PWM fan hub no longer controlling speed

Teuton

Teuton

Teuton

Associate
Joined
16 Dec 2013
Posts
310
Location
Staffordshire
Hi guys,

so with a recent rebuild into a new case I swapped out some SP120s for 4 Silverstone/OcUK fans. Both are 3 pin.

Before, I had my corsair fans and the pump (PWM) plugged into a Phanteks PWM fan hub, which in turn is plugged into the 4-pin CPU fan header on my motherboard, a Gigabyte X99m gaming 5.

With this setup, I could control the speeds using the gigabyte EZ tune utility, running from around 300 RPM to 1300 or so on the fans.

Now with the rebuild, exact same setup, I can't change the fan speed at all. When the software calibrates the fans run at the exact same speed at all %s. What gives?

I know it's probably not ideal to be looking to run the pump and fans off the same PWM signal, but as far as I'm aware this is the only PWM header on the board.

The fans are running at a constant 1100-1300 RPM which is pretty loud and the CPU is really cool so I'd like to turn them down to around 500-600 to get silence.

Help appreciated.
 
Pump is PWM controlled .. and so is the Phanteks PWM controlled fan hub. It is not a PWM fan hub for PWM fans. It is a PWM controlled fan hub to run 3-pin variable voltage fans on.
 
Glanza said:
Disconnect your pump from the fan hub then rerun the calibration with just fans connected.
Click to expand...

I'll try that. Don't see why I'd have to do that now though when it worked before.

Doyll, I know that, but what difference does it make if I plug 3 of the 4 pump pins into a PWM fan hub?
 
Anyone know what speed an EK PWM D5 will run if it's plugged into a voltage control 4-pin mobo header? Maybe that's a bad idea.

I might try plugging the pump directly into the CPU header and run the fan controller off another header.
 
Last edited:
Teuton- zero rpm more than likely as at 23W it'll blow the fan header control circuit.

Re the PWM hub try without the pump and see what happens. Are the fans and pump positioned in the same positions on the hub?
 
Tealc said:
Teuton- zero rpm more than likely as at 23W it'll blow the fan header control circuit.

Re the PWM hub try without the pump and see what happens. Are the fans and pump positioned in the same positions on the hub?
Click to expand...

I thought about that and rejigged them about before I posted this thread, tried the pump in different positions on the hub and it still won't let me change anything.

I'm confused as to why it let me change RPMs with Corsair fans but not with these Silverstone ones.
 
Teuton said:
I thought about that and rejigged them about before I posted this thread, tried the pump in different positions on the hub and it still won't let me change anything.

I'm confused as to why it let me change RPMs with Corsair fans but not with these Silverstone ones.
Click to expand...

This might not help, but in my Enthoo Evolv Manual it says that if the PWM is plugged into the CPU_Fan header then plug it into the PSU as well. But if it's plugged into another mobo header then you should unplug it from the PSU, if not the fans will run 100%. So might be worth disconnecting the Enthoo PWM from the PSU.
 
I wouldn't recommend unplugging 12v power from the PSU to the fan hub as with a pump and 4 fans it would put too much load on the CPU header, unless of course the pump is separately powered from a Molex and uses just the PWM signal to operate.

Have you, at any time, loaded the pump to the FAN1 position on the PWM fan hub? Or installed a fan with the connector upside down?

I've built similar circuits to the Phanteks implementation (pic in my sig) and full speed is a failure mode of the circuit if the input transistor is blown. If I recall correctly the input transistor sitting on the PWM signal on Fan1 is a tiny SOT-23 package and could have been blown short circuit. This transistor is responsible for providing the PWM signal to the PNP transistor for Fan 1 and also the P-ch MOSFET IC that controls the other fan positions. I can't remember if there is any input protection on the circuit. I put a 1k resistor on mine so to limit the current.

It might not be this of course.

Are you able to try the Corsair fans again?
 
Last edited:
Tealc said:
I wouldn't recommend unplugging 12v power from the PSU to the fan hub as with a pump and 4 fans it would put too much load on the CPU header, unless of course the pump is separately powered from a Molex and uses just the PWM signal to operate.

Have you, at any time, loaded the pump to the FAN1 position on the PWM fan hub? Or installed a fan with the connector upside down?

I've built similar circuits to the Phanteks implementation (pic in my sig) and full speed is a failure mode of the circuit if the input transistor is blown. If I recall correctly the input transistor sitting on the PWM signal on Fan1 is a tiny SOT-23 package and could have been blown short circuit. This transistor is responsible for providing the PWM signal to the PNP transistor for Fan 1 and also the P-ch MOSFET IC that controls the other fan positions. I can't remember if there is any input protection on the circuit. I put a 1k resistor on mine so to limit the current.

It might not be this of course.

Are you able to try the Corsair fans again?
Click to expand...

Thanks for the response. I have tried the pump in FAN1 and another position. The pump is powered by molex so it's just the 4 fans needing power, should I try unplugging the PSU power cable to the fan hub?

I can indeed try the corsair fans again and I will, but it's in a Parvum case so taking the back panel off and on again is quite a job so saving it until I have the time and patience lol!
 
If it is Molex powered then the pump probably hasn't destroyed the fan hub, however it could have still blown.

You don't have to fit all the fans, just bung one or two on and try a few arrangements and see what happens. Often with these things it is a process of elimination. I'd have been in there with my multimeter if it were mine.
 
You must log in or register to reply here.
Back
Top Bottom