Finished my system, well - almost.

[Z3RO_0]

Hey, I have a few questions for the experts on this, I am not sure exactly how much gas I am meant to put in for a static charge: I put in about 80 psi - any more and I would get frost right up to the compressor.

My capillary tube is *approximately* 2 meters, I just kept chopping some off it until it started working.

Now, this is a bad way to do it I think, so maybe Gary or one of the other phase change gods could tell me:


What are some normal running pressures for a small system like the ones we build
What pressure I should statically charge it to
A good capillary tube length for R290 and 150W of heat, or a capillary calculator/chart.


Thanks

[Redwolf]

Quote from teampuss
I would like to propose a rule for sizing cap tubes (single stage systems). Based on what I have been seeing in the forums, I would propose the following:

For R404A/R507:

Using a center point of 150 watts load I would guestimate 10 feet (3 M) of .028 inch (.7 mm) cap tube.

For each increase of 15 watts, I would shorten the cap tube by 1 foot (30 cm).

For each decrease of 15 watts, I would lengthen the cap tube by 1 foot (30 cm).

Then for R22/R290 (propane) shorten 1 foot (30 cm).

For .026 cap tube, multiply by .7.

For .031 cap tube, multiply by 1.6.

It is best to stay between 5 feet (1.5 M) and 16 feet (5 M), with center range of 10 feet (3 M) being ideal.

[Z3RO_0]

Ok, so 2m should be fine until I need to fine-tune it.

Now, what pressure should I charge it to? Just some rough numbers would be fine. :)

[Redwolf]

Depends...
Do you want to run in a vacuum or stay positive on the low side?

If Vacuum between 5 and 10 inches of hg. If positive about 1psig

[100]

Welcome Zero!

I would say (depending on your compressor volume), you should use 40psi of propane. you could top it up with an ethene gas if you are unhappy with temps. importance is not to overcharge the system, as you'll be getting a lower overall heat capacity in your system (i think?). I wouldn't head over 100psi personally from the sounds of things, but I lack experience with propane systems.

[Z3RO_0]

At the moment its running the low side at 12" hg and the high side about 100psi from memory but I'm not sure.

My K type thermal probe says -25 degrees C on the evap, which at this stage is just a 3meter ish length of 1/4" copper pipe.

The frost is only about 20" from the capillary tube, and the bottom of the compressor is not cold when its running, but its not hot either.

The static pressure is about 60psi when its not running and after it has equilized.

So, should I lower it to 40psi static and lengthen the capillary tube to get some better temps?

[Redwolf]

Oh so you dont even have a block on it or mounted to a chip.
I wouldn't worry about it till then.

[kayl]

what size compressor are you using, you will need a good compressor to handle 10HG (low side), about 1PSI would be better. my system with R290 has 70PSIG static charged.
under load mine runs 5PSIG and 145PSIG but each system is different.
also are you making a chillier or a DD
any pics, also have you insulated yet?

[Z3RO_0]

The compressor is a 1/4hp one, I'm making a direct die system.

I'm going to put my good evap on tomorrow and test it with a fake cpu heat load :)

I think i'll change the capillary to 3m and let some gas out untill it runs at about 10" hg vacume on the low side.

[Epsilon]

Like gary says:
Do not charge to low or high side pressure. But to best temperature under load.

3 meter of captube is fine, maybe a little long.

Fill the system to 30psi (?, not sure about this pressure) static pressure and then fire up the system. Load the evap.

Now put little gas in, in hushes. Let the system stabelise after each hush. Keep adding gas until the temperature stops dropping :).

[100]

zero what valve have you got, and where is it on your system?

what model is your compressor too?

p.s. nice project epsilon! i like the neat cut on the bottom of the prommy condensor!

[Z3RO_0]

I'll check the model tomorrow, I don't have the unit in my room at the moment.

I have access valves on the high and low side, the compressor has a strange process port that I haven't been able to use, so I am charging through the low side.

[100]

The 'process port' I **tink** maybe just a standard compressor service port, if it is brazed shut about 4 inches from the compressor that is probably it. You can charge your oil up into that tube.

[Gary Lloyd]

Forget about static pressure. This is something that is used in charging cascade low stage systems for safety reasons.

BTW, telling us the pressures is meaningless if we don't know what refrigerant you are using.

And do NOT add ethane.

[100]

He talks about propane, he's using propane?

Would ethene/ethane turn the system into a bomb?

[Z3RO_0]

Yes, I am using propane or R290. The service port on it is strange, its like a screw in connector but it has no core in it - I think you need a tool to use it. That is no problem though because I can charge fine through the suction.

I wouldn't add ethane, and if I was stupid enough to do that I don't have any anyway. :p


So, to charge it I should do it like so:

Add a small charge of R290 to 20psi static, then startup the system with my 100w heat load on the evaporator and continue adding gas until the temps won't drop any more. Giving it 10 min after each charge to stabilize.

[kayl]

or ya could static charge to 80PSIG start up (will pull down quicker)
then you can add/remove r290 as needed, 100PSIG will take ages on intimal pull down
you will find that if ya charge without a load wont need much gas at all.
Then when you slap it on a load temps swan dive and you need to add more refrigerant.

[Z3RO_0]

Thanks - I guess I had better make up that load tester :p

[100]

What kind of pelt will you be using ? I'm in the same situation as you as regards to precise load testing abilities (I use a manson power supply rigged to a marlow 100w pelt).

[Z3RO_0]

I am not going to use a pelt. It's a ciruit that loads a transistor about the size of an amd core.

[100]

You lost me. How will you load your system?

[Z3RO_0]

Originally posted by Z3RO_0
It's a ciruit that loads a transistor about the size of an amd core.

Umm, how do I explain this better: You know when you use a transistor to drive a load, it will get hot? This basicley does that - and makes a transistor get very hot. I would then mount this on my evap. :)

[kayl]

yes they do get hot.
i guess you could also use a overvoltage voltage regulator of some sort
apply like 24v to it and stick the collector onto the heatsink.
those babies get real hot.

[100]

zero ygpm

[Z3RO_0]

100: It is *NOT* a peltier, is is a TIP142 NPN Epitaxial Silicon Darlington Transistor, to be precise.