TPS40200-HT: Soft Start pin voltage regulator is failing

Hey Mitchell,

I have contacted the person who originally answered the question to see if there was any resolution.

He is currently on vacation, but hopefully I will get a response monday.

In the mean time is a schematic something you would be able to provide?

Thanks,

Daniel

Here is the schematic that is failing:

Seems to be a problem with the reply text box. The pasted schematic looks fine in the reply box but then does not show up in the forum reply

Hey Mitchell, would you be able to provide the current you are seeing the issue at as well?

Thanks,
Daniel

800 to 900 ma on the +12V output

Hey Mitchell,

I did a quick sim using your conditions and saw that the device may be hitting its over current condition using your set up.

Given the min threshold is 35 mV (although given the commercial min threshold is 65 mV which may be more accurate for a "25 C" measurement)

I found the threshold you are reaching with 800 to 900 mA is around 70 mV

This would lead me to believe that the reason some of the devices are failing and not all of them is the varied thresholds that you are seeing between the devices.

Could you try decreasing the resistance value of the current limiting resistor and seeing if this solves the issue?

Thanks,

Daniel

Your suggested "fix" of changing the sense resistor to a smaller value might work around this problem but will not fix it.  I claim this for the following reasons:

1. The "failing" boards worked fine for over 200 hours before starting to fail.  (If this was a chip to chip variance in the over-current threshold, the boards would have failed from the start)

2. The failing controller chips' voltage output on the SS pin drops by 100mv to 150 mv from 8V as VDD changes from 28V to 32V.  Replacing the chip on the failing board with a new chip, the voltage on the SS pin does not change at all when VDD changes 28V to 32V

3. The failing chips have a random 100 mv glitch coming OUT of the FB pin that is not present when the chip is replaced with a new chip.

We have sent some of these chips out to a company that does failure analysis.  I am waiting on their results.  I was hoping TI had seen this problem before and could give me some guidance on what might cause the chips to get damaged to start failing in this way.

Hey Mitchell,

This value can drift heavily over temperature and life.
This is part of why the specification is so wide in the HT (it is already pretty wide for the commercial device)

Are you able to provide the temperature the device is at for the 200 hours?

Thanks,
Daniel

I am waiting on the failure analysis of the third party company before making any circuit changes. I was hoping TI has some insight that thye would share about the failure mechanisms of this IC. I guess not.