LM26001: output voltage transient

tom greive

Part Number: LM26001

We have designed in a LM26001. It supplies 5 volts at about 50mA, from a 24 volt power supply. It works most of the time flawlessly, but if power is removed for a short duration, I'm getting a voltage transient that peaks at 23 volts and doesn't return to 5 volts for 40 mS. It seems to occur if the power is applied as the 5 volt line drops to about 4 volts. I have already burned up 3 CPU's. Please let me know what I should try next.

thanks
tom

over 6 years ago

0 Samuel Jaffe over 6 years ago

TI__Mastermind 25590 points

Tom,

Please share schematic, layout, and scope shots which demonstrate this issue. We will investigate from there.

Thanks,
-Sam

0 tom greive over 6 years ago in reply to Samuel Jaffe

Prodigy 70 points

lm26001_transient.pdf

0 Samuel Jaffe over 6 years ago in reply to tom greive

TI__Mastermind 25590 points

Tom,

That definitely looks abnormal. I will contact the product experts to see what might cause this.

In the meantime can you please share scope shots of:

The switch node while VOUT is rising past desired VOUT
FB during this event
COMP during this event
SS during this event

Thanks,

-Sam

0 tom greive over 6 years ago in reply to Samuel Jaffe

Prodigy 70 points

I'm not sure what part of the wave you are interested in but I kept the top (yellow) trace on the 5v line for reference. I had to change the volt/div for the blue trace, and time scale on some of the pics to see more detail.Sw_pin.pdf FB_pin.pdf Comp_pin.pdf ss_pin.pdf Enable ckt.pdf

0 Samuel Jaffe over 6 years ago in reply to tom greive

TI__Mastermind 25590 points

Tom,

Thank you for the detailed report. This is very unique and will require some testing on our end which may take some time.

I am glad you found the EN workaround. Please continue to use this until we are able to come back with more information on this issue.

Thanks and you will be hearing from us in the coming days.

-Sam

0 Katelyn Wiggenhorn over 6 years ago in reply to Samuel Jaffe

TI__Genius 11040 points

Hello,

Thank you for sharing the detailed reports on your issue. I am ordering an EVM, and I will test your conditions on the EVM.

In addition what is the nominal load during this condition?

Thanks,
Katelyn

0 tom greive over 6 years ago in reply to Katelyn Wiggenhorn

Prodigy 70 points

My 5v draw with the CPU, display, and logic draws about 50 mA, so I setup a 110Ω resistive load for my tests. This way I could repeat the test without frying my CPU.

0 Katelyn Wiggenhorn over 6 years ago in reply to tom greive

TI__Genius 11040 points

Hello,

Thank you for sharing the loading condition. I apologize for the delay due to the holidays. I expect my EVM to arrive by Monday, and I will continue with the test on my side.

Best Regards,

Katelyn

0 Katelyn Wiggenhorn over 6 years ago in reply to Katelyn Wiggenhorn

TI__Genius 11040 points

Hello,

I have tried to replicate your test conditions on my side on the EVM. With the EVM as designed, when VIN transitions from 4.5V to 24V, I do not see the output voltage following the input voltage - as seen in your oscilloscope waveforms. I also tried to change the COMP network and inductor to match your design. Again, I do not see the output voltage following the input voltage. I do see a 2 V overshoot on the output with the COMP network from your design. Could you try changing the COMP network to match the EVM and redo this test as well?

Do you see this problem across multiple boards? Does this problem happen on every input voltage transition as you show or is this behavior sporadic? What is power supply - is it a DC source? Could you also share your board layout?

Thank you,
Katelyn

0 tom greive over 6 years ago in reply to Katelyn Wiggenhorn

Prodigy 70 points

Katelyn,

This is significantly better. I don’t know where my values for the comp. caps came from, but they were quite a bit off. The transient still gets up to 8.4v (3.4 above set point), but it only lasts about 30 µS.

Do you see this problem across multiple boards?

We have only build 2 board, but we have evidence of this on both.

Does this problem happen on every input voltage transition as you show or is this behavior sporadic?

The problem is completely repeatable. I have 24 v being supplied through a relay that is controlled with a function generator. Once I get the timing right it happens every time.

What is power supply - is it a DC source?

A bench type adjustable 0-32v 3 Amp

Could you also share your board layout?

If you still want it I can send it to you.

Thanks

tom

lm26001_comp_caps.pdf

0 Katelyn Wiggenhorn over 6 years ago in reply to tom greive

TI__Genius 11040 points

Hi Tom,

Thank you for sharing the updated results. This look much better now. Is this transient acceptable to your downstream devices?

Thank you,
Katelyn

0 Katelyn Wiggenhorn over 6 years ago in reply to Katelyn Wiggenhorn

TI__Genius 11040 points

Hi Tom,

Due to inactivity, I am going to close this thread, but if any questions remain, please open a new thread and the team will quickly address your additional questions.

Thank you,
Katelyn

Power management

Power management forum

LM26001: output voltage transient