better than reostats for fan speed control

[phantompunisher]

hey ive been looking for a decent reostat for a week around here and cant find any localy. then i saw a this post:

Originally posted by macklin01
Not necessarily. You can also combine a lower-wattage potentiometer with an NPN transistor to attain the same effect.

I made a fanbus based on the principle shown here

My implementation (which modifies the above design) can be found here.

Also note that you can increase capacity by using NPN transistors in parallel. -- Paul

i followed the links to the temp controled units and made some. they work realy well and were cheap. for hard core overclocking i would recomend the normal style without the temp sensers for max cooling control.

[dimmreaper]

I find that fixed voltage regulators(I like to use the beefy 5A parts) with a trimmer pot on the ground pin work well myself. The comparator in the regulator compares the voltage between the ground pin and the output pin, so changing the voltage at the ground pin forces the regulator to change the output voltage.

I'd like to build a step-up/step-down switching supply for 12V input with a 6-16V range, but can't seem to find the time to hunt the parts down.

[phantompunisher]

yea a switching powersupply would be awsome. but these smaller ones are so simple and effective. they bleed a little juice but if your using them the idea is quite the system down so .5-1.3 volts missing isnt goign to hurt you. post if you ever make the switcher. that would be nice and worthy of a fancy prodject box. my little circuts are just wrapped in electrical tape and duct taped to the inside of my case

[surlyjoe]

Originally posted by phantompunisher
. post if you ever make the switcher. that would be nice and worthy of a fancy prodject box. my little circuts are just wrapped in electrical tape and duct taped to the inside of my case

hmm, I thought that was fancy

[dimmreaper]

Well the point of using a switcher supply is not so much for efficiency, but rather it is the only way to my knowledge to make a step-up supply. I'd generate some high frequency pulsating DC, force it through a step-up transformer and end up with more voltage on the other side. From there I could either use a comparator to control the pulse width modulator, or I could just have the pulse widths fixed and step the voltage down with a linear setup for reduced components and less development time.

[Áedán]

The following will get you 16V from an input that's between 11 and 13v. It's designed to operate at 100KHz, and will require some heatsinking to get rid of the 2W or so of heat that the switcher will produce. It'll produce about 16W (1A@16V)

I *think* you should be able to adjust the feedback resistors to bring the voltage up or down.

There's quite a few components in this design, so I'll post a boost converter instead of a flyback one next.

[Áedán]

Boost converter. There's no way this will go below 12V due to the design. Again, this is a 16W output converter.

[Holst]

if you head over to www.bit-tech.net and look up macromans PWM fan controllers.
I think there the best you can get if a little more complex than a standard rheostat.

[Áedán]

I'd disagree with you there, but only for one reason.. Using something like a flyback regulator, you can get above 12V. In fact, you can go as high as you like, but I'd recommend staying with 16V as the maximum. Some of my PAPST fans quite clearly state that they'll operate at 16V, despite being designed for 12V.

AidanII

[dimmreaper]

Originally posted by AidanII
There's no way this will go below 12V due to the design.

Sure there is, you just have to lower the switching frequency enough so that the inductor can't pull the voltage up.

Those are some nice designs, though I'm looking at needing more like 1.5A. The two 120mm fans on my radiator draw .51A at 12V, they will draw more at 16V. I'd rather use a nice fat 5A MOSFET

Those designs may not be clean enough for fan overvolting, as while the RMS voltage would be 16V the ripple might be enough to trigger the overvoltage protection built in to many motor controllers. It's hard to say how much of an effect motor inductance would have on the ripple too. But if they are not clean enough it wouldn't take much to clean them up.

[dimmreaper]

Originally posted by AidanII
Some of my PAPST fans quite clearly state that they'll operate at 16V, despite being designed for 12V.

Panaflow 12v fans use a 24V controller IC . . .