1/2" elect switch help?

[[XC] mysticmerlin]

I am making a chiller but it is for backup if the water cooling loop can't keep up. So what I am looking for is a silenoid that will switch the water to the chiller loop when the thermostat turns the chilling loop on. Anyone seen such a thing? I don't want to run it in line as I will loose most of the cool out the radiator. All help appreciated. Other idea welcome also.
Thanks, Scott

[Starkiller42]

I'm sure somebody's made one! But I'm afraid I personally don't know of it. You best bet would be to temperature control the chiller directly to the loop temp, and if the loop temp gets over x*C in the reservoir, then turn on the compressor, and turn it off once it reaches (x-n)*C.

If you don't mind me asking, why not just chill the whole loop? Even a relatively small compressor should keep a fairly large loop at or near ambient (ie better than your rad, so you could chuck it). If your compressor is a little bigger, and you're worried about condensation, you could use that same temperature controller and a big res to hold your loop temps to within a degree or two of your choosing.

But should you take up my first suggestion, I'd recommend this loop order:

res-pump-blocks-rad-HX-res

or to just have an in-res evap. Any path you go a large res is a good idea.

Hope that helps!

[[XC] mysticmerlin]

It will be on the 2 Q6600's in my sig. I have inline HX and it is for back-up if say the A/C in the house dies. It is a 1/16 compressor that I have and a TXV w/ Temp control thermostat.
I have 2 120x3 rads on it now. Running the 2 Q6600's a 7950Gx2 and one North chip set. They both run 24/7 100% load on WCG less game time.
K back on topic. I have seen mechanical switch valves on cloths washers but I can't seem to find anything I can adapt for my needs. And the 1/16 hp comp. will not hold that kind of load even @ ambient.

[[XC] 2long4u]

You could use a heater control valve but they are either manual or vacuum control.

[Starkiller42]

hmmmmm, your rads should hold your load without too much difficulty, but I definitely agree that the compressor is likely too small to completely cool the system.

That being said, I'd still recommend just temp-controlling the compressor and have an in-res HX; that way you don't need 2 pumps (one for regular, one for chilled). At that kinda load, with a lot of the heat being sent through the rads, the chiller system will just be focused on bringing the loop temp down to a reasonable level, so you won't lose any cool through the rad. Ie, if your at an ambient of 25, and your liquid at full load is running at 40, you set it to turn on at 39 and off at 35. That give your rads at least a delta of 10*C over ambient, but will keep your components from getting to toasty. And with a nice big res you could control that down to one or two degrees .

[[XC] mysticmerlin]

sry for the quick hack but so you get what I am after maybe

[[XC] 2long4u]

I wouldn't do it that way. Because the compressor is so small I'de run it through the radiator first before the HX. That way you are just adding to the cooling capacity you already have. Then you wouldn't even need a valve just turn it on when you need the extra cooling.

[DetroitAC]

A TXV with a 1/16hp compressor? Do you have a picture of the TXV? I have not seen one that small.

I'd suggest you always run coolant through the HX, but a pick a low restriction type, such as a commercial SLHX, they are designed to keep suction gas restriction very low, that should keep your water pressure losses low. Then you can control the refrigeration unit off a normal thermostat pressure control.

[Starkiller42]

I wouldn't do it that way. Because the compressor is so small I'de run it through the radiator first before the HX. That way you are just adding to the cooling capacity you already have. Then you wouldn't even need a valve just turn it on when you need the extra cooling.

Pretty much the solution I was trying to suggest earlier. Glad I'm not the only one who thinks that!

[Xeon th MG Pony]

SGHX would not work well as an evap due to the internal structer, unles you ran the liquid through the proper liquid ports and evaped inside the suction gas side how ever with doing so there will be no room for error as the slightest freeze up will rupture the exchanger!

A compresssor that tiny will not work as txvs start at 3,000 BTU and up.

[Starkiller42]

Well, the compressor would work, he just couldn't txv it, right?

[[XC] mysticmerlin]

I do have a 1/2hp Roto so I can still do it. This was to be a seperate system is why I was wanting to make it "switch" not run through it all the time. So I can play with the oc's or bench with it but still have it as a back up so I don't fry 2 quads because my wife gets cold or what ever.
Thanks all for you input so far, Keep it coming. Still sounds like a fun project even if I have to go back to cap tube.

[Starkiller42]

But you can still put it on a switch; just wire the switch in series with a controller, and you're good to go! Having an in-line HX or a in-res won't hurt you in any way when they are not active; the inline will just add restriction. If you can deal with that, then you have no worries! Go for it!

That rotary should bring you down to sub-zero temps. It would be more hardcore, but more of a pain to run it (noise & electricity). Let us know if you decide to go for it!

[fart_plume]

if you still want a switch valve look in a grainger catalog under hydralics, electrically controlled spool valve. you can get control voltages ranging from 12v to 440v units different flow rates and they have different input output sizes for what ever your needs are. Be warned though these valves are not cheap they are industrial units and made to last.