Fake fan tach signal?

[med]

So here's a thread that's a bit over a year old that shows precisely what I think I'm after. I have a Dell drive enclosure where I just want to pretend as though the jet-engine fans are still present. I'm not an electronics guy and don't trust myself to put together a device like pictured in the thread:
Generate Fan RPM Signal W/O Fan

Right now I literally just have a spinning fan core generating the signal to prove the concept-- but of course it's not entirely silent. And it's a bit ghetto

After googling around like a newb, I came across a couple items on ebay-- would something like either of these work?
Adjustable PWM Signal Generator Module * Output adjustable * based on NE555
1PC Nano Signal Generator Module 555 pulse frequency adjustable square wave PWM

I think people intend to use something like this to control the PWM lead / speed of the fan, but obviously my need is different. I'm not sure of the actual specs needed (V, A, frequency etc)

Do people sell something for this exact purpose? I'd gladly buy half a dozen, but just can't find the right thing.

 

[Fenris_Ulf]

The PWM signal generators adjust the WIDTH of the signal (how much it's "on" vs "off"). The second one listed allows frequency adjustment, which is what you need. Run it off of 5v only and hook the output to the fan signal header. You may need to play with the settings to get the right minimum frequency.

 

[ChRoNo16]

I think RPM is just a resistance measurement, IE If I am right (Which I may not be) grounding out that pin should make the board think the fan is 100% but I would double check my info.

 

[EchoWars]

What Fenris is saying is that you don't need to generate a full-blown PWM signal...a plane square wave from the 555 will do, since the speed detection on the motherboard simply catches the rising edge of the pulse coming back from the fan to calculate fan speed. Higher fan speed = more pulses per second which is translated into a bigger number. The PWM is the signal coming back into the fan for speed control, and is not the same as the signal sent from the fan to the MB that is used to determine the fan RPM. Two different things.

The common cooling fans used in computers use standardized connectors with two to four pins. The first two pins are always used to deliver power to the fan motor, while the rest can be optional, depending on fan design and type:

• Ground – common ground
• Power – nominally +12 V, though it may be variable depending on fan type and desired fan rotation speed
• Sense output from fan – outputs a signal that pulses twice for each rotation of the fan as a pulse train, with the signal frequency proportional to the fan speed
• Control input – a pulse-width modulation (PWM) input signal, which gives the ability to adjust the rotation speed on the fly without changing the input voltage delivered to the cooling fan

This would be a pretty simple build. Your first link shows a decent attempt, but it could probably be made even smaller.

Edit: Look here. Scroll down to the section on 50% duty cycle. Two resistors, two small caps, and the formula for frequency (and you don't need an adjustable potentiometer). Since a real fan outputs twice per rotation, then if you know the fan speed you want to emulate, just double that number and that's your frequency target. For instance, R2=1K and C1= 0.1µf, then f=7.2KHz and would emulate a fan running at about 3600 RPM.

Edit #2: Assuming that you don't want to fool with all that, then something like this would work, but he (and others selling the same thing) don't say anything about the frequency range. Shoot 'em a message and ask.

FWIW, the 555 timer is capable of sourcing or sinking 200mA, so if you don't mind all the fake fans reading the same thing, you should be able to connect the the output of one of these boards to as many fan sense points on a motherboard as you like.

 

[SuperFastCars]

My issue:
"Alert - Rear Fan Failure" when BIOS starts up the PC.

I looked online, and found out my Optiplex 755 is supposed to have a 5.25 inch hard drive fan enclosure ("Dell NH645 HDD Caddy"), with a oddball 12V 0.36A fan called "AVC model BN06015B12H". I tried to find the RPM everywhere, but then I found a substitute fan, the Foxconn PVA092G12M , which gives a maximum rpm of 3,200 rpm.

I found a link online for the dell pinout for their weird 5 pin layout:

1. Tach/Sense (White/Yellow) - Two pulses per revolution, fan gives it's speed.
2. 12V (Red) - Power Positive, +/- 1.2 Volts
3. Ground (Black) - Power Negative
4. PWM/Control (Blue) - This is a 21-28 kHz, 5.35 Volt signal from the motherboard, 20 to 100% duty cycle.
5. Unused

*** Make sure to verify with a volt meter which pin is +12V from ground, I read online that if you plug it in wrong, it can fry the motherboard. ***

I bought a real cheap dell 5-pin fan on ebay for $4, and a NE555 timer circuit for $3. (I checked on Texas Instruments' website, and the NE555 is capable of 4.5-16 volts, max frequency of 0.1 MHz, at 2 mA drain.) They sell the timer circuit on banggood real cheap, but they're shipping is very slow. (3-4 weeks)

Once they arrive, I'm going to first tweak the timer chip using an Arduino. The stock rear fan on my Optiplex 755 runs at 3200 rpm, so I'll need to generate a 50% duty cycle signal at 6400 rpm, with a period of 0.15625 microseconds. (That's what the variable resistor is for.)

Then I'm going to rip out the wire from the cheap dell fan, and solder it directly into the NE555 timer circuit. That way it will plug in exactly the right way.

Finally I can plug it into the PC, and see if the "Alert - Rear Fan Failure" message goes away. I'll probably wrap it in electrical tape to prevent any short circuits.

Once I get the OS loaded up, I can probably use a program like speedfan to fine tune the speed to exactly 3200 rpm.