Peltier/Water Overclocking

[jagojago12]

Originally posted by Pyro
Yes, always apply AS3 on both sides of the TEC. So it would be like this:
waterblock-->AS3-->peltier-->AS3-->coldplate-->AS3-->CPU


Also, after reading some of those overclocking forums posts, some people say I need to put a load on the 5V rail to actually get the 12V rail working consistently. When I bought the Antec TruePower, it says that it has dedicated circuitry for each rail, so it can achieve max power output (or something like that) reliably. Does this hold true?

I too have read about loading the 5v rail. I'm not quite sure. I'd post at oc-forums and see what they have to say. I'll do some research myself and see what I can come up with then post it here.

Thanks a bunch man . This is some really good info, and it's helping me out a great deal .

[number1sixerfan]

I wouldn't look into a nb waterblock, because the heat from the TEC may warm the waterblock instead of cooling it. A crystal orb and as3 would suffice.

[Pyro]

No problem. I'm glad I could help out.

I'm glad you provided that link about the power supply mod. I'm going to use that on my Antec 400 watt PSU to power a the Swiftech's 80watt TEC and the monster 120mm fans. That should keep my 12v on my Enermax 431 watt running flawlessly.

Did you add the switch to the PSU, or do you simply unplug the power cable. It looks like I may have to do alittle soldering to connect the leds to the switch.

If you did use a switch, what colored wires did you connect?

[jagojago12]

We did it slightly different when putting the power supplies inside the case.

We took some radio shack copper wires from the Enermax's green (power on) and black (ground) cables and connected it onto the Antec's green and black cables on the ATX12V connector. So when we press the power button on the case both power supplies are turned on simultaneously. I guess that is an alternative to the "switch" thing that website was talking about. It may not potentially be as flexible but is an easier modification I think . Once the power to the motherboard is turned off both power supplies are turned off as well since both of them are now sharing the same power on current signal.

[medunaweb]

yeah all you need to do is strip back some of the jacket on the green wire going to the mobo on both PSU connect those with a wire. Then do the same with one black wire that way they will share the same grnd with the motherboard for a circuit. The green wire is the rst on the schematics which is the power on/off line. I did that same mod in my case too it works great. Very simple just two jumpers needed. I am doing this with jagojago by the way and if i remember correctly the peltier is clamped down by hex screws so we should be able to get a good deal of torque on those screws.

[Pyro]

jagojago12 and medunaweb, thanks for providing the tips and links. Its helped out alot. Guys, good luck with your peltier overclocking adventure.

[jagojago12]

If anyone knows, we are using a CPUfx pump that is similar to this except instead of using a 145GPH pump, ours has 500GPH.

Link: http://www.cpufx.com/Merchant2/merch...tegory_Code=R1

When we were testing the water pump/resoiver by laying it to the side (horizontal instead of vertical) the pump's noise decreased a lot and looked like the water that was being pumped has slowed down greatly. We guessed that the water was moving slower because we still had a few tiny bubbles of air left in the system and the bubbles moved slower when the pump was laid on its side than vertically.

I don't know if this is some basic law of thermodynamics that we're breaking or if it's just not pumping water as fast (or even if it's pumping) or maybe it is pumping just as fast and we just don't realize it.

[Big_Mac]

Originally posted by jagojago12
If anyone knows, we are using a CPUfx pump that is similar to this except instead of using a 145GPH pump, ours has 500GPH.

Link: http://www.cpufx.com/Merchant2/merch...tegory_Code=R1

When we were testing the water pump/resoiver by laying it to the side (horizontal instead of vertical) the pump's noise decreased a lot and looked like the water that was being pumped has slowed down greatly. We guessed that the water was moving slower because we still had a few tiny bubbles of air left in the system and the bubbles moved slower when the pump was laid on its side than vertically.

I don't know if this is some basic law of thermodynamics that we're breaking or if it's just not pumping water as fast (or even if it's pumping) or maybe it is pumping just as fast and we just don't realize it.

To test your flow rates, grab 2 buckets, fill one with water, and time how long it takes to transfer the water from one to the other.

Then change the orientation of the pump, and time it again.

That's the easiest way to compare flow rates.