PWM Fan Controller Increase base speed

[tfixrepairs]

Hi,

I am looking to build a Pulse Width Modulator for a fan to increase the fan base speed at around 40 - 50% so that when temperature increases the pulse sent via the signature wire would increase the current signal hence making the fan spin faster than it was intended by original controller.

So far I was going about using a 555 timer to mimic the current fan controller output but is there a way to control the 555 timer frequency by original signature wire so that when original signature increases the timer would amplify the signal accordingly. I hope you understand what I am on about.

Maybe someone has built something before? Any help would be much appreciated.

Regards

 

[4pyros]

making the fan spin faster than it was intended by original controller

A fan will only spin so fast. Is the original controller limiting it some how? What is the original controller? Andy

 

[#12]

Yes but...there is so much information you did not provide.

 

[tfixrepairs]

Basically the device does not get sufficient cooling due to fan controller not controlling the fan correctly via the Pulse Width Modulator. The fan is a 12V with a signature max 3.3V wire controlling how fast it spins. Now an original controller increases the fan speed when temperature rises and what I want to do it make fan spin faster from the beginning and to preserve the original controllers ability to control the speed when temperature rises.

So I need to install some kind of signal amplifier on the signature wire. I hope now it makes more sense.

Help would be much appreciated as this is driving me nuts

Regards

 

[4pyros]

What is the original controller?
What kind of fan?
Whats it for?
How do you know it is PWM?
What is a 3.3 volt "signature" wire?
We can not give you advice on how to modify somthing without knowing what it is.
Andy

 

[#12]

A start

This is a start. You will still neeed to amplify the 3.3 volt signal to get more range, but it's a start.

 

[MikeMl]

If the fan is not cooling; get a bigger or faster (noisier) fan. Trying to run the existing fan faster will likely overload the motor inside the fan, and you will be replacing it anyway...

 

[tfixrepairs]

Ok. The system is gaming console Playstation 3 which has original controller built into the motherboard.
The fan has 3 wires in: +12V, -12V and a third wire which controls how fast the fan spins.
The modification would be to restrict the fan running lower than 40% speed.

 

[Reloadron]

Basically the device does not get sufficient cooling due to fan controller not controlling the fan correctly via the Pulse Width Modulator. The fan is a 12V with a signature max 3.3V wire controlling how fast it spins. Now an original controller increases the fan speed when temperature rises and what I want to do it make fan spin faster from the beginning and to preserve the original controllers ability to control the speed when temperature rises.

So I need to install some kind of signal amplifier on the signature wire. I hope now it makes more sense.

Help would be much appreciated as this is driving me nuts

Regards

OK, if I may suggest something here. I agree with many of the post in that more information is needed. What you refer to as a signature wire I have never heard of pertaining to fan speed control. You may want to give this link a read and see if any of the shown circuits pertain to what you have. What is on this 3.3 volt line? The way you describe it, it seems to be an analog signal (you mention 3.3 volt max). Are you saying this signal determines fan speed as in 0 to 3.3 volts = low to max fan speed?

Additionally in the configuration you have there should be a sensor somewhere used to sense temperature and send that signal to the fan speed controller. Sometimes it can be as simple as relocating the sensor to a hotter zone. Additionally making sure if for example the sensor was bonded to a heat sink it has not become loose.

As to fans themselves, there are a few flavors used in cooling electronics. There are two wire fans, there are 3 wire fans and there are 4 wire fans, all covered in the link I provided. In the case for example of a typical 3 wire fan the extra lead is generally a tach signal out of the fan. A typical 4 wire fan may include a tach out and allow for PWM signal in. The width of the PWM pulse determines the fan speed. Again, people need to know exactly what you have to make suggestions.

<EDIT> OK, I see it is a Play Station 3 with a 3 wire fan. You are absolutely sure the third wire is for speed control and not a tach output as I outlined? </EDIT>

Ron

 

[tfixrepairs]

The PS3 fan is PWM fan. This means it gets 12V all the time through the red and black wires. The grey wire supplies a pulse of electricity to the fan which in turn determines how fast it spins. The faster the cycle of pulses, the faster it spins.

 

[MikeMl]

The PS3 fan is PWM fan. This means it gets 12V all the time through the red and black wires. The grey wire supplies a pulse of electricity to the fan which in turn determines how fast it spins. The faster the cycle of pulses, the faster it spins.

Well, then you need to increase the ON time of the PWM signal, not the amplitude. If when the fan is running full-on now, and the PWM duty cycle is approaching 100%, there is NO way to make the existing fan run any faster.

 

[tfixrepairs]

Thanks for reply. When the fan starts it runs at speed level 1, when temperature rises it runs at speed level 2, when temperature reaches max fan runs at speed level 3. What I want to is to increase the fan speed at each current level.

 

[MikeMl]

So hang a scope on the PWM signal, and measure the duty cycle at each level. If the Duty Cycle is 100% at the third level, replace the fan!

 

[tfixrepairs]

There aren't any problems with any fans. I want this modifications to be installed in all future Playstations fixed by our company.

 

[Reloadron]

The PS3 fan is PWM fan. This means it gets 12V all the time through the red and black wires. The grey wire supplies a pulse of electricity to the fan which in turn determines how fast it spins. The faster the cycle of pulses, the faster it spins.

God, the more I read about those PS3 units Sony could have done a much better job of engineering the cooling.

OK, I found this particular read interesting. The guy designed a nice simple little PWM circuit to drive the fan.

What you need to understand is what PWM actually is, those pulses of electricity. In this case with a fan, PWM is just turning the fan on and off at a very fast rate. During a given period of time the pulse is on for a percentage of the period and off for a percentage of the period. If we look at the start of any pulse to the start of the next pulse that is the period. The greater the on time (the pulse width) during that period the faster the fan will spin.

Short of buying some of the external cooling devices I have seen to force more air through the thing there is no easy way to increase fan speed short of what the link has come up with.

One method might be to apply a constant 3.3 volts to the control line which in theory would run the fan full speed constantly. Simple but likely also noisy. I don't own a PS3 so have no clue how loud it would be.

Another method would maybe be to build a small "Pulse Stretcher" circuit to extend the on time of the pulses.

The last method would be to build your own PWM temperature controlled circuit like the one I linked to. This would require some soldering skills and maybe about $10 USD in components. The cost of building a pulse stretcher would be about the same and likely do less.

Maybe some others have better thoughts to share but that would be it for me.

Ron

 

[Reloadron]

There aren't any problems with any fans. I want this modifications to be installed in all future Playstations fixed by our company.

Now I see this. Since you repair these things then I would think pulse stretcher. Obviously you want a mod to fix an inherent problem. I had no clue this was for a business and commercial application. Would have been nice if you mentioned that. There is a hell of a difference between screwing around with a personal project and modifying OEM equipment.

Ron

 

[tfixrepairs]

I guess I should have mentioned it was intended for Playstation System. So now once you guys figured out what I need, how about some advice how to get one built.

I was looking into this circuit:

Pulse width modulator by LM3524 | Circuit Project Electronic

 

[MikeMl]

You missed the obvious suggestion I made. Use a scope to look at the PWM signal in the existing circuit. If at the third level, the signal is steady flat line at 3.3V, there is nothing you can do to make the existing fan run faster. If you see a pulse train, measure both the period and the duty cycle. Report back with what you see.

 

[tfixrepairs]

I might not be able to express myself properly because I don't think you still get what am I trying to build here, no offence.

I would also like to note that this kind of modification was made by couple people over the net and they are charging money for the mod. I don't want to pay for something that can be built myself.

I am also unfortunate enough not to own a scope as yet.

I would also for a millionth time would like to explain that I not trying to run the fan faster than it's maximum capacity. What I want to do is to increase the speed curve of the fan according to temperature.

Take an example that at 40C fan spins at 1000RPM, at 60C fan spins at 2000RPM, at 80C fan spins 3000RPM. I am trying to achieve the increase of fan spinning at 2000RPM at 40C, at 3000RPM at 60C and at 3000RPM at 80C.

Regards

 

[Reloadron]

OK, lacking a scope, try as Mike suggest using a meter. When at what should be step 3 see if the reading is a steady 3.3 volts. Though a scope would be best for this. Steady means steady as in the meter reading is steady and not erratic.

Next if the PWM period I mentioned earlier were known (back to a scope) this would be easier. If for example at stage 3 as we called it, the signal is not at a constant stable 3.3 volts then possibly a pulse stretcher may be incorporated.

Ron