SPCR

Hi all,

I want to make something like this
http://www.modthebox.com/review34_1.html

in the simple way, it's safe just to use potentiometer without using any additional component like capacitor or the ic?

thanks

amenxyz wrote:in the simple way, it's safe just to use potentiometer without using any additional component like capacitor or the ic?

No. Get a Zalman Fanmate1.

yes, it's good to buy fan controller like zalman. but i want to build it simple and from whatever component i got.
here's another link i found
http://bit-tech.net/article/56/
still my question is, what's the different between this and using additional component? I just find out the ic is an adjustable regulator (LM317)
http://www.national.com/pf/LM/LM317.html

oh, and there's additional faq at Cliff's FanBus
http://www.fanbus.com/faq/faqredirect.html

A rheostat can handle much more power than a pot, but has to be able to dissipate whatever the max the draw. It is usually more expensive than a voltage controller, most of which use the LM317. The latter usually does not allow 12V to be passed, as the circuit itself takes bit of power. Usualyy ~11V is the max -- about the same as 12V in real-world apps. The main advantage of a voltage controller is that it is independent of the load -- ie, use any fan -- and often cheaper in parts cost.

Hi MikeC & others,

When push comes to shove, a $10 to $15 fan controller sounds nice. I am not familiar with the Zalman Fanmate, but I did review the Computuning fan controller @ http://www.modthebox.com/review34_1.html
and find it would do a reasonable job.

So the Computuning control knob is ugly. Put it on the rear panel - no-one will see it. No center position - oh well. Heatsink could be too small - heatsink for how many fans? Unshielded bottom PCB fan controller circuit
- fix with some insulating paper.

As a techy, I have built up quite a few circuits and a fan controller would be a breeze (excuse the pun). One transistor, a trimpot, a handfull of other components, a printed circuit design and power connectors, I could design one that would rival most controllers on the market.

Having said that, I doubt if I could build one and market for less than the $10 to $15 fan controller mentioned above. The LM317 does the job most admirably.

My 2-bits worth.

TerryW

MikeC wrote:most of which use the LM317. The latter usually does not allow 12V to be passed, as the circuit itself takes bit of power. Usualyy ~11V is the max -- about the same as 12V in real-world apps.

Mike, LM317 "drop-out voltage" is 1.5-2.25V typ. (depending on output current and temperature), therefore I wouldn't suggest this regulator when maximum airflow is required and the fan draws more than ~100mA. You may end up with only 10-10.5V output that is probably too less in some cases.

A low drop-out regulator (like LM2941) would be better. It only needs a larger output capacitor and, of course, costs slightly more than a LM317. That is why it isn't used in cheap commercial controllers.

As a perfectionist, I prefer the closed-loop (temp. controlled) solution. Well, we can't beat commercially available temp. controllers (economics of scale), but for those interested in building their own, I have designed and operated one. If SPCR is interested, I could post an article about it. It isn't that useful as it seems (as I said, similar products from China have a lcd display, might be temptingly cheap and need much less hand work) but it offers the pleasure of creativity

What do you think?

This looks very interesting, although it doesn't set the fanspeed, just turns them on at specified temps. Has lots of other nifties too. Sorry folks, tried and tried but can't find the manufacturer.

ATrueReview BC-9901

BC-9901 Computer Temperature Controller <- for sale at ntcw Vancouver BC

I have just built one of these: http://www.cpemma.co.uk/ef.html

Works very well and is cheap. Since i'm only going to use it with panaflo fans (2-3), i think a heatsink on the PNP-transistor would be overkill.

Another way to go is the NoiseMagic controller sold by Silicon Acoustics http://www.siliconacoustics.com/nmt2.html. This controller is designed to mount on the fan housing. But it is easy to modify. In my case I have put standard 3 pin leads on it and put the thermister on a longer wire lead. Eventualy I intend to use this to install a thermal controlled fan on my CPU. I have not got that far yet but I have tested this unit with my mods and it is working. One nice feature of this controller is that it applies full votage to the fan for about 3 seconds at startup. So you don't have startup problems like you might have with other controllers. At $13.00 it is worth having a look.

Hal

The noisemagic and zalman are all good and dandy but the DIY in the first post goes down to 3v - something I desire. How hard is it to assemble and use once the kit arrives?

-Ken

the DIY in the first post goes down to 3v -

GXcad, looks like it's preassembled, but I wonder if the comment about consistent voltage startup at 3V is for real. In general, I have found that ~4.3V is about where Panaflo 80s stop starting consistently. I think for a fan to start consistently at 3V it has to be a much higher power device (read: louder, more airflow) than a Panaflo L.