New PSU blew fan controller

JessieD

JessieD

JessieD

Associate
Joined
20 Jul 2017
Posts
564
I bought an EVGA G3 850W psu for my rig last week. It seems to be fine, sooo quiet. But when I initially connected everything up an alarm went off in my fan controller & smoke started coming out of it. I turned everything off & removed the fan controller. The fan controller is old & I'm not going to bother replacing it. My only concern is that my PSU & rig are ok.

Without meaning to expect you to have a crystal ball was it just a coincidence do you think?
 
Can you see from which part of fan controller smoke came out and what controller is it?
And what fans it had connected and how many per channel?
Not all fan controller tolerate that much load and fans likely take more current during spin up.
If you can take good (not potatophone) picture where components and markings are well visible that might tell something.


If PC works PSU there shouldn't be anything wrong in PSU.
But if you have multimeter you could have easily checked voltages by jumpstarting PSU (green wire to black) but you can thank Corsair for ******* up all PSUs with all black cable fashion ****.
(orange 3,3V: red 5V: yellow 12V)
With colour coding all that checking would be easy and fast.
Now you have to count pins:
https://en.wikipedia.org/wiki/Power_supply_unit_(computer)#Wiring_diagrams


*** Swearing should be fully starred - thanks, Armageus ***
 
Last edited by a moderator:
I haven't got a multimeter. I did take photos right after it happened is this any use? https://postimg.org/image/glk2t05bh/

It had four 120 mm fans connected. One per channel. No idea which part was smoking it just billowed out. Not sure on brand sorry it was quite old & I threw it away after I took the photos
 
Kaze Master Ace
http://www.scythe-eu.com/en/products/pc-accessory/kaze-master-525.html
Googling for that VFD5.25 gave it up.

http://www.fudzilla.com/component/k2/19079-scythe-kaze-master-ace-525-tested?start=2
I would keep that 1A ampere current per channel as let's say "optimistic" value when there's no heatsinks.
But can't see any sure signs of burned component or smoke on that side of PCB and single fan per channel shouldn't have stressed those.

Other side of PCB might have shown something.
Inadequate "autopsy"...

BTW, was election of pope successful?
https://en.wikipedia.org/wiki/Papal_conclave#Smoke_colors
:D
 
EsaT said:
Kaze Master Ace
http://www.scythe-eu.com/en/products/pc-accessory/kaze-master-525.html
Googling for that VFD5.25 gave it up.

http://www.fudzilla.com/component/k2/19079-scythe-kaze-master-ace-525-tested?start=2
I would keep that 1A ampere current per channel as let's say "optimistic" value when there's no heatsinks.
But can't see any sure signs of burned component or smoke on that side of PCB and single fan per channel shouldn't have stressed those.

Other side of PCB might have shown something.
Inadequate "autopsy"...

BTW, was election of pope successful?
https://en.wikipedia.org/wiki/Papal_conclave#Smoke_colors
:D
Click to expand...
You are very good:D
If you zoom in you can see burnt liquidy stuff around the yellow doober on the bottom right Also there is blackening on the right metal frame on the same side
 
Last edited:
That could be smoke, though they've done later hand soldering which leaves "melting" signs to protective laquer.
Shows as "dirty" looking surface near soldering points also in this:
http://www.fudzilla.com/images/stories/Reviews/coolers/ScytheKazeMasterAce/panel-treca-big.jpg
Not enough light on that side of image.

That yellow thing is miniature transformer, or it could be also 1:1 isolator.
At least there's no smoke signs in its device center side "legs" with clear soldering surface visible.

Inadequate documentation when user guide doesn't have circuit diagram...
At least they've clearly taken seriously filtering voltage for that big likely main IC with those capacitors near power connector.
 
EsaT said:
Can you see from which part of fan controller smoke came out and what controller is it?
And what fans it had connected and how many per channel?
Not all fan controller tolerate that much load and fans likely take more current during spin up.
If you can take good (not potatophone) picture where components and markings are well visible that might tell something.


If PC works PSU there shouldn't be anything wrong in PSU.
But if you have multimeter you could have easily checked voltages by jumpstarting PSU (green wire to black) but you can thank Corsair for ******* up all PSUs with all black cable fashion ****.
(orange 3,3V: red 5V: yellow 12V)
With colour coding all that checking would be easy and fast.
Now you have to count pins:
https://en.wikipedia.org/wiki/Power_supply_unit_(computer)#Wiring_diagrams
Click to expand...


You get a tester connector with the EVGA G3 which does the same thing as jumping the green and black wires, so the op can use that instead.
 
Last edited by a moderator:
JessieD said:
I haven't got a multimeter. I did take photos right after it happened is this any use? https://postimg.org/image/glk2t05bh/

It had four 120 mm fans connected. One per channel. No idea which part was smoking it just billowed out. Not sure on brand sorry it was quite old & I threw it away after I took the photos
Click to expand...

If you want to test the psu with the components disconnected you can use the connector that you get with it.
 
Without knowing circuit diagram and location of part which gave up that magic smoke it's hard to say but this could be one scenario:
Some capacitor had dried up but was still working just enough while having voltage.
But when you shut down PC and switched PSUs it "cooled" and didn't work at all stressing some other part.


That's also one failure mode for especially cheap capacitor PSU:
Keeps working and booting while being connected to power, but once disconnected from wall 5V standby cools down and voltage regulation/ripple goes completely haywire.
(and after multiple booting attempts/enough warming up time PC can boot again...)
 
EsaT said:
Without knowing circuit diagram and location of part which gave up that magic smoke it's hard to say but this could be one scenario:
Some capacitor had dried up but was still working just enough while having voltage.
But when you shut down PC and switched PSUs it "cooled" and didn't work at all stressing some other part.


That's also one failure mode for especially cheap capacitor PSU:
Keeps working and booting while being connected to power, but once disconnected from wall 5V standby cools down and voltage regulation/ripple goes completely haywire.
(and after multiple booting attempts/enough warming up time PC can boot again...)
Click to expand...
So can I put it down to the fan controller being at the end of its life & the PSU is fine? Would that be a reasonable assessment? Oh & I managed to find it in the bin but there is an lcd on the otherside of the board so I can't see any damage, if it is there.
 
If PC doesn't complain about voltages or have stability problems likelyhood is very good for PSU being correctly working.
Heck, some motherboards can have BIOS giving some voltage complaints for lot less.

And all good PSUs monitor their outputs for over/undervoltage shutting down if tolerances are exceeded.
(those protections are one of the things they leave out from absolutely cheapest PSUs)
 
You must log in or register to reply here.
Back
Top Bottom