Help with PSU (Temp control fan, load bank, & PWM circuit)

[jocanon]

You know what they say. If you can't fix it feature it. What you might try is a resistor from the fan control to +12. This would make it run at full speed sooner. Pick the size to max it out at say 35 to 40 amps at room temperature.

That really might not be a bad idea, even if just to prevent the case from getting so hot that you can't touch it.

 

[ronv]

Actually it is a resistor to ground. My bad. You were faster than my edit I bet.

Out of curiousity it would be interesting to know what the part # of the componet is on the heatsink and how hot the heatsink is with just pin 4 grounded. We might be able to estimate how close it is to the spec.

 

[jocanon]

Sure, I will see when I get home if I can find the part number. What component should I be looking for will it be obvious from what is attached to the heat sink?

 

[ronv]

I would guess something like the FETs you used in the load box.

 

[jocanon]

I wonder what is the best way to get a temperature reading on these components. I have an infrared thermometer that is just point and shoot, but is there a more accurate way to measure?

 

[Dusey52]

A resistor from pin 4 to ground will set a new minimum fan speed...

 

[jocanon]

Hm, is there a way to keep the min fan speed as is, but just lower the temperature at which it reaches max speed, in other words make the thermal control more sensitive.

 

[ronv]

That's kind of what we did with your circuit.

 

[jocanon]

yeah, I guess you are right, but how difficult would it be do you think to modify the existing thermal control without having to add another thermistor and op-amp? That's more what I am getting at now.

 

[ronv]

It would probably be pretty easy if we had a schematic of it. Might be difficult to trace it out from the board.
I wouldn't give up to soon on the resistor to ground. Without checking it is hard to say how much higher the low speed would be to get the top speed sooner.

 

[jocanon]

Appearently there are two pots on the one daughter board according to a guy in the rcg forum. Once I find what he is talking about, I can try and see what those adjust.

 

[jocanon]

I took apart a broken unit. Should I get part numbers for the big black one (it's the bridge rectifier and you can see the part number of that one in the picture, it's GPJ15M) in the first picture and the white ones in the second pic. Anything else you can see in the pictures I should get part numbers for? I will be offline for a while but back on later tonight.

 

[jocanon]

looks like 150C is the max operating temperature of the bridge rectifier.

https://www.datasheetarchive.com/GPJ15M-datasheet.html

 

[jocanon]

same thing for the fets, 150C (the white ones...turns out the white part is just to screw them to the heat sink, they are really just regular MOSFETs)

https://www.infineon.com/dgdl/SPP_I...7003b&fileId=db3a304318f3fe290119090054e32a20

 

[jocanon]

There is also this one in there, but it's the same thing, 150C...

https://www.infineon.com/dgdl/SPP_A...21528&fileId=db3a304314dca389011536a41121152a

 

[jocanon]

I think it is well within spec. I ran it for about 20 minutes on 50 amps and it was as hot as it was going to get. The heat sink was only 45C, the transformer was a little hotter at 110C. Would you agree with me that it is well within spec and no further mod is needed at this point other than shorting pin 4 to ground?

 

[ronv]

The bridge is probably the easiest. Just rough numbers..... If we get 50 amps at 12 volts the bridge needs to be able to provide 5 amps at 120 volts plus say 15% for losses - 6 amps. If you go down the derating curve it can be 130C and provide 6 amps. This is absolute max.
Actually the things that will probably fail are the 470 Ufd. capacitors right next to the heatsink. Electrolytic capacitors don't like heat especially when they are working hard as they do in switchers.
I'm guessing these are good ones rated for 105C.

 

[jocanon]

yep, 105C

 

[ronv]

I think you are ok.

 

[()blivion]

Electrolytic caps used to be the most common point of failure for SMPS's. A lot of units "had cancer" do to a major Taiwanese manufacturer using a bad electrolyte formula. You can read about it *HERE*. Not really relevant to your setup, ("HP reportedly purged its product line in 2004") but interesting read none the less. The problem still persists to this day, but is much less frequent.