What's the coldest chiller someone's made?

[dangerous_squid]

Alot of threads talk about what liquid/pumps are good to keep liquid as low as -80 C, but I haven't seen any threads about a chiller keeping a cpu near those temps even unloaded. Is it possible to make a water chiller that can compete with a ss direct die in terms of temps?

Does liquid simply not have the thermal capacity of refridgerant?

[n00b 0f l337]

Not really without cascading something. Its just direct die has much less "heat exchangers" to deal with. Chillers have many more.

[Kunaak]

NesLabs, sells chillers that are advertised to reach well below 100C.

but thats chillers roughly the size of a fridge, and meant for medical labs.

I can't find any of the monsters at the moment, but heres one of the controllers for some of the neslabs ones.

http://cgi.ebay.com/NESLAB-cryotrol-...QQcmdZViewItem

[eshbach]

the coldest one i saw on the forums was pc ice's at about -55C i think...

unfortunantly all his posts are gone.

[SexyMF]

I only achieved -27C loaded temps (coolant). Unloaded I reached under -50C. There is always the tradeoff between capacity and low temps.

You can't compete with DD rigs. The waterblocks aren't optimised for such cold coolant.

[ixtapalapaquetl]

The waterblocks aren't optimised for such cold coolant.

Interesting. What physical changes would be necessary for such an optimization?

[SexyMF]

At a guess I'd think a thinner base and lower flow rate, less turbulent path may benefit.

[[XC] MarioMaster]

the coldest one i saw on the forums was pc ice's at about -55C i think...

unfortunantly all his posts are gone.

I saw that Pc-Ice was banned, how come?

[catkicker]

I saw that Pc-Ice was banned, how come?

Probably best to stay on the topic.


Would be cool to see someone design a block for a chiller.

[Sneil]

I saw that Pc-Ice was banned, how come?

actually I'm interested as well. He was a good guy and as amaizing at phase change as chilly1. What happened? Or is it a taboo subject..

On topic, I think he want's to know if a cascade chiller is possible and if anyone's attempted one. What liquid could you use that goes to -80 or lower?

[M.Beier]

actually I'm interested as well. He was a good guy and as amaizing at phase change as chilly1. What happened? Or is it a taboo subject..

On topic, I think he want's to know if a cascade chiller is possible and if anyone's attempted one. What liquid could you use that goes to -80 or lower?

Isnt that what misteroaster got??

[[XC] MarioMaster]

Hmm....a cascade chiller....I never really thought about the idea, some people use denaturated alcohol as that stays thin at low temps.

[dangerous_squid]

As far as liquids there are specially engineered liquids that can go down to -110 C, mantain good viscosity, and are much better heat transferers. Of course, they cost a bundle.

More specifically I guess I was curious about if/where there are inefficentcies in liquid HX.

Also putting 3 evaporators into a resevior, would triple it's heat removing capacity (pulldown), but not change the max lowest temperature, correct??

[[XC] MarioMaster]

You are correct in that statement, but I believe what he was referring to was to have a cascade phase-change system cooling the liquid. The first phase-change unit would cool the second one, making the second stage be able to produce a lower temperature. People do this all the time in direct die (DD) systems but is uncommon in chillers.

[boshuter]

Here is a pic of one of the Neslab cascade chillers Kunnack was talking about. I have 2 of these units, they are rated for -110c and are immersion chillers. They have a "wand" on the end of the hose that is made to be immersed in the liquid. Both of mine are going to be converted to direct die, a very simple task

[Pirate_Freder]

Probably best to stay on the topic.


Would be cool to see someone design a block for a chiller.

well if someone could direct me to information to figure out or inform me of what would be optimal for a waterblock at such cold temps i might be able to come up with somthing . i have been studieing engineering for the last 4 years sooo...

[epion2985]

well if someone could direct me to information to figure out or inform me of what would be optimal for a waterblock at such cold temps i might be able to come up with somthing . i have been studieing engineering for the last 4 years sooo...

-all copper
-no seals (brazed shut)
-1/2" inlet/outlet
-profile that is easy to insulate
-low restriction
-high magnitude of heat transfered from block to water per unit of time (forget the technical term, been a while since my last class )

[Pirate_Freder]

all righty . also, lower flow rate would be better right?(i recall reading somthing). make it for tons of heat transfer over more time

[fatty]

I use an old maze 4 block all copper when I test my chillers but so far only looking at -30 ish but I think that was more to do with the fact my eheim 1250 kept slugging up

[epion2985]

all righty . also, lower flow rate would be better right?(i recall reading somthing). make it for tons of heat transfer over more time

Well you want to remove as much heat as fast as possible. Simply the goal is to keep an object under a heat load X as cold as possible. You have almost 4 years of engineering schooling on me, so you tell me how to go about it.

Correct me if I am wrong but forcing the coolant though a restrictive maze at high pressure is good right? Something to do with the way the coolant contacts/impacts the block internals?

[ilkkahy]

Here is the waterblock i intend to use with dryice chilled mehtanole:



Originally i meant it to be tested with 2x172w peltiers but i will try it out with methanole when i get other parts made for the system too. The inlet diameter is 18mm and waterblock surface area is over 600cm^2. After the projects i will use it in my watercooler.

[Xeon th MG Pony]

be care full with methanole it will absorb through the skin and is leathal in high dossages, just in case you didn't know same with ethylen Glycole.

[epion2985]

What I really want to try is the storm block. With coolant that wont get thick around -40C like denatured alcohol and a powerfull high pressure pump it should work well. Only downside with denatured alcohol is it takes forever to pull down.

I guess my search for ideal coolant is still on.

[Pirate_Freder]

yeah, high pressure is good IF! the liquid is thin, if gooey well...never really cared enough to think about that :P, since i want to avoid that. i was just saying i recalled mention on these forums of slower being better, to my knowledge this is false but i wanted to check, probabaly was just said because they were running anti-freeze in the mix. i've got some ideas running through my head though, hopefully will get some 3d models, or at least 2d done and posted soon. and on my schooling, i didn't say engineering shooling, i have been learning drafting/modeling/engineering in one class for 4 years and have dabbled outside of it to with other education and some nifty designs and such. don't wan tyou thinking i'm some pro-engineering guru :p, should still come up with some good stuff though

[ilkkahy]

If -60c temperatures are enough then i suggest freezium to be used. However if i remember right aluminium shouldnt be used in the loop and system should be rid of air completely.

However with dryice i will use methanole since it is easy to get and doesnt freeze in -79c. Acetylene would make cold mixture with dryice but im under impression it makes worse fumes than methanole.

About the high pressure pump. I think benefits from effective pump are mostly due the turbulence which increases the effectiveness of certain type of waterblocks. Also if there is very low resistance in block there might get warm spots to some areas of the waterblock. My pump power can be varied between 30-45-60w so there will be pretty huge flow with this system due the big inlets and pump.