I have my whole rig apart, and therefore decided to collect some data.

In my wiring thread DW asked about my quick connects.

These are quick couplers that seal at both ends when seperated.
I installed them to allow easy access for filling and flushing.

There were 2 female ends on the case floor, 2 male ends on the reservoir and 2 male ends on the filling / flushing gear.



Here is the setup during a flush.



Here are the male ends on the res.

This system worked really well, but I always suspected severe flow restrictions due to the quick connects.

Here is the (very crude) data:

All other fittings are eliminated (no water block, no rad, no res).

With quick couples - 45 GPH
without quick couples - 90 GPH

The pimpy little couplings gave me a 50% restriction of flow!!!

I have now worked a new solution. This gives me all the advantages of the quick connects, but only a 12% reduction in flow (79 GPH):



The two high points are the male quick connect pieces. During normal operations the water flow bypasses the quick connects (they are essentially air traps).

When it is time to fill or flush, I just hook the female ends (on the fill / flush gear) to the waiting male ends, turn that sexy little brass tap........and I have the same access as before!

That whole contraption imaged above will be mounted on the back side of my mobo tray.

Nice work Putwig. I like it, I didn't want to give up the flow rate for them, but they are a fine piece of gear. Nice thinking on it. I love simple solutions.

Hey putwig thats all cool and stuff but whats with the SODA bottle(pepsi bottle?) as a resevoir or whatever.

If I'm not mistaken, that's a dubro fuel tank. Got it's little brother on my RC heli.

AntiM is correct. The second pic shows the res, and it is made out of a Dubro fuel tank for RC aircraft.

The pepsi bottle to the left of the first pic is part of my filling and flushing set up. Other then fluid change time, it is hidden in the closet.

Is water cooling similar to a car in that you have to limit the flow through the radiator to give it sufficient time to cool off? By this I mean if you pull the thermostat out of a car it can overheat because the coolant goes through the radiator too quickly to cool off.

A thermostat in a car is designed to let the engine warm up as quickly as possible by limiting coolant flow. Overheating is generally caused by a stuck thermostat...not by removing it.

But most radiators used for cooling cpu's do have a "sweet spot" at a relatively low flow rate. This doesn't mean you should run at a very low flow rate. A block generally works better at removing heat at a faster flow rate than is ideal for a radiator. The best flow rate is a compromise of the two. The block being the more important component.

I do know what I am talking about with cars as I am a VW and Audi master technician with over 20 years of experience. It may not overheat on a cold day, but put it under load on a hot day and it probably will overheat. You are half right in the job of a thermostat but it also does restrict flow compared to having it removed. If the coolant does not spend enough time in the radiator it will not cool enough.
As you said there is a sweet spot for the radiator at a low flow rate. I believe this confirms what I thought. As you also said too low is not good for the block so a compromise must be reached. It probably isn't fair compairing a car to a computer because you want the engine in a car to run at a certain temp whereas you want the cpu to be as cool as possible.

Didn't mean to offend dude. But it did sound like a totally n00b question. I've lived in the desert all my life. Generally there is one overwhelming reason a car over heats. A stuck thermostat. The easiest quick fix is to remove it. I had a '69 Chevelle that actually ran cooler without a thermostat. Doesn't mean I didn't replace it when I got home. It's a safe assumption that more modern car designers pay closer attention to flow rate vs. heat dissapation in their design. They would naturally use just a big enough radiator to do the job...it's cheaper that way. Where I live cars are fitted with larger radiators than you would get with a new car elsewhere. That gives you an idea of how much thought goes into cutting costs nowadays.

You're absolutely right. A car's cooling system is not the same as a comps. They both have a closed loop system maintaining equal pressure and flow rate throughout the system. They are both water based. Radiator...pumps...hoses. After that the similarities fall apart.

The point is...the smaller the radiator used. The more critical flow rate becomes. Too slow and it can't move heat from the block efficiently enough. Too fast and the radiator can't dissapate the heat efficiently enough.

So simply bolt a '69 Chevelle radiator to the side of your case and you won't be able to pump water through it to fast to cool well!

No offense taken. Just trying to explain that I knew about cars. I am a n00b when it come to water cooling a comp. I was just wondering about the similarities.