3 pin fans: What are the voltages used by the tacho (speed) line?

[tgm1024]

Ok, I'll just list things out instead of getting all wordy-smurdy because we engineers typically like stuff like that.

1. I have a cooling pad for a laptop.
2. Both 2-pin 5v Fans (70x70x15 mm) were failing.
3. The only 2 pin 5v fans I could find at that size are too loud.
4. Only quality fans at that size are 12v 3 pin. No worries on the connections: I have pin connector wire that'll do the job just fine.
5. A 12v fan driven at 5v would likely be too slow (if I simply use the red/black wires).

Ok, how do I hardwire the tacho line so that the fan behaves as a 5v fan? Are there are 3 distinct voltage levels handed to it?

Note: I'm clearly not talking about PWM fans.

THANKS guys!!!

 

[richaron]

Tacho wire is just a feedback from the fan for the controller to sense; It doesn't control the fan speed. Fan speed on 3 pin fans is controlled by varying the voltage. So you can make a 12V fan behave like a 5V fan by... giving it 5V

If you want it faster you'll have to give it more Volts or buy a faster fan.

Edit: Or you could put a pot (or fixed resistor) on the noisy 5V fans you mentioned to slow them down

 

[tgm1024]

Ah, ok, THANKS(!), my mistake.

I'll have to keep looking I suppose. These things I put in now are horrendously "whir"y.

 

[tgm1024]

Edit: Or you could put a pot (or fixed resistor) on the noisy 5V fans you mentioned to slow them down

Could I simply wire the two fans in series? Or is there a risk of having one of them fail, hog the current, and potentially damage the computer's USB port?

 

[richaron]

Could I simply wire the two fans in series? Or is there a risk of having one of them fail, hog the current, and potentially damage the computer's USB port?

Good question...

I can't give any advice on that, I simply don't know. I like the idea in theory and it should do what you want, but I simply don't know much about standard USB over current/voltage protection. Here's hoping someone else can chime in.

Edit: I didn't make that image above, it came from a quick search. I just noticed the 1-2-3 pin numbers and colours are all messed up, heh. Sif trust me

 

[Paperdoc]

You can't really use 12 VDC 3-pin standard computer case fans at 5 VDC. 5 VDC is actually about the MINIMUM voltage that a computer mobo will send out to a fan, because below that it may stall and not re-start. So it may not even start up with only 5 VDC. As a fan wears, it starts to slow down due to increased bearing friction, so any fan that even starts and runs on 5 VDC will fail early for this reason, long before it would fail in a "normal" situation.

Control of the speed of a 3-pin computer fan is done entirely by varying the voltage on Pin #2 from 12 down to 5. These fans are all "brushless", meaning that they do not have any commutator and brush system that switches the current to a series of windings. In fact, they have their magnets in the rotor, and the windings are all stationary as field coils. Instead they use an electronic "simulated commutator" system.There is a small sensor that generates electrical pulses (two per evolution) to sense the exact position of the rotor, and that is fed into the motor's circuit board to synchronize the switching of electrical current to the field coils with the position of the rotor, no matter what its speed. That same circuit sends out a version of that on Pin #3 as a series of 5 VDC pulses, two per revolution, that the mobo can count to generate a speed reading. The mobo does not actually use that for speed control, though. It DOES use that to detect fan failure - no signal, or a signal too slow, generates a failure warning.

 

[tgm1024]

Ok, thanks because that information regarding 3-pin fans is incredibly illuminating.

But again, given my two 5v "noisy" fans, is it safe to run them wired in series? The 5v supplied by the USB would presumably be hacked in half resulting in a much slower fan speed (and hopefully much less noise).

The thing about that config that I don't understand is what happens when a fan starts to fail: does the other fan start to hog the current, or does the resistance fly upward enough to damage the USB?

Just trying to think in terms of what I don't even know I don't know about this.

 

[Paperdoc]

Yes, you could wire both 5V fans in series, rather than parallel. Then a failure of either fan would simply open the circuit and neither would work. Worst case might be if one of then shorts out completely so that it is effectively removed from the circuit, leaving the other the only load. But if you are talking about wiring the two of them into a 5 VDC series configuration, then the remaining fan will still only get 5 VDC and work.

There are two downsides to this. One is that each fan will only receive 2.5 VDC, and hence run very slowly - you would get LESS that half the cooling you are getting now, and that might be of little use to you. The other is that such a low voltage may not be sufficient to start up those fans from stopped. This latter MAY be a worse effect because the reason for the noisy condition of the fans now may be that their bearings already are badly worn and causing excess friction.

Unfotunately your best solutions are:
1. Find matching 5 VDC fans to replace them with - maybe from the manufacturer of the original cooling pad.
2. If that's not possible or too expensive, buy a new cooling pad.

The other options, of course, are to tolerate the problem, or to quit using that pad.

 

[mindless1]

You can't really use 12 VDC 3-pin standard computer case fans at 5 VDC. 5 VDC is actually about the MINIMUM voltage that a computer mobo will send out to a fan, because below that it may stall and not re-start. So it may not even start up with only 5 VDC. As a fan wears, it starts to slow down due to increased bearing friction, so any fan that even starts and runs on 5 VDC will fail early for this reason, long before it would fail in a "normal" situation.

You are overgeneralizing. It is true that "some" 12V fans won't spin up at 5V, but many will, reliably. The most troublesome are those which are thicker because the torque needed to spin is higher.

It's generally the opposite about fan wear. A 12V fan spinning at 5V wears MUCH slower, so much so that the 12V fan would have worn out its bearings many years sooner than the smaller amount of wear had any significant effect on it running at 5V.

Come to think of it, I don't recall ANY fan I've ever ran at 5V failing. Granted if it was a a sleeve bearing model, I lubed it generously before deployment and practically speaking I can't think of much gear I've ran for more than 15 years... At least talking computers there was one old server, then a battery charger, soldering rework station, and a couple stereo amps are the only things that come to mind at the moment which were still used at 15 y/o.

 

[mindless1]

This is probably not a difficult thing to solve. Most likely all that was needed for the loud 5V fans was a diode in series on the positive lead. A resistor would work too, offering even better adjustment of speed but it becomes a bit fiddly trying different resistor values till the ideal is found, for a one-off project anyway. The resistance range to look at is a few tens of ohms and you can calculate (and then derate for cooler operation) the resistor wattage or just overshoot and use a 2W rated resistor.

I have both a rheostat and multi-position switch with increasing #s of diodes in series that I built around 20 years ago in order to more easily test fan speed control before there were better options (only to determine the # of diodes or resistance needed then pick the right value discrete components), but even in some modern devices or DIY projects there is no variable fan speed control so doing it old school still helps pick the optimal fan speed, and increase fan lifespan, decrease noise and dust from reducing the RPM. Plus it only costs a few cents to use a resistor or diodes. If You Can Solder.

 

[tgm1024]

Lots to ponder, thanks all!!!

The reason I'm interested in fixing this instead of replacing it are basically two fold:

1. It is of really heavy aluminum and of high quality construction, like I've never seen elsewhere.
   
   
2. I like fixing things.

So I'll have a go at it and if I get stuck, I'll drag meself back here and beg for assistance...

 

[aigomorla]

Other then that, in a 3 pin the yellow line is just a rpm sense line, so u wont even use that at all.

You are overgeneralizing. It is true that "some" 12V fans won't spin up at 5V, but many will, reliably. The most troublesome are those which are thicker because the torque needed to spin is higher.

It's generally the opposite about fan wear. A 12V fan spinning at 5V wears MUCH slower, so much so that the 12V fan would have worn out its bearings many years sooner than the smaller amount of wear had any significant effect on it running at 5V.

A Lot of PC fans will not spin at 5V... u will require 7V to initate spin, and then can downvolt to 5V, but a fan motor has a min start voltage.

Usually the quite ones have a built in internal resistor, so when you try to start the fan at 5V in actuality your supplying less then that, which causes the motor not to spin.

to be on the safe side you will probably need a usb fan, which have motors meant to start at 5V like these:
https://www.amazon.com/s/ref=nb_sb_noss_1?url=search-alias=electronics&field-keywords=usb+120mm+fan

 

[mindless1]

^ There are more than you're implying that will spin at 5V. The tricky part in this case is that tgm1024 is looking for 70mm X 15mm, a fairly rare size compared to 60/80/92/120mm.

As far as super-low RPM fans that use an internal series resistor to achieve that super-low RPM at 12V, you wouldn't pick one of those as the airflow would be too low at 5V. A typical target would be something rated for around 1600RPM at 12V, or if the cooling pad fan openings can be enlarged to 80mm diameter -or- going the opposite direction if there is sufficient depth to use a 20-25mm fan then add a 10mm gasket with a 60mm x 20-25mm, either of which opens up more fan make/model options.

You linked some 120mm fans. If there is a more common size than 70mm x 15mm that'll work, I can test one of the several different fans I have to make sure it'll run at 5V. However I mostly buy major brand fans at deep discount, if I find a few that do fine at 5V but they are only available now for $15 each, $30 for a couple fans seems sort of steep.

I would look at what can be done to get larger fans in. If there's any way to shoehorn 80x25mm in, that makes things a lot easier, including getting sufficient airflow at 5V.