Author Topic: Partially "dead" TVS - diode?  (Read 1350 times)

0 Members and 1 Guest are viewing this topic.

Offline masta_k74Topic starter

  • Regular Contributor
  • *
  • Posts: 55
  • Country: de
Partially "dead" TVS - diode?
« on: November 16, 2022, 11:07:02 am »
I have removed a TVS diode from a defective switching power supply unit to check it and when I measure it with my power supply unit @ 32V (current limitation to 50mA and series resistor) it shows the normal value (0.64V) in one direction and a value of 15.6V in the reverse direction, although this diode is specified for 250V according to the datasheet?
When measuring with the DVM, however, the display shows 0.35V in both directions.
Have any of you had similar cases?

The power supply was just pulsating and this TVS reached over 130°C in a few seconds according to the Flir camera.
The TVS was here with some more passive components and a normal diode as a snubber on the primary winding of the transformer.

So far, I have only known complete continuity or completely open with defective diodes / Zener diodes, and a TVS functions more or less like a Zener diode. :-//
Are TVS diodes also subject to an ageing process like varistors?
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: Partially "dead" TVS - diode?
« Reply #1 on: November 16, 2022, 10:38:24 pm »
partial failure, replace it.

But suspect that the TVS fail is an effect and root cause lies elsewhere


j
An Internet Dinosaur...
 

Online thm_w

  • Super Contributor
  • ***
  • Posts: 7522
  • Country: ca
  • Non-expert
Re: Partially "dead" TVS - diode?
« Reply #2 on: November 16, 2022, 10:40:48 pm »
Yes, I've seen similar issues before with partial failure, not dead short.
I don't think its aging, seems more likely repeated overload or thermal related, which could indicate another concern.

Whats the part number?

Your measurements are a bit interesting though, 0.35V with the DMM and probably 1mA current. While the power supply is 16V at 50mA. Maybe try to set power supply to 1mA.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline timenutgoblin

  • Regular Contributor
  • *
  • Posts: 189
  • Country: au
Re: Partially "dead" TVS - diode?
« Reply #3 on: November 17, 2022, 12:53:14 am »
I have removed a TVS diode from a defective switching power supply unit to check it and when I measure it with my power supply unit @ 32V (current limitation to 50mA and series resistor) it shows the normal value (0.64V) in one direction and a value of 15.6V in the reverse direction, although this diode is specified for 250V according to the datasheet?
When measuring with the DVM, however, the display shows 0.35V in both directions.
Have any of you had similar cases?

The power supply was just pulsating and this TVS reached over 130°C in a few seconds according to the Flir camera.
The TVS was here with some more passive components and a normal diode as a snubber on the primary winding of the transformer.

So far, I have only known complete continuity or completely open with defective diodes / Zener diodes, and a TVS functions more or less like a Zener diode. :-//
Are TVS diodes also subject to an ageing process like varistors?

As already mentioned, a typical DMM in the diode range has a test current of 1mA. Also, the open circuit test voltage for a DMM in the diode range is typically 3V, but some DMMs have a higher test voltage such as Fluke 17X (7V) and then there's the EEVBlog 121GW which can be selected between 3V and 15V. (See attachment).

I think your best approach is to use your bench power supply for testing TVS diodes and zener diodes. The fact that your DMM measured 0.35V in both directions is an indication that the TVS has failed in at least one mode of failure.
 
The following users thanked this post: masta_k74

Offline masta_k74Topic starter

  • Regular Contributor
  • *
  • Posts: 55
  • Country: de
Re: Partially "dead" TVS - diode?
« Reply #4 on: November 18, 2022, 10:42:27 am »
Part number of the diode is p6ke250.
The diode is installed in the device (battery charger, designed as a flyback converter) as a snubber on the primary winding of the transformer. As far as I can tell from the traces, there is only one fast switching diode, with opposite polarity, in the signal path. 4 x 400k resistors are connected in parallel across both diodes.
At this point, I have only known snubbers consisting of diode, resistor and capacitor. (Using a TVS for this was unknown to me).
I have not been able to find any other defects, despite a thorough check of the components.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16385
  • Country: za
Re: Partially "dead" TVS - diode?
« Reply #5 on: November 18, 2022, 01:35:00 pm »
Just replace it, likely it just failed from the number of avalanche cycles it had to do, as they do have a lifetime number of pulse rating.

Replace with the new part, p6KE250A, and it should work, though check the ESR of small electrolytics on the primary side as well, they likely have gone high by now.
 
The following users thanked this post: thm_w, masta_k74

Offline masta_k74Topic starter

  • Regular Contributor
  • *
  • Posts: 55
  • Country: de
Re: Partially "dead" TVS - diode?
« Reply #6 on: November 18, 2022, 08:29:41 pm »
I have already ordered the P6KE250A, they should arrive here tomorrow or Monday. Of course I still have to check the capacitors on the primary side. I have already done this on the secondary side, as I initially suspected a short circuit on the secondary side. (Physically speaking, it's exactly the same with this flyback converter).

Edit:

Everything is great, the charger works perfectly again. For a 120€ device, the effort and the 27ct for the TVS was definitely worth it.
The capacitors still have values like fresh from the factory, so I don't think it's necessary to replace them yet.
I will now do an extensive test, but I have to discharge my battery first, as it is an automatic charger and only works with the battery connected. With a full battery, it always switches off after a few seconds.
« Last Edit: November 19, 2022, 12:03:57 pm by masta_k74 »
 
The following users thanked this post: thm_w


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf