Author Topic: Hot-Swapping killing IC's  (Read 1182 times)

0 Members and 1 Guest are viewing this topic.

Offline Apn15Topic starter

  • Contributor
  • Posts: 10
  • Country: us
Hot-Swapping killing IC's
« on: January 20, 2022, 04:20:51 pm »
I have a project where I am using a TI TPS7B6933DBVR to power an EEPROM and a NAND gate that is used for an external IC BOOT circuit.

This regulator is power by a LiFePO4 battery at 24V, but whenever I remove and re-apply power, it appears that the regulator and other IC's have shorted to ground.

I have found a few articles online about this issue, but none seem to have any solutions. Is there anything I could add to the circuit to help or prevent this from happening.

See below for shots of the circuits.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8320
  • Country: fi
Re: Hot-Swapping killing IC's
« Reply #1 on: January 20, 2022, 04:55:40 pm »
With the regulator specified up to 45V transient, this is not necessarily the right answer, but anyway:

The large 10uF MLCC capacitor causing inrush voltage peaking with inductive supply wiring. This can easily go up to twice the input voltage, and sometimes even beyond.

Try the trivial fix: solder any 100uF or so electrolytic capacitor (which, by definition, always have significant ESR) in parallel with the 10uF MLCC.

See https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf  for more information.
« Last Edit: January 20, 2022, 04:57:49 pm by Siwastaja »
 
The following users thanked this post: SeanB, Ian.M, tooki

Offline Apn15Topic starter

  • Contributor
  • Posts: 10
  • Country: us
Re: Hot-Swapping killing IC's
« Reply #2 on: January 20, 2022, 07:08:40 pm »
Thanks for the suggestion and the Article.

I am going to order some different capacitors and may try the series resistor circuit that was in the article.
 

Offline Apn15Topic starter

  • Contributor
  • Posts: 10
  • Country: us
Re: Hot-Swapping killing IC's
« Reply #3 on: August 22, 2022, 04:14:13 pm »
I am returning to this project and have tried several solutions to this problem with limited results.

The article previously posted suggested a 30V TVS, while it did limit the voltage, the spike was still around 35-38V in practice which is still too high.

I have seen several articles for calculating snubber circuits, but they are all for buck or fly-back converters.

The oscillation frequency of the transient is 82.64kHz. Is there any way to calculate the snubber circuit I need?
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8320
  • Country: fi
Re: Hot-Swapping killing IC's
« Reply #4 on: August 22, 2022, 04:49:38 pm »
Did you try the electrolytic capacitor? Anything from the junk box, really, as long as C is significantly larger than the 10uF. Which is why I said at least 100uF.
 

Offline PeteH

  • Supporter
  • ****
  • Posts: 114
  • Country: ca
Re: Hot-Swapping killing IC's
« Reply #5 on: August 22, 2022, 05:50:25 pm »
Why would you not just add a series ~10 to 100 ohm resistor before any capacitance. You don't really care about the voltage drop as the supply is linear anyway. Feel free to add a bunch of series resistance to dampen out the inrush event, limit the peak current to less than 1-2A during hot-swap...

Will also help spread the heat out from the LDO.
 

Offline Apn15Topic starter

  • Contributor
  • Posts: 10
  • Country: us
Re: Hot-Swapping killing IC's
« Reply #6 on: August 22, 2022, 05:59:29 pm »
I did try an 100uF electrolytic capacitor and it did fix my problem. Its a little larger of a solution than I wanted.
 

Offline PeteH

  • Supporter
  • ****
  • Posts: 114
  • Country: ca
Re: Hot-Swapping killing IC's
« Reply #7 on: August 22, 2022, 06:00:28 pm »
Series resistance would result in the smallest solution size.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf