unique waterblock question

[r0ckstarbob]

http://www.pcrivals.com/projects/computerator/images/nipplet.jpg

http://www.pcrivals.com/projects/computerator/images/heatsink.jpg

http://www.pcrivals.com/projects/computerator/images/cpu.jpg

http://www.pcrivals.com/projects/computerator/images/cpufrost.jpg

now here's the question

why would this be a bad design for a waterblock?

maybe it's a silly question but i can't seem to get a straight answer from the other forums i'm on. no this isn't my *ahem*waterblock (the term waterblock is being used loosely here. in fact, these images are from a really facinating article where these two guys went to watercooling with alchohol at around -17C by installing their their computer in a cube fridge. you can check it out here
http://www.pcrivals.com/projects/computerator/index.shtml

now keep in mind, the system i'm designing has a 175gph pump quite literally about 2 inches from my CPU. travel distance from the pump to the CPU is actually 5 inches horizontally - no vertical climb whatsoever. (its in a desktop case)

I'm pumping some VERY cold liquid (alchohol at -27F or so) directly FROM the reservoir TO the CPU, and then from the CPU back into the reservoir which stays at around -27F or so. - (travel distance back to the 3 gallon reservoir is approximately 15 inches horizontally. again, no vertical lift whatsoever)-

so anyways, whaddaya think? is it a bad design, a good design and why? why are the other waterblocks better? i'm brand new to watercooling at this level (subzero that is) so am asking alot of questions before i do it.

thanks

RSB

[Burning Phoenix]

Looks to me that it should work fine except i would hope the bottom of the copper cap is perfectly flat for got contact between the it and the cpu. I'm not sure if it is also important to completely cover the die with the copper cap for more heat absorption by the waterblock. I wonder though how it is being held securely to the cpu for it may be heavy using glue.

[sfa ok]

It's not bad design, it's just not the best design. If you got a Danger Den (http://www.dangerden.com/) Maze 2 block, the water will be pumped directly above the cpu die, which would be really nice with water that's chilled that much.

Remember to protect yourself from condensation, especially on the tubes.

[Warlord2]

I always thought the more surface area the better

I was thinking of that same design though but I didnt know if it would be good enough

my goal is the beat the maze2 block=/

its going to take me a while but I WILL DO IT!!!!

[Patchmaster]

I would think the only potential problem would be with how the coolant flows through the chamber. This design looks a lot like the beehive cooler (http://www.spodesabode.com/view.php?pageid=block) created by Dr. Surlyjoe. But his has the connectors arranged to create a whirlpool effect inside the cooler. This ensures there are no dead spots where water isn't actively circulating. I'm not sure the top/side connector locations are the best for elminating dead spots. They could be okay, but without some serious testing tools it would be hard to know for sure.

BTW, water is MUCH better than alcohol at transferring heat. I gladly bow to your superior experience in this area, but you might get even better performance if you used a water/alcohol mix. (Obviously, you need enough alcohol to keep the water from freezing.)

[Phil]

It looks a lot like one of surlyjoes designs

Beehive Waterblock (http://www.spodesabode.com/view.php?pageid=sdblock)

[r0ckstarbob]

Originally posted by Patchmaster
BTW, water is MUCH better than alcohol at transferring heat. you might get even better performance if you used a water/alcohol mix. (Obviously, you need enough alcohol to keep the water from freezing.)

yeah... i know. if you can point me the way to something better then alcohol that won't freeze at -27F (-33C) i'm all about it believe me. something that has a freezing point of -50 to -70C... if you know of something and can send me a link or some info or whatever, you would definately have my eternal gratitude. i mean i've been racking my brain on this one for weeks trying to think of a better solution that wasn't

A.) illegal
B.) didn't require me to sign over my first born child or take out a morgage on the house to pay for it.

alcohol is it as far as i can tell. water just freezes, point blank. at that temp even with the alcohol mixed in... you get icy rubbing alchohol. i thought about salt water but ruled it out almost immediately for it's corrosive properties. maybe bleach or ammonia? i dunno. i'm sunk here...

ideas?

wait... don't they use pure ammonia for industrial refrigeration units? hey, wow, that might be the ticket...

hmmm... things to think about

oh and sfa_ok...

check it out. i've got the condensation problem licked and its using a technique that i've never heard of anyone else using. go here and check it out. i think you'll really be impressed.

http://forums.overclockers.ws/vb/showthread.php?s=&postid=237118#post237118

and it works. without fail.

thanks guys, for all your input.

RSB

[Phil]

I'm not sure how low exactly but I know that the freezing point of water is lower when it is flowing, so experiment with differant alchohol/water mixes to see how low you can get the temp before the water does start to crystalise

[dimmreaper]

You know they call anti-freeze "anti-freeze" for a reason. You should try it sometime. I've had it mixed 1:1 with water in my truck at temps of -40F and you know what? It didn't freeze! Perhaps mixing it 1:1:1 with water and alcohol would work well even in sub -50C temps . . .

[William]

Originally posted by dimmreaper
You know they call anti-freeze "anti-freeze" for a reason. You should try it sometime. I've had it mixed 1:1 with water in my truck at temps of -40F and you know what? It didn't freeze! Perhaps mixing it 1:1:1 with water and alcohol would work well in sub -50C temps . . .


yeah, i was going to say that. There is a chem formula that I can dig up to calculate how much antifreeze and water you need it to keep it from freezing at X temp. Give me temps and a volume and I will figure it out(i hope)!

[joshiieeii]

Originally posted by dimmreaper
You know they call anti-freeze "anti-freeze" for a reason. You should try it sometime. I've had it mixed 1:1 with water in my truck at temps of -40F and you know what? It didn't freeze! Perhaps mixing it 1:1:1 with water and alcohol would work well even in sub -50C temps . . .

I agree with Dimmreaper, If you use a 6:2:1 ratio of Water, Anti-freeze, and Red Line's Water Wetter. I think you could really achieve something.

Water wetter is a very cool additive, if you haven't already tried it, I reccommend it. It boasts 20C cooler temp than just regular anti-freeze, improves the heat transfer, and helps solve that corrosion problem.

*shrugs* I would think that even in Antartica they are using some kind of fluid to run around the cylinder heads to distribute the heat (;) heh, what little there is at -60C)

[Patchmaster]

if you can point me the way to something better then alcohol that won't freeze at -27F (-33C) i'm all about it believe me. something that has a freezing point of -50 to -70C... if you know of something and can send me a link or some info or whatever, you would definately have my eternal gratitude. Like the others said, I'd give standard antifreeze a try. At the rate you're pushing the coolant around the only place you're likely to have any problem at all with freezing is in your reservoir. In the hoses and waterblock the water's probably moving too fast to have a chance to crystalize. Antifreeze should work fine. It will reduce the overall cooling capacity, but something tells me that isn't your biggest problem.

As for more exotic coolants, I came across a web site a couple weeks ago for a distributor of coolants for extreme situations (mostly very cold or very hot). I don't recall the URL. I think I was looking for the specific heat values for various coolants. You might try searching for something like that. Anyway, water was far and away the most efficient coolant they listed. Most of the exotic coolants were well less than half as efficient as water.

[r0ckstarbob]

hey guys, just wanted to say thanks for all the input. currently i'm doing some research into all this stuff, but you guys have definately given me some good stuff to go on.

oh, and just a note

dimmreaper antifreeze works absolutely terrible for heat transfer. in fact it's alot like liquid styrofoam in that it changes temperature very little and very slowly. yes it will keep water from boiling or freezing, but it's more of an insulator then anything. its most attractive qualities are that you can mix it with water and it's temperature resistant. thats oversimplifying things a bit but thats the jist of it. run a car on nothing but antifreeze and it'll overheat before you can get to the grocery store. mix it with water and you can use it fine. the more antifreeze to water ratio though, the less effective water becomes.

water IS the best heat transfer liquid out there. treated with WaterWetter and the heat transfer efficiency of water increases %20-30 if you can believe it.

for all of you running watercooled systems at all, you should definately go down to your local automotive store and find this product. you are definately going to increase the efficiency of your setup.

joshiieeii - the more i read and learn the more i think you're on the money when you're talking a 6:2:1 ratio of water-antifreeze-waterwetter but i want to find something better then antifreeze. i'm back to thinking rubbing alcohol instead of antifreeze but i'm not set on anything. i'm doing ALOT of looking (clocked almost 10 hours already just today on the web just reading and looking). water has to comprised of at least 20% antifreeze to keep it from freezing. but that means that 20% of the liquid isn't doing anything except insulating the water from going to ice. if i can find a 3rd compound that will not only reduce the freezing temperature of water but also affect SOME measure of heat transfer in the process, i think i'll probably go with it.

William i am going to take you up on your offer for a gunk to water ratio!!! after doing a BUNCH of reading, i want as much water in my system as possible and want to use as little additives (alcohol, WaterWetter) as i have to.

(for those of you curious as to the architecture of this little monster) heres what we're looking at.

it's got a sandwich architecture and everything is mounted horizontally. the copper evaporator coils (the refrigeration coils) are on the very bottom of the sandwich. sitting directly on top of those refrigeration coils is the copper reservoir. on top of the copper reservoir is the motherboard mounted flat against it.

William : The reservoir measures out at 370mm long x 260mm wide x 130mm tall and will hold 3.4 US Gallons at (ideally) -27F. so if you're doing the calculations, shoot for -40 or -50 F if you can.

the warm water leaves the CPU and enters the reservoir at the back end, furthest away from the pumps. inside the reservoir there are a bunch of copper walls (or ducts) that will absorb heat from the liquid and route the liquid back and forth and back and forth... continuing to absorb heat the whole way, until it reaches the front of the reservoir where the pumps are housed. once it hits the pumps, it gets shot back into the CPU waterblock and the process repeats.

the whole system runs at -27F or so (-33C). remember, that copper reservoir sits right on top of those big ol copper refrigeration coils.

using one RIO 120GPH pump (sorry, i know i said 175 before, i was wrong, my bad) it takes 90 seconds to cycle all the liquid in the reservoir which is actively being cooled by the refrigeration coils beneath it.

*whew*

well this turned into a lengthy thanks, but i'm all full of words and nobody i know personally gives a sh** about any of this and thinks i'm insane so i have nobody to talk to. just a big notebook with 130 pages of notes and concept sketches. so this is the only place i get to talk about it. i won't lie. i'm stoked to be working on this. i'm learning ALOT about all KINDS of things. at 25 years of age, without being in some kind of school, there aren't alot of ways for a person to do that on their own ya know.

anyhow, thank you very very much for your input. i'll post something when i find a solution. keep the ideas comming. they're very helpful.

RSB

[r0ckstarbob]

oh, and waterwetter lowers the freezing point of water from 32F to 31F. no appreciable difference.

[r0ckstarbob]

alrighty, after 14 hours of searching the web and comming up with zilch i finally emailed the guys at WaterWetter (redline.com) to see if their product will work in a water/ethanol solution. Ethanol is Grain Alcohol and is alot more potent then either Isopropyl (rubbing alcohol) or Methanol (wood alcohol) so i shouldn't have to use as much of it per gallon of water.

freezing temps

Ethanol = -173 F
Isopropyl = -126 F
Methanol = -114 F

we'll see what they say

do we have any chemistry majors in the house? can anyone come up with a ratio of Ethanol to Distilled Water in a 3.4 gallon tank that will keep the freezing point down to around -40 or -50 F? i want to keep as much water in the system as i can so don't want to use any more alcohol then i have to. water is some of the best stuff out there for heat transfer. my system is only going going to go to -30F or so but want the coolant to have about a 10-20 degree margin and not operating AT it's freezing point if that makes sense.

thanks

RSB

i love my little obsessions... thanks for indulging me

[joshiieeii]

I still say water would be a better choice.... I mean look at the amount of heat water can hold Vs. the amount of heat any of the alcohols can hold. Water wins hands down. Plus you are not going to have to worry about a leak that can BLOW :eek: up your puter, hehe.

[r0ckstarbob]

no, you misunderstand me. i DO want to use water. as much of it as i can. but have to cut it with alcohol just enough to get the freezing point down below -30F. thats why i'm going to take william up on his offer if it still stands so i can get that ratio and not use any more then i have to.

and besides, as long as it's less then 120 proof i'll be fine in the volitility field! :D :D

never had a flaming doctor peper before???

*chuckle*

**shudder**

maybe i should just do the whole darn thing with Vodka...:eek: