As promised, heres some pictures of the work in progress..
This is the box made up from styropor... and rezized picture... sorry bout dat.
http://i52.photobucket.com/albums/g2.../theboxJPG.jpg
Printable View
As promised, heres some pictures of the work in progress..
This is the box made up from styropor... and rezized picture... sorry bout dat.
http://i52.photobucket.com/albums/g2.../theboxJPG.jpg
here is the box on its frame plus cooling system...
http://i52.photobucket.com/albums/g2...framecomp1.jpg
Pipes for cooling system is all spread out. No system as of yet...
please scale those pictures.
man, drop your pic size they are insane
High pressure side finished, missing back plate to cover it up... http://i52.photobucket.com/albums/g2...6/boxback2.jpg
inside box, cooling pipes pretty getto style there...
http://i52.photobucket.com/albums/g2...lingpipes1.jpg
Back plate in place to ensure proper air flow, kind of... air enters to the right and passes over the pipes, then drawn into mainbord habitat. Will probably do some ducting to direct the air flow towards essential parts..
http://i52.photobucket.com/albums/g2...insidebox1.jpg
Still have som serious painting to do, a lot of adaption of covers and stuff. luckily thursday is holiday here in Norway, so I might get something done by then.
looks cool, although if you plan to put a computer in there that dinky compressor and convection cooled condensor aren't going to be able to handle the heatload
Planning to put inn some fans to help cooling. Just hope the compressor will cope...
look like you hacked apart your dorm fridge, if not it probably been easier just to buy one.;)
looks like its in desperate need of a condenser replacement in order to be able to cool anything close to a modern pc heatload.
I hope the surface area of the radiator is going to be sufficient combined with added fans. Otherwise Ill have to get hold of the new M2 systems with 35 watt cpu...
If it doesent handle the load, im in big trouble as the cost of a radiator plus all the brazing equipment and valves would put my project way ahed in the future...
I have alot of fans tho, but would prefer to use as few as possible due to noize.... guess i just have to test it, and see if added fans does the trick....
.....trembling in agony and pain......
there are no "radiators" in a refrigeration loop. There are condensers and evaporators. I assume you are talking about the condenser.Quote:
Originally Posted by mindnumb1
And those dinky condensers on the back of refrigerators are really bad. Pick up a decent one from one of the members here or on flebay. You can get one for $25-$45.
Yes , I am talking about the condenser.
And for the compressor: how are they rated? I guess there are many factors involved, like the cooling medium and cooling capasity in condenser side.
My compressor is only labled with cooling medium ( r600) and frequenzy and voltage, ( 50-60 hz, 230-250 v) cant see any marks on amps draw, or displacement.
I guess if displacement was known one could figure out what its good for, but for now, I have to test it under load to see. I am corrently thinking about putting in some kind of load to see how it behaves. any tips on a proper load source, my wifes hair dryer would be a little over the top.. :)
For heatload tester read jins advice, it works very well.
http://www.xtremesystems.org/forums/...ad.php?t=99979
As far as the compressor rating, it depends on what you want to rate. If you want to rate the displacement its ussually done in Cubic Centimeters (cc). If you want to rate the voluemetric efficiency I forget what its rated in. If you want to rate the power (energy over time) horsepower and watts are the common measures of power. The capacity is ussualy rated in btu's, One Btu is equal to the amount of heat required to raise the temperature of one pound of liquid water by 1 degree Fahrenheit at its maximum density, which occurs at a temperature of 39.1 degrees Fahrenheit. Take your pick. Most people here use hp or watts. I prefer watts, makes other calculations more intuitive.
hmmm, 100 watt light bulb=100w,150bulb=150w
you and your light bulbs lolQuote:
Originally Posted by wdrzal
what could be better or easier? you already know the wattage,the selection is big and the "price is right" sounds like a good name for a game showQuote:
Originally Posted by epion2985
Yea light bulbs should work excellent for this, but for direct die it's a lil bit risky :)
@ mindnumb1: Did you braze up the system yourself or did you rip it out of a freezer/cooler? Looks kinda fun but your compressor seems a bit small.
Perfect ! Gonna try it out !!!
Cant believe I didnt I didnt think of that !
AASMAUKR :
I just ripped it out, no brazing involved. Thats why the pipes looks so getto, its not possible to route them properly... but I intend to cover them up anyways, so its not gonna look too bad. (or so I hope) :)
Now . getting close to an update here... Pictures tonigt i guess.
Have some serious concerns tho..
I did a load test with a 100 watt lightbulb. It barely gets by that load. Ouch. I really dont know the heat load of the system, but im pretty sure it is well over 100 watts. It seems I might have to get hold of a new cooling system, unless I can alleviate some of the load by optimizing air flow over the condenser.
I did however find out how to adjust the shutoff / on point for the thermostat. I will post pictures on how its done.
Does anyone know how the operation speed of the compressor works? Is it fixed (frequenzy controlled by the powerline freq.) or does it vary? seems to be fixed, so I am guessing added cooling cant do much difference exept lowering pressure and thus comp. load on condenser side... Please tell me it is not so !!! :)
Yay, another Norwegian into this phase-change stuff.
Skal sammen med aasmaukr prøve å fikse et community på adressen www.fasekjoling.no.
Fett !!!
Er dere i oslo eller ?
Og Big sturl, du er vel ikke tilfeldigvis Sturla?
Hehe, jo :) Det vil si, jeg heter Sturla. Forøvrig finnes det også én annen Sturla som driver på med fasekjøling, nemlig Stoolman. Da både navnet Sturla og fasekjøling er nokså sjeldne ting, så tror jeg ikke du finner de to i kombinasjon flere steder (hvis jeg ikke tar feil).Quote:
Originally Posted by mindnumb1
And for those of you that don't speak Norwegian, we aren't talking about anything of interest.
ok, hehe, kjenner Stoolman via kona hans som er venninde med ekskona til broren til kona mi.... LOL ! Men holder du til i Oslo?
Vi tar det over Pm, ok? Orker ikke whinete mods.
Ok, mr. moderator, we'll take it over PM.
sorry, have to say it but you said "stoolman" and in english that means poo man
yes, thats his nick.... his kind of humor.. :)
Aiiiiiieieeeee! Stop with the Norweigish! I can't understand you!
heres some picture of the progress. Below you see the covers painted, the box itself painted, front cover of insulated box with the foam seals mounted with silicone, and a homemade cable extention for mobo power hookup, the socket was ripped from an old mobo by heating it on the stove and ripping the socket out... smelled really good ! then soldering on the cables ripped from an old powersupply...
http://i52.photobucket.com/albums/g2...rontcovers.jpg
http://i52.photobucket.com/albums/g2...boxpainted.jpg
http://i52.photobucket.com/albums/g2...solatedbox.jpg
http://i52.photobucket.com/albums/g2...eshomemade.jpg
http://i52.photobucket.com/albums/g2...lesandmobo.jpg
Tomorrow Ill probably will be done with the last gluing stuff, and mounting the different switches. also need to make an extention for the aux power cable, and routing them together with all the other cables that go to the motherboard... hopefully ill be ready for a testrun wednesday, but... its slow... :)
i just thought about your idea with heatload testing and think about lightbulbs, not all 100watts of a 100 watt lightbulb are turned into heat
Well, most of it is, as the part that is light turns to heat as it is absorbed by the surroundings. In a closed box, no light of the spectrum we see escape, but a really, really minute amount might be part of the spectrum that can pass throug walls of some thickness. This is most certanly less than 0,01 %.
Another thing is that the voltage might change from place to place, and hence the effect, but the difference is less than +-10 % i think.
I thought 95% of the light bulb is wasted as heat
k nevermind, just popped into my head
did a load test with my current setup ( athlon xp/ sempron) abit nf7-s and radeon 9800 pro in the rig. I have some fluctuations in temperatures, but it seems to hold off the load, and I noticed that the cooling capasity increased a bit when I managed to seal up the "suction box" ( where my fans sit and pull air over the condensers)
hadnt hooked up the ide disk correctly, so I did not get a full load scenario, just booted repeatedly...
had -11C in the box ( temp probe is flushed with air from the cpu hs/fan) and mobo reported 15-16C degrees on cpu.
It operates quite silently, the most noize comes from the powersupplies.
Im gonna put in a bottle of Cooling liquid mixed to freeze at -12C to get a little buffer for higher load over time. Tonight I might get a better picture of how its going to work. Crossing fingers.... :)
I think 99.999% of the energy going in to the lightbulb is relesed as heat, rest as sound. I am not sure if light falls in your standart energy group. On one hand photons have mass and can have a velocity. Hence they can have kenetic energy. But that has nothing to do with actuall light. I should ask my physics professor tomorrow.Quote:
Originally Posted by iboomalot
He he !! Light is a subject you can study for quite some time !
It might not be considered "energy" itself, but it most certainly is a good carrier of energy. We do for instance have a rater large orb in the center of the solar system (the sun, I think its commonly named) from witch we get almost more energy than we can handle, all in the form of light, or radiation.
Depending on the amount of energy each photon carries, it has a corresponding "wavelenght" and also the speed is relative to this. Likewise the mass corresponds to the amount of energy.
Depending on the wavelength of the photon, it has a certain chance of being stopped in, or, deflected off the medium its hitting. Some mediums, like air and glass lets some wawelenghts pass, and others be absorbed or deflected. Thats why we are not toasted, and still get some light down here.
Now:
1. All heat radiation is happening in the form of Photons leaving the body radiating the heat.
2. All bodies with a temperature above absolute zero radiates heat.
With the examble of a lightbulb, it is safe to calculate with 100 % energy going into the box, unless there is a window in it.
FYI Didn't every one notice the compressor is using an ebil flammible gass ie Butane ;).
I think you're going to have some load issues.
Yep, I think there might be some load issues... I hope to get enough buffer headroom to get some decent continous low temps.
Getting closer to do a real test, pretty much finished now, only minor adjustements to do before I can take it down to my office/ kitchen :)
And... its in my living room, testing it out...
Had some windows issues ( from earlier on) , but they should be resolved now, so I can start testing the system for real pretty soon.
Had it up an running on a Spec viewperf benchmark for an hour or so to test how it handles load, and it seems to be kind of ok. Tempratures gets stable in the -7.5 to -8 area. This was with just a mild overclock, and a little extra voltage, so I am not sure how it will react to more oomph.
My delta t between air temp and cpu temp is somewhere near 25 degrees, so, cpu temps is less than impressive at the moment :) but I think I can get it a little better by some adjustments.
Temp regulation of the system is kind of difficult, as the thermostat shuts on and off with a delta T of 15 degrees. I have to take a look at this to see if I can get that window a little narrower. As it is now, it shuts off at - 25, and on at -10. I dont want it to get much colder than -25, but my buffer Ice has a freezing temp of - 12, so i would like the system to start at -15 or lower.
Now , for tonight im gonna test some more, and also try switching the fans over to the 12 volt rail to see if added air flow has any effect.
Ill try an post some pictures soon :)
if u find it unsatisfactory when done mod it in a submerged system!!!
gretings,
q
I have thougt about that, and find it kind of intruiging, but its kind of messy...
Thinking about making my own heatpipe cooler (mostly to test the principle) for the cpu... hehe..
Anyone tried making their own heat pipe?
You'd need a vacumer to be able to produce a heatpipe wouldn't you? That kinda stuff is expensive to say the least. I know overklokk has one, he might be able to help you (with that, and pretty much anything relativ to phase cooling).
Nice to see more Norwegians here ~_^ I'l keep this english out of respect for the other readers however.
I havent quote made it into the phase cooling scene yet, but I'm pretty confident I can surpass a lot of you phase cooling guys with my new übar TEC/water system thats soon done =)
-Stigma
Looking pretty good.
Maybe in complexity, but not in temps ;DQuote:
Originally Posted by Stigma
Well... I beg to differ. In a 30-45 days or so I can probably give you some proof ~_^
And please note however, I did say I'd beat -some- (EDIT: well no, to be presice I said "a lot") of the people using phase cooling. Not -most- and certainly not -all-. I am quite aware of the limitations of TEC cooling, but I do expect to be able to hit full load temps of somewhere between -15 to -25 (calculating with an overclocked conroe as load). I've seen plenty of DIY phase cooling projects that don't hit sub-zero temps at load on the other hand.
Theres more to my new cooling system than just a regular water loop + a TEC, otherwise I wouldn't be mouthing off ~_^
-Stigma
Quote:
Originally Posted by Stigma
good luck !
Boldy spoken stigma... please let us know of your results :) Maybe those 400w tec might do the trick :)
Stigma, best of luck! Been thinking abot pelts a lot, but found them to be a little powerhungry.
did some testing yesternight , and going from 5v to 12 in the exhaust fans dropped the load temp to -15. I then upped the voltage and fsb to 1,75 and 222, and the temps got stable in the - 10 area . Cpu temp did a big jump, abot 10 degrees.
When I get the cpu load resonably stable, Im gonna start optimizing the gfx card.
Also gonna put in a couple of extra fans, and run them on low woltage to get up air flow.
Thanks for the support. I'l try my best to put my money where my mouth is =P As I said, hopefully in a little over a month I should have most of the stuff in place. if not, then there will atleast be put up a worklog detailng the stuf I have ready so far.
mindnumb1: Yea, they ARE power hungry. Not only that but also terribly inefficient hehe. I would have gone for direct phase-cooling for my CPU if not for the fact that it isn't possible for me to physically fit the stuff through my wall. You see on the other side I have my server and all the stuff that makes noise, and breaking the perfect silence (Project Freeflow will have no moving parts except for the heavily padded harddrive), with the buzz of a compressor is just not an option.
Maybye at some later point I will take the step up to phase-cooling some day(In fact the likelyhood of that is quite high), but that will require a bit of "modding" to my wall hehe, and I'm not quite ready for that just yet =)
-Stigma
Well how about a quazi phase change, Heat pipe cooled pelts? I am planning to do some experiments on homemade heatpipes, ( wich is phase change cooling, Woo hoo!) Ill let you know if its possible to make it work eficiently.