Autocascade

[XC] gomeler · 07-01-2006, 11:36 AM

I will admit that in regards to phase cooling I am still quite the nub, single stages I can do, but that is about it. I also know there are tons of other people just like me, so to satisfy all our un-asked questions, I'm going to start and hopefully compile all of this stuff. I may end up jumping into the autocascade competition if I feel like I can cut it with all the pro's, we will see. So with that said..questions, be kind

1) For an autocascade what would be the minimium HP/watts required to cascade C02 with r290 under a decent load for a modern processor? 1/3 HP, 1/2HP?

2) Since r290 carries mineral oil so well and CO2 doesn't, besides a phase seperator do I need to do anything to seperate the oil from the CO2?

3) Within a phase seperator should there be any form of baffles such as steel wool to slow down the gasses and seperate them?

4) Will I run into issues with the CO2 forming dry ice within the evap/suction line and if so what are the proper ways to go about avoiding this?

I am sure I will have more questions than this as I go through preparing for this. I know people might

about me not extensively using the search bar, but the more clear and precise answers I get, the easier it'll be for me to build one of these, document it, and make something permanent for this forum.

[486] · 07-01-2006, 11:54 AM

i also have a question that doesnt deserve a new thread
can argon be used as high side gas or too high pressure?
and if argon cant be used can r23 be bought in small [5 pound] bottles in the US

1.ive seen 1/2 HP autocascades
2. i think you just can use the phase sep as an oil sep too
3.no baffles in the one im building [or any others that i know of]
4.people get dice in their metering device but maybe putting a filter dryer after the h/x might help

edit: was looking at the 4 stage cascade thread and argon is 5 th stage gas

Metaxas · 07-01-2006, 12:38 PM

Man I wish you could get gases in smaller cylinders...the smallest I have been able to find of any gases (except 134a) is 20lbs.

They are EXPENSIVE too...oh well, I guess its the price that we pay for this hobby...

n00b 0f l337 · 07-01-2006, 12:41 PM

Quote:

1) For an autocascade what would be the minimium HP/watts required to cascade C02 with r290 under a decent load for a modern processor? 1/3 HP, 1/2HP?

I think technically its not our hardware, its our tuning right now. 1/3hp-1/2hp techinically should work, but I don't think many have done 1/2hp except for with rotary's and then again, most havent done even a 1hp autocascade.

Quote:

2) Since r290 carries mineral oil so well and CO2 doesn't, besides a phase seperator do I need to do anything to seperate the oil from the CO2?

Oil will seperate (its a liquid) in the phase seperator.

Quote:

3) Within a phase seperator should there be any form of baffles such as steel wool to slow down the gasses and seperate them?

Copper wool, not steel. And yes thats what we use. :P

Quote:

4) Will I run into issues with the CO2 forming dry ice within the evap/suction line and if so what are the proper ways to go about avoiding this?

Oh yes you will. Proper way? Large capillary line (ID wise), or other metering device, and look at a PT chart and tune it away from the solid phase.

n00b 0f l337 · 07-01-2006, 12:42 PM

Quote:

can argon be used as high side gas or too high pressure?

Your confusing what you read in the s7' autocascade thread. The argon is wayyyyyyyyyyyyyyyyyy to far from condensing, go look at its critical pressure/temperature.

Quote:

and if argon cant be used can r23 be bought in small [5 pound] bottles in the US

Generally no, but then again, go to a Air-Liquid with your license and ask.

[XC] gomeler · 07-01-2006, 01:45 PM

Interesting. I think I'm going ot have to brave trying this compressor because I can't currently warrant dropping $150 on a compressor for a system that could potentially not work. If autocascade doesn't work then I'll just make a nice single stager. I think I might use really wide diameter cap tube and perhaps a filter/dryer before the captube for the CO2 line. Going to begin parts gathering within the next week or two. Might be the only 1/3 HP compressor in the autocascade competition =)

ps- thanks for the answers n00b of l337, was afraid I'd just get flames from the crowd.

n00b 0f l337 · 07-01-2006, 01:48 PM

Filter drier would be frozen, it needs to be before HX. If you've never built a single stage, dont even try an autocascade. Your going to get yourself hurt if you do.

Gray Mole · 07-01-2006, 01:59 PM

Agreed on both counts.

I made the mistake of using a solid core filter in the frozen part of the cascade I made, and it took me a while to figure out what was going on

ah well, you learn as you go.

Autocascades are the most difficult cooler to make. The most. Even when you think you have it, it can go wrong for no reason you can figure out. It's difficult for experienced guys to deal with.

The only guy who was here that had no problems with Autocascades was Cryotek, mainly because he made them for a living. Otherwise they are NOT easy, NOT simple, and to be honest, don't perform as well as a regular cascade unless you know exactly what you're doing. They're also 10x more dangerous to work on as the pressures can 'get you' before you even realise there's a problem.

If you're going to be dumb enough to skip past a SS and go directly to -100 cooling (My first real custom was a cascade, so I suppose I'm one of the dumb ones too) then go with Cascade, not Auto. Unless you have a lot of prior related experience (I've done Auto aircon for about 15 years

which I hate), or are willing to DEVOTE yourself to about 3-4 months of solid sticky-reading and research (also guilty

) then seriously it's not a good idea to get into the heavy stuff too fast.

But if you do anyways, build slow and ask a LOT of questions and post a LOT of pics along the way, so you can get the help you need, and stay safe doing it

Gray

n00b 0f l337 · 07-01-2006, 02:05 PM

Talking to Gomeler now, seems he's built a few SS...

Gray Mole · 07-01-2006, 02:10 PM

Cool, then the one thread I would recommend above any other is GCLG2000's worklog on his cascade.

http://www.xtremesystems.org/forums/...ad.php?t=54312

probably the best cascade worklog, and one of the ones I used to pick up on a lot of the details I couldn't find anywhere else.

Still wouldn't recommend Auto's, but if done well and with access to enough of the right gasses, then can and do outperform regular cascades.

Cheers

Gray

[XC] gomeler · 07-01-2006, 02:25 PM

I've built a few SS's in my time here, I just haven't published any worklogs of them =) I've looked around a bit and there is definetly a lot of stuff that I'll need to be careful of. I'm considering incorporating a sort of emergency expansion tank w/ some sort of solenoid to vent the high-side in case the system goes fubar quickly. If I have any questions I will definetly be hitting this place up, I'll be working slowly and safely while documenting every step. If I can pull this off within the next 3 months and get it 24/7 stable then I'll be proud of myself. If not, well it was a lofty goal.

[486] · 07-01-2006, 07:07 PM

wow, this thread got big quick!

and could i go without the copper wool, if not than crap i have to tear my phase sep again

thanks

[XC] gomeler · 07-02-2006, 09:46 PM

For filter/driers is it possible to oversize them? I'm considering a liquid line filter/drier for both the r290 and Co2 lines that will sit between the phase separator and the metering devices. I just want to make sure getting too large of a filter/drier wouldn't be an issue. The larger filter/driers also seem to all be threaded, I guess this is to avoid burning up the desiccant?

jinu117 · 07-02-2006, 10:31 PM

I've seen quite large filters with sweat? In fact, if you walk out to your A/C at home, chances are it is brazed up :P

[XC] gomeler · 07-03-2006, 10:49 AM

Hmmm k, I'm going to grab a decent sized liquid line filter/drier w/ a 1/4" or 3/8" inlet/outlet and I'll just wrap it in wet rags to keep from burning it up along with a shielding gas flowing inside to protect the magical stuff inside.

404Power · 07-03-2006, 11:21 AM

Quote:

Originally Posted by Gray Mole

The only guy who was here that had no problems with Autocascades was Cryotek, mainly because he made them for a living. [...]

Hmm ... do you mean is or was?

Anyone knows what's going on @ cryo-tek's place?

Tried to contact him 2 times ... then I realized that ht wasn't online here @ XS since 8th of june!

[486] · 07-03-2006, 05:16 PM

with the cascade before auto, i'm using used compressors so wouldn't a auto be better because only 1 compressor to fail? and pressures would be easier to watch because only 1 loop? right?

[XC] gomeler · 07-03-2006, 08:31 PM

I am just making a guess here but I would bet that an autocascade is more stressful on a compressor than a cascade. My reason for this would be the potentially higher pressures and increased heatload. Once I actually have my system built I'll be able to compare it to one of my similar single stage units. That is just my guess, but it could be completely wrong.

Thought I would add that I have settled on the compressor I will be using. 1/3 HP hermetic and best of all it comes with all the electricals meaning I won't have to deal with that right now. If someone has had a bad experience with this compressor please speak up, I'll be ordering it on Thursday or Friday.

[486] · 07-04-2006, 07:01 AM

i have a 1 hp from a beefy window a/c,
in a cascade if the 1st stage compressor dies and your pressure switch fails the 2nd builds pressure till it pops [extremely violently]

autocascades have 2 pressures to monitor and only 1 commpressor so if it dies the system is just "turned off"

cascades have 4 and most people cant pay attention to something thats not supposed to be moving

[XC] gomeler · 07-04-2006, 08:10 AM

Ahh I see, now that makes sense. Either way though there is a good chance the computer would be toast =D

[486] · 07-04-2006, 09:35 AM

not if you set the bios to turn the computer off at 60C

[XC] gomeler · 07-04-2006, 11:30 AM

Some bios' get the temperatures fuxored. My MSI Neo2 registers temperatures above 60*C when it is under phase =\

[486] · 07-04-2006, 04:14 PM

hmmm, ya learn somethin every day...

but you could have a relay that will cut the power to computer and compressor and have 2 pressure sensors [on the high side], one that closes the circuit at 50+ psi and another that opens the circuit at 400+psi so if there is too much pressure or too little nothing will run, and have an override switch on the low pressure switch so you can start the computer