GW Instek PSP-2010 Power Supply Fan Problem

[larrybud]

I bought one of these power supplies non-working off ebay and got it going with some cap replacements (wish I saw the thread on here sooner, would have saved me some time!)

It's been working fine, but yesterday I put a larger load on it, and the fan tried to kick on but made some clicking noises. Looks like it was barely trying to turn. I pulled the fan out and tested it with my other 12V supply and it wouldn't turn. I had a fan in my bin that is a close match and the same physical size:

Old: 12V 0.16A
New: 12V 0.24A

I put the new fan in, and while it spins and changes speed as I increase the load (I have it hooked to my electronic load tester), it makes an odd buzzing sound (not mechanical).

I put my scope on the fan connector and speed is controlled via a PWM signal. Couple of questions:

• Can I put this slightly larger wattage fan in without an issue (close enough?)
• Are some fans not compatible with a PWM signal, which would cause this buzzing?
• Or do I have an additional underlying problem in the fan circuitry that I haven't found?

Note the PWM signal is not exactly "clean" on my scope. The caps in that area check fine for ESR, no leakage, bulging, nothing out of temp. Diodes and transistors in that section check ok.

I also put a large cap across the fan to smooth the signal and it run seemingly around full speed and very smoothly.

[m k]

Maybe, yes, maybe.

Seems that your new fan is a DC fan.

[benj38]

If I read between the lines correctly, both the old and the new fan have only 2 wires, and the circuit supplies a PWM waveform to these two wires, with an increased duty cycle to get higher speed.

If the above assumptions are correct, then the old fan was brushed, and the new one in brushless.
Brushless DC motors are not intended to be powered by a PWM signal, as you keep turning on and off the supply to the electronic controller inside the fan!
If you ignore this, they will buzz and be unhappy, and the controller may fail prematurely.

Brushless fans that can accept a PWM signal for speed control usually have 4 wires: GND, +VDC (constant voltage!), Tachometer out, PWM signal in.

[wraper]

I put the new fan in, and while it spins and changes speed as I increase the load (I have it hooked to my electronic load tester), it makes an odd buzzing sound (not mechanical).

You must have a newer HW version with a single PCB + front panel board. No separate large CPU board, just a lot of empty space. Older version had a normal analog fan voltage control. In newer HW version some dumbass decided that it's OK to PWM fan power directly (it's not). I don't recall schematic I made, but I added a few resistors and capacitor to convert it into a normal voltage control.

Are some fans not compatible with a PWM signal, which would cause this buzzing?

None of brushless DC fans are compatible with PWM (unless over PWM control wire). Doing so is a fan abuse.

[wraper]

Here are modification photos. Transistor replaced with Darlington TIP122, 3 SMD resistors and capacitor added, existing electrolytic cap connected differently, one trace cut. Large cap is 100uF, smaller one is 47uF. Don't recall if larger 10R resistor is original or not.

[wraper]

If the above assumptions are correct, then the old fan was brushed, and the new one in brushless.

No, it's a DC fan that gets abused. In my PSU original fan was emitting strange noises at low PSU load / fan speed and fan replacement did not help. I had older HW version PSU too which did not have such an issue. Then i found out newer one was PWMing the fan directly.

[benj38]

If the above assumptions are correct, then the old fan was brushed, and the new one in brushless.

No, it's a DC fan that gets abused. In my PSU original fan was emitting strange noises at low PSU load / fan speed and fan replacement did not help. I had older HW version PSU too which did not have such an issue. Then i found out newer one was PWMing the fan directly.

And I assumed they knew better 
I suspected that they may have abused a brushless fan since I didn't see a brushed 12V DC fan in ages, but I said to myself "Nah, they wouldn't"...

[wraper]

Here is the schematic I recreated.