TEC Cooling Project - Need help with my electronics =]

[killermouze]

Hi everyone, figured I'd start this thread over here to double my chances of solving my electrical pwm motherboard problem.

I've been posting on overclock.net and 4Ryan6 has done a great job of helping me out.
http://www.overclock.net/t/1521325/water-air-combo-peltier-cooler-above-ambient#post_23173180

Anyways, my question involves using the signal from an x99 asus deluxe pwm (cpu_opt) as a signal for a power amplifier to adjust the wattage going to a thermal electric:
http://www.accuthermo.com/products_5.asp?pid=12420101032578117703&pname=FTX700D+H-Bridge+Amplifier+for+TEC+Temperature+Controller+Systems

Do you think that would work where I could control the power via my asus suite by hooking the amplifier's pwm leads to the motherboards pwm? I know the asus suite can be tricked into thinking some pumps are fans.

Thanks!

Mouze

 

[4Ryan6]

Hi Welcome to Toms Hardware Forums!

I just wanted to say I'll be happy to help you all I can, however I only share information that I am 100% positive about, I don't get into the hypothetical or assumptive type of questions that I have not personally tested and the answer is not solid.

So when I jump in to help it is with tried and tested information that I am 100% confident giving you.

Since I am not using the type of electrical control you're asking about I'll just keep an eye on the thread until I have something to add regarding it.

I opted not to use any type of voltage regulation after discovering some of the peltier peculiarities, meaning the things that can cause a peltier to stall such as the cold side getting too cold and having a chilling effect on the top side which will stall the peltier, or vice versa, the hot side getting too hot and warming the cold side resulting in stalling out the peltiers performance.

Don't forget the peltier is only about 1/8th inch thick, hot/cold temperatures do seriously affect the peltiers output.

I had discovered with there being a need for a hot/cold balance to keep the peltier outputting the needed cold, that voltage control could possibly offset that balance and stall the peltier.

It's kind of like once you find something in your experimentation that works you go with it, and that's what I did, even though along the experimentation trail, many had suggested I run some type of voltage control over the peltiers.

I assume their intentions were to save electricity, but really if you're going to run a peltier at all, especially for higher overclock purposes, saving electricity is kinda already out the window.

Once again, Welcome! Ryan

 

[4Ryan6]

Quoted from your thread at OCN.

I saw your post with the single 12v rail with 2/3rds out, and the pci 6 pin. That's no problem. I have an excellent EVGA SuperNOVA 1200 P2 Power Supply. What I'm considering is whether it is possible to hook up the TEC to it and run all my other components from it. Currently, my other components use 650ish during stress, so with a 150-200 watt TEC, I should be fine, but I am worried about how the TEC actually will draw power. Will it just suck up every single watt it can until it basically destroys itself, or does it actually limit itself to its rating? Do I need some type of power regulator between the TEC and power supply?

Someone actually tried to run the whole thing off one power supply and it failed!

I suggest 2 power supplies one for the computer itself which can be a multiple 12v rail and control your pumps and fans with it as well, but the peltier power supply needs to be a single 12v rail power supply, as the peltiers draw their constant amperage load immediately upon powering the power supply.