Author Topic: SMPS Controller shotguns relay across the pcb with magic smoke  (Read 1095 times)

0 Members and 1 Guest are viewing this topic.

Offline PorenbetonTopic starter

  • Newbie
  • Posts: 4
  • Country: de
Just thought I'd share this somewhat funny failure on the pcb i am currently repairing from the control board of my dishwasher.
You see that tiny hole in the side of this IC?



Yeah not much to look at, something just shorted out internally, now the high voltage DC side is shorted to the 12V DC output side.
Fortunately a fuse blew, so nothing much besides this controller is toast.
However i wouldnt have even seen this hole if not for the (though only superficial) devestation it did to the pcb.
First I though that a trace vaporized from some really high overcurrent, but it looked too directional.
Only with a magnifier could I see that it actually came from the IC.



Look how much magic smoke dispursed on this board from this tiny hole. There was a transformer in the way that i removed, it shot right past it,
but didnt actually even taint it.
The funniest thing is, the distance it impacted with the relay was about 50mm, the dispersion diameter however on it was only like 4mm.



That magic smoke puffed out of this hole hellish fast, wouldnt be surprised if it went supersonic for a split second even haha.
« Last Edit: June 25, 2020, 01:45:22 am by Porenbeton »
 

Offline TerraHertz

  • Super Contributor
  • ***
  • Posts: 3958
  • Country: au
  • Why shouldn't we question everything?
    • It's not really a Blog
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #1 on: June 25, 2020, 11:14:00 am »
Wow. A magic smoke laser.

Shorting HV to LV side that time may not have been a disaster, but in other systems using that part, maybe it could be.
Collecting old scopes, logic analyzers, and unfinished projects. http://everist.org
 

Offline station240

  • Supporter
  • ****
  • Posts: 967
  • Country: au
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #2 on: June 27, 2020, 10:12:50 am »
Ah it's one of those controller+mosfet in a DIP 87 package, with the missing pin.
The blow hole is between the Source and Drain pins of the mosfet, so the mosfet itself blew up.

I'd imagine the smoke was quite high pressure by the time the epoxy was finally breached.
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7333
  • Country: gb
  • Electronics Hobbyist & FPGA/Embedded Systems EE
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #3 on: June 27, 2020, 10:48:12 am »
A great deal of these appliances use non-isolated converters because there is no user access to any exposed metal parts, connections, etc.  This is fine until the controller IC shorts I suppose.  If there is a transformer no harm can be done to the LV side.
 
I would expect a lot of components on that board to have been damaged by having the 12V bus momentarily jump towards 300V+.
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8525
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #4 on: June 27, 2020, 08:51:17 pm »
Those ICs are used because they're cheap, and are rather sensitive to input power surges.
I suspect that's what caused yours (and countless others) to blow up.
 

Offline PorenbetonTopic starter

  • Newbie
  • Posts: 4
  • Country: de
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #5 on: June 28, 2020, 04:38:03 pm »
A great deal of these appliances use non-isolated converters because there is no user access to any exposed metal parts, connections, etc.  This is fine until the controller IC shorts I suppose.  If there is a transformer no harm can be done to the LV side.
 
I would expect a lot of components on that board to have been damaged by having the 12V bus momentarily jump towards 300V+.

After just changing the controller + fuse everything works again as it should.
I think since the internal FET shorted and the fuse therefore (which was inline with the HV side of the transformer) blew probably almost instantly no high voltage actually reached the 12V side.
Just from measuring the FB pin there was a dead short to the HV side. Maybe the short developed right after the mains got disconnected by the fuse from the heat still trapped inside the IC for a short amount of time.
« Last Edit: June 28, 2020, 04:41:09 pm by Porenbeton »
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7333
  • Country: gb
  • Electronics Hobbyist & FPGA/Embedded Systems EE
Re: SMPS Controller shotguns relay across the pcb with magic smoke
« Reply #6 on: June 28, 2020, 04:46:59 pm »
I think you got lucky!  Maybe the capacitance on the 12V side combined with the high resistance of the FET (often these are 1-2 ohm devices) gave it just enough time that the fuse saved parts. Or maybe the MOSFET failed initially in a higher resistance mode which still led to destruction but meant the process took longer.

But good work keeping another appliance out of the e-waste ... that reminds me, we need to figure out why our coffee maker died.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf