharge
All pins
V(ESD)
V
Charged-device model (CDM), per AEC
Corner pins (1, 16, 17, 26, 27, 42, 43,
and 52)
Q100–011
±750
(1) AEC Q100-002 indicates that HBM stressing shall be in accordance with the ANSI/ESDA/JEDEC JS-001 specification
6.3 Recommended Operating Conditions
Unless otherwise noted: voltages are in respect to SOURCE connected to reference ground.
MIN NOM MAX UNIT
VDD
Supply voltage
(Maximum switching frequency derated for VVDD
< 9 V)
7.5
0
12
5
18
V
Input voltage
IN
18
38
V
A
ID(RMS) Drain RMS current
Positive source current
LDO5V
25 mA
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
5
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.3 Recommended Operating Conditions (continued)
Unless otherwise noted: voltages are in respect to SOURCE connected to reference ground.
MIN NOM MAX UNIT
RRDRV RDRV to SOURCE resistance from external slew-rate control resistor
CVNEG VNEG to SOURCE capacitance from external bypass capacitor
LBBSW BBSW to SOURCE inductance from external buck-boost inductor (1)
0
1
3
500
10
kΩ
µF
4.7
10 µH
(1) > 1-A current rating is recommended.
6.4 Thermal Information
LMG3522R030
THERMAL METRIC(1)
RQS (VQFN)
52 PINS
0.28
UNIT
RθJC(top)
Junction-to-case (top) thermal resistance
°C/W
(1) For more information about traditional and new thermal metrics, see the Semiconductor and IC Package Thermal Metrics application
report.
6.5 Electrical Characteristics
Unless otherwise noted: voltage, resistance, capacitance, and inductance are in respect to SOURCE connected with
reference ground; –40 ℃≤TJ ≤125 ℃;
VDS = 520 V; 9 V ≤VVDD ≤18 V; VIN = 0 V; RDRV connected to LDO5V; LBBSW = 4.7 µH
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNIT
GAN POWER TRANSISTOR
VIN = 5 V, TJ = 25°C
26
45
3.6
5
35
mΩ
mΩ
V
RDS(on)
Drain-source on resistance
VIN = 5 V, TJ = 125°C
IS = 0.1 A
Third-quadrant mode source-drain
voltage
VSD
IS = 20 A
3
V
VDS = 650 V, TJ = 25°C
VDS = 650 V, TJ = 125°C
VDS = 400 V
1
µA
µA
pF
IDSS
Drain leakage current
Output capacitance
10
235
COSS
CO(er)
Energy related effective output
capacitance
320
pF
VDS = 0 V to 400 V
CO(tr)
QOSS
QRR
Time related effective output capacitance
Output charge
460
190
0
pF
nC
nC
Reverse recovery charge
VDD –SUPPLY CURRENTS
VDD quiescent current
VVDD = 12 V, VIN = 0 V or 5 V
700
1200
20
µA
VVDD = 12 V, fIN = 140 kHz, soft-
switching
VDD operating current
15.5
mA
BUCK BOOST CONVERTER
VNEG output voltage
VNEG sinking 50 mA
V
A
–14
Peak BBSW sourcing current at low peak
IBBSW,PK(l current mode setting
(peak external buck-boost inductor
0.3
0.8
0.4
0.5
1.2
ow)
current)
Peak BBSW sourcing current at low peak
IBBSW,PK( current mode setting
(peak external buck-boost inductor
1
A
high)
current)
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
6
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.5 Electrical Characteristics (continued)
Unless otherwise noted: voltage, resistance, capacitance, and inductance are in respect to SOURCE connected with
reference ground; –40 ℃≤TJ ≤125 ℃;
VDS = 520 V; 9 V ≤VVDD ≤18 V; VIN = 0 V; RDRV connected to LDO5V; LBBSW = 4.7 µH
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNIT
High peak current mode setting enable –
IN positive-going threshold frequency
280
420
515
kHz
LDO5V
Output voltage
LDO5V sourcing 25 mA
4.75
25
5
5.25
100
V
Short-circuit current
50
mA
IN
VIN,IT+
VIN,IT–
Positive-going input threshold voltage
Negative-going input threshold voltage
Input threshold hysteresis
1.7
0.7
0.7
100
1.9
1.0
0.9
150
2.45
1.3
V
V
1.3
V
Input pulldown resistance
VIN = 2 V
200
kΩ
FAULT, OC/ZVD, TEMP –OUPUT DRIVE
Low-level output voltage
Output sinking 8 mA
0.16
0.2
0.4
V
V
Output sourcing 8 mA, measured as
High-level output voltage
0.45
V
LDO5V –VO
VDD, VNEG –UNDER VOLTAGE LOCKOUT
VVDD,T+
VDD UVLO –positive-going threshold
voltage
6.4
6
7
6.5
7.6
7.1
V
V
(UVLO)
VDD UVLO –negative-going threshold
voltage
VDD UVLO –input threshold voltage
hysteresis
510
mV
V
VNEG UVLO –negative-going threshold
voltage
–13.6
–13.0
–12.3
–12.1
VNEG UVLO –positive-going threshold
voltage
V
–13.3 –12.75
GATE DRIVER
From VDS < 320 V to VDS < 80 V, RDRV
disconnected from LDO5V, RRDRV = 300
kΩ, TJ = 25℃, VBUS = 400 V, LHB current
= 10 A, see 图7-1
20
V/ns
V/ns
V/ns
MHz
From VDS < 320 V to VDS < 80 V, RDRV
tied to LDO5V, TJ = 25℃, VBUS = 400 V,
LHB current = 10 A, see 图7-1
90
Turn-on slew rate
From VDS < 320 V to VDS < 80 V, RDRV
disconnected from LDO5V, RRDRV = 0 Ω,
TJ = 25℃, VBUS = 400 V, LHB current = 10
A, see 图7-1
150
VNEG rising to > –13.25 V, soft-switched,
maximum switching frequency derated for
VVDD < 9 V
Maximum GaN FET switching frequency.
2
FAULTS
DRAIN overcurrent fault –threshold
current
IT(OC)
60
75
70
90
80
A
A
DRAIN short-circuit fault –threshold
current
IT(SC)
105
di/dt threshold between overcurrent and
short-circuit faults
di/dtT(SC)
150
A/µs
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
7
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.5 Electrical Characteristics (continued)
Unless otherwise noted: voltage, resistance, capacitance, and inductance are in respect to SOURCE connected with
reference ground; –40 ℃≤TJ ≤125 ℃;
VDS = 520 V; 9 V ≤VVDD ≤18 V; VIN = 0 V; RDRV connected to LDO5V; LBBSW = 4.7 µH
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNIT
GaN temperature fault –postive-going
threshold temperature
175
°C
GaN temperature fault –threshold
temperature hysteresis
30
185
20
°C
°C
°C
Driver temperature fault –positive-going
threshold temperature
Driver temperature fault –threshold
temperature hysteresis
TEMP
Output frequency
4.3
9
82
14
kHz
%
GaN TJ = 150℃
58.5
36.2
0.03
64.6
40
70
43.7
6
%
GaN TJ = 125℃
GaN TJ = 85℃
GaN TJ = 25℃
Output PWM duty cycle
%
3
%
IDEAL-DIODE MODE CONTROL
Drain-source third-quadrant detection –
VT(3rd)
0
0.15
V
–0.15
threshold voltage
0
0
0.2
A
A
0℃≤TJ ≤125℃
–40°C ≤TJ ≤0°C
–0.2
Drain zero-current detection –threshold
current
IT(ZC)
0.35
–0.35
6.6 Switching Characteristics
Unless otherwise noted: voltage, resistance, capacitance, and inductance are respect to SOURCE connected with reference
ground; –40 ℃≤TJ ≤125 ℃; 9 V ≤VVDD ≤18 V; VIN = 0 V; RDRV connected to LDO5V; LBBSW = 4.7 µH
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNIT
SWITCHING TIMES
From VIN > VIN,IT+ to ID > 1 A, VBUS = 400
V, LHB current = 10 A, see 图7-1 and 图
7-2
td(on)
Drain-current turn-on delay time
Turn-on delay time
28
33
42
54
4.3
69
22
24
ns
ns
ns
ns
ns
ns
(Idrain)
From VIN > VIN,IT+ to VDS < 320 V, VBUS
400 V, LHB current = 10 A, see 图7-1
and 图7-2
=
td(on)
tr(on)
td(off)
tf(off)
From VDS < 320 V to VDS < 80 V, VBUS
400 V, LHB current = 10 A, see 图7-1
and 图7-2
=
Turn-on rise time
2.8
48
From VIN < VIN ,IT– to VDS > 80 V, VBUS
400 V, LHB current = 10 A, see 图7-1
and 图7-2
=
Turn-off delay time
From VDS > 80 V to VDS > 320 V, VBUS
400 V, LHB current = 10 A, see 图7-1
and 图7-2
=
Turn-off fall time(1)
VIN rise/fall times < 1 ns, VDS falls to <
200 V, VBUS = 400 V, LHB current = 10 A,
see 图7-1
Minimum IN high pulse-width for FET
turn-on
STARTUP TIMES
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
8
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.6 Switching Characteristics (continued)
Unless otherwise noted: voltage, resistance, capacitance, and inductance are respect to SOURCE connected with reference
ground; –40 ℃≤TJ ≤125 ℃; 9 V ≤VVDD ≤18 V; VIN = 0 V; RDRV connected to LDO5V; LBBSW = 4.7 µH
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNIT
From VVDD > VVDD,T+
(UVLO) to FAULT high, CLDO5V = 100
nF, CVNEG = 2.2 µF at 0-V bias linearly
decreasing to 1.5 µF at 15-V bias
t(start)
Driver start-up time
310
470
µs
FAULT TIMES
Overcurrent fault FET turn-off time, FET
VIN = 5 V, From ID > IT(OC) to ID < 50 A, ID
di/dt = 100 A/µs
toff(OC)
115
65
170
100
250
ns
ns
ns
on before overcurrent
Short-circuit current fault FET turn-off
time, FET on before short circuit
VIN = 5 V, From ID > IT(SC) to ID < 50 A, ID
di/dt = 700 A/µs
toff(SC)
Overcurrent fault FET turn-off time, FET
turning on into overcurrent
From ID > IT(OC) to ID < 50 A
200
Short-circuit fault FET turn-off time, FET
turning on into short circuit
From ID > IT(SC) to ID < 50 A
80
380
50
180
580
ns
µs
ns
IN reset time to clear FAULT latch
From VIN < VIN,IT– to FAULT high
250
t(window)
Overcurrent fault to short-circuit fault
window time
(OC)
IDEAL-DIODE MODE CONTROL TIMES
VDS < VT(3rd) to FET turn-on, VDS being
discharged by half-bridge configuration
inductor at 5 A
Ideal-diode mode FET turn-on time
50
65
ns
ID > IT(ZC) to FET turn-off, ID di/dt = 100
A/µs created with a half-bridge
configuration
Ideal-diode mode FET turn-off time
55
76
ns
ns
Overtemperature-shutdown ideal-diode
mode IN falling blanking time
150
230
360
(1) During turn-off, VDS rise time is the result of the resonance of COSS and loop inductance as well as load current.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
9
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.7 Typical Characteristics
95
160
140
120
100
80
-40C
0C
25C
85C
125C
-40C
0C
25C
85C
125C
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
60
40
20
0
50 100 150 200 250 300 350 400 450 500
0
50 100 150 200 250 300 350 400 450 500
RDRV Resistance (k)
RDRV Resistance (k)
图6-2. Turn-On Delay Time vs Drive-Strength Resistance
图6-1. Drain-Current Turn-On Delay Time vs Drive-Strength
Resistance
140
-40C
0C
120
25C
85C
100
125C
80
60
40
20
0
0
50 100 150 200 250 300 350 400 450 500
RDRV Resistance (k)
图6-4. Turn-On Slew Rate vs Drive-Strength Resistance
图6-3. Turn-On Rise Time vs Drive-Strength Resistance
600
6.5
-40C
0C
6
500
25C
85C
5.5
125C
400
5
4.5
4
-40C
300
25C
125C
OC limit
200
100
0
3.5
3
0
5
10
15
20
25
30
35
40
45
50
0
5
10
15
20
25
30
35
Drain-Source Voltage (V)
Source Current (A)
图6-5. Drain Current vs Drain-Source Voltage
IN = 0 V
图6-6. Off-State Source-Drain Voltage vs Source Current
Copyright © 2023 Texas Instruments Incorporated
10
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
6.7 Typical Characteristics (continued)
2
1.8
1.6
1.4
1.2
1
1750
1500
1250
1000
750
500
250
0
-40C
0C
25C
85C
125C
0.8
0.6
0
50 100 150 200 250 300 350 400 450 500
Drain-Source Voltage (V)
-40 -20
0
20
40
60
80 100 120 140 160
Junction Temperature (°C)
图6-8. Output Capacitance vs Drain-Source Voltage
图6-7. Normalized On-Resistance vs Junction Temperature
250
250
Vds = 0 V
Vds = 50 V
Vds = 400 V
Vds = 0 V
Vds = 50 V
Vds = 400 V
Vds = 400 V
Vds = 400 V
200
200
150
100
50
150
Vds = 50 V
Vds = 50 V
100
Vds = 0 V
Vds = 0 V
50
0
0
0
400
800
1200
1600
2000
0
400
800
1200
1600
2000
IN Switching Frequency (kHz)
IN Switching Frequency (kHz)
VDD = 12 V TJ = 25 °C
VDD = 12 V TJ = 125 °C
图6-10. VDD Supply Current vs IN Switching Frequency
图6-9. VDD Supply Current vs IN Switching Frequency
80
70
60
50
40
30
20
10
0
0
50 100 150 200 250 300 350 400 450 500 550
Drain-Source Voltage (V)
图6-11. Repetitive Safe Operation Area
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
11
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
7 Parameter Measurement Information
7.1 Switching Parameters
图 7-1 shows the circuit used to measure most switching parameters. The top device in this circuit is used to re-
circulate the inductor current and functions in third-quadrant mode only. The bottom device is the active device
that turns on to increase the inductor current to the desired test current. The bottom device is then turned off and
on to create switching waveforms at a specific inductor current. Both the drain current (at the source) and the
drain-source voltage is measured. 图 7-2 shows the specific timing measurement. TI recommends to use the
half-bridge as a double pulse tester. Excessive third-quadrant operation can overheat the top device.
DRAIN
Slew
Rate
Direct-Drive
SOURCE
GaN
RDRV
IN
VDD
LDO5V
TEMP
VNEG
OCP, OTP,
UVLO,
TEMP
Current
LDO,
BB
LHB
FAULT
OC
VBUS
+
–
CVDC
SOURCE
DRAIN
Slew
Rate
Direct-Drive
SOURCE
GaN
RDRV
IN
+
VDS
VDD
LDO5V
TEMP
VNEG
OCP, OTP,
UVLO,
TEMP
Current
LDO,
BB
_
PWM input
FAULT
OC
ID
SOURCE
图7-1. Circuit Used to Determine Switching Parameters
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
12
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
50%
50%
IN
ID
td(on)(Idrain)
1A
td(on)
td(off)
tr(on)
tf(off)
80%
VDS
80%
20%
20%
图7-2. Measurement to Determine Propagation Delays and Slew Rates
7.1.1 Turn-On Times
The turn-on transition has three timing components: drain-current turn-on delay time, turn-on delay time, and
turn-on rise time. The drain-current turn-on delay time is from when IN goes high to when the GaN FET drain-
current reaches 1 A. The turn-on delay time is from when IN goes high to when the drain-source voltage falls
20% below the bus voltage. Finally, the turn-on rise time is from when drain-source voltage falls 20% below the
bus voltage to when the drain-source voltage falls 80% below the bus voltage. Note that the turn-on rise time is
the same as the VDS 80% to 20% fall time. All three turn-on timing components are a function of the RDRV pin
setting.
7.1.2 Turn-Off Times
The turn-off transition has two timing components: turn-off delay time, and turn-off fall time. The turn-off delay
time is from when IN goes low to when the drain-source voltage rises to 20% of the bus voltage. The turn-off fall
time is from when the drain-source voltage rises to 20% of the bus voltage to when the drain-source voltage
rises to 80% of the bus voltage. Note that the turn-off fall time is the same as the VDS 20% to 80% rise time. The
turn-off timing components are independent of the RDRV pin setting, but heavily dependent on the LHB load
current.
7.1.3 Drain-Source Turn-On Slew Rate
The drain-source turn-on slew rate, measured in volts per nanosecond, is the inverse of the turn-on rise time or
equivalently the inverse of the VDS 80% to 20% fall time. The RDRV pin is used to program the slew rate.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
13
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8 Detailed Description
8.1 Overview
The LMG3522R030-Q1 is a high-performance power GaN device with integrated gate driver. The GaN device
offers zero reverse recovery and ultra-low output capacitance, which enables high efficiency in bridge-based
topologies. Direct-drive architecture is applied to control the GaN device directly by the integrated gate driver.
This architecture provides superior switching performance compared to the traditional cascode approach and
helps solve a number of challenges in GaN applications.
The integrated driver ensures the device stays off for high drain slew rates. The integrated driver also protects
the GaN device from overcurrent, short-circuit, undervoltage, and overtemperature. Regarding fault signal
reporting, LMG3522R030-Q1 provides different reporting method which is shown in 表 8-1. Refer to Fault
Detection for more details. The integrated driver is also able to sense the die temperature and send out the
temperature signal through a modulated PWM signal.
Unlike Si MOSFETs, GaN devices do not have a p-n junction from source to drain and thus have no reverse
recovery charge. However, GaN devices still conduct from source to drain similar to a p-n junction body diode,
but with higher voltage drop and higher conduction loss. Therefore, source-to-drain conduction time must be
minimized while the LMG3522R030-Q1 GaN FET is turned off.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
14
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8.2 Functional Block Diagram
TEMP
DRAIN
VDD
Third-Quadrant
Detection
GaN
LDO
LDO5V
UVLO
(+5V, VDD, VNEG)
FAULT
Series
Si FET
Die
Temp
Sensing
Thermal
Shutdown
Short Circuit Protection
Overcurrent Protection
OC
IN
Gating logic control
& level shifting
Buck-Boost
Controller
BBSW
VNEG
SOURCE
RDRV
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
15
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8.3 Feature Description
The LMG3522R030-Q1 includes advanced features to provide superior switching performance and converter
efficiency.
8.3.1 GaN FET Operation Definitions
For the purposes of this data sheet, the following terms are defined below. The SOURCE pin is assumed to be at
0 V for these definitions.
First-Quadrant Current = Positive current flowing internally from the DRAIN pin to the SOURCE pin.
Third-Quadrant Current = Positive current flowing internally from the SOURCE pin to the DRAIN pin.
First-Quadrant Voltage = Drain pin voltage –Source pin voltage = Drain pin voltage
Third-Quadrant Voltage = SOURCE pin voltage –DRAIN pin voltage = –DRAIN pin voltage
FET On-State = FET channel is at rated RDS(on). Both first-quadrant current and third-quadrant current can flow
at rated RDS(on)
.
For LMG3522R030-Q1 in On-State, GaN FET internal gate voltage is held at the SOURCE pin voltage to
achieve rated RDS(on). The GaN FET channel is at rated RDS(on) with VGS = 0 V because the LMG3522R030-Q1
GaN FET is a depletion mode FET.
FET Off-State = FET channel is fully off for positive first-quadrant voltage. No first-quadrant current can flow.
While first-quadrant current cannot flow in the FET Off-State, third-quadrant current still flows if the DRAIN
voltage is taken sufficiently negative (positive third-quadrant voltage). For devices with an intrinsic p-n junction
body diode, current flow begins when the DRAIN voltage drops enough to forward bias the p-n junction.
GaN FETS do not have an intrinsic p-n junction body diode. Instead, current flows because the GaN FET
channel turns back on. In this case, the DRAIN pin becomes the electrical source and the SOURCE pin
becomes the electrical drain. To enhance the channel in third-quadrant, the DRAIN (electrical source) voltage
must be taken sufficiently low to establish a VGS voltage greater than the GaN FET threshold voltage. The GaN
FET channel is operating in saturation and only turns on enough to support the third-quadrant current as its
saturated current.
For LMG3522R030-Q1 in Off-State, GaN FET internal gate voltage is held at the VNEG pin voltage to block all
first-quadrant current. The VNEG voltage is lower than the GaN FET negative threshold voltage to cut off the
channel.
To enhance the channel in off-state third quadrant, the LMG3522R030-Q1 DRAIN (electrical source) voltage
must be taken sufficiently close to VNEG to establish a VGS voltage greater than the GaN FET threshold voltage.
Again, because the LMG3522R030-Q1 GaN FET is a depletion mode FET with a negative threshold voltage, this
means the GaN FET turns on with DRAIN (electrical source) voltage between 0 V and VNEG. The typical off-
state third-quadrant voltage is 5 V for third-quadrant current at 20 A. Thus, the off-state third-quadrant losses for
the LMG3522R030-Q1 are significantly higher than a comparable power device with an intrinsic p-n junction
body diode.
8.3.2 Direct-Drive GaN Architecture
The LMG3522R030-Q1 uses a series Si FET to ensure the power IC stays off when VDD bias power is not
applied. When the VDD bias power is off, the series Si FET is interconnected with the GaN device in a cascode
mode, which is shown in the Functional Block Diagram. The gate of the GaN device is held within a volt of the
series Si FET's source. When a high voltage is applied on the drain and the silicon FET blocks the drain voltage,
the VGS of the GaN device decreases until the GaN device passes the threshold voltage. Then, the GaN device
is turned off and blocks the remaining major part of drain voltage. There is an internal clamp to make sure that
the VDS of the Si FET does not exceed its maximum rating. This feature avoids the avalanche of the series Si
FET when there is no bias power.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
16
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
When LMG3522R030-Q1 is powered up with VDD bias power, the internal buck-boost converter generates a
negative voltage (VVNEG) that is sufficient to directly turn off the GaN device. In this case, the series Si FET is
held on and the GaN device is gated directly with the negative voltage.
Comparing with traditional cascode drive GaN architecture, where the GaN gate is grounded and the Si
MOSFET gate is being driven to control the GaN device, direct-drive configuration has multiple advantages.
First, as the Si MOSFET does need to switch in every switching cycle, GaN gate-to-source charge (QGS) is lower
and there’s no Si MOSFET reverse-recovery related losses. Second, the voltage distribution between the GaN
and Si MOSFET in off-mode in a cascode configuration can cause the MOSFET to avalanche due to high GaN
drain-to-source capacitance (CDS). Finally, the switching slew rate in direct-drive configuration can be controlled
while cascode drive cannot. More information about the direct-drive GaN architecture can be found in Direct-
drive configuration for GaN devices.
8.3.3 Drain-Source Voltage Capability
Due to the silicon FET’s long reign as the dominant power-switch technology, many designers are unaware
that the headline drain-source voltage cannot be used as an equivalent point to compare devices across
technologies. The headline drain-source voltage of a silicon FET is set by the avalanche breakdown voltage.
The headline drain-source voltage of a GaN FET is set by the long term reliability with respect to data sheet
specifications.
Exceeding the headline drain-source voltage of a silicon FET can lead to immediate and permanent damage.
Meanwhile, the breakdown voltage of a GaN FET is much higher than the headline drain-source voltage. For
example, the breakdown voltage of the LMG3522R030-Q1 is more than 800 V.
A silicon FET is usually the weakest link in a power application during an input voltage surge. Surge protection
circuits must be carefully designed to ensure the silicon FET avalanche capability is not exceeded because it is
not feasible to clamp the surge below the silicon FET breakdown voltage. Meanwhile, it is easy to clamp the
surge voltage below a GaN FET breakdown voltage. In fact, a GaN FET can continue switching during the surge
event which means output power is safe from interruption.
The LMG3522R030-Q1 drain-source capability is explained with the assistance of 图 8-1. The figure shows the
drain-source voltage versus time for a GaN FET for a single switch cycle in a switching application. No claim is
made about the switching frequency or duty cycle.
VDS(tr)
VDS(off)
VDS(switching)
t0 t1
t2
图8-1. Drain-Source Voltage Switching Cycle
The waveform starts before t0 with the FET in the on state. At t0 the GaN FET turns off and parasitic elements
cause the drain-source voltage to ring at a high frequency. The peak ring voltage is designated VDS(tr). The high
frequency ringing has damped out by t1. Between t1 and t2 the FET drain-source voltage is set by the
characteristic response of the switching application. The characteristic is shown as a flat line, but other
responses are possible. The voltage between t1 and t2 is designated VDS(off). At t2 the GaN FET is turned on at a
non-zero drain-source voltage. The drain-source voltage at t2 is designated VDS(switching). Unique VDS(tr), VDS(off)
and VDS(switching) parameters are shown because each can contribute to stress over the lifetime of the GaN FET.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
17
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
The LMG3522R030-Q1 drain-source surge voltage capability is seen with the absolute maximum ratings
DS(tr)(surge) and VDS(surge) in the Specifications where VDS(tr)(surge) maps to VDS(tr) in 图8-1 and VDS(surge) maps to
V
both VDS(off) and VDS(switching) in 图 8-1. More information about the surge capability of TI GaN FETs is found in A
New Approach to Validate GaN FET Reliability to Power-line Surges Under Use-conditions.
8.3.4 Internal Buck-Boost DC-DC Converter
An internal inverting buck-boost converter generates a regulated negative rail for the turn-off supply of the GaN
device. The buck-boost converter is controlled by a peak current mode, hysteretic controller. In normal operation,
the converter remains in discontinuous-conduction mode, but can enter continuous-conduction mode during
start-up. The converter is controlled internally and requires only a single surface-mount inductor and output
bypass capacitor. Typically, the converter is designed to use a 4.7-μH inductor and a 2.2-μF output capacitor.
The buck-boost converter uses a peak current hysteretic control. As shown in 图 8-2, the inductor current
increases at the beginning of a switching cycle until the inductor reaches the peak current limit. Then the
inductor current goes down to zero. The idle time between each current pulse is determined automatically by the
buck-boost controller, and can be reduced to zero. Therefore, the maximum output current happens when the
idle time is zero, and is decided by the peak current but to a first order is independent of the inductor value.
However, the peak output current the buck-boost can deliver to the –14-V rail is proportional to the VDD input
voltage. Therefore, the maximum switching frequency of the GaN that the buck-boost can support varies with
VDD voltage and is only specified for operation up to 2 MHz for VDD voltages above 9 V.
VNEG_avg
ûVNEG
VNEG
IDCDC,PK
Smaller
inductor
Larger
inductor
Idle time
Buck-boost
inductor current
1/(buck-boost frequency)
图8-2. Buck-Boost Converter Inductor Current
The LMG3522R030-Q1 supports the GaN operation up to 2 MHz. As power consumption is very different in a
wide switching frequency range enabled by the GaN device, two peak current limits are used to control the buck-
boost converter. The two ranges are separated by IN positive-going threshold frequency. As shown in 图 8-3,
when switching frequency is in the lower range, the peak current is initially set to the lower value IBBSW,M(low)
(typically 0.4 A). When switching frequency is in the higher range, the peak current is raised to the higher value
IBBSW,M(high) (typically 1 A) and requires a larger inductor. There is a filter on this frequency detection logic,
therefore the LMG3522R030-Q1 requires five consecutive cycles at the higher frequency before it is set to the
higher buck-boost peak current limit. The current limit does not go down again until power off after the higher
limit is set. Even if the switching frequency returns to the lower range, the current limit does not decrease to the
lower limit.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
18
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
IBBSW,PK(high)
IBBSW,PK(low)
IBBSW,PK
IN
图8-3. Buck-Boost Converter Peak Current
As the peak current of the buck-boost is subject to two different peak current limits which are 0.4 A and 1 A for
low and high frequency operation (see Internal Buck-Boost DC-DC Converter), so the inductor must have a
saturation current well above the rated peak current limit. After the higher limit is established by switching at a
higher frequency, the current limit does not go back to the lower level even when GaN device is then switched at
a lower frequency. Therefore, selecting an inductor according to the higher 1-A limit is recommended.
8.3.5 VDD Bias Supply
Wide VDD voltage ranges from 7.5 V to 18 V are supported by internal regulators which supply the bias supplies
needed for the internal circuits to function. TI recommends to use a 12-V unregulated power supply to supply
VDD.
8.3.6 Auxiliary LDO
There is a 5-V voltage regulator inside the part used to supply external loads, such as digital isolators for the
high-side drive signal. The digital outputs of the part use this rail as their supply. No capacitor is required for
stability, but transient response is poor if no external capacitor is provided. If the application uses this rail to
supply external circuits, TI recommends to have a capacitor of at least 0.1 μF for improved transient response.
A larger capacitor can be used for further transient response improvement. The decoupling capacitor used here
must be a low-ESR ceramic type. Capacitances above 0.47 μF will slow down the start-up time of the
LMG3522R030-Q1 due to the ramp-up time of the 5-V rail.
8.3.7 Fault Detection
The GaN power IC integrates overcurrent protection (OCP), short-circuit protection (SCP), overtemperature
protection (OTP), and undervoltage lockout (UVLO).
8.3.7.1 Overcurrent Protection and Short-Circuit Protection
There are two types of current faults which can be detected by the driver: overcurrent fault and short-circuit fault.
The overcurrent protection (OCP) circuit monitors drain current and compares that current signal with an
internally set limit IT(OC). Upon detection of the overcurrent, the LMG3522R030-Q1 conducts cycle-by-cycle
overcurrent protection as shown in 图 8-4. In this mode, the GaN device is shut off and the OC pin is pulled low
when the drain current crosses the IT(OC) plus a delay toff(OC), but the overcurrent signal clears after the IN pin
signal goes low. In the next cycle, the GaN device can turn on as normal. The cycle-by-cycle function can be
used in cases where steady-state operation current is below the OCP level but transient response can still reach
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
19
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
current limit, while the circuit operation cannot be paused. The cycle-by-cycle function also prevents the GaN
device from overheating by overcurrent induced conduction losses.
The short-circuit protection (SCP) monitors the drain current and triggers if the di/dt of the current exceeds a
threshold di/dt T(SC) as the current crosses between the OC and SC thresholds. It performs this di/dt detection by
delaying the OC detection signal by an amount tOC,window and using a higher current SC detection threshold. If
the delayed OC occurs before the non-delayed SC, the di/dt is below the threshold and an OC is triggered. If the
SC is detected first, the di/dt is fast enough and the SC is detected as shown in 图 8-5. This extremely high di/dt
current would typically be caused by a short of the output of the half-bridge and can be damaging for the GaN to
continue to operate in that condition. Therefore, if a short-circuit fault is detected, the GaN device is turned off
with an intentionally slowed driver so that a lower overshoot voltage and ringing can be achieved during the turn-
off event. This fast response circuit helps protect the GaN device even under a hard short-circuit condition. In
this protection, the GaN device is shut off and held off until the fault is reset by either holding the IN pin low for a
period of time defined in the Specifications or removing power from VDD.
During OCP or SCP in a half bridge, after the current reaches the upper limit and the device is turned off by
protection, the PWM input of the device could still be high and the PWM input of the complementary device
could still be low. In this case, the load current can flow through the third quadrant of the complementary device
with no synchronous rectification. The high negative VDS of the GaN device (–3 V to –5 V) from drain to source
could lead to high third-quadrant loss, similar to dead-time loss but for a longer time.
For safety considerations, OCP allows cycle-by-cycle operation while SCP latches the device until reset. By
reading the FAULT and OC pins, the exact current fault type can be determined. Refer to Fault Reporting for
detailed information.
IT(OC)
Inductor current
VSW
Input PWM
图8-4. Cycle-by-Cycle OCP Operation
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
20
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
toff(SC)
toff(OC)
IT(SC)
< t(window)(OC)
IT(OC)
ID
IN
OC
Cycle by cycle OCP
Latched SCP
图8-5. Overcurrent Detection vs Short-Circuit Detection
8.3.7.2 Overtemperature Shutdown
The LMG3522R030-Q1 implements two overtemperature-shutdown (OTSD) functions, the GaN OTSD and the
Driver OTSD. Two OTSD functions are needed to maximize device protection by sensing different locations in
the device and protecting against different thermal-fault scenarios.
The GaN OTSD senses the GaN FET temperature. The GaN FET can overheat from both first-quadrant current
and third-quadrant current. As explained in GaN FET Operation Definitions, a FET can prevent first-quadrant
current by going into the off-state but is unable to prevent third-quadrant current. FET third-quadrant losses are a
function of the FET technology, current magnitude, and if the FET is operating in the on-state or off-state. As
explained in GaN FET Operation Definitions, the LMG3522R030-Q1 has much higher GaN FET third-quadrant
losses in the off-state.
When the GaN FET is too hot, the best protection is to turn off the GaN FET when first-quadrant current tries to
flow and turn on the GaN FET when third-quadrant current is flowing. This type of FET control is known as ideal-
diode mode (IDM). When the GaN OTSD trip point is exceeded, the GaN OTSD puts the GaN FET into
overtemperature-shutdown ideal-diode mode (OTSD-IDM) operation to achieve this optimum protection. OTSD-
IDM is explained in 节8.3.10.
The Driver OTSD senses the integrated driver temperature and trips at a higher temperature compared to the
GaN OTSD. This second OTSD function exists to protect the LMG3522R030-Q1 from driver thermal-fault events
while allowing sufficient temperature difference for OTSD-IDM to operate. These driver thermal events include
shorts on the LDO5V, BBSW, and VNEG device pins. When the Driver OTSD trip point is exceeded, the Driver
OTSD shuts off the LDO5V regulator, the VNEG buck-boost converter, and the GaN FET. Note that OTSD-IDM
does not function in Driver OTSD. This is why the Driver OTSD must trip higher than the GaN OTSD function.
Otherwise, GaN FET third-quadrant overheating cannot be addressed.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
21
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
Besides the temperature difference in the GaN OTSD and Driver OTSD trip points, further temperature
separation is obtained due to the thermal gradient difference between the GaN OTSD and Driver OTSD sense
points. The GaN OTSD sensor is typically at least 20°C hotter than the driver OTSD sensor when the device is
in GaN OTSD due to GaN FET power dissipation.
The FAULT pin is asserted for either or both the GaN OTSD state and the Driver OTSD state. FAULT de-asserts
and the device automatically returns to normal operation after both the GaN OTSD and Driver OTSD fall below
their negative-going trip points. During cool down, when the device exits the Driver OTSD state but is still in the
GaN OTSD state, the device automatically resumes OTSD-IDM operation.
8.3.7.3 UVLO Protection
The LMG3522R030-Q1 supports a wide range of VDD voltages. However, when the device is below UVLO
threshold, the GaN device stops switching and is held off. The FAULT pin is pulled low as an indication of UVLO.
The LDO and buck-boost are turned on by the rising-edge of the VIN UVLO and shuts off around 5 V to 6 V.
8.3.7.4 Fault Reporting
The FAULT and OC outputs form a fault reporting scheme together. The FAULT and OC outputs are both push-
pull outputs indicating the readiness and fault status of the driver. These two pins are logic high in normal
operation, and change logic according to 表8-1.
表8-1. Fault Types and Reporting
NORMAL
UVLO, OT, and RDRV-OPEN
OVERCURRENT
SHORT-CIRCUIT
FAULT
OC
1
1
0
1
1
0
0
0
FAULT is held low when starting up until the series Si FET is turned on. During operation, if the power supplies
go below the UVLO thresholds or the device temperature go above the OT thresholds, power device is disabled
and FAULT is held low until a fault condition is no longer detected. If RDRV is open, FAULT is also held low. In a
short-circuit or overtemperature fault condition, FAULT is held low until the fault latches are reset or fault is
cleared. The OC pin is held low if there is a short-circuit or overcurrent fault. The signals help notify the controller
the exact type of faults by reading the truth table. If a combined reporting of the faults on a single pin is desired,
one can short the OC pin to ground during power up. All faults assert the FAULT pin then and the OC pin is not
used. Please note: internal protection happens regardless of the connection of the pin outputs, which means that
the protection features continue to operate even if fault reporting is ignored.
8.3.8 Drive-Strength Adjustment
The LMG3522R030-Q1 allows users to adjust the drive strength of the device and obtain a desired slew rate,
which provides flexibility when optimizing switching losses and noise coupling.
To adjust drive strength, a resistor can be placed between the RDRV pin and SOURCE pin. The resistance
determines the slew rate of the device, from 20 V/ns to 150 V/ns, during turn-on. On the other hand, there are
two dv/dt values that can be selected without the resistor: shorting the RDRV pin to ground sets the slew rate to
150 V/ns, and shorting the RDRV pin to LDO5V sets the slew rate to 100 V/ns. The device detects the short to
LDO5V one time at power up. Once the short to LDO5V condition is detected, the device no longer monitors the
RDRV pin. Otherwise, the RDRV pin is continuously monitored and the dv/dt setting can be changed by
modulating the resistance during device operation. The modulation must be fairly slow since there is significant
internal filtering to reject switching noise.
备注
Parasitic power loop inductance can influence the voltage slew rate reading from the VDS switching
waveform. The inductance induces a drop on VDS in the current rising phase before voltage falling
phase, if this drop is more than 20% of the VDC, the voltage slew rate reading can be influenced. Refer
to 节 9.5.1.2 for the power loop design guideline and how to estimate the parasitic power loop
inductance.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
22
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8.3.9 Temperature-Sensing Output
The integrated driver senses the GaN die temperature and outputs the information through a modulated PWM
signal on the TEMP pin. The typical PWM frequency is 9 kHz with the same refresh rate. The minimum PWM
pulse width is around 30 ns, which can be observed at temperature below 25°C. The target temperature range is
from 25°C to 150°C, and the corresponding PWM duty cycle is typically from 3% to 82%. Following equation can
be used to calculate the typical junction temperature TJ,typ in °C from the duty cycle DTEMP
:
TJ,typ (°C) = 162.3 * DTEMP + 20.1
The tolerances of typical measurement are listed in 表8-2.
表8-2. Typical Junction Temperature Measurement Based on TEMP Signal and Tolerance
Typical TJ Measurement Based on TEMP Signal (°C)
25
85
125
Tolerance (°C)
±5
±6
±10
At temperatures above 150°C, the duty cycle continues to increase linearly until overtemperature fault happens.
When overtemperature happens, the TEMP pin is pulled high to indicate this fault until the temperature is
reduced to the normal range. There is a hysteresis to clear overtemperature fault.
8.3.10 Ideal-Diode Mode Operation
Off-state FETs act like diodes by blocking current in one direction (first quadrant) and allowing current in the
other direction (third quadrant) with a corresponding diode like voltage drop. FETs, though, can also conduct
third-quadrant current in the on-state at a significantly lower voltage drop. Ideal-diode mode (IDM) is when an
FET is controlled to block first-quadrant current by going to the off-state and conduct third-quadrant current by
going to the on-state, thus achieving an ideal lower voltage drop.
FET off-state third-quadrant current flow is commonly seen in power converters, both in normal and fault
situations. As explained in GaN FET Operation Definitions, GaN FETs do not have an intrinsic p-n junction body
diode to conduct off-state third-quadrant current. Instead, the off-state third-quadrant voltage drop for the
LMG3522R030-Q1 is several times higher than a p-n junction voltage drop, which can impact efficiency in
normal operation and device ruggedness in fault conditions.
To improve device ruggedness in a GaN FET overtemperature fault situation, LMG3522R030-Q1 devices
implement a GaN FET overtemperature-shutdown ideal-diode mode (OTSD-IDM) function as referenced in
Overtemperature Shutdown. The OTSD-IDM function is described in more detail in the following section.
8.3.10.1 Overtemperature-Shutdown Ideal-Diode Mode
Overtemperature-shutdown ideal-diode mode (OTSD-IDM) is implemented in LMG3522R030-Q1. As explained
in Overtemperature Shutdown, ideal-diode mode provides the best GaN FET protection when the GaN FET is
overheating.
OTSD-IDM accounts for all, some, or none of the power system operating when OTSD-IDM is protecting the
GaN FET. The power system may not have the capability to shut itself down, in response to the LMG3522R030-
Q1 asserting the FAULT pin in a GaN OTSD event, and just continue to try to operate. Parts of the power system
can stop operating due to any reason such as a controller software bug or a solder joint breaking or a device
shutting off to protect itself. At the moment of power system shutdown, the power system stops providing gate
drive signals but the inductive elements continue to force current flow while they discharge.
The OTSD-IDM state machine is shown in 图 8-6. Each state is assigned a state number in the upper right side
of the state box. The OTSD-IDM state machine has a similar structure to the OP-IDM state machine. Similar
states use the same state number.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
23
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
1
FET OFF
OTSD-IDM waiting for IN
falling-edge blank time to
expire
IN falling-edge blank
time expired
IN falling edge
IN falling edge
First-quadrant drain
current
2
3
FET OFF
FET ON
OTSD-IDM looking for negative
VDS
OTSD-IDM looking for first-
quadrant drain current
Negative VDS due to
third-quadrant drain
current
Enter OTSD
图8-6. Overtemperature-Shutdown Ideal-Diode Mode (OTSD-IDM) State Machine
1. The LMG3522R030-Q1 GaN FET always goes to state #1 if a falling edge is detected on the IN pin. OTSD-
IDM turns off the GaN FET in OTSD-IDM state #1. OTSD-IDM is waiting for the IN falling edge blank time to
expire. This time gives the opposite-side FET time to switch to create a positive drain voltage. After the blank
time expires, the device moves to OTSD-IDM state #2.
2. For OTSD-IDM state #2, OTSD-IDM keeps the GaN FET off if it is coming from OTSD-IDM state #1 and
turns the GaN FET off if it is coming from OTSD-IDM state #3. OTSD-IDM is monitoring the GaN FET drain
voltage in OP-IDM state #2. It is looking for a negative drain voltage which means third-quadrant current is
flowing. This is also the starting state when the device enters OTSD. After a negative GaN FET drain voltage
is detected, the device moves to OTSD-IDM state #3
3. OP-IDM turns on the GaN FET in OTSD-IDM state #3. OP-IDM monitors the drain current in this state. If
first-quadrant drain current is detected, the device moves to OP-IDM state #2.
State #1 is used to protect against shoot-through current. The state #1 in the OTSD-IDM state machine waits for
a fixed time period before proceeding to state #2. The fixed time period is to give the opposite-side switch time to
switch and create a positive drain voltage. A fixed time is used to avoid a stuck condition for cases where a
positive drain voltage is not created.
State #1 will help protect against shoot-through currents if the converter continues switching when the
LMG3522R030-Q1 enters OTSD. Meanwhile, if the converter initiates switching with the LMG3522R030-Q1
already in OTSD, shoot-through current protection can be obtained by switching the OTSD device first to force it
to progress though state #1. For example, the synchronous rectifier in a boost PFC can go into OTSD during
initial input power application as the inrush current charges the PFC output cap. A shoot-through current event
can be avoided if converter switching begins by switching the synchronous rectifier FET before switching the
boost PFC FET.
If there is no IN signal, the state machine only moves between states #2 and #3 as a classic ideal-diode mode
state machine. This allows all the inductive elements to discharge, when the power system shuts off, with
minimum discharge stress created in the GaN FET.
Note that the OTSD-IDM state machine has no protection against repetitive shoot-through current events. There
are degenerate cases, such as the LMG3522R030-Q1 losing its IN signal during converter operation, which can
expose the OTSD-IDM to repetitive shoot-through current events. There is no good solution in this scenario. If
OTSD-IDM did not allow repeated shoot-thru current events, the GaN FET would instead be exposed to
excessive off-state third-quadrant losses.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
24
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8.4 Start-Up Sequence
图8-7 shows the start up sequence of LMG3522R030-Q1.
Time interval A: VDD starts to build up. FAULT signal is initially pulled low.
Time interval B: After VDD passes the UVLO threshold VVDD,T+(UVLO), both LDO5V and VNEG start to built up. In a
typical case where CLDO5V = 100 nF and CVNEG = 2.2 μF, LDO5V reaches its UVLO threshold earlier than VNEG
.
The start-up time may vary if different capacitors are utilized. If VDD has some glitches and falls below UVLO
threshold VVDD,T-(UVLO) in this time interval, LDO5V and VNEG will stop building up and only resume when VDD
goes above VVDD,T+(UVLO) again. A longer start-up time is expected in this case.
Time interval C: After LDO5V and VNEG both reach their thresholds, the FAULT signal is cleared (pulled high)
and the device is able to switch following the IN pin signal.
VVDD,T+(UVLO)
VDD
VNEG
VLDO5V
t(start)
FAULT
(A)
(B)
(C)
图8-7. Start-Up Timing Diagram
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
25
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
8.5 Safe Operation Area (SOA)
8.5.1 Repetitive SOA
The allowed repetitive SOA for the LMG3522R030-Q1 (图 6-11) is defined by the peak drain current (IDS) and
the drain to source voltage (VDS) of the device during turn on. The peak drain current during switching is the sum
of several currents going into drain terminal: the inductor current (Iind); the current required to charge the COSS of
the other GaN device in the totem pole; and the current required to charge the parasitic capacitance (Cpar) on
the switching node. 310 pF is used as an average COSS of the device during switching. The parasitic
capacitance on the switch node may be estimated by using the overlap capacitance of the PCB. A boost
topology is used for the SOA testing. The circuit shown in 图 8-8 is used to generate the SOA curve in 图 6-11.
For reliable operation, the junction temperature of the device must also be limited to 125°C. The IDS of 图 6-11
can be calculated by:
IDS = Iind + (310 pF + Cpar) * Drain slew rate at peak current
where drain slew rate at the peak current is estimated between 70% and 30% of the bus voltage, and Cpar is the
parasitic board capacitance at the switched node.
Q1,Q2:
LMG3522R030
D
Q1
GaN
FAULT
IN
RDRV
Vbus
520 V
L
Temp, Current
L
O
A
D
Cout
S
D
Q2
Iind
Cpar
Vin
GaN
FAULT
50 pF
IN
RDRV
Temp, Current
S
图8-8. Circuit Used for SOA Curve
Refer to Achieving GaN Products With Lifetime Reliability for more details.
8.6 Device Functional Modes
The device has one mode of operation that applies when operated within the Recommended Operating
Conditions.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
26
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9 Application and Implementation
备注
以下应用部分中的信息不属于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。
9.1 Application Information
The LMG3522R030-Q1 is a power IC targeting hard-switching and soft-switching applications operating up to
520-V bus voltages. GaN devices offer zero reverse-recovery charge enabling high-frequency, hard-switching in
applications like the totem-pole PFC. Low Qoss of GaN devices also benefits soft-switching converters, such as
the LLC and phase-shifted full-bridge configurations. As half-bridge configurations are the foundation of the two
mentioned applications and many others, this section describes how to use the LMG3522R030-Q1 in a half-
bridge configuration.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
27
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.2 Typical Application
Vdc
D
LMG352x
CDRVH
RDRVH
RDRV
IN
Isolator
ISO77X1F
RINH
PWM_H
CNEGH
TEMP_H
VNEG
BBSW
TEMP
CINH
ROCH
LBBSWH
ZVD/OC_H
ISO
RFTH
/FAULT_H
CFTH COCH
ZVD/OC
/FAULT
LDO5V
VCC2
CLDOH
SGND
VDD
PWR
Transformer
driver
SN6505
CVDDH
GND
SGND
S
D
Cdecoupling
SW
LMG352x
CDRVL
RDRVL
RDRV
IN
Isolator
ISO77X1F
RINL
CINL
PWM_L
CNEGL
TEMP_L
ZVD/OC_L
/FAULT_L
VNEG
BBSW
TEMP
LBBSWL
ZVD/OC
/FAULT
LDO5V
CLDOL
VDD
PWR
Transformer
driver
SN6505
CVDDL
GND
SGND
S
图9-1. Typical Half-Bridge Application With Isolated Power Supply
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
28
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
Vdc
D
LMG352x
CDRVH
RDRVH
RDRV
IN
Isolator
ISO77X1F
RINH
PWM_H
TEMP_H
CNEGH
VNEG
BBSW
TEMP
CINH
ROCH
RFTH
LBBSWH
ZVD/OC_H
/FAULT_H
CFTH COCH
ZVD/OC
/FAULT
LDO5V
CLDOH
SGND
VDD
CVDDH
Dbootstrap
Rbootstrap
GND
S
D
DZ
SW
Cdecoupling
RDRVL_start
CDRVH
LMG352x
RDRV
IN
RDRVL
Isolator
ISO77X1F
RINL
CINL
PWM_L
TEMP_L
CNEGL
VNEG
BBSW
TEMP
LBBSWL
ZVD/OC_L
/FAULT_L
ZVD/OC
/FAULT
LDO5V
CLDOL
VDD
PWR
Transformer
driver
SN6505
CVDDL
GND
SGND
S
图9-2. Typical Half-Bridge Application With Bootstrap
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
29
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.2.1 Design Requirements
This design example is for a hard-switched boost converter which is representative of PFC applications. 表 9-1
shows the system parameters for this design.
表9-1. Design Parameters
DESIGN PARAMETER
EXAMPLE VALUE
200 VDC
400 VDC
20 A
Input voltage
Output voltage
Input (inductor) current
Switching frequency
100 kHz
9.2.2 Detailed Design Procedure
In high-voltage power converters, circuit design and PCB layout are essential for high-performance power
converters. As designing a power converter is out of the scope of this document, this data sheet describes how
to build well-behaved half-bridge configurations with the LMG3522R030-Q1.
9.2.2.1 Slew Rate Selection
The slew rate of LMG3522R030-Q1 can be adjusted between approximately 20 V/ns and 150 V/ns by
connecting a resistor, RRDRV, from the RDRV pin to GND. The RDRV pin is a high-impedance node if a large
RRDRV resistor is used. Therefore it can be susceptible to coupling from the drain or other fast-slewing high-
voltage nodes if it is not well-shielded. This will manifest itself as an unstable switching dv/dt and in extreme
cases transient faults due to the RDRV being detected as open. Shielding the pin in the layout should be a
priority, however if this coupling is still a problem, a cap of up to 1 nF from RDRV to GND can be added to
stabilize the pin voltage.
The slew rate affects GaN device performance in terms of:
• Switching loss
• Voltage overshoot
• Noise coupling
• EMI emission
Generally, high slew rates provide low switching loss, but high slew rates can also create higher voltage
overshoot, noise coupling, and EMI emissions. Following the design recommendations in this data sheet helps
mitigate the challenges caused by a high slew rate. The LMG3522R030-Q1 offers circuit designers the flexibility
to select the proper slew rate for the best performance of their applications.
9.2.2.1.1 Start-Up and Slew Rate With Bootstrap High-Side Supply
Using a bootstrap supply introduces additional constraints on the start-up of the high-side LMG3522R030-Q1.
Prior to powering up, the GaN device operates in cascode mode with reduced performance. In some circuits, a
slower slew rate can be required for the start-up of a bootstrap-supplied half-bridge configuration. More
information can be found in 节9.4.2.
9.2.2.2 Signal Level-Shifting
In half-bridges, high-voltage level shifters or digital isolators must be used to provide isolation for signal paths
between the high-side device and control circuit. Using an isolator is optional for the low-side device. However,
using and isolator equalizes the propagation delays between the high-side and low-side signal paths, and
provides the ability to use different grounds for the GaN device and the controller. If an isolator is not used on the
low-side device, the control ground and the power ground must be connected at the device and nowhere else on
the board. For more information, see Layout Guidelines. With fast-switching devices, common ground
inductance can easily cause noise issues without the use of an isolator.
Choosing a digital isolator for level-shifting is important for improvement of noise immunity. As GaN device can
easily create high dv/dt, > 50 V/ns, in hard-switching applications, TI highly recommends to use isolators with
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
30
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
high common-mode transient immunity (CMTI) and low barrier capacitance. Isolators with low CMTI can easily
generate false signals, which could cause shoot-through. The barrier capacitance is part of the isolation
capacitance between the signal ground and power ground, which is in direct proportion to the common mode
current and EMI emission generated during the switching. Additionally, TI strongly encourages to select isolators
which are not edge-triggered. In an edge-triggered isolator, a high dv/dt event can cause the isolator to flip states
and cause circuit malfunction.
Generally, ON/OFF keyed isolators with default output low are preferred, such as the TI ISO77xxF or ISO67xxF
series. Default low state ensures the system will not shoot-through when starting up or recovering from fault
events. As a high CMTI event would only cause a very short (a few nanoseconds) false pulse, TI recommends a
low pass filter, like 300-Ωand 22-pF R-C filter, to be placed at the driver input to filter out these false pulses.
9.2.2.3 Buck-Boost Converter Design
图 9-3 and 图 9-4 show the buck-boost converter efficiency versus load current with different inductors and peak
current modes. A minimum inductance value of 3 µH is preferred for the buck-boost converter so that the di/dt
across the inductor is not too high. This leaves enough margin for the control loop to respond. As a result, the
maximum di/dt of the inductor is limited to 6 A/µs. On the other hand, large inductance also limits the transient
response for stable output voltage, and it is preferred to have inductors less than 10 µH.
85
80
75
70
LBBSW = 3.3 H
LBBSW = 4.7 H
65
60
55
50
LBBSW = 10 H
0
10
20
30
40
50
60
70
80
90
Buck-Boost Load Current (mA)
VDD = 12 V
TJ = 25 °C
VDD = 12 V TJ = 25 °C
图9-4. Buck-Boost Efficiency vs Load When
图9-3. Buck-Boost Efficiency vs Load When
IBBSW,PK = IBBSW,PK(high)
IBBSW,PK = IBBSW,PK(low)
9.2.3 Application Curves
VDS (50 V/div)
ID (2.5 A/div)
VDS (50 V/div)
ID (1.25 A/div)
VOUT = 400V
IL = 5 A
RDRV = 40 kW
VBUS = 400 V
IL = 5 A
RDRV = 40 kW
Time (5 ns/div)
Time (5 ns/div)
D006
D007
图9-5. Turn-On Waveform in Application Example 图9-6. Turn-Off Waveform in Application Example
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
31
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.3 Do's and Don'ts
The successful use of GaN devices in general, and LMG3522R030-Q1 in particular, depends on proper use of
the device. When using the LMG3522R030-Q1, DO:
• Read and fully understand the data sheet, including the application notes and layout recommendations.
• Use a four-layer board and place the return power path on an inner layer to minimize power-loop inductance.
• Use small, surface-mount bypass and bus capacitors to minimize parasitic inductance.
• Use the proper size decoupling capacitors and locate them close to the IC as described in Layout Guidelines.
• Use a signal isolator to supply the input signal for the low-side device. If not, ensure the signal source is
connected to the signal GND plane which is tied to the power source only at the LMG3522R030-Q1 IC.
• Use the FAULT pin to determine power-up state and to detect overcurrent and overtemperature events and
safely shut off the converter.
To avoid issues in your system when using the LMG3522R030-Q1, DON'T:
• Use a single-layer or two-layer PCB for the LMG3522R030-Q1 as the power-loop and bypass capacitor
inductances is excessive and prevent proper operation of the IC.
• Reduce the bypass capacitor values below the recommended values.
• Allow the device to experience drain transients above 600 V as they can damage the device.
• Allow significant third-quadrant conduction when the device is OFF or unpowered, which can cause
overheating. Self-protection features cannot protect the device in this mode of operation.
• Ignore the FAULT pin output.
9.4 Power Supply Recommendations
The LMG3522R030-Q1 only requires an unregulated VDD power supply from 7.5 V to 18 V. The low-side supply
can be obtained from the local controller supply. The supply of the high-side device must come from an isolated
supply or a bootstrap supply.
9.4.1 Using an Isolated Power Supply
Using an isolated power supply to power the high-side device has the advantage that it works regardless of
continued power-stage switching or duty cycle. Using an isolated power supply can also power the high-side
device before power-stage switching begins, eliminating the power-loss concern of switching with an unpowered
LMG3522R030-Q1 (see Start-Up and Slew Rate With Bootstrap High-Side Supply for details). Finally, a
properly-selected isolated supply introduces less parasitics and reduces noise coupling.
The isolated supply can be obtained with a push-pull converter, a flyback converter, a FlyBuck™ converter, or an
isolated power module. When using an unregulated supply, the input of LMG3522R030-Q1 must not exceed the
maximum supply voltage. A 16-V TVS diode can be used to clamp the VDD voltage of LMG3522R030-Q1 for
additional protection. Minimizing the inter-winding capacitance of the isolated power supply or transformer is
necessary to reduce switching loss in hard-switched applications. Furthermore, capacitance across the isolated
bias supply inject high currents into the signal-ground of the LMG3522R030-Q1 and can cause problematic
ground-bounce transients. A common-mode choke can alleviate most of these issues.
9.4.2 Using a Bootstrap Diode
In half-bridge configuration, a floating supply is necessary for the high-side device. To obtain the best
performance of LMG3522R030-Q1, TI highly recommends Using an Isolated Power Supply. A bootstrap supply
can be used with the recommendations of this section.
In applications like a boost converter, the low side LMG3522R030-Q1 always start switching while high side
LMG3522R030-Q1 is unpowered. If the low side is adjusted to achieve very high slew rate before the high side
bias is fully settled, there can be unintentional turn-on at the high side due to parasitic coupling at high slew rate.
The start-up slew rate must be slowed down to 100 V/ns by changing the resistance of RDRV pin of the low side.
This slow down can be achieved by controlling the low side RDRV resistance with the high side FAULT as given
in 图9-1.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
32
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.4.2.1 Diode Selection
The LMG3522R030-Q1 offers no reverse-recovery charge and very limited output charge. Hard-switching
circuits using the LMG3522R030-Q1 also exhibit high voltage slew rates. A compatible bootstrap diode must not
introduce high output charge and reverse-recovery charge.
A silicon carbide diode, like the GB01SLT06-214, can be used to avoid reverse-recovery effects. The SiC diode
has an output charge of 3 nC. Althought there is additional loss from its output charge, it does not dominate the
losses of the switching stage.
9.4.2.2 Managing the Bootstrap Voltage
In a synchronous buck or other converter where the low-side switch occasionally operates in third-quadrant, the
bootstrap supply charges through a path that includes the third-quadrant voltage drop of the low-side
LMG3522R030-Q1 during the dead time as shown in 图 9-7. This third-quadrant drop can be large, which can
over-charge the bootstrap supply in certain conditions. The VDD supply of LMG3522R030-Q1 must be kept
below 18 V.
DRAIN
VDD
VF
SOURCE
+
œ
DRAIN
VDD
VF
SOURCE
图9-7. Charging Path for Bootstrap Diode
As shown in 图9-8, the recommended bootstrap supply includes a bootstrap diode, a series resistor, and a 16-V
TVS or zener diode in parallel with the VDD bypass capacitor to prevent damaging the high-side LMG3522R030-
Q1. The series resistor limits the charging current at start-up and when the low-side device is operating in third-
quadrant mode. This resistor must be selected to allow sufficient current to power the LMG3522R030-Q1 at the
desired operating frequency. At 100-kHz operation, TI recommends a value of approximately 2 Ω. At higher
frequencies, this resistor value must be reduced or the resistor omitted entirely to ensure sufficient supply
current.
DRAIN
+12 V
VDD
VF
SOURCE
图9-8. Suggested Bootstrap Regulation Circuit
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
33
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.5 Layout
9.5.1 Layout Guidelines
The layout of the LMG3522R030-Q1 is critical to its performance and functionality. Because the half-bridge
configuration is typically used with these GaN devices, layout recommendations are considered with this
configuration. A four-layer or higher layer count board is required to reduce the parasitic inductances of the
layout to achieve suitable performance. 图 9-9 summarizes the critical layout guidelines, and more details are
further elaborated in the following sections.
Vdc
Use RC circuits if
SGND is floating to
FET signal ground
for CMTI
Use high-side signal
ground plane
(connected to SW)
to shield traces
D
LMG352x
CDRVL
RDRVH
Minimize power
loop to mitigate
transient
RDRV
IN
overvoltage
Isolator
ISO77X1F
RINH
PWM_H
TEMP_H
CNEGH
VNEG
BBSW
TEMP
CINH
ROCH
LBBSWH
ZVD/OC_H
RFTH
/FAULT_H
CFTH COCH
ZVD/OC
/FAULT
LDO5V
CLDOH
SGND
VDD
PWR
Transformer
driver
SN6505
CVDDH
GND
SGND
S
D
Recommend to parallel a
cap to reduce the noise
Minimize the loops of
these caps to mitigate
switching noises
Use Kelvin connection
between high side signal
ground and SW
SW
Cdecoupling
Use low-side signal
ground plane
(connected to GND)
to shield traces
LMG352x
CDRVL
RDRV
IN
RDRVL
Isolator
ISO77X1F
RINL
CINL
PWM_L
CNEGL
TEMP_L
ZVD/OC_L
/FAULT_L
VNEG
BBSW
TEMP
LBBSWL
ZVD/OC
/FAULT
LDO5V
CLDOL
VDD
PWR
Transformer
driver
SN6505
CVDDL
GND
SGND
S
Minimize the loops of
these caps to mitigate
switching noises
Use Kelvin connection
between low side signal
ground and GND
PGND
图9-9. Typical Schematic With Layout Considerations
9.5.1.1 Solder-Joint Reliability
Large QFN packages can experience high solder-joint stress. TI recommends several best practices to ensure
solder-joint reliability. First, the instructions for the NC1 and NC2 anchor pins found in 表 5-1 must be followed.
Second, all the LMG3522R030-Q1 board solder pads must be non-solder-mask defined (NSMD) as shown in the
land pattern example in Mechanical, Packaging, and Orderable Information. Finally, any board trace connected
to an NSMD pad must be less than 2/3 the width of the pad on the pad side where it is connected. The trace
must maintain this 2/3 width limit for as long as it is not covered by solder mask. After the trace is under solder
mask, there are no limits on the trace dimensions. All these recommendations are followed in the Layout
Example.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
34
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.5.1.2 Power-Loop Inductance
The power loop, comprising of the two devices in the half bridge and the high-voltage bus capacitance,
undergoes high di/dt during switching events. By minimizing the inductance of this loop, ringing and
electromagnetic interference (EMI) can be reduced, as well as reducing voltage stress on the devices.
Place the power devices as close as possible to minimize the power loop inductance. The decoupling capacitors
are positioned in line with the two devices. They can be placed close to either device. In Layout Examples, the
devices are placed on the bottom layer and the decoupling capacitors are placed on the top layer. The PGND is
placed on the top layer, the HVBUS is located on top and third layer, and the switching node is on the top layer.
They are connected to the power devices on bottom layer with vias. Area of traces close to the devices are
minimized by bottom layer in order to keep clearance between heatsink and conductors.
The power loop inductance can be estimated based on the ringing frequency fring of the drain-source voltage
switching waveform based on the following equation:
1
L
=
(1)
pl
2
2
4π f
C
ring ring
where Cring is equal to COSS at the bus voltage (refer to 图 6-8 for the typical value) plus the drain-source
parasitic capacitance from the board and load inductor or transformer.
As the parasitic capacitance of load components is hard to character, it is recommended to capture the VDS
switching waveform without load components to estimate the power loop inductance. Typically, the power loop
inductance of the Layout Example is around 2.5 nH.
9.5.1.3 Signal-Ground Connection
The LMG3522R030-Q1's SOURCE pins are internally connected to the power IC signal ground. The return path
for the passives associated to the driver (for example, bypass capacitance) must be connected to the SOURCE
pins. Local signal ground planes must be connected to SOURCE pins with low impedance star connection. In
Layout Examples, local signal ground planes are located on third layer to act as the return path for the local
circuitry, and connected to the SOURCE pins with vias between the third layer and the bottom layer.
9.5.1.4 Bypass Capacitors
The gate drive loop impedance must be minimized to obtain good performance. Although the gate driver is
integrated on package, the bypass capacitance for the driver is placed externally. As the GaN device is turned
off to a negative voltage, the impedance of the path to the external VNEG capacitor is included in the gate drive
loop. The VNEG capacitor must be placed close to VNEG and SOURCE pins. In the 节 9.5.2, the bypass
capacitors, C3 and C13, are located in the top layer and are connected to VNEG pins with vias and SOURCE
pins through the local signal ground plane.
The VDD pin bypass capacitors, C1 and C11, must also be placed close to the VDD pin with low impedance
connections.
9.5.1.5 Switch-Node Capacitance
GaN devices have very low output capacitance and switch quickly with a high dv/dt, yielding very low switching
losses. To preserve this low switching losses, additional capacitance added to the output node must be
minimized. The PCB capacitance at the switch node can be minimized by following these guidelines:
• Minimize overlap between the switch-node plane and other power and ground planes.
• Make the GND return path under the high-side device thinner while still maintaining a low-inductance path.
• Choose high-side isolator ICs and bootstrap diodes with low capacitance.
• Place the power inductor as close to the GaN device as possible.
• Power inductors must be constructed with a single-layer winding to minimize intra-winding capacitance.
• If a single-layer inductor is not possible, consider placing a small inductor between the primary inductor and
the GaN device to effectively shield the GaN device from the additional capacitance.
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
35
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
• If a back-side heat-sink is used, use the least amount of area of the switch-node copper coverage on the
bottom copper layer to improve the thermal dissipation.
9.5.1.6 Signal Integrity
The control signals to the LMG3522R030-Q1 must be protected from the high dv/dt caused by fast switching.
Coupling between the control signals and the drain can cause circuit instability and potential destruction. Route
the control signals (IN, FAULT and OC) over a ground plane placed on an adjacent layer. In Layout Example, for
example, all the signals are routed on layers close to the local signal ground plane.
Capacitive coupling between the traces for the high-side device and the static planes, such as PGND and
HVBUS, could cause common mode current and ground bounce. The coupling can be mitigated by reducing
overlap between the high-side traces and the static planes. For the high-side level shifter, ensure no copper from
either the input or output side extends beneath the isolator or the CMTI of the device can be compromised.
9.5.1.7 High-Voltage Spacing
Circuits using the LMG3522R030-Q1 involve high voltage, potentially up to 650 V. When laying out circuits using
the LMG3522R030-Q1, understand the creepage and clearance requirements for the application and how they
apply to the GaN device. Functional (or working) isolation is required between the source and drain of each
transistor, and between the high-voltage power supply and ground. Functional isolation or perhaps stronger
isolation (such as reinforced isolation) can be required between the input circuitry to the LMG3522R030-Q1 and
the power controller. Choose signal isolators and PCB spacing (creepage and clearance) distances which meet
your isolation requirements.
If a heat sink is used to manage thermal dissipation of the LMG3522R030-Q1, ensure necessary electrical
isolation and mechanical spacing is maintained between the heat sink and the PCB.
9.5.1.8 Thermal Recommendations
The LMG3522R030-Q1 is a lateral device grown on a Si substrate. The thermal pad is connected to the source
of device. The LMG3522R030-Q1 can be used in applications with significant power dissipation, for example,
hard-switched power converters. In these converter, TI recommends a heat sink connected to the top side of
LMG3522R030-Q1. The heat sink can be applied with thermal interface materials (TIMs), like thermal pad with
electrical isolation.
Refer to the High Voltage Half Bridge Design Guide for LMG3410 Smart GaN FET application note for more
recommendations and performance data on thermal layouts.
Copyright © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
36
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
9.5.2 Layout Examples
Correct layout of the LMG3522R030-Q1 and its surrounding components is essential for correct operation. The
layouts shown here reflect the GaN device schematic in 图 9-1. These layouts are shown to produce good
results and is intended as a guideline. However, it can be possible to obtain acceptable performance with
alternate layout schemes. Additionally, please refer to the land pattern example in Mechanical, Packaging, and
Orderable Information for the latest recommended PCB footprint of the device.
The the top-layer layout, mid-layer and bottom-layer layout are shown. The mid-layer layout includes the outlines
of the top level components to assist the reader in lining up the top-layer and mid-layer layouts.
Minimize the loops of C1,
C3, C11, C13.
Isolation boundary
图9-10. Half-Bridge Top-Layer Layout
Kelvin connections between
Low side signal
Source and signal ground plane
ground plane for
shielding
High side signal
ground plane for
shielding
Minimize the power loop by
using vertical structure and
nearest signal layer for return
图9-11. Half-Bridge Third-Layer (Light Grey) and Bottom-Layer (Dark Grey) Layout
Copyright © 2023 Texas Instruments Incorporated
Submit Document Feedback
37
Product Folder Links: LMG3522R030-Q1
English Data Sheet: SNOSD97
LMG3522R030-Q1
ZHCSNE5C –OCTOBER 2020 –REVISED JUNE 2023
www.ti.com.cn
10 Device and Documentation Support
10.1 Documentation Support
10.1.1 Related Documentation
• Texas Instruments, High Voltage Half Bridge Design Guide for LMG3410 Smart GaN FET application note
• Texas Instruments, A New Approach to Validate GaN FET Reliability to Power-line Surges Under Use-
conditions
• Texas Instruments, Achieving GaN Products With Lifetime Reliability
• Texas Instruments, Direct-drive configuration for GaN devices
10.2 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。
10.3 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 的《使用条款》。
10.4 Trademarks
FlyBuck™ is a trademark of Texas Instruments.
TI E2E™ is a trademark of Texas Instruments.
所有商标均为其各自所有者的财产。
10.5 静电放电警告
静电放电(ESD) 会损坏这个集成电路。德州仪器(TI) 建议通过适当的预防措施处理所有集成电路。如果不遵守正确的处理
和安装程序,可能会损坏集成电路。
ESD 的损坏小至导致微小的性能降级,大至整个器件故障。精密的集成电路可能更容易受到损坏,这是因为非常细微的参
数更改都可能会导致器件与其发布的规格不相符。
10.6 Export Control Notice
Recipient agrees to not knowingly export or re-export, directly or indirectly, any product or technical data (as
defined by the U.S., EU, and other Export Administration Regulations) including software, or any controlled
product restricted by other applicable national regulations, received from disclosing party under nondisclosure
obligations (if any), or any direct product of such technology, to any destination to which such export or re-export
is restricted or prohibited by U.S. or other applicable laws, without obtaining prior authorization from U.S.
Department of Commerce and other competent Government authorities to the extent required by those laws.
10.7 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。
11 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 © 2023 Texas Instruments Incorporated
English Data Sheet: SNOSD97
38
Submit Document Feedback
Product Folder Links: LMG3522R030-Q1
PACKAGE OPTION ADDENDUM
www.ti.com
21-Jul-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
2000
250
(1)
(2)
(3)
(4/5)
(6)
LMG3522R030QRQSRQ1
LMG3522R030QRQSTQ1
XMG3522R030QRQSTQ1
ACTIVE
VQFN
VQFN
VQFN
RQS
52
52
52
RoHS-Exempt
& Green
NIPDAU
Level-3-260C-168 HR
Level-3-260C-168 HR
Call TI
-40 to 150
-40 to 150
-40 to 150
LMG3522R030Q
LMG3522R030Q
Samples
Samples
Samples
ACTIVE
ACTIVE
RQS
RoHS-Exempt
& Green
NIPDAU
Call TI
RQS
250
TBD
(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.
Addendum-Page 1
PACKAGE OPTION ADDENDUM
www.ti.com
21-Jul-2023
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.
OTHER QUALIFIED VERSIONS OF LMG3522R030-Q1 :
Catalog : LMG3522R030
•
NOTE: Qualified Version Definitions:
Catalog - TI's standard catalog product
•
Addendum-Page 2
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 的销售条款或 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:
©2020 ICPDF网 联系我们和版权申明