Using a cube fridge as a supplemental heat exhanger for WC

[Boilerhead]

I have a 1.4/266 berd on the way to use as a dyno-mule for testing the new system's capabilities so I am curious as to temps vs. performance and where you hit the wall of diminishing returns within the system.

It will be wc with a Koolance (arrived at my office in Buffalo today) so I really -shouldn't- have problems with higher temps, but I am curious as to the actual performance benefits of further temp lowering.

I have a cube fridge which isn't doing anything atm. I may decide to add a second radiator to the system inside the fridge to further lower the coolant temps if this will allow me to oc further.

It seems to me that it would be fairly simple to run the coolant line from the Koolance rad outlet into the fridge, through another rad and then out to the CPU block. I can vary the coolant temp with fridge temp, pump velocity, in-fridge rad fan speed or in-fridge radiator size if I need to/want to and I should at least be able to signifigantly lower the coolant temp.

It's rated at 191 BTU/hr and in tandem with the in-case Koolance rad I would think it should be capable of dropping the coolant temps 10 to 20c depending on how I do the heat exchanger setup.

I have the spare fridge and the room to screw around so it seems to me that for the price of nice heater core, some tubing and clamps and the effort of punching a couple of holes in this fridge this project is too easy -not to do-

In the summer the humidity around here can get pretty high, when do I have to start worrying about condensation in the case? Should I plan on an insulation program now or possibly wind tunneling to accelerate evaporation? I know that a lot of people turn to wc for decible reasons, but frankly I could care less. After years of working in an equipment room full of screaming fans that noise doesn't bother me in the least, hell it even makes the crappier MP3's sound better :P

Anyone have an idea how low I might get coolant temps with this type of setup? Are sub-zero temps possible with this with some farting around and extra insulation all over the place including in the fridge? I can even make a cooling box out of closed-cell foam board (I have 1", 2" and 4" thick boards for my model RR scenery) inside the fridge to enclose the freezer plate and heat exchanger to minimize losses.

I have also been wondering about coolant velocities with the Koolance system. I am curious as how close to optimum their dual pump setup is. I've got blocks for everything so that should slow the coolant down a bit from a CPU block only system. Does anyone know or has experimented with increasing pump volume and/or velocity on a Koolance? I would think that there would be an advantage to having a high volume variable speed pump setup to fine tune the heat transfer abilities. There's a finite point past which coolant which is a certain temperature will move through a block too fast to draw off the maximum amount of heat and it seems to me that finding that point should be a wc project goal. That would also apply to the radiator end of the system which it seems to me should also be tuned for maximum heat transfer at a given velocity.

I have a DigiDoc5 on the way so I can monitor coolant temps on either side of the heat exchangers to see what's going on.

I should have the Koolance in my hands here in Port Clinton on Friday or Sat so I can lap the blocks and start proofing the integrity of the stock system and maybe do some crude heat transfer experiments with artifical heat sources.

I'll post my progress and observations as I go along as well as pics of anything notable.

Any opinions on the cube fridge concepts? Seems like a good cheap OC enhancement to me and full of potential for tweaking stuff around.

Henry

[Tiger]

I have a water chiller under test that is basically of this kind of concept.
Regarding enhanced air cooling using a second rad in the fridge. It is possible but I'll put some ideas forward for you to ponder. If you take the secondary air cooling route with a rad you are faced with the facts that this route requires 3 heat exchanges. Water/rad then rad/air then air/evaporator. You know that air is not as good a conductor of heat as water. The cooling effect therefore is not as effective as water itself.
Using the direct route of cooling the water directly using the evaporator is a more effective route.
I am using a 110W compressor based system and do not know its cooling capacity but I do know that the Duron I am ocing is producing 70W. With the current configuration the system is just short of the kind of cooling I need in order to hold proc temps constantly in sub zero conditions. The kind of specs that I think would be needed for your rig would be in the order of 200W compressor. This kind of power is generally not found in fridge/freezers. The kind of figures you need for this kind of heat exchange are in the order of 2000BTU/hr which are typically produced by small air conditioning plants.
So from what I have said I don't think in current form it will produce sustainable results.
If you follow the route it is possible to attain low temps with coolant running sub-0 and CPU temps in the sub-10 range and it is very quiet. The temps would creep up probably at the rate of 1C per hour. My system will cool things at the rate of 3C/hr when the system is switched off
The next possibility is running direct phase change on the proc itself by making a new evaporator that can be attached to the proc.
The venture into the sub-zero arena brings with it all the condensation prevention requirements as well.
From an ocing point of view I have gained an extra 75MHz which equates to a 10% gain in performance using this route. That does not seem a lot but the quietness of the system is probably worth every penny I spent on it, £15 all in all, which is what I paid for my FOP-32-1.

[Boilerhead]

I'd like to design some sort of interface which will maximize the heat transfer between the coolant and the freezer plate. My latest idea is a labrynth box which cycles the coolant back and forth over the top of it and then back again on the bottom of it.

I know sub-zero would asking too much, but its eems to me that using this as a secondary cooling system in addition to the existing Koolance II radiator I ought to be able to reap signifigant benefits from it.

To what extent I will have to play the insulation game has yet to be determined.

Henry