Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Computer fans again. Tach signal.

Status
Not open for further replies.

DirtyLude

Well-Known Member
The problem is, the tach signal only works when the power is on.

The tach signal 'on' pulse is actually on for half of the fan rotation and off the other half, so 50% duty cycle based on one fan rotation.

If I use PWM to drive the fan, and connect the tach signal to a timer input on my microcontroller, when the PWM goes low, it shuts off the tach signal, giving false signals.

I'm looking for a way around this problem. Will a low pass filter on the driver PWM smooth out the driver power enough to keep the tach signal up when the PWM is low? Or maybe the same thing on the tach side to filter out the smaller signal drops from the PWM low signals?
 
the way around it is not using PWM, or switch to a tach separate from the fans power, like a little reedswitch glued to the outside of the fan frame and a tiny magnet glued to one of the blades

most of your generic computer fans actually give two tachometer pulses per revolution, make sure you have a pull-up resistor on the tach line.

instead of pwm, look into variable voltage control, this way the fan is constantly receiving power, not just short pulses of power.

not sure how you do it directly with a MCU, i'm using some specialized chips from Maxim to do it
 
Thanks. I saw the Maxim chips when I did a search earler. Unfortunately Digikey lists the max6651 for $7.95. That's way too rich for me and my low buck PC water cooling rig.

I'll do some more experimintation and see what I can do. A low pass filter for some basic voltage control should work. It just seems like it takes hours of experimentation to get anything to work.
 
"a little reedswitch glued to the outside of the fan frame and a tiny magnet glued to one of the blades"

This will make a lot of noice and destroy the fan in a short time!

Ante :roll:
 
Not sure if this helps...

If the microcontrolLer is of your design, why not also sample the PWM signal, and only "clock" the Tach signal while the PWM is high? This could likely be done with a spare input, and a minor software change.

You may also be able to "hack " the fan itself and supply power to the Hall sensor circuit or what ever they are using, giving you a steady TAch signal.
 
Thanks.

I thought about that as well; sample only when the pwm is high. It would add a level of programming complexity that would be easier to get around with a simple filter if possible.

If it's not possible, to get around this problem with a simple hardware change, then I'll probably fall back on the software option.
 
Instead of using a 0 / 12V PWM signal, use a 3or 5 / 12V, so that the fan is never completely off. Then your tacho signal will work.
 
petesmc said:
Instead of using a 0 / 12V PWM signal, use a 3or 5 / 12V, so that the fan is never completely off. Then your tacho signal will work.
That's a good idea.

I have a 3v line and a 12v line. If I put a diode across them, the fan should have 3v while the transistor is off, and 12v when it's on The diode will stop the 12v from leaking into my 3v line.

I'm not very good with electronics theory, but I think that should work.

I could never turn the fans down lower than that 3v supply, but I don't see that ever happening anyway. Actually these diodes have a voltage drop, don't they? I guess my 3v would turn into 1 or 2v.

Hmm... I'd have to make sure my 3v regulator can supply enough current to those fans and that it doesn't add alot of noise to the 3v line. I have an RF transmitter on that line, which is very sensitive to noise.

I'll try the filter first, that's just as simple a solution, if it works.
 
If by "filter" you mean to put a capacitor on the PWM output to the fan, won't you just be delivering a variable DC level, in effect just making a simple switching supply?

This end could be achieved by just using a LM317.

What about putting an IR LED pair on opposite sides of the fan, and just dividing the count by the number of fan blades. You could get a IR LED pair from an old VCR. The transmitting IR led is often very tiny, and couild be fit into the side of the fan housing, same with the receiver LED.
 
Hey Mark, Check your PM's!
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top