Power management forum - Recent Threads

RE: LM25019: LM25019 Stability Issues

Niranjan Nair — Mon, 10 Aug 2026 08:27:15 GMT

Hi Shivam

Ok, let me know how that goes

Regards,

Niranjan

LM25019: LM25019 Stability Issues

Shivam Chakraborty — Fri, 24 Jul 2026 06:40:00 GMT

Part Number: LM25019

Hi,

I am using the LM25019 COT buck IC from TI to step down 36V (Vin nominal) to 12V. The input voltage range is from 32V (Vmin) to 40V (Vmax).

Below is the attached schematic for reference.

The designed switching frequency of the inductor is 350 kHz, and a type 3 feedback ripple circuit is selected for min ripple at the output.

With the above attached circuit, I am getting stability issues. There are multiple bursts of pulses in close succession, followed by a long off time, and the pattern repeats. This causes the output to fluctuate from 10V to 0V.

Could you please review my design and give proper suggestions for its improvement? I require 12.8V at the output for the operating input voltage range. The output load varies from 0.45W to 1W.

Thanks and regards,

Shivam

TPSM365R15FEVM: TPSM365R15FRDNR with incorrect output

Jeff Krupp — Sun, 09 Aug 2026 21:53:04 GMT

Part Number: TPSM365R15FEVM

Hello. I am using the TPSM365R15FRDNR in fixed-output 5V configuration. It seems to regulate at 2.49V both loaded or unloaded. Vcc is at 3.2V. I've tried it both in fixed-output and reworked with an external voltage divider (40k+10k), but both give me 2.49V (interestingly half the intended Vout). I've also tried changing C45 to a 10uF instead of 100uF

Any other ideas?

RE: TPSM365R15FEVM: TPSM365R15FRDNR with incorrect output

Niranjan Nair — Mon, 10 Aug 2026 08:26:18 GMT

Hi,

That is a strange behaviour. Can you try out a few things :-

1) Does this behaviour have any variation with the input voltage?

2) What is the voltage at the FB pin (when you have the 40k+10k resistor divider)?

3) Does this happen with multiple ICs? Can you swap the IC out for a new one and check?

Regards,

Niranjan

LP5018: LP5018 I2C rise time

qinjiang zhu — Tue, 04 Aug 2026 00:25:15 GMT

Part Number: LP5018

In my application of the LP5018, I've encountered an I2C timing issue. Specifically, I'm unable to meet the I2C rise time requirement, which must be within 300ns. My measured I2C rise time is around 580ns, and my I2C clock rate is 183kHz.

--While the rise time limit for 400kHz is 300ns and for 100kHz is 1000ns, there's no standard for intermediate rates. I'm wondering if, for instance, 200kHz might allow for a rise time of up to 650ns?

--I've identified the source of this delay. There's a 500mm cable between my responding slave device and the master, and it's adding roughly 300ns of delay.

--I attempted to reduce the rise time by lowering the pull-up resistor value to around 2k, which did bring the rise time within the 300ns limit. However, this caused a new issue: the low-level output voltage (VOL) on the bus was raised to about 0.5V. Since the LP5018's maximum allowable low-level input voltage (VIL) is 0.4V, this unfortunately exceeds the device's specification.

--Could you please advise on any recommended measures to optimize the I2C bus rise time, while also keeping cost considerations in mind?

RE: LP5018: LP5018 I2C rise time

Felix Wang — Mon, 10 Aug 2026 08:16:42 GMT

Hi qinjiang,

For the question 2 you may need to contact Sienna for support, as I'm mainly responsible for LED driver support.

For the question 1, yes, as the output low level on A side is 0.2V.

For the question 3, from my thought there should be no difference to place the buffer on either side. As the purpose of adding the buffer is strengthening the pull-up capability and controlling the output low level.

Best regards,

Felix

TPS7A20L: 6-8mV noise is present in LDO output

Anal E S — Mon, 10 Aug 2026 08:08:17 GMT

Part Number: TPS7A20L

Hi TI Team,

We are currently using the MIC5365-1.8YMT-TZ LDO from Microchip Technology in one of our products. The application has a high dependency on power-supply noise and output stability.

As part of an improvement study, we evaluated the TPS7A20L18PDQNR from Texas Instruments as a potential replacement. Based on the datasheet specifications, the TPS7A20 appears to offer better performance in several areas, particularly in terms of output noise, PSRR, quiescent current, and load capability.

We therefore replaced the MIC5365 with the TPS7A20L18PDQNR and tested it on our existing PCBA. Both devices are footprint-compatible, so we were able to evaluate them on the same PCB without changing the surrounding circuit.

However, we are seeing an unexpected result in our practical application.

Test Conditions

PCB: Same PCBA
Load: Approximately 30 mA
Input voltage: Same for both tests
Output voltage: 1.8 V
External components: Same PCB components and layout
Only the LDO was changed: TPS7A20L18PDQNR ↔ MIC5365-1.8YMT-TZ

TPS7A20L18PDQNR Result

The attached oscilloscope waveform shows the output of the TPS7A20L18PDQNR. We are observing considerably more noise/ripple and less stable output compared with the MIC5365.

MIC5365-1.8YMT-TZ Result

As a part of the same test, we replaced the TPS7A20L18PDQNR with the original MIC5365-1.8YMT-TZ and tested the same PCBA under the same load conditions.

Interestingly, the MIC5365 produces a noticeably cleaner and more stable 1.8 V output with lower observed noise.

This is contrary to what we expected based on the published specifications of the two devices.

Our Question

Could you please help us understand why the TPS7A20L18PDQNR may be exhibiting higher output noise in this particular application, despite its datasheet specifications indicating significantly lower noise and better PSRR?

Since both devices were tested:

On the same PCBA
With the same PCB layout
With the same input voltage
With the same 30 mA load
Using the same surrounding components

we are trying to determine whether there could be a TPS7A20-specific stability or layout consideration that we are missing.

We would appreciate any feedback or recommendations from TI regarding the expected behavior and what parameters we should investigate.

We can also provide capacitor part numbers, oscilloscope settings if required.

Thank you in advance for your support.

UCC25661: UCC256612: Duty cycle asymmetry at startup (Vcap_div ≠ Vin/2) and causing output overvoltage

yathavakannan viswa — Fri, 07 Aug 2026 12:32:57 GMT

Part Number: UCC25661

Hello TI Support Team,

During the initial startup phase, the controller exhibits a severely uneven duty cycle, which destabilizes the resonant tank and causes a high-voltage spike on the output.

Instead of starting with symmetric, high-frequency pulses, the controller commands an extremely asymmetric duty cycle. Specifically, we are observing either the High-Side or Low-Side FET remaining ON for extended durations exceeding 5µs during this transient state.

Because of this uneven ON time, the resonant capacitor divider node fails to establish and settle at the required V_in/2 midpoint voltage. As a result, the full V_in voltage is effectively applied directly across the transformer primary, resulting in a output overvoltage condition.

Yellow Trace (C1): Switch Node (FET switching node)
Green Trace (C2): Capacitor Divider Node

swnode vs cap div at 220Vin.zip

RE: UCC25661: UCC256612: Duty cycle asymmetry at startup (Vcap_div ≠ Vin/2) and causing output overvoltage

yathavakannan viswa — Mon, 10 Aug 2026 07:56:15 GMT

Hi Sougata,

Please find the SCH and design calculator documents.e2e.ti.com/.../UCC256612.zip

TPS2065: Looking for SOIC-8 1.5A Power Switch with Active-High Enable

Yuxi Liu(Tony) — Mon, 10 Aug 2026 07:55:07 GMT

Part Number: TPS2065

Hi Team,

My customer is currently using the TPS2065DR in an SOIC-8 package. They need to upgrade to a 1.5A current limit version because their actual continuous current has reached 1.2A.

However, they are facing a pin-to-pin migration issue:

The TPS2068DR matches their package and current requirements, but its EN logic is active-low, which is opposite to the TPS2065DR's active-high logic.
The TPS2069 matches the required active-high logic and 1.5A rating, but it does not appear to have an SOIC-8 package option.

Could you please help clarify the following questions:

Is there a TPS2069DR available in the SOIC-8 package, or any plan for it?
For the TPS2068DR, can the EN pin be left floating? Does this chip have an internal pull-up resistor on the EN pin?
If neither option works, do we have any other 1.5A, active-high, pin-to-pin compatible power switches in the SOIC-8 package to recommend?

Thanks and best regards,
Tony Liu

RE: TPS62875-Q1: Requisition to verify the use of parallel inductor.

Dinesh Bala Surya — Mon, 10 Aug 2026 07:43:45 GMT

Hi Tanvee,

Thank you for your support and clarification.

Regards,
Dinesh Bala Surya

TPS62875-Q1: Requisition to verify the use of parallel inductor.

Dinesh Bala Surya — Thu, 06 Aug 2026 09:46:59 GMT

Part Number: TPS62875-Q1

Hi,

This is Dinesh Bala Surya from iWave Global. We have a plan to use TPS62875QWRZVRQ1 this regultor for generating 0.8V. But as per our required package size and current, the inductor of 0.1uH is not available as suggested in webench. We have inductor with 0.22uH and 10A. Could you please verify that shall we use 0.22uH, 20A inductor x2 in parallel for this regulator instead of 0.1uH x1?

Looking forward to your response.

Regards,
Dinesh Bala Surya

BQ24618: TTC pull Low ,disable Termination function

gary hsu — Mon, 10 Aug 2026 07:36:26 GMT

Part Number: BQ24618

Hello,

TTC pull Low ,disable Termination function (sense resistor =10m ohm), When the current is less than approximately 125mA, it enters nonsynchronous mode. During this mode, a large current ripple occurs (Ip-p = 900mA), accompanied by unusual noise. After a period of time, it returns to a normal ripple.
Questions:
1. Is this entry into nonsynchronous mode normal for application?
2. Does this IC provide a condition where the battery cut-off current is < 50mA?

Thank you.

Best regards,

TPS717: Request forinformation - TPS71711DCKT IC

Kateryna Krupa — Mon, 10 Aug 2026 07:35:20 GMT

Part Number: TPS717

Dear TI team,

I would like to ask your urgent support. We currently have a significant quantity of ICs (TPS71711DCKT) in stock with date code 1831, and we would like to confirm whether these components can still be used in production. Could you please provide us with information regarding:

whether ICs with this datecode are still suitable for production use
the recommended method for checking wether the components are still in good condition and meet the required specifications
Any specific tests, inspections, qualification procedures you recommend before using them in production
Whether there are any storage-life limitations or other factors we should consider for these parts?

Wewould appreciate your guidance on how to verify the condition and reliability of these components before releasing them for production

Thank you in advance for your support.

Kateryna Krupa

RE: TPS62875-Q1: Requisition to verify the use of parallel inductor.

Tanvee — Mon, 10 Aug 2026 07:34:46 GMT

Hi Dinesh,

Unfortunately, we do not have AEQ graded module.

Currently, only possible solution I see on our end is to use 2 x TPS62873-Q1 in stack operation.

Best regards,

Tanvee

UCC33421-Q1: obsolescence report for the components listed below

Hardware Intern — Fri, 07 Aug 2026 12:30:39 GMT

Part Number: UCC33421-Q1

Hello Team,
"UCC33421QDHARQ1 " This the newly launched component of TI , can you please share Grade(Automotive/Industrial/Military) , Storing Tempreture and obsolescence report for the components listed below. Our goal is to ensure long-term availability for a minimum of 10 years.

To assist with our internal tracking, could you please categorize each component according to the following criteria:

Category A: EOL > 10 years
Category B: EOL > 5 years

Components for Review:

RE: UCC33421-Q1: obsolescence report for the components listed below

Prateek Verma — Mon, 10 Aug 2026 07:27:07 GMT

Hi,

TI typically will not EOL a device if there are any customers still ordering. We do not plan to EOL this part.

So it should fit in Category A : EOL > 10Years

Thanks,

Prateek

LM5050-1: M5050-1 gate drive reduces during IEC 61000-4-4 EFT causing output voltage dip

Kamal Din — Thu, 16 Jul 2026 15:49:34 GMT

Part Number: LM5050-1

Hi,

I am using the LM5050-1 as an ideal diode controller for a 12 V redundant power supply.

The PSU passes EFT when tested on its own. However, when the LM5050 ideal diode stage is connected, the system fails IEC 61000-4-4 EFT.

Circuit

The circuit closely follows the LM5050 datasheet:

LM5050-1
N-channel MOSFET (Infineon IPA029N06NXKSA1)
R18 = zero ohm
Standard IN, OUT and VS connections as shown in the datasheet
OFF pin left floating (R2 not fitted)
C13 fitted between IN and OUT
VS filtered by R1/C3 as per the reference circuit

Investigation performed

I have carried out a number of investigations:

Replaced the busbar with twisted pair wiring.
Added common-mode chokes.
Added Y capacitors.
Tried different grounding arrangements.
Measured the PSU output before the ideal diode stage.
Shorted the MOSFET drain and source (problem disappears).
Measured Gate, VGS, OUT, IN and VS during EFT.

Measurements

The measurements show:

PSU output before the LM5050 remains clean.
LM5050 IN remains essentially constant.
LM5050 VS tracks the output with only a small disturbance.
During EFT, the LM5050 OUT (gate drive) reduces by several volts.
- MOSFET gate voltage follows the LM5050 OUT pin (R18 = zero ohm).
- VGS reduces by several volts for a few milliseconds before recovering.
- The output voltage only dips slightly (approximately 0.5 V) before recovering.
The gate drive reduction appears to be initiated by the LM5050 rather than by the MOSFET itself.
Additional observation

I also tested the TI LM5050 evaluation board previously under similar EFT conditions and observed similar behaviour, which makes me think this may not be PCB-layout related.

Questions
1. Is this behaviour expected during IEC 61000-4-4 EFT?
2. Are there any known recommendations to improve EFT immunity with the LM5050-1?
3. Is there any recommended modification to the standard application circuit?
4. Is there a newer TI ideal diode controller recommended for redundant 12 V power supplies with improved EFT immunity?
Any guidance would be greatly appreciated.

1 - AV5233_EPD_Rev4.pdf

2-Layout Top

2-Layout middle 1

3 -Layout middle 2

5 - Bottom

Plot ( Vo and Vgs)

Evaluation board plots ( Output and Vg)

RE: LM5050-1: M5050-1 gate drive reduces during IEC 61000-4-4 EFT causing output voltage dip

Shiven Dhir — Mon, 10 Aug 2026 07:22:58 GMT

Hi Kamal,

After some investigation we have found that that EFT pulse is causing the reverse current mechanism to activate. It is replicable on our EVM as well because the device is reacting as it is supposed to. When EFT burst is applied, VIN will get a jitter and cause reverse current condition.

Probing VIN may not show is since EFT burst is usually high frequency and low amplitude.

There can be only 2 ways to fix it.

1. Add more CIN capacitors to decouple the burst.

2. Increase value of C in this work around further.

Regards,

Shiven Dhir

RE: LM7480-Q1: LM74800 + LM74801 cascaded reverse polarity protection issue

Shiven Dhir — Mon, 10 Aug 2026 07:18:59 GMT

Closing this thread to avoid duplicity.

Regards,

Shiven Dhir

LM7480-Q1: LM74800 + LM74801 cascaded reverse polarity protection issue

MUHAMMED HASHIR M — Sat, 08 Aug 2026 10:57:22 GMT

Part Number: LM7480-Q1

Hi TI team,

I am using two reverse-polarity/reverse-current protection stages in series:

24 V Input -> LM74800 – Stage 1 -> LM74801 – Stage 2 -> 24 V Fan (0.18 A

My requirement is to use the first stage for reverse-polarity protection and the second stage to prevent regenerative current from the motor controllers from flowing back to the other 24 V loads.

I tested the circuit in two conditions:

1. Stage 2 removed:The circuit works continuously for several hours without any issue.

24 V → LM74800 → fan (24 V / 0.18 A)

2. Stage 2 connected:

24 V → LM74800 → LM74801 → fan (24 V / 0.18 A)

The circuit initially works correctly, but after approximately 1–2 minutes, the HGATE of the first-stage LM74800 falls to 0 V and the output voltage collapses.

After the failure, I measured:

EN/UVLO = 23.48 V
OV = 0.809 V
VS = 23.48 V
CAP − VS = 4.3 V
DGATE − A = 0.17 V
HGATE − OUT = 0 V
OUT = 0.24 V

The second-stage MOSFET has also been checked and appears healthy.

Since the first LM74800 works normally for hours when the second stage is removed, I suspect the second-stage LM74801 or its startup/charge-pump behavior may be interacting with the first-stage LM74800.

Could you please clarify:

Is cascading LM74800 followed by LM74801 recommended/supported?
What could cause the first-stage CAP − VS voltage to fall from its normal ~12 V to only 4.3 V after the second stage is connected?
Since EN/UVLO and OV are within normal levels, what condition could cause the first-stage HGATE to turn OFF?
Could the LM74801 second stage or its charge pump/input capacitance cause this behavior in the first-stage LM74800?

I have attached my schematic for reference.

TPS62A01: -25C 0.6Vout voltage droop

Cameron Wutzke — Fri, 07 Aug 2026 01:30:27 GMT

Part Number: TPS62A01

Hello!

curious for opinions on a few angles we are debugging right now. We see VOUT droop about 30mV at cold temp and high temp. It's spurratic which feels like something marginal to me:

1) anything marginal worth checking? We are right on cusp of datasheet VOUT

2) would our SPICE model be accurate in recreating this at a -25C simulation?

open to debugging ideas. Thanks!

RE: TPS62A01: -25C 0.6Vout voltage droop

Andreas Mueller — Mon, 10 Aug 2026 07:18:53 GMT

Hello Cameron,

thanks for using E2E.

30mV voltage drop at Vout=0.6V is too much. The datasheet is specifying ±9mV feedback accuracy. I could help reviewing the schematic and layout if needed.

The PSPICE model does not support the simulation of temperature dependencies.

Let me know if you have any questions or comments.

Best regards,

Andreas.

LM7480-Q1: Can two LM7480 reverse polarity protection circuits be connected in series?

MUHAMMED HASHIR M — Mon, 03 Aug 2026 11:39:59 GMT

Part Number: LM7480-Q1

Hi TI Team,

Can two LM7480-Q1 reverse polarity protection circuits be connected in series?

My configuration is:

The first reverse polarity stage powers only a few small loads (less than 3 A) and also feeds the input of the second reverse polarity stage. The high-current motor controllers are connected only after the second reverse polarity stage.

Is this configuration supported? Will the first LM7480 have any impact on the operation or current capability of the second LM7480, or are there any design considerations when cascading two reverse polarity protection stages?

I am facing an issue where, when the motor load is connected after the second RP stage, the output voltage of the second RP stage fluctuates. At the same time, both DGATE and HGATE voltages of the second LM7480 varying

Thank you.

RE: LM7480-Q1: Can two LM7480 reverse polarity protection circuits be connected in series?

Shiven Dhir — Mon, 10 Aug 2026 07:17:49 GMT

Hi Muhammed,

Do you have an oscilloscope?

When stage 2 is connected, and load is turned on, please probe the behavior of Stage 1 device. Please probe VCAP-VS, DGATE, HGATE, VIN etc. Do you see any oscillations?

Since the VCAP-VS is only 4.3V, HGATE will stay off. VCAP-VS needs to ~6.6V for HGATE to operate. Also, if you use 2xLM74801, did you find the same issue?

Regards,

Shiven Dhir