[Solved] Giving RPM to Mobo AND Fan Controller?

[angelkiller]

Hi all,

I bought a NZXT Sentra 1 which is a cheap fan controller. However it makes my Scythe S-Flex's buzz. Luckily I wasn't expecting much. (I mostly got it for the blue LCD )

So now, I want my motherboard to control my fans. But I still want my fan controller to monitor the fans' rpm. So both the mobo and fan controller will get the rpm signal.

I'm not sure how safe or possible this is electrically. Can I just split the RPM wire going to the motherboard and wire it to the fan controller? Here's a quick and dirty illustration.


Possible? If not how? (or how not) Thanks for any help!

[angelkiller]

Hmmm. Does everyone understand what I'm trying to do? I hope I'm explaining it clearly enough. Another picture:


I want to know if I can split the RPM cable. Electrically speaking, will it work? If not, how can I get both the motherboard and fan controller to read the RPM of a single fan? Thanks!

[Namheul]

Try it out and see

I have no idea if it will work or not.

[Fallen Kell]

I really don't know if it will work. I am not sure how the signal is sent down the 3rd wire, i.e. if it is pulse/time based, or if it is wattage based. If the signal is simply pulse based, then a simply Y splitter will work. If it is voltage based, well, when you split the wire with a Y splitter, you will be splitting the wattage. I suspect it is not wattage based as long wires will also cut the wattage down do to resistance in the wire's length, but I really don't know.

[Felger Carbon]

I really don't know if it will work. I am not sure how the signal is sent down the 3rd wire, i.e. if it is pulse/time based, or if it is wattage based. If the signal is simply pulse based, then a simply Y splitter will work. If it is voltage based, well, when you split the wire with a Y splitter, you will be splitting the wattage. I suspect it is not wattage based as long wires will also cut the wattage down do to resistance in the wire's length, but I really don't know.

The tach signal transmitter from the fan is an open-collector transistor. The receiver at the mobo is specified to be a 2 or 2.5ma load when pulled low (I forget which). What you want to do here is put two loads on the tach signal open-collector, which means it has to drive 4 or 5ma when low. While in a specific case it might work, I wouldn't depend on it in general.

[Vicotnik]

But does the motherboard really need the RPM signal in order to control the fan? I only have experience with the Asus Q-fan feature on a few of my old boards, but with those no RPM signal was required. The output voltage was regulated based on CPU temp, that's all.

[brsanders]

Angelkiller - did the signal split work ? It would be nice to have my fan controller and Speedfan know the speed. Please let us know

Thanks

[angelkiller]

Angelkiller - did the signal split work ? It would be nice to have my fan controller and Speedfan know the speed. Please let us know

Thanks

Ok, I've been really busy lately and this was low on my list, but I got around to doing this. Sorry for the delay.

Anyways, I didn't want to cut any wires on my S-Flex's so I pulled the fan off one of my many Intel stock heatsinks. Then I made this:


So I connected everything and it worked! Speedfan saw the fan spinning at ~1500rpm. My fan controller doesn't report actual rpms, but it showed that it detected a fan.

The tach signal transmitter from the fan is an open-collector transistor. The receiver at the mobo is specified to be a 2 or 2.5ma load when pulled low (I forget which). What you want to do here is put two loads on the tach signal open-collector, which means it has to drive 4 or 5ma when low. While in a specific case it might work, I wouldn't depend on it in general.

Honestly, I have no idea what this means. Now that I've tried it and it works in this case, what's the technical conclusion?