g72x fan blue pwm wire help

I have a g72x and my gpu was hitting 93 c constantly when I am gaming while my cpu barely goes up to 50 (I repasted both cpu and gpu a short while ago). Then I noticed that my fan was no where near top speed especially compared to when I used intelburn or prime95 to get my cpu up to 58 to 60 c. It feels like the fan's controlled by the cpu's temperature.

So I decided to add a spst switch to the blue wire (pwm control) so that I can turn off pwm while I'm gaming and use regular fan control when I'm surfing the web, etc. But I was being a dumb*** and didn't use soldering iron(I knew that I should use one but I don't have enough room at home and I wasn't going to use it again for years probably). Instead I got those cheap wire glue... After applying the glue and connecting the wires, everything was fine for a day. The next day when I turned on my laptop the fan wasn't spinning, so I took out the switch and reconnected the pwm wire but no go.

Now the fan can only stay at maximum with the pwm wire disconnected. When the pwm wire is connected, the fan just stops. Is there anyway to rescue this fan or should I just go to ebay and buy a new one?

 

Hi relluickuo,

It's hard to say for sure, but it is possible that your motherboard
PWM driver is damaged, or the TACH feedback signal is not
working properly. Most PWM fans have an internal pullup resistor
on the PWM input. If the PWM input is not driven, the input
floats high, and defaults to 100% duty cycle. Since your fan
runs at 100%, when the wire is disconnected from the motherboard,
it suggests that the fan is okay. The fact that the fan stops running,
when the wire is connected to the motherboard, points toward the
driver is no longer modulating the signal (stuck at ground). It is
possible that the driver got zapped due to switch bounce, or a
short to other circuitry (if not properly insulated). Mechanical switches
tend to bounce for many milliseconds before settling to their final position.
These oscillations can cause voltage spikes, if not limited/handled properly.
Voltage spikes beyond the rating of the driver will usually damage the part.
The good news that it is probably just a simple transistor, which drives
the signal. It can be easily replaced, if you can trace the signal back from
the fan connector. Here is an easy test, to see if the fan is okay:

1. Power the fan, with the PWM input floating.
2. Using a clip lead or loose wire, short the fan ground wire (black?) to the
PWM input.
3. If the fan slows down, the PWM input is likely fine, and so is the fan.
4. If the fan speed does not change, the fan may be damaged.

One other possibility is the TACH signal. Some fan controllers will stop
driving the fan, if the TACH signal is not seen. You might want to check
the cable/connector for any loose connections, that might prevent the
TACH signal from reaching the fan controller.

Good luck,

Sir Robin

 

Thanks for the advice, I'll try that tomorrow afternoon when I get the time. Hopefully it's just the fan that's busted, that way at least I can get a replacement if I wanted to.

 

It turned out the fan only turns on when it reaches certain temperature. When I turned it on in the morning the temperature is way below the lower limit which is why the fan stayed off... Today I found out it was working b/c I turned it off after coming on here to reread the post, then reconnected wire to make sure it wasn't working and when I turned on the laptop the fan started spinning... I was confused so I turned off the computer again, left it for an hour and turned it back on and the fan didn't start again. That's when I realized there might be a temperature limit.

It was good that I didn't destroy the fan driver on the laptop. I might reconnect the spst switch but after reading about the possible voltage spikes, I think I will leave it as is and deal with the high temperature.

 

That's great news relluickuo!

Yeah, PWM driving circuits are usually a bit more forgiving
than that. I was a little surprised that hot opening/closing the
switch blew the driver, but that's how it was acting.

You may still be able to use the switch trick. It depends on how
the circuit is designed, and the component types. Alternately, you
can try using 3 wire control. The easiest way to do this, is to use
an aftermarket fan controller, and hack it into the chassis somewhere.
You leave the PWM input floating, and adjust the fan voltage, to
change the speed. The fan controller will include a potentiometer or
switch, to set the voltage/speed.

It does sound like the primary fan is controlled by the processor temp.
Are there any BIOS controls, to set the limits? If so, you can try setting
lower limits, for the processor. Also, some BIOS's pass ACPI info to the
OS, detailing how to control the fans. If this is the case, you can
probably run one of the fan control applications, like Speedfan, to override
the automatic fan control, when you are going to run a game. Lastly,
though it's a little counter intuitive, you can try disabling some of the
processor throttling options and/or overclock, so the processor heats up
more, when playing games. This will increase your performance, and make
the fan spin faster.

Good luck,

Sir Robin