low budget cascade

[supercoolin]

I have jumped off the deep end and decided to build a low budget, low temp cascade. I have found it very difficult to SS a -60c core temp system, but even harder to maintain long term moisture control with a sub -60c system, so I decided to go for the middle. A little system that uses a 1/2hp rotary for the "sub-cooler", to get the 2nd stage condenser temps around -25C and then a 1/3hp to cool the processor to -80c with a core temp target of -60c. Then to top it off, I still want to shove it in my computer case.

First project was a small heat exchanger. As I did my research into the pro and cons of tube vs plate, I chose to go with a tube unit. In order to get a lot of surface area, I decided to coil a 1/4" inside another 1/4" and fit in into a 2" dia shell. The incoming feed is routed inside the small coil and then it will surround all areas before exiting the bottom of the outside coil. Hopefully I did approach this in a logical manner. Evap control will be with a Danfoss expansion valve with a #3 orifice.
http://members.cox.net/gorillasnawt/cascade1.jpg
http://members.cox.net/gorillasnawt/cascade2.jpg
http://members.cox.net/gorillasnawt/cascade3.jpg
http://members.cox.net/gorillasnawt/cascade4.jpg
http://members.cox.net/gorillasnawt/cascade5.jpg

Next was the oil separator. Again I wanted small so I cut a 4" piece of 2" dia tubing, cut out twenty five stainless steel screens, and made a central feed tube. The supply will feed through the 1/2" dai tube towards the bottom of the separator, be feed up through the screen array and exit at the top of the camber. Oil will collect at the bottom and be returned to the suction through a 36" long .041 cap tube. Crude but cheap and hopefully efficient.

http://members.cox.net/gorillasnawt/cascade6.jpg
http://members.cox.net/gorillasnawt/cascade7.jpg
http://members.cox.net/gorillasnawt/cascade8.jpg

I will update as I progress and any input and suggestions are appreciated.

[aenigma]

Looks very nice, I really like the pipework! You built your HX almost exactly like mine. I used two 16' lengths of 3/16" tube in tandem wrapped around a 3/4" capped pipe and put it in a 2" pipe. It worked really good. Just recently though I decided to try a plate HX since it should be easier to insulate and my HX was way too large for my needs and required a lot of refrigerant to fill the HX. To give an idea of how good it was working and how oversized it was I was getting about -20c at a little below the halfway point of the HX and the liquid line on 2nd stage was -40c.

Also, when I first brazed it years back I didn't run protection gas through it, and tried to clean it out as best I could. Turns out my attempt at cleaning it was worthless because I cut it open a few days ago and it looked like the insides had been dropped in a fire! Still worked just great though, that design should work just fine for you! :up:

[tiborrr]

I just loooove your pipework, especially oil separator looks lovely! :up: Keep it up!

Best Regards,
N.

[[XC] Lead Head]

Your oil separator, has a surface are of around 31 square inches, with a 150 PSI static charge, that is 4700 lbs of pressure, tying to push it apart.

I can't tell how big your HX is, but it looks to be about 7" or so. With that same 150 PSI static charge, you have 7500 lbs pushing outwards on that.

I'm not trying to be a jerk here, but do you feel confident that your brazing and those pieces of pipe can handle that kind of force?

[Ozzfest05]

it looks great I like the oil sep but how are you going to fit it in your case what kind of case is this lol

[Kepons]

Won't 03 orifice be too big?

[ruffus]

ya i was thinking the same thing i thing he should start off with a 00 orfice and then if he finds it to small to go out as orfices are cheap u can get them for like $12

[sdumper]

Lead Head;3663847']Your oil separator, has a surface are of around 31 square inches, with a 150 PSI static charge, that is 4700 lbs of pressure, tying to push it apart.

I can't tell how big your HX is, but it looks to be about 7" or so. With that same 150 PSI static charge, you have 7500 lbs pushing outwards on that.

I'm not trying to be a jerk here, but do you feel confident that your brazing and those pieces of pipe can handle that kind of force?

Im confused I dont get the same numbers you did..

http://www.online-calculators.co.uk/volumetric/cylindervolume.php

Im getting 1875 lbs and 3300 but maybe I calculated incorrectly :(

Diameter = 2 so radius = 1 and length = 4 and 7 respectively...

[wdrzal]

http://www.copper.org/applications/plumbing/techref/cth/tables/cthindex_table.html

Look @ table 3 for working pressure of M,L,K & ACR copper tube. There is a lot of other usefull information on copper tube there. just googled that page.

[sdumper]

Thanks Walt so if I'm reading this correctly copper pipe made of drawn copper with a 2 inch diameter has a burst pressure of 3915 psi and the Brazed joints if made of http://www.under-the-ice.com/product_info.php?cPath=36&products_id=83 {DynaFlo Alloy HB} has roughly a burst pressure of 805psi for a 2 inch diameter brazing area at 100f.

Im probably using the charts wrong but thats how I read it.

[supercoolin]

Lead Head;3663847']Your oil separator, has a surface are of around 31 square inches, with a 150 PSI static charge, that is 4700 lbs of pressure, tying to push it apart.

I can't tell how big your HX is, but it looks to be about 7" or so. With that same 150 PSI static charge, you have 7500 lbs pushing outwards on that.

I'm not trying to be a jerk here, but do you feel confident that your brazing and those pieces of pipe can handle that kind of force?

The areas of concern is the pressure against the caps, not the entire surface area of the cylinder, 2" type L copper is rated to 1200psi working pressure, the connection (end cap) is where you need to be concerned. The silver solder joints should be good to 500psi, the brazed joints should be good to 700psi. Prep, heat control and good flow insures that the entire joint surface has been bonded.

[wdrzal]

Thanks Walt so if I'm reading this correctly copper pipe made of drawn copper with a 2 inch diameter has a burst pressure of 3915 psi and the Brazed joints if made of http://www.under-the-ice.com/product_info.php?cPath=36&products_id=83 {DynaFlo Alloy HB} has roughly a burst pressure of 805psi for a 2 inch diameter brazing area at 100f.

Im probably using the charts wrong but thats how I read it.

You use Working pressures in a system design, not burst pressures.

[[XC] Lead Head]

Im confused I dont get the same numbers you did..

http://www.online-calculators.co.uk/volumetric/cylindervolume.php

Im getting 1875 lbs and 3300 but maybe I calculated incorrectly :(

Diameter = 2 so radius = 1 and length = 4 and 7 respectively...

Surface area = Circumference times height + area of the top and bottom end caps.

Circumference of a circle = Diameter x Pi, Area of a circle = pi(Radius^2)


The areas of concern is the pressure against the caps, not the entire surface area of the cylinder, 2" type L copper is rated to 1200psi working pressure, the connection (end cap) is where you need to be concerned. The silver solder joints should be good to 500psi, the brazed joints should be good to 700psi. Prep, heat control and good flow insures that the entire joint surface has been bonded.

With a 150 PSI static charge, which is not unreasonable for a cascade, there will be nearly 500 pounds pushing on those end caps. That is cutting it a bit close there then IMO.

[supercoolin]

Lead Head;3664395']Surface area = Circumference times height + area of the top and bottom end caps.

Circumference of a circle = Diameter x Pi, Area of a circle = pi(Radius^2)



With a 150 PSI static charge, which is not unreasonable for a cascade, there will be nearly 500 pounds pushing on those end caps. That is cutting it a bit close there then IMO.

Again we have maximum 500psi working pressure for the joint, so the caps are good for 1600lbs of actual force against them [(?)psi x area]. The pressure against the caps/end doesn't increase as the cylinder gets longer, the total force on the joints is only the internal pressure per square inch times the total square inches of surface exposed on the end cap. Now if we exceed 500psi working pressure within the system, then the joints are not safe. On the second stage, where the oil separator will be, I will install a high pressure bypass valve at 400psi since I will be using HC-502A as my coolant. I am contemplating a 450psi blow-off valve, but that will be added only if I see the HP bypass being triggered often.

[aenigma]

Lead Head;3663847']Your oil separator, has a surface are of around 31 square inches, with a 150 PSI static charge, that is 4700 lbs of pressure, tying to push it apart.

I can't tell how big your HX is, but it looks to be about 7" or so. With that same 150 PSI static charge, you have 7500 lbs pushing outwards on that.

I'm not trying to be a jerk here, but do you feel confident that your brazing and those pieces of pipe can handle that kind of force?


Lead Head;3664395']Surface area = Circumference times height + area of the top and bottom end caps.

Circumference of a circle = Diameter x Pi, Area of a circle = pi(Radius^2)



With a 150 PSI static charge, which is not unreasonable for a cascade, there will be nearly 500 pounds pushing on those end caps. That is cutting it a bit close there then IMO.

Just wanted to let you know that my HX I mentioned before your post was 2" (or was it 2 1/8"?) in diameter and well over 12" long and I have seen static pressures much higher than that since it's in an autocascade, and have had no problems at all. My receiver in this same unit is made from the same pipe and has been up around 400psi and at one point I had it steady at 350psi for quite some time and I have had no problems.

[supercoolin]

Well have managed to get all the pieces here, the case is on a UPS truck and a last few of the homemade parts are finished so the assembly begins. I have been reading all the different ideas for sub-coolers and different ways to cool the cap tubes so I tried something I think is different and am going to see how it works. I am tired of the running the cap tube from a spun filter/drier through the insulation and then running it around the suction line for a little pre-cooling, so after thinking about I decided to run the cap tube the full lenght of the suction line from my accumulator. Then I thought, what the heck I will jamb the spun filter inside the accumulator shell and that way I get a sub-cooler and accumulator in one package that will be a lot easier to insulate and I won't have to mess with the cap tube sticking out of the suction line insulation. I also decide to use the Freeze Pack SS controller from Under-the-Ice.com with a couple of Eurotherm temp controllers for the 2 stage and the twin compressor 1st stage.

http://members.cox.net/gorillasnawt/cascade9.jpg
http://members.cox.net/gorillasnawt/cascade10.jpg
http://members.cox.net/gorillasnawt/cascade11.jpg
http://members.cox.net/gorillasnawt/cascade12.jpg
http://members.cox.net/gorillasnawt/cascade13.jpg
http://members.cox.net/gorillasnawt/cascade14.jpg

On the 1st stage I am using to compressor that will stage in a out as needed when the processor is under load. When I am not rendering, I don't need max cooling so one of the 1st stage compressors can drop off and stand by until I need the additional cooling on the system under full load.

[GuGaCoSa]

Nice subcooler, in fact i would add another liquid line filter, those filter will have hard time filtering as it will be immerse in evaporating liquid, blocking the moisture retaining.
But very good idea, i would like to try it sometime.
Regards

[supercoolin]

I have been searching fro a small condenser, but finding one around 8" x 11" (size limit) has been unsuccessful so I ended up building one out of a larger core. One of the thing that is very important to me is having the "rifling" of the tubes. I haven't found one small unit that offered the rifling. One thing I can say is that this is sure taking a lot more time than I originally imagined.


http://members.cox.net/gorillasnawt/cascade15.jpg

http://members.cox.net/gorillasnawt/cascade16.jpg

http://members.cox.net/gorillasnawt/cascade17.jpg
The rifling of the tubes really improves efficiency
http://members.cox.net/gorillasnawt/cascade18.jpg
THIS SHOULD SUCK SOME AIR!!!
http://members.cox.net/gorillasnawt/cascade19.jpg

http://members.cox.net/gorillasnawt/cascade20.jpg

http://members.cox.net/gorillasnawt/cascade21.jpg

http://members.cox.net/gorillasnawt/cascade22.jpg

http://members.cox.net/gorillasnawt/cascade23.jpg

I am hoping to have the suction line and evaporator made up today also.

[bazx]

you have some fine hand skills and make good reading

good luck with the build

ps: looks like you are in EvOcarlos workshop

[[XC] gomeler]

Did you just coil copper tubing and cut it for the u-bends or do they sell u-bends like that?


ps: looks like you are in EvOcarlos workshop

First thing I noticed :p:

[supercoolin]

you have some fine hand skills and make good reading

good luck with the build

ps: looks like you are in EvOcarlos workshop

Thanx Bazx, but I have seen your work and I'm not even in the same league.

EvOcarlos has a clean room compared to mine,..hehe...


gomeler;3701981']Did you just coil copper tubing and cut it for the u-bends or do they sell u-bends like that?

First thing I noticed :p:

I raid A/C shops for old condensers and evaporators and cut all the 180's off old coils, been able to collect different radius bends for just in case, like this build. Use my tubing flare tools to "re-size" the bends on the ends to slip back into the coil.

[sdumper]

Very nice work!!!

[supercoolin]

Got the accumulator/drier/filter combo finally fitted up with the evap. It sure is a lot cleaner with the cap tube completely out of the way. Looks like the 2nd stage will fit neatly into the bottom of the original case with the compressor, the accumulator and the oil separator. Once I attach the second matching case side by side, I can start installing the other compressors and HX. With nothing but the 1st stage going into the second case, there will be tons of room in that one.


http://members.cox.net/gorillasnawt/cascade24.jpg

http://members.cox.net/gorillasnawt/cascade25.jpg

[teyber]

this is VERY tastefully done...

where are you located? just curious

[supercoolin]

this is VERY tastefully done...

Thanx


where are you located? just curious

Las Vegas, Nevada

[teyber]

Chilly1 is still around and about, just no on the forums. hes from los vegas and he has a couple storage units full of parts and hes trying to sell them cheaply... If your interested i can pm you his number. last time i talked he was trying to get rid of some non floated temprite oil seps for cheap and some evaps for very cheap.

Cheers

[Dualist]

Excellent work all round dude :up:

But your condensor is beautifully done, nice one :)

[supercoolin]

Got everthing for the 2nd stage temp fitted into case 1 with room for everything else, now I just have to sit back and wait until the 2nd matching case gets here to install the 1st stage compressors and pipe everything up. The 2nd stage was where it got hairy for room and I had to cram it all together, but once the two matching computer cases are connected side by side all the other stuff will go a lot quicker. Boy I sure hope this all works.......

http://members.cox.net/gorillasnawt/cascade26.jpg

http://members.cox.net/gorillasnawt/cascade27.jpg

[GuGaCoSa]

Very good man, congrats, i'm only concerned about the reciever joint, looks like you soldered it, if so could it resist the pressure?
Keep up the good work

[supercoolin]

..........looks like you soldered it, if so could it resist the pressure?
................

All joints are silver or silver alloy brazed joints, NO "soldered" joints, just different temp alloys. I use the Silvaloy where I can clean it with acid and remove the oxidation and/or flux and 99% silver solder where I can only pre-etch with acid, and must do final clean with just no residue cleaning chemicals.

[Kepons]

Very nice build, but tell me guys why you use those expensive silver rods instead of cheap Silfoss (bronze ones)?

[[XC] gomeler]

Very nice build, but tell me guys why you use those expensive silver rods instead of cheap Silfoss (bronze ones)?

Silver looks better and is easier to work with. The price difference isn't too bad either to insure a leak-free seal.

[GuGaCoSa]

Nice, 99% silver solder?Never heard before, you used that one in the receiver joint?looks amazing.

[supercoolin]

Nice, 99% silver solder?Never heard before, you used that one in the receiver joint?looks amazing.

It is not 99% silver, it got it nick-name "silver 99" because it was 99% pure alloy solder, no flux or impurities. Now they make it 100% pure alloy. With these pure alloy solders, you have to use a tinning flux that leaves a residue that must be cleaned afterwards, but the joints are rated upwards of 600 PSI working pressures. Silvabrite 100 (silver solder) is 95.6% tin, 4.0% copper and 0.4% silver and can be used with a MAPP Gas bottle torch.

http://members.cox.net/gorillasnawt/solder99.jpg

[supercoolin]

Is there a methodology for determining the capacity of my new condenser? Here is a picture of the Asetek unit @ 300W and the new condenser. The core is 8" x 8" x 3" with 24ea. 3/8" rifled tubes.

http://members.cox.net/gorillasnawt/cascade28.jpg

[teyber]

figure it out yourself i would say, im sure it would be less work. find at what heat load does the condenser reject liquid at a delta of 15c between the liquid and the ambient temp... or you could just say its problably good enough :D

[tiborrr]

I'd say 0.7-0.8kW @ 15K deltaT.

[Kepons]

Man, according to what you're saying, the solder you're using is just a soft solder and shouldn't be used in this application IMO.

[tool_462]

It is not 99% silver, it got it nick-name "silver 99" because it was 99% pure alloy solder, no flux or impurities. Now they make it 100% pure alloy. With these pure alloy solders, you have to use a tinning flux that leaves a residue that must be cleaned afterwards, but the joints are rated upwards of 600 PSI working pressures. Silvabrite 100 (silver solder) is 95.6% tin, 4.0% copper and 0.4% silver and can be used with a MAPP Gas bottle torch.




Man, according to what you're saying, the solder you're using is just a soft solder and shouldn't be used in this application IMO.

600 PSI seems enough to me :p:

[Kepons]

600 PSI seems enough to me :p:

The reason the connections in refrigeration units are done with braze (not soft solder) is the strenght of the connection. Maximum working pressure is one thing and the other thing is the joint ability of resisting the stress during operation in low or high temperatures.

[supercoolin]

The reason the connections in refrigeration units are done with braze (not soft solder) is the strenght of the connection. Maximum working pressure is one thing and the other thing is the joint ability of resisting the stress during operation in low or high temperatures.

It is in the application, not the rating. Where you have drilled into a piece of pipe and there is very little material to grab onto, of course you would braze with silvaloy/brazing rod, or in a high vibration area. But where you have a swedged connection or a standard fitting, the silver alloy solder is more than adequate. Even with a soft solder and a swedged connection under 1/2", the actual tubing is going to break long before joint is,but if you used soft solder where you ran a 1/4" tube through the side of a 1" tube, then all bets are off. Also with brazing, you can get the joint to hot and it will fail long before a low temp soldered joint.

[piotres]

WOW 2 stages in Lian-li V2000 - first time sb built sth like that I bet :clap: .

What gases are You planing to use ?

Can Freeze pack cope with temperature lower than -60C ? :rolleyes:

Good luck :)

Peter

[supercoolin]

WOW 2 stages in Lian-li V2000 - first time sb built sth like that I bet :clap: .

Actually two of them side by side


What gases are You planing to use ?

HC-502a, ethylene


Can Freeze pack cope with temperature lower than -60C ? :rolleyes:

not sure until we try, but my target is -60C working temps


Good luck :)

Peter

Thanks

[supercoolin]

Finally starting to come together. Spent most of the time painting the case. 2nd stage is all fitted up and triple vac'd with a light charge. Fired her up using the super-deheater as the condenser and she hovered at -62 on 502a. Should have already gotten the 2nd case but the web store I purchased it from were drop shippers and they never found the case to have shipped to me. After pulling my chain for over two weeks I canceled the order and picked up a used one on eBay. Since I am painting them metallic Cobalt Blue, I just needed any color shell. Once I get the 2nd case and paint it, everything else is drop in a braze up.

http://members.cox.net/gorillasnawt/cascade29.jpg

http://members.cox.net/gorillasnawt/cascade30.jpg

[tiborrr]

That's a very tidy insulation job, supercoolin. Keep up the good work! :up:

[SexyMF]

Yeah, nice work. But I'm not sure about the low budget bit of this build. :confused:

[teyber]

Yeah, nice work. But I'm not sure about the low budget bit of this build. :confused:

the compressors are $20 a pop shipped :up:

[supercoolin]

Yeah, nice work. But I'm not sure about the low budget bit of this build. :confused:


yep $19.95 ea shipped x 3, everything else was hand built (i.e. oil sep, accum, receiver, condensers, de-heaters). The Freeze Pack controller represents the big bucks on this build.

[Jor3lBR]

Looks very clean! Can't belive you fitting a cascade in the case and still have space for mobo installation, I thought you were using the case for the cascade only. You said you were in California?

[sjg0]

yep $19.95 ea shipped x 3, everything else was hand built (i.e. oil sep, accum, receiver, condensers, de-heaters). The Freeze Pack controller represents the big bucks on this build.

You hand built those condensers?

[supercoolin]

You hand built those condensers?

I believe you understand the point I was trying to make. I found old pieces of condenser coil and cut out the sizes that I needed and re-built them. Free except for the Silvaloy brazing rods to braze up the relocated 180 bends. Shrouds and end caps were made out of sheet metal and plate aluminum scraps laying around the garage, $5.00 vs $200.00 if bought off the shelf.

[supercoolin]

You said you were in California?

Las Vegas, Nevada

[tiborrr]

Keep up the good work, supercoolin, and don't be afraid to show more progress! We are waiting! :yepp:

[supercoolin]

Well as I wait impatiently for the second case I bought off eBay, I went ahead and started the condenser for the 1st stage. I took and old 3KW unit with 3/8" rifled tubes and cut it down to 6.5" x 18" x 3". This will go in the top of the case with 4 115cfm 120mm fans pulling air through it. Cost for this, 2 hrs labor, 25cents with of acid and two sticks of Silvaloy brazing rod @ $1.15 ea. Oh and 2.00 worth of 24ga sheet metal for the shroud. I post another pic of it mounted in the case when its done.

http://members.cox.net/gorillasnawt/cascade31.jpg


Also when I evac'd the 2nd stage, I decided to run a little freon through it and then change out the filter before final charge to make sure everything was mostly clean. Since R-134a is pretty good at picking up about any and all types of residue or residuals, I charged it with full 15oz. can full. It pulled down to -31C in less than 30 sec with 13" vac low side and 110PSI high side. As the accumultator began to chill down to -10C, the evaporator went to -45. When it hit -45 on R-134a, I had to conclude that mounting the spun filter/drier inside the accumulator and having the cap tube completely enclosed within the suction line was acting as a sub-cooler or even as far as the system was acting as an autocascade. This could end up pretty interesting once it is all put together, heck it might even end up being a autocascade on top of a full cascade (say 3 stage?) This is turning to be pretty damn fun.

BTW the oil seperator is working like a champ, oil level has stayed within 25cc's of the initial pre-charge and there has been no flucuation of the low side pressure that would occurs if the oil level drops below the auto re-feed port. I am pumped!!!

[sjg0]

I believe you understand the point I was trying to make. I found old pieces of condenser coil and cut out the sizes that I needed and re-built them. Free except for the Silvaloy brazing rods to braze up the relocated 180 bends. Shrouds and end caps were made out of sheet metal and plate aluminum scraps laying around the garage, $5.00 vs $200.00 if bought off the shelf.

Omg no actually I didnt realize that. Thats amazing bro! I said that because the pics of the condensers you posted looked like you bought them, upon closer inspection i can see now they have been made by you, lol mucho props! They look really great!

[supercoolin]

Well as I impatiently wait for the second V2000a case, I went ahead and finished up the 1st stage condenser and started to fit up the piping so I can just braze it up once I mount everything in the case.

http://members.cox.net/gorillasnawt/cascade32.jpg

2nd stage is running with R134a and a 75W load and staying at -42. I have been running it a couple of days and have been pulling it down to check oil seperator flow and then recycling the charge. So far after about 40 hrs of run time, it looks like my guess on cap tube size for the return was dead bang on, staying at proper level and no added oil. Absolutely pure damn luck.

http://members.cox.net/gorillasnawt/cascade33.jpg

Looks like everything will fit pretty tightly, but fit it will. Coolant is actually runner through the HX to test the systme and oil sperator operation. Full sight glass of R134a at 95PSI high side..

http://members.cox.net/gorillasnawt/cascade34.jpg

[teyber]

AHAHA how low of a vacuum is your second stage in? :D

looks great, i really love it!

[supercoolin]

AHAHA how low of a vacuum is your second stage in? :D

looks great, i really love it!

THANX.....

12.6 in.Hg. I would have never believed I could pull those kinds of temps with R134a, I will have another tank of HC502a on Tuesday, so I will have see what that will do before I final charge it with R1150.

[Odium]

Very impressive!

[supercoolin]

been fitting up the plumbing waiting for the new case to arrive. It's on the UPS truck and scheduled for delivery tomorrow. Hopefully we will be making frost with the 1st stage this weekend (have a little right now on the 2nd stage piping..hehe)


http://members.cox.net/gorillasnawt/cascade35.jpg

[supercoolin]

Well not sure if I should post the pictures and progress that has been deleted during this last "Outage" but have been doing testing and tune and she seems to be pretty healthy. While I am waiting for the ethylene to arrive I charged the second stage with R507. Three separate start up and she holds -83 with the R507 and no insulation on all the piping. Looking forward to see what happens with it charged with the R1150. Hope I will get into the -100 club.


http://members.cox.net/gorillasnawt/cascade46.jpg
http://members.cox.net/gorillasnawt/cascade44.jpg
http://members.cox.net/gorillasnawt/cascade30.jpg

[teyber]

wow nice! and you got tons of room for a third sta.... haha just playing. it looks seriously sweet!

[sdumper]

Amazing how compact and clean everything is!!!
Great work!!!!!!

[supercoolin]

wow nice! and you got tons of room for a third sta.... haha just playing. it looks seriously sweet!


Next one will be a real 3 stage...this was my 1st attempt at a cascade to see if I could even build one and just get a stable -60 under load. It works much better than I had hoped so now I think I can tackle a real system.

[supercoolin]

Amazing how compact and clean everything is!!!
Great work!!!!!!


thanks for the kind words but it was a lot of compromises to make things fit and sacrific clean lines.

[bazx]

thanks for the kind words but it was a lot of compromises to make things fit and sacrific clean lines.

working to a set size is very difficult if not the hardest way to build

and you have carried it off very well

i think that you will enjoy a bigger build with more room to work

3 stage cascade is a great deal of fun to build good luck

[supercoolin]

working to a set size is very difficult if not the hardest way to build

and you have carried it off very well

i think that you will enjoy a bigger build with more room to work

3 stage cascade is a great deal of fun to build good luck

It was a challenge build and the most exciting part is the oil separator. I couldn't find anything small enough to work in my tight spaces so it was "Build-your-own" time. It is working flawlessly without any floats or valves, it is fully automatic and actually works!!! Everything else is so basic in design (Accum, Reciever, cond., evap.) that I wasn't worried about function, the oil separator was the real challenge on this project. And I have achieved my -60 loaded cpu target with only 52db.

[supercoolin]

When I joined this forum and started looking through some of the incredible projects here, I never dream I would be able to take three $20.00 compressors I bought off ebay and throw together a bunch of home-made parts and get -90 on straight R507. After a little adjustment of the gases in both stages and a 20min run with 50watts load, she held -88 with a low of -90. This stuff is totally cool...............Thanks guys for all the systems you have shared on this forum and the inspiration I have gotten readin about them, to even attempt anything like this.

[supercoolin]

http://members.cox.net/gorillasnawt/cascade47.jpg


Using R507 on the 1st stage 0psi low 260psi high 8K superheat
R507 on the 2nd stage -28inHg low +0psi high 5K-11K superheat (moves around over 5 minutes, need smaller fan @ super de-heater coil

HX temps
at the expansion valve -55, out of the HX -44 @bulb
2nd stage condsensor coil in 23, liquid out -35
2nd stage accum/sub-cooler combo -68
cpu evap with 50Watts -88


friggin can't wait for the R1150

all three little rotary's are running about 45-50, considering adding water coils around each to add a little compressor cooling.

[Jack]

Looking really nice and compact! What load are you aiming for with the R1150?

And why water coils? Those compressor don´t run hot at all... well, not yet anyway (the ethylene will make the compressors work harder).

[supercoolin]

Looking really nice and compact! What load are you aiming for with the R1150?

And why water coils? Those compressor don´t run hot at all... well, not yet anyway (the ethylene will make the compressors work harder).

I am shooting at -60C with a 130W load (Phenom II at 4.8Ghz)

I am going to try and add a little water cooling to the compressors because I believe if I can drop the compressor temps 10-15C, it should add 200-300W of capacity to the 1st satge.

[Fitseries3]

i love this build. very nice.

are you using the 2 compressors in the 1st stage and a 3rd in the second stage? so thats ~10k btu in the 1st stage and 5200btu in the second?

(sorry for the noobness, still learning here) :)

[supercoolin]

Well the tune is done and the mix works. I still have a bunch more on the controls but she starts with a single button, full auto startup including the mobo with full Windows shutdown. This is with a 175W load.

Ambient Temp 28.5C
1st Stage R-507
245 PSI Static
-48C at the HX
3psi low / 235psi high

2nd Stage R1150 92%/ Propylene 8%
295psi Static
-29.3C out of the HX
-116C at the temp plate (1/4" plate between evap and load cell)
-8inHg Low/275psi high

1523W load reading on the Kill-a-Watt so i'm curious how it compares to the power used by other cascades (is it efficient or crappy)

THANX GUYS FOR ALL THE HELP!!!!!!!!!!!!!!!!

http://members.cox.net/supercoolin/116below.jpg

[paul_]

Wow:shocked: :shocked: :shocked: I want one!

[[XC] gomeler]

Awesome stuff, what sort of running pressures does the 2nd stage see? -116C is nuts for R1150. I'd check temps with a proven accurate hand-held thermometer, but no matter what that is an awesome project :up: Can we get some more pictures of it all?

[supercoolin]

gomeler;3833529']Awesome stuff, what sort of running pressures does the 2nd stage see? -116C is nuts for R1150. I'd check temps with a proven accurate hand-held thermometer, but no matter what that is an awesome project :up: Can we get some more pictures of it all?


Checked with a hand held and the temps were even lower, but bios reports -91C with Phenom II 9950 @ 4.73GHZ idle : 1.65V

2nd stage -17inHg low - 245PSI high 280PSI static w/ 100% ethylene.

Problem is that within 25min I have frost outside the CPU enclosure on the Mobo. I have everything formal coated, but it is just to cold to run for long periods of time so, as I had originally planned, I will start mixing gases (propylene now) until I get -65C to -70C idle temps. Heck I never dreamed this little system would ever even get to -70C, alone get me into the -100C club.

Once I get the controls finished and the water cooling sections back in, I will shoot some overall pcitures. Still have a little more "stuffing" of parts into the case before she is really done.

[EliTE22]

Checked with a hand held and the temps were even lower, but bios reports -91C with Phenom II 9950 @ 4.73GHZ idle : 1.65V

2nd stage -17inHg low - 245PSI high 280PSI static w/ 100% ethylene.

Problem is that within 25min I have frost outside the CPU enclosure on the Mobo. I have everything formal coated, but it is just to cold to run for long periods of time so, as I had originally planned, I will start mixing gases (propylene now) until I get -65C to -70C idle temps. Heck I never dreamed this little system would ever even get to -70C, alone get me into the -100C club.

Once I get the controls finished and the water cooling sections back in, I will shoot some overall pcitures. Still have a little more "stuffing" of parts into the case before she is really done.

whats the load temps?

[supercoolin]

When I started checking the "tune" on the Eurotherm controllers I bought off Ebay used, I found that they had a -12 offset programmed into the one I was monitoring the evap temp with. Since I changed from a PT100 thermocouple to a K-type, I re-calibrated it, verified it with my handheld and this seems to be the more accurate Evap temp at 75-100W load.

BTW: How do I verify for the -100C club.

http://members.cox.net/supercoolin/103below.JPG

[tiborrr]

Show the thermometer sumberged into glass of ice & water slush showing 0°C on the one pic and then show us frozen evap head with thermometer reading sub or equal -100°C.

Contact mod, e.g. RunMC, to assign you appropriate '-100°C club' rank.

[[XC] gomeler]

Do as tiborrr says and you can also boil the sensors in water to see how close they read to 100 Celsius. -103C is much closer to what I was expecting out of this build but it could still be off by a few degrees.

[supercoolin]

AS suggested I re-calibrated the Eurotherm controller with it's auto tune feature and verified it in the ice water. I had to slip the thermo couple into a copper sheath to protect it from the water so it only came down to 1C after about 3 minutes. If I touched the ceramic bead directly on the ice it would fluctuate between 0 and 1 so I will have to trust that it is accurate. After 30 minutes of running it held stable at -105C with about a 100W load. Once I put the full system back together, I am pretty confident that the Phenom II would idle in the -100C range. This is a long way from 10 years ago when I started messing with phase change and water cooling on Slot 1 and Slot A processors in my quest for +1Ghz. Here is a link to my old Home page ten years ago.


Thermal Acceleration Extreme (http://members.cox.net/supercoolin/home.html)

AN OLD WEBPAGE (http://members.cox.net/supercoolin/Copper%20Spacers.html)

http://members.cox.net/supercoolin/cal.JPG

http://members.cox.net/supercoolin/105below.JPG

[tiborrr]

Good enough, welcome to -100°C, brother! :up:

[godmod]

Congratulations.

Can you test with a higher load?

[Buckeye]

Congratz supercoolin !!! :clap::up:

Not only on -100c club but for building an fantastic unit, very nice ! :up:

[Rise]

this is a great thread, do you have any pictures of your full system up and running? i'd love to see what you've got it hooked up to and how that setup is arranged as well. either way, great job man!