Fan RPMs measuring

[crmaris]

In some reviews there are charts with the fan RPMs according to the load.

How it is possible to take the fan RPMs through the PSU? I haven't made any research on this subject so please bear with my question if it sounds silly :)

[rafal_iB_PL]

Tachometer. You typically put a reflexive tape on one of the blades and passes are counted by the device. Of course you have to postition the tachometer so that sensor is in line with the tape. With units having a wire fan grill, it is possible to stick the tape even without opening the unit. I'm yet to try peeling though.

[crmaris]

thanks. I think that some reviewers take in parallel the PWM signal from the fan but I would like to see how they do this exactly. They use a converter or they simple stick a thin wire in the PWM pin?

Since I have some unused ports left in my LabJack I could easily log the fan speeds and make a nice looking graph.

[Elledan]

I'd recommend using reflective dye or so instead of tape to prevent unbalancing the fan blades :) Would kind of mess up noise measurements :p

[Zero82z]

I'd recommend using reflective dye or so instead of tape to prevent unbalancing the fan blades :) Would kind of mess up noise measurements :p
A small piece of tape doesn't have enough mass to unbalance a fan.

[crmaris]

the thing is that the PSU will be inside the hot box and the tachometer must be vertical to take the readings. So I have two problems, one is the small (in height) hot box and the other that I must find a way to mount the tachometer inside the box in vert position above the PSU.

[Tator Tot]

You could always split the wire and lead the RPM Sense wire to a fan controller that can read out RPM's.

That's a rather crude method; but it would work.

[Zero82z]

You could always split the wire and lead the RPM Sense wire to a fan controller that can read out RPM's.

That's a rather crude method; but it would work.
PSU fans generally don't have RPM sense wires.

[Tator Tot]

PSU fans generally don't have RPM sense wires.
It depends on the model I know, but once you have the cable made it's easy after that.

[Makalu]

if the fan doesn't have rpm circuitry in it then you're out of luck and if the fan does have it then it'll already have an rpm wire.

I don't think you can mount a tachometer in a permanent position and expect a piece of reflective tape on a bottom mounted fan blade to align up since psu and fan sizes are different and fan located offset differently and different fan guards that could block the beam.

The rear mount fans are easy for me...bottom mount I have to open the hotbox side door and hold the tachometer with both hands and brace my grip against the side of the hotbox and crane my neck upside down to look up at the fan and get the beam on the tape and then hold steady to look down at the rpm reading. I can do it in about 7 seconds though...from door off to door on if I'm trying to maintain temps.

[Olle P]

Tachometer. ... you have to postition the tachometer so that sensor is in line with the tape.At what distance? Going too close and it will interfere with the airflow, forcing the fan to speed up, and thus providing a false result.

Using a stroboscope one must just have the fan visible. (Directly or via some CCTV.)

thanks. I think that some reviewers take in parallel the PWM signal from the fan ...That's just to get a voltage reading, right, not the actual speed?

[crmaris]

actually you can easily measure the fan speed via fan's sense cable.

Look at this very informative article that Seafalco wrote.

Is in Greek but you will easily understand the general concept.

http://www.thelab.gr/air-cooling/fan-speed-measurement-92394.html

[rafal_iB_PL]

At what distance? Going too close and it will interfere with the airflow, forcing the fan to speed up, and thus providing a false result.

I measured at approx 20 to 30 cms, and somatically there was no somatically appreciable difference. What could make much bigger difference would be poorly "installing" the tape, so that it hits the grill as fan spins, not only altering actual RPMs, but also skewering acoustic themselves.