Hacking into Fan speed switch...

hey, ive got an idea to control my HVAC via carputer and need to discuss some circuitry with you guys.

Firstly, I have 5 fan settings : 0 1 2 3 4
These are controlled by a rotary switch. The fan harness has 5 wires: one for each speed and ground. When in position 1, wire 1 and ground and connected. So on and so forth. Obviously in 0, nothing is connected.

To control the fan speed, i was going to use a Crystalfontz 633, which can control 4 PC fans. I was going to use each fan output as a ground for a relay or something which would connect the appropriate fan speed wire to ground. I would have to be careful though and make sure no two fan outputs were on at the same time, because that would short some **** out.

Someone was mentioning to me that they wished they had more control over the fans, as they liked more air than 1, and less than 2. I thought about it for a little bit, and realized that the Crystalfontz also does PWM to the fans...

so, what if i used fan speed 4, which would make the fans blow maximum, but used PWM. That way, i also wouldnt have to worry about shorting anything out and i also have better control over the fanspeed.

How would i make the circuit though, i would need something to handle all the current of the blowers, and i dont think a relay could handle fast switching of PWM...so is this even possible? I would like to make at least a test circuit, to see how the blowers respond to PWM. I know some fans/blowers make extra noise at certain PWM levels.

thanks for your time

see..... i am going to control my HVAC controls using my carputer but since its vacuum controlled the only way i can think of is use high torque servos and wire them into a controller and make a little program in VB that has the actual vent symbols and temp select and it will tell it to turn the knobs a few degrees.What do u think?

Bah, sorry, i dont know what the hell I was smoking when I posted that up. The fan switch connects the GROUND side of each circuit to ground. Not 12V to GND.

Also, i measured current draw up to fan speed 3, and its 10A. I couldnt measure fan speed 4 because my meter can only handle up to 10A and i didnt want to blow its fuse.

In addition i dont think PWM would be very good for the blowers or the electrical system. Maybe im wrong but switching ~15A on and off rapidy cant be good for the system. Your input?

CodesterZ24: im guessing youve got a cavy. Ive seen your HVAC controls but i dont know how to interface with the vacuum stuff...mine is all wheel/cable/gear driven.

Quote:

---

Originally Posted by Dexter

I would have to be careful though and make sure no two fan outputs were on at the same time, because that would short some **** out.

---

You can always use a 1-out-of-10 decoder chip. Only one of the output can be selected via an address. Is a good idea because it maintains the compatability of the interface and the fan electrics.

Quote:

---

Originally Posted by Dexter

How would i make the circuit though, i would need something to handle all the current of the blowers, and i dont think a relay could handle fast switching of PWM...so is this even possible? I would like to make at least a test circuit, to see how the blowers respond to PWM. I know some fans/blowers make extra noise at certain PWM levels.

---

You cant use a relay for PWM, that be killed in no time.

You really need a power driver, a MOSFET perhaps together with a fast diode to clamp any crap comming back into from the motor. Dont run the PWM frequency too high as you will stress out the drivers, reduce its efficiency and generate more heats. Running at some frequency may become annoying like 1Khz or so, you will hear that winning noise.

Ill stick with the relays and some logics if I where you.

much thanks for reading and responding to my post.

Yeah, I think i will just stick with using the 4 fan outputs and relays.

good idea/effort though. i have a vacuum based setup. i saw a guy that actualy used vacuum actuators for his setup and it worked. i'll have to dig that thread up.

also: i dont know if anyone would know this, but i sorta destroyed the temperature sensor to my HVAC. I'm pretty sure it's just a little thermoresistor. dont ask how, its embarassing ;)

anyways, anyone know where i can get one other than a salvage yard? its a 95 thunderbird. i didn't know if thermoresistors were pretty standard and i'd get a pretty accurate reading/correlation with my hvac if i just picked one up from radioshack.

Well that depends if you're talking about the external sensor or the internal one I guess. 95 Thunderbird LX, currently saving up for a new engine >_< I thought it was a puddle, not a lake :-P

Anyway the internal one is a thermisitor, although you could have destroyed the circuitry if this is a voltage-type destruction. If it's a physical thing then the internal one is behind the dash here:

http://www.cupofnoodles.com/~silver/...tbird_temp.jpg

Though since thermisitors generally don't have any value listing on them and the only other way to match it would be to test it (if it were good) I'd highly recommend just going to a junkyard and pulling the piece out.

Is the selector switch connected to any intermediate control circuits or directly to the coils on the fan

Not sure; are you asking me?

I used the decoder chip as suggested by Chris31, and made a circuit board, with transistor driven relays.

The OEM circuit only connects the blower to a resistor pack (to change "speeds") and then ground. There is only one intermediate wire that comes from the resistor pack to the A/C control circuit, which turns the A/C on the fans are on, so its used essentially as a "fan on" sense.

Please, post one electric schematic for see your idea.. ;)


Mauri