Thermatic fan (convert single-speed fan to high/low speeds)

[chimpboy]

I thought this might interest some people.

I have a rover V8 transplanted into my car, and for space reasons there's no longer a mechanical fan, only a 16" electric fan in front of the radiator.

I'm fitting air con to the car and this means I want some airflow whenever the a/c is switched on; usually cars do this with an auxiliary fan but I only have room for one fan. When it's on, it's FARKING ON... noisy and moving a lot of air, which is fine if it cools the engine but not something I want going on 100% of the time.

Long story short - less than ten bucks at a pick-a-part type wreckers gets you a high wattage resistor from an Audi or BMW that they use to enable a slow speed on their electric fans. Find them in 80s and later vehicles, usually bolted so they sit in front of the fan (they benefit from the air flow keeping them cooler). They are approx 0.5 - 1 ohm resistors with nice metal housings and screw down terminals that should be good for 80 plus watts.

I grabbed one from an Audi and it does exactly what I wanted, runs the fan at a lower speed that is almost silent but still definitely pushes air through the radiator and a/c condenser.

Of course if you have room for two fans you can just run one instead of both for the "low" function. But note that one 16" fan moves as much air as two 14" fans anyway.

[stuee]

Good idea.

**Boring tidbit on**

I had a thought of playing around with pwm control of a thermo fan using high-powered mosfets or bjt's. No reason for doing it, just a geeky thing to do and we'd done similar motor control projects at uni. Idea was use PID control so that the temperature is as close to constant as possible, rather than bouncing between the two temperature limits. Could be used in conjunction with an electric water pump too.

But alas it all proved futile as the basic bang bang control is much more simple and easy to use. Also very reliable and effective at what it does. I don't think there is any benefit of running a motor at any exact temperature anyway.

**Boring tidbit off**


Out of curiosity what sort of heat does the resistor give off when its in use. is it warm or burning hot??.

[chimpboy]

It gets hot, but for the period I tried it (before I mounted it in front of the fan), it never got too hot to touch or anything.

Funny, I thought about a pwm control as well but it looked too hard and too expensive, and I really just wanted two fixed speeds not variable speed.

[-Scott-]

How much do these fans draw? I recently discovered switch mode plug-in replacements for LM78xx series regulators (TO220 package) - with a little bit of extra circuitry you could probably set one of these up to vary the output to the fan, without wasting too much power anywhere else.

Thinking more about it, they're probably horribly noisy suckers which would interfere with radio reception, so forget that.

Since the resistor is "throwing away" power anyway, why not use a linear voltage regulator mounted in the same position, with temperature feedback (NTC resistor?) to vary voltage to the fan? A sort of intermediate solution - infinitely variable speed control, without the hassle (and noise) of digital switching circuits?

[chimpboy]

How much do these fans draw? I recently discovered switch mode plug-in replacements for LM78xx series regulators (TO220 package) - with a little bit of extra circuitry you could probably set one of these up to vary the output to the fan, without wasting too much power anywhere else.

Thinking more about it, they're probably horribly noisy suckers which would interfere with radio reception, so forget that.

Since the resistor is "throwing away" power anyway, why not use a linear voltage regulator mounted in the same position, with temperature feedback (NTC resistor?) to vary voltage to the fan? A sort of intermediate solution - infinitely variable speed control, without the hassle (and noise) of digital switching circuits?

A 16" fan will allegedly draw around 20 amps.

I think if you try to go too slow with just a series resistor or voltage drop, you get stalling problems... that is something I read anyway. I got the impression that the PWM approach (there are stacks of them for sale now as it's the latest thing in HHO gas nonsense) is preferred because it maintains torque while reducing fan speed. Would this be right? I didn't look into it that hard.

I assume the "low speed" resistor approach I am using reduces the fan motor's torque, but not enough to cause stalling concerns.

Not sure.

It's funny though, as minor as this problem is, I was fretting about it for ages because I couldn't work out how to fit a second, auxiliary fan. Then the penny dropped... I am a bit slow sometimes.

[DamTriton]

I believe SPAL do a two speed controller for fans, tweakable at both points for high and low speed (when full speed not needed), and interlinked to A/C for low speed.

I'll try to find out more info for you...

Edit: here it is

http://jaycorptech.com/showproduct.aspx?productid=155

[chimpboy]

I believe SPAL do a two speed controller for fans, tweakable at both points for high and low speed (when full speed not needed), and interlinked to A/C for low speed.

I'll try to find out more info for you...

Edit: here it is

http://jaycorptech.com/showproduct.aspx?productid=155

Thanks... I think that's kind of the gold option, but I am pretty happy with the free option as a bronze-standard way to do it.

[MightyMouse]

Beware of transient currents with PWM controllers - particularly on initial startup.

Its not unusual for a continuous rating of 20A to have to withstand 100A surges. Your design needs to be capable of handling this otherwise........

The sort of specs are normal for fan drivers - although 20A is the absolute max i've seen a fan draw, most are somewhat less.

[-Scott-]

I think if you try to go too slow with just a series resistor or voltage drop, you get stalling problems... that is something I read anyway. I got the impression that the PWM approach (there are stacks of them for sale now as it's the latest thing in HHO gas nonsense) is preferred because it maintains torque while reducing fan speed. Would this be right? I didn't look into it that hard.

I assume the "low speed" resistor approach I am using reduces the fan motor's torque, but not enough to cause stalling concerns.

Yes, low voltage will cause stalling problems, when you get to the point that motor torque is about the same as the start-up drag. Theoretically, each pulse from a PWM circuit will allow the motor to produce full start-up torque - but if the pulses become too narrow I would expect the same stalling problems, but probably at a lower DC equivalent voltage.

Yes, your resistor approach is reducing the voltage at the motor, which limits the current the motor can draw, which limits the torque it can produce - slower speed.