Massive Water Chiller

[Xeon th MG Pony]

All so a bit late to point out but the sensors will not at all be effected by oil being pushed into them, and at a 90 degree angle little oil would be pushed into the sensing tube, often the line will be brazed right into the pipe with no concern for any sort of riser. with that said it is all ways wise to put a stub tube in for futur repairibility.

[Stewie007]

Well, the original design was more the issue. I had it blowing straight through a tee with the sensor on the opposite end and the suction port on the branch.

When I re-thought it.... I over-thought it. But either way it works... hehe.

[Stewie007]

Little bit of work tonight. What I really wanted to do was get the check valve in place so that I could start tomorrow at the same point that I left the previous week. I had a dilemma though. The valve was too long to simply splice into place. I had to sweat a tube and braze the stub of the valve into the elbow. That meant I had to use phoscopper. Lucky for me the new monster torch tip made for easy heating and no issues arose.

Once the valve was sealed I tested the system for differential. At first I made the mistake of pressurizing the high size opposite the check valve. As a result, the initial pressurization of the low side blew pressure through the compressor into the high side behind the check valve. Once I bled all pressure and reconnected to the discharge service port and pressurized it was a different story. I pressurized and equalized to 200PSI and then pressurized to 50 PSI on the low side. No bleed through occurred. Shot one is initial. Shot 2 is after 15 minutes.

What's this mean? I gotta finish the water piping and do the electrical work. The issue I'm running into is that my junction box is pretty big. The only place it seems to fit is along side the condenser reservoir that now would sit behind it. That isn't really a problem, but it changes the design a bit. Hopefully tomorrow goes well.

[Stewie007]

Continuing from yesterday... The biggest part I got done today was the shroud coil. I had fewer tubes than I wanted, but I think it will work out alright. All it really has to do is build up a good frost inside the shroud to provide somewhat cool air. I don't want to condense TOO much water. I gotta find a way to drain it, though. Its pretty much as I designed it, save a few details. I cut some holes in the shroud for the tubing. Each has a flared end. I have to redo the reservoir inlet, however. It "conflicts" with the pressure sensor.

[Stewie007]

I then insulated the chest. Not that it was a big pain, but I wish I had some properly formed insulation pieces for a hotel pan. Would have been nice. :p It works out fine though.

[Stewie007]

I installed the reservoir to the suction inlet. Its a little off center, and I didn't take any pics. Needless to say I've got to do some testing. Aside from that I have to clean out the evaporator of all the junk before filling it. I'll have to get a siphon after I do that to get the water out of the heat exchangerVacuuming the system down again. This time I'm using both pump ports.

[Stewie007]

Finally, this is what happens when you do something stupid.... like use a hole saw, with your hand in its kick back range.... without a glove on!! I didn't take a shot of the cut, but luckily it was a somewhat minor surface abrasion on the back side of the hand next to the thumb. It bled until I cleaned it with Hydrogen-Peroxide... Sizzled and then stopped.. :-p Not smart...

[Buckeye]

Ouch man !!!!
Remember safety first !!!!

Glad your ok

[Crabby]

Dang, sorry about your hand man! Hey what is the "shroud coil" thing for?

[Stewie007]

Its intent is to cool the air passing through the radiators. The water won't be flowing through constantly. Instead two solenoids will control the flow for a temporary bypass to cool the coil down and then allow the frost to do the cooling. One solenoid is normally open and the other is normally closed. Thus, they will both be energized to bypass only. I gotta remember to put a riser after the coil to prevent it from draining even a little bit after the solenoid's close.

On a side note, no work will be done next week. I will be on call for work, so I won't be able to do any work.

[Stewie007]

I've been thinking. I've got to control the condenser side with the compressor shut off. I was considering using the same control that the solenoid uses (thermostat) to control the condenser pump and fans. Does anyone know where I can find the appropriate distribution block for the additional wires. I'll be splitting one source into three devices.

I know what I could use, but I'm not sure where to find them. I'm going to look later tonight, but since I'm on call for work I need to be as efficient as possible. So any help would be appreciated.

[Stewie007]

You'll have to forgive me for the size.

I've created a diagram for the wiring of the compressor system. The Low Pressure switch has been designated as the control for the compressor. However, for a short period at start-up it will also control the oil solenoid. I will use a timing relay that has a "Delay on Make" setting so that when the low pressure switch activates the compressor, the relay will also be fed power, and after a set period will turn on for several seconds to open the solenoid. Then the contacts will open and the solenoid will de-energize (thus closing). It will literally be a fraction of an am load, so it won't matter much.

I am going to try fitting all of the blocks inside the junction box.

I am working on the second diagram.

[Stewie007]

Control set # 2 is set up to control several devices. This will be set to a dual pole relay. A switch will control the Chill Pump & Computer circuit. When that energizes, the thermostat is given power. Saved settings are in memory. This entire circuit is isolated from the compressor circuit. I will need a 20A circuit for the computer and condenser equipment.

This is how it works. The Thermostat controls the condenser side. When the contacts close, it activates the condenser water pump, the radiator fans, and the high side solenoid. The cold coil solenoids are connected to a relay that runs and then repeats. It will only do so when there is demand for cooling and the condenser system is energized. When the contacts open, each is deactivated and it returns to a power down state. The computer continues to run with the chilled pump. The compressor system turns off on its own given the pump down activation.

Should I deactivate power to the computer, the thermostat will not be able to start. The system will be off completely. This way, I cannot turn the computer on without activating the system. The idea is that if I for whatever reason instinctively walk into the room and touch the computer power button nothing will happen without at least the chilled pump running at the very least. And if the chilled pump is running, that means the thermostat can activate the system.

[Stewie007]

Hoping to have my relays by the end of the week. I've got a better picture of the controls now.

I'm going with a 120V SPST switch to control the output side which will be wired to the accessories as shown above. The terminal blocks on the chill pump and computer side will move to the opposite side of the relay such that the coil is a part of that circuit, but not in series with the loads. The coil doesn't control anything on its own, kind of my mistake when drawing. The one that I listed is an Omron G7L-1A-TUBJ-CB100/120 or similar.

The second relay is an Interval Time Delay Relay. This one will control the oil solenoid. It will energize when the system starts up (Compressor) as stated above. It will disconnect power to the solenoid when the count down expires to reset upon the next energize sequence. Part is a Dayton 1EGC8 or similar.

The next one is a repeat cycle timer. Macromatic TD-88162 or similar. Its pretty much as stated. On for X Seconds/Minutes, off for X seconds/minutes... on...off... etc etc... Water solenoids are wired to it.

All in all, should be able to wire it up soon. Just gotta find a drain pan for the cold coil condensation. Any suggestions? Gotta be 15 x 15 and I'm having trouble finding that hehe

http://www.mouser.com/ProductDetail/...osryxXgCxrI%3d
http://www.macromatic.com/search/pro...oduct=TD-88162
http://www.grainger.com/Grainger/DAY...GC8?Pid=search

[Stewie007]

More work, and getting closed to wiring. I wish I had more time to work on the thing... I'd be finished by now... but it is what it is eh?

Condenser reservoir is all connected and in place. I'm quite pleased with the way that it worked out. Its nice and snug. I'm also much more pleased with the braze quality on this one over the last one. I was able to use the hole saw to achieve a much better fit. The slip fitting does nicely.




Moving on, I've completed some work on the water inlets for the chilled side. The coil outlet is protected from draining by a vertical run such that the cold water will remain after the solenoid valves revert to the normal flow. I'm going to place the solenoid valves on the outside of the system. Frankly, its getting so cramped inside that there isn't a heck of a lot of room for more parts. Thankfully, I'm almost finished with piping.





I started with the soft tubing. Unfortunately, my original design produced too much tension with this stuff so I elected to go another way. I'll alternate flow between the smaller and larger radiators.

[Stewie007]

That's it so far. I've got only some piping to run and I'll be wiring it soon. The base of the chest is the next thing after wiring that I'll work on so that I can get the piping done before testing. I won't be building the chest until the system is tested vigorously.

I've gotten all of my relays set as appropriate. Just gotta figure out how to set up the junction box.

[bwanaaa]

Finally,

............

Am I stupid or just befuddled. Why are there two separate radiators in this thing.?

A regular refrigerator has a compressor->radiator (to cool the compressed refrigerant)->capillary tube->evaporator (where the liquid refrigerant becomes a gas and in the process absorbs a ton of heat)->back to compressor.

So the evaporator is where the coldness happens. That's what usually is on the cpu no?

Unless you are designing this like a PWR fission reactor with two separate loops-using the evaporator to cool a separate water cooling loop for the cpu. But since the 'core' is not radioactive, I dont see the benefit.

[Stewie007]

I would be happy to answer all of your questions..... When you decide to be a bit less rude.

[limpkorn]

I may be wrong and if I am I would love to hear an explanation as well so I can better understand your setup stewie.

It seems like you have the heat exchangers with the water loop acting almost as if they were cooling towers correct? You have the chiller barrel and the condensing barrel where the water that is cooled off by the heat exchangers goes in tot he condensing barrel to condense the high pressure gas back to liquid to make a trip to the chiller barrel to cool off the glycol solution.

Is that the jist of it or am I off?

Thanks,

p.s. I was polite

[Stewie007]

I may be wrong and if I am I would love to hear an explanation as well so I can better understand your setup stewie.

It seems like you have the heat exchangers with the water loop acting almost as if they were cooling towers correct? You have the chiller barrel and the condensing barrel where the water that is cooled off by the heat exchangers goes in tot he condensing barrel to condense the high pressure gas back to liquid to make a trip to the chiller barrel to cool off the glycol solution.

Is that the jist of it or am I off?

Thanks,

p.s. I was polite

They do indeed act as cooling towers.

There is a temporary exchange of energy between the chilled loop and condenser loop, though. The coil inside the shroud receives flow when the solenoids at the inlets are energized and water is redirected through the coil for a set time. That gives the radiators a bit of help and hopefully makes the discharge air from the chiller be a bit cooler than it otherwise would be. It's set at an interval that attempts to keep it frosty, but not a block of ice.... I've gotta make a drain pan for that, though.

I'll give a better explanation when I get it up and running. I'll probably just make a video.... hehe.

[limpkorn]

Can't wait to see it brother thanks for the explanation!

[Stewie007]

I'm quite eager myself. I have very little left to do but connect all of the water piping and wire it. Although, the wiring is a project in itself... Gotta make sure I check my work very carefully.....

[limpkorn]

I'm quite eager myself. I have very little left to do but connect all of the water piping and wire it. Although, the wiring is a project in itself... Gotta make sure I check my work very carefully.....

Tell me about it. I'm not as experienced as you are as you added quite a few controls to your setup, but even wiring my simple water chiller took some thought and double checking to make sure I wasn't going to fry anything.

[Stewie007]

I added too many controls..... mostly because of the water cooling option and oil separator. The pump down and oil separator were suggestions from some folks on here. The additional timers and relays were a necessity given the design of the chilled loop and the fact that I wanted my set-up to be fool proof.

At the very least, no computer power unless the pump is on.

[Daddeo2]

Great looking project Stewie! It's quite obvious you've put a lot of time and effort into it. I can't wait to see the results once that bad boy is tuned.