Exploring Below Ambient Water Cooling

[4Ryan6]

I'm not sure that cooling method is he using, but I´m not sure if can be easy get a 0°C with chilled water.

You can actually go below 0c but to do it requires a Rock salt solution same as makes Ice Cream, but a salt solution would not be good to be pumping through a system, so that was a bad reference on my part, and could not be long term maintained either even with a rad slush box setup, like rubix_1011 has previously used, My Bad!

I can however take my TEC cooling down to 0c, if I added anti-freeze and insulated the motherboard.

 

[thequn]

LOL funny because that was my original intent when making this system but it kinda failed because my ice chest were too small for my radiator. Dam them being too small on the inside... yeah I have 5 pound of ice cream rock salt sitting in my closet from when I was gonna do that. so I guess I forgot to order tubing. All i got what my lite and the new houseing for the pump to remove then other bay rez from the system.

 

[thequn]

Oh yeah I added UV dye Sorry Ryan I know you hate dye LOL.

 

[4Ryan6]

Oh yeah I added UV dye Sorry Ryan I know you hate dye LOL.

It's only because of a leak or a spill and what some additives can do to carpet, floors, and desktops, and animals that discover the leak and lap it up before you can clean it up.

Steam distilled water doesn't hurt anything and at one time I was running 10 gallons of water in my cooler and had to keep a constant eye on it's fittings, it was outside of the computer so no problems there with a possible leak, but wouldn't help the flooring that's for sure.

What are your future plans, now that you know this works for you?

 

[thequn]

I guess make it bigger. honestly I have no idea. I think ill keep this until it get annoying to refill the ice all the time. then jump in to a new cooling experiment.
i have been wanting to try braking my Air conditioner open but ill wait till the winter when I don't need it for my room... lol.

I also have a mini fridge I can use outside hummmm....

 

[4Ryan6]

I see you've already found the TEC cooling thread, I was going to recommend you take a look at it if you hadn't.

Using the TEC cooling solved the never ending Ice swapping monotony, in the beginning it's not a problem but does get old after a while, it also dropped the monthly electrical use by $50.00 as I was using 2 freezers to freeze the ice.

TEC cooling also yields more solid dependable temperatures especially if say you're gaming and have not noticed your ice has melted, but your overclock is still pumping heat into your reservoir that has now gone way above ambient.

TEC cooling is always operating, always cooling, so the ice worries are a thing of the past, but the ice cooling is a very functional cooling and allows stable overclocks you just cannot expect traditional radiator water cooling to handle.

Just some points to consider.

 

[4Ryan6]

Does thegun have a name you'd like to share?

You've already taken some first steps and discoveries with the ice cooling that will put you in a position of knowing what it can do for you.

Additionally you're experimenting in territory that allows much more cooling capability than air or traditional radiator water cooling does.

Most that have seen this thread take the Ohhh and Ahhh approach, like cheerleaders at a game, but never actually do it, so they never learn what you've already learned to get to this point of your own cooling discoveries.

That's why I asked what your future plans are, it's exciting to me actually seeing someone use this information.

So will you be satisfied going backwards from this point, after what you've already discovered?

 

[thequn]

I been looking at the math... lol I cant count... lol
Am I am gonna order a second D5 pump and put the gpu on a rad and just use the ice for the Cpu. This should normalizes the the process and stop my refrigerator from freaking out. lol. i also need to insulate the dam tubes I keep forgetting to.
Then maybe in a month or two ill drop a TEC I have a Noctua NH-D14 sitting in my Amd rig i can move over would need to find an old pos water off of ebay it.

Yeah I have read every post on both of theses I wish I could find Moto post about his Pelter cooler but I have not located it yet.

 

[thequn]

oh dear so i split the loop, gpu is now own own pump and rad.


I mounted my cpu pump to the fan up top lol.



so i finally after taking the gpu off the ice line finally the temps dropped to low 🙁((((

Time for insulation

 

[thequn]

Ps, I learned to thumb nail photos.

 

[4Ryan6]

@ thegun, A D5 pump should always be located below the reservoir so coolant is always supplied to it.

It is not designed to run dry to self prime itself like you have it setup now.

You're risking burning up the ceramic bearing it is never supposed to be run dry at all and your new setup has it pulling the coolant into the intake, that's not good!

The D5s are circulation pumps not self priming pumps, never ever run them dry.

 

[thequn]

LOL, @ this Ryan, Don't worry I know what i am doing on this part, I just sucked the water though before ever turning on the pump. it works fine, and yes your right for the non experienced user. LOL

 

[4Ryan6]

LOL, @ this Ryan, Don't worry I know what i am doing on this part, I just sucked the water though before ever turning on the pump. it works fine, and yes your right for the non experienced user. LOL

Good!

My concerns are gone!

Pretty Smart on your part! Ry

 

[thequn]

Finished my little set up all insulated with kneaded eraser as well other stuff now no more condensation. did notice that my gpu block is not getting perfect contact with the die for some reason, fixed as best as possible unable to clean inside due to one of the screws wont budge.

 

[thequn]

ps took me 3 hours to remove the indigo xtream metel TIM I suggest not using it.

 

[4Ryan6]

ps took me 3 hours to remove the indigo xtream metel TIM I suggest not using it.

Regarding liquid metal, should have read the Thermal Compound Roundup I did.

Indigo Extreme is same, same!

Plus the temperatures you have to run it at for it to flow and seat properly are temperatures my TEC chilled water cooling won't even allow you to reach.

 

[thequn]

oh its worse then tat Ryan it didn't even set correctly lol

LOL i totally agree with your statement about liquid metal TIM's


i am currently using Arctic Silver Ceramique 2. i love that stuff, seems to be the favorite for extreme cooling solutions i consider this a bit extreme LOL

 

[4Ryan6]

If you need a thicker but easy to spread TIM extreme heat durable like for use on the hot side of a peltier, Timtronics Grey Ice 4200.

If you need a thinner consistency TIM that compresses to the thinnest layer achievable, absolutely perfect for GPU dies, Artic MX-4.

These are the best of the best through all of my testing after the thermal compound roundup was completed and further testing was done.

These are the thermal compounds I spend my money on.

 

[saint19]

I think that you would get better results with Gelid GC2 or Gelid GC Extreme...just my two cents 😉

 

[4Ryan6]

I think that you would get better results with Gelid GC2 or Gelid GC Extreme...just my two cents 😉

I actually have some Gelid GC Extreme on hand an unopened package I have never used, they sent me the TIM when they shipped me the fans for the cooling fan roundup. It's not like I'm short on thermal compound even though I have given away quite a bit of the leftovers from the thermal compound roundup.

If you say it's that good German, I'll have to open up the package and give it a try.

 

[saint19]

I think that you would get better results with Gelid GC2 or Gelid GC Extreme...just my two cents 😉

I actually have some Gelid GC Extreme on hand an unopened package I have never used, they sent me the TIM when they shipped me the fans for the cooling fan roundup. It's not like I'm short on thermal compound even though I have given away quite a bit of the leftovers from the thermal compound roundup.

If you say it's that good German, I'll have to open up the package and give it a try.

Go ahead, you will get very good results. I use those for LN2 and the bets thing is: Not cure time need...

 

[4Ryan6]

If you need a thicker but easy to spread TIM extreme heat durable like for use on the hot side of a peltier, Timtronics Grey Ice 4200.

If you need a thinner consistency TIM that compresses to the thinnest layer achievable, absolutely perfect for GPU dies, Artic MX-4.

These are the best of the best through all of my testing after the thermal compound roundup was completed and further testing was done.

These are the thermal compounds I spend my money on.

I think that you would get better results with Gelid GC2 or Gelid GC Extreme...just my two cents 😉

I actually have some Gelid GC Extreme on hand an unopened package I have never used, they sent me the TIM when they shipped me the fans for the cooling fan roundup. It's not like I'm short on thermal compound even though I have given away quite a bit of the leftovers from the thermal compound roundup.

If you say it's that good German, I'll have to open up the package and give it a try.

Go ahead, you will get very good results. I use those for LN2 and the bets thing is: Not cure time need...

I recommended the above Timtronics Grey Ice 4200 and Artic MX-4, because I had comparative tests all the way back to the thermal compound roundup, do you have any actual test results comparing the TIM from Gelid with other know TIMs.

The only way to actually know how good any TIM is besides our own opinions is actually testing it against others?

Gelid was actually invited to participate in the Thermal Compound Roundup but they failed to submit a testing sample.

The one thing I learned from the Thermal Compound Roundup is that other major name brand companies that are actually producers of thermal compound are whats actually in the tubes labeled as other companies, which means some of the companies like possibly Gelid don't even manufacture thermal compound at all.

The same as some power supply major brands are not actually building the power supplies, it would be interesting to discover who is supplying what to whom, because if you look at the thermal roundup which the tests were conducted on an even controlled environment, the TIMs are all very close in conduction performance.

Just remember this German, there is no substitute for "Kraft Miracle Whip Salad Dressing!" :lol:

 

[thequn]

LOL i seen a lot of test of a lot of tim's and the funny thing is every test is dramatically different esp the Mayo test that beets 90% of the TIM out there.

 

[cbrunnem]

LOL i seen a lot of test of a lot of tim's and the funny thing is every test is dramatically different esp the Mayo test that beets 90% of the TIM out there.

thats because most of the test you see online are basement test of sorts. they are statistically inaccurate as the sample sizes used are normally not high enough and the methods are inherently innaccurate. to get a good result you would need to use something more accurate then an actually cpu and heatsink. now doing something like that would not give you real word number but rather comparative numbers against other TIMs. now thats not to say that most online results you see arent good just not accurate or precise. theres no way you could be precise with a cpu/motherboard/heatsink combo.

 

[4Ryan6]

LOL i seen a lot of test of a lot of tim's and the funny thing is every test is dramatically different esp the Mayo test that beets 90% of the TIM out there.

thats because most of the test you see online are basement test of sorts. they are statistically inaccurate as the sample sizes used are normally not high enough and the methods are inherently innaccurate. to get a good result you would need to use something more accurate then an actually cpu and heatsink. now doing something like that would not give you real word number but rather comparative numbers against other TIMs. now thats not to say that most online results you see arent good just not accurate or precise. theres no way you could be precise with a cpu/motherboard/heatsink combo.

What really matter to us, the users of the products, are the results from cpu/motherboard/heatsink combo testing, in hopes that we get the same results for ourselves.

Even those results can be different if the application of the TIM is done improperly simply because the user is clueless as to what the TIM can and cannot do regarding how it is applied.

Too many are just ignorant as to how to apply TIM in the first place as is proved by whats under some graphics card heat sinks applied to the GPU die, some look like the TIM was applied with a caulking gun.

And these are the ones that are supposed to know what they're doing, applying thermal compound from the graphics card manufacturer!

All along TIM is only to fill the microscopic imperfections between the 2 contacting surfaces, it was never designed to be applied as a full layer, and the best results have everything to do with the TIMs application.

Even when we're talking the same TIM tested by different testers with varying results, pertaining to using too much in the application, uncontrolled ambient for air cooling tests, and uncontrolled water temperatures for the water cooling tests.

You can trust the tests I ran to give you the same results as long as the sentence above is the same, and I know they were not laboratory conducted, but usually that's the same excuse the brand holders give when their products don't rock the world in reality, when all their tests were strictly controlled to their marketing advantage.

Which does happen!

Since we're on this subject you might find this interesting.

http://www.tomshardware.com/forum/1738434/last-read

So cbrunnem, How have you been doing?

It's been a while since you've posted in this thread, my Peltier cooling has been in operation now for over 8 months and is working very well.

Hope you and your family are all well! Ryan