How hard would it be to make a circuit that decreases resistance by temperature?

Nazo

Nazo

2[H]4U
Joined
Apr 2, 2002
Messages
3,672
My dad wants to rig up a cooling system on his home theatre cabinet. Rather than just fans, what he'd really like is some kind of temperature based speed control. Something that could increase resistance to the fan (or fans, not sure yet) as temperatures go lower so they slow down, or maybe just a simple switch that changes to a different circuit by temperature. Since it has to go down as temperature goes up, I don't have any idea how to do something like that myself. If it's so complicated it's going to need a bunch of chips and such though it's just not worth the effort.
 
no need for a big complicated circuit. there are discrete components called thermistors (these are what is used below a CPU as a temperature reading). some increase resistance as temperature increases and some are the inverse.

here's a whole slew of them at digikey: http://www.digikey.com/scripts/DkSearch/dksus.dll?Criteria!!!!!!!794&Site=US&Cat=34014355

have fun.
 
The link didn't help much. I found two thermistors when I ran a search (couldn't tell what they did) but they are out of stock. For one thing, I need to know what kind can handle the sort of power that would run through. I have a fan controller I played around with on the old fan I had handy, and it started to smell after a short period of time. The fan I'm using (which I'm going to give to him if it's not too loud) is running 3.84W @12V, so it has to handle that and I think it has to run up to a max of 30K-ohm at, let's say 25C down to 0 at 32C or so (doesn't have to be exactly this, but, the closer, the better. Maybe a 25K-ohm max?) Any ideas what fits that criteria? Sorry, I just don't know how to read this stuff. BTW, there's a ton of them at Mouser, though I find their online server to be very inconvenient since it forces you to look at the details via a seperate very hard to read PDF file for each seperate one. I've never once worked with thermistors before so I'm afraid I don't know much about them or how they work.

EDIT: Thought it was 30K on the fan controller I tried. I guess it's actually 350 ohm? Don't know how I got the wrong scale (carelessness I suppose.)
 
One "easy" way to do it, is set up a 555 timer in astable mode at around 100Hz. Use a NTC thermistor, tied to Vcc, and a comparable sized resistor, tied to ground, as a voltage divider, the center of which is tied to pin 5 of the 555, control. This will provide about 50% duty cycle at 25C, increasing as the temperature does.

555ntcpwm.jpg


Edit: The value of C6 is such that this will NOT produce 100Hz, it should be increased. I just threw this together as an example.
 
Adrenaline said:
One "easy" way to do it, is set up a 555 timer in astable mode at around 100Hz. Use a NTC thermistor, tied to Vcc, and a comparable sized resistor, tied to ground, as a voltage divider, the center of which is tied to pin 5 of the 555, control. This will provide about 50% duty cycle at 25C, increasing as the temperature does.


Edit: The value of C6 is such that this will NOT produce 100Hz, it should be increased. I just threw this together as an example.
Click to expand...
are you suggesting making an astable multivibrate to generate a PWM signal to drive the fan? Fundamentally, that'd work but he specified he wanted it to be easy and 555 timer circuits are notorious for being picky and finicky about their C and R values. I figured he could just put a simply thermistor (probably in line with a few standard resistors and maybe a pot to control magnitude) and let that take care of it.
 
Yeah, kind of too much for such an amateur as myself. Anyway, I didn't know they had fans with builtin temperature controls. I knew there were some that could do it via software, but, not entirely independantly. The links vanilla_guerilla gave should pretty well get things done well enough for this job and if they end up costing more, not by very much at all considering how those components get overpriced.
 
vanilla_guerilla said:
there are many fans you can buy that have built in thermostats, auto controlling fan speed by temp. antec comes to mind, and probably thermaltake.
Click to expand...
This is what you're looking for. You'll spend less on the fan itself than you will on the components (plus shipping) for a nice PWM circuit. PWM would give you a little more flexibility to meet your exact requirements, but commercially-available TC fans are your best bet.

 
Oh, he already ordered the thermaltake 120mm listed there (the antec was probably just as good, maybe even a bit quieter, but, it had no details and was at a place he doesnt know compared to newegg which he does.) Ive already rigged up an adjustable adaptor with 12V to an old molex connector, so now as soon as I go down there next he should be set up with a case fan in minutes without even voiding the waranty or anything (I wont have to cut the fan wires to plug it up thanks to the extra molex plug I had on hand.)

It is too bad that I cant get something exactly to my requirements, but, it does occur to me that if all else fails, the fan probably just uses a thermistor or something like that to increase resistance, so I can probably just switch the plug to 9V or maybe even 7.5V and the fan at its highest and lowest settings will be quieter.
 
There is an easier circuit using one transistor and a few other bits. Have a look at the first circuit in this thread http://hardforum.com/showthread.php?t=1025895

Remove the led and R2 and put your fan in there instead. Make R1 about 1.5k ohm. Use a good size transistor rated at about 25W.

Get a thermistor and connect it from +12V to the "Uproc Pin" end of R1 (ignore the Uproc, that's only for his circuit). Connect a 100k pot between GND and "Uproc Pin" end of R1. Vary the pot till the fan speeds up enought when it gets hot.

Note: If the fan speeds up when it gets cold it means I'm being dyslexic again. Swap the thermistor and pot over.
 
You must log in or register to reply here.
Back
Top