Chilled Water Cooling Version 3.0

Silver Surfer · Jul 10, 2017

http://www.overclock.net/t/1633988/chilled-water-cooling-vs-3-0-build-log

Silver Surfer here, OCAddict at overclock.net, check out my new Chilled Water Cooling 3.0 Build Log at OCN.

The goal of version 3.0 was simply to make a better cooling solution than what I was already running with version 2.0, and that goal has been realized!
Version 3.0 uses 1/2 the electrical power that 2.0 did when all 3 TEC assemblies were powered up.
Version 2.0 with all 3 TECs powered pulled 600w at 60a, version 3.0 pulls 300w at 24a.
Version 2.0 took 25 minutes from cold boot to reach the set 8.9c cutout point of the temperature controller, 3.0 takes 6 minutes and 15 seconds.
Version 3.0 is so quiet operating I had to keep checking to see if it was running.
Version 3.0 exhausts a lot less heat into the room.
Version 3.0 can support the 7700K overclocked to 5.2ghz with hyper threading enabled running Intel Burn Test and still lower the coolant temperature while the test is running all the way down to the cut out point at 8.9c.
Version 3.0 uses approximately 1/6th of the coolant volume that version 2.0 did.
Electrical load during surfing the net 2.0 used 400w constantly, 3.0 uses 100w with the 2nd 100w TEC kicking in occasionally for about a minute or so.
Version 1.0 and 2.0 both used a 2nd power supply to power the chilled water cooling, version 3.0 uses only one power supply to power everything.
Plus the chilled water cooling of version 3.0 is all in one case not 2 separate cases.
3.0 is definitely a better chilled water cooling solution.

Check it out. Ryan

Reefa_Madness · Jul 28, 2017

Hello Ryan,

Good to see your chiller continues to be refined. I've took a peek at your previous thread (ver2) over a OCN a while back. Nice job with merging the two cases and with your other refinements. I guess that I'll have to go check out the update to see where you've gone with this.

For the members here on OCF, Ryan's project is worthy of a look...you might just find yourself thinking outside the box a bit.

Robert17 · Jul 28, 2017

Nice work.

Silver Surfer · Jul 29, 2017

Reefa_Madness said:
Hello Ryan,

Good to see your chiller continues to be refined. I've took a peek at your previous thread (ver2) over a OCN a while back. Nice job with merging the two cases and with your other refinements. I guess that I'll have to go check out the update to see where you've gone with this.

For the members here on OCF, Ryan's project is worthy of a look...you might just find yourself thinking outside the box a bit.

Hey it's great to hear from you, I hope all is well with you?

Version 3.0 has some major improvements and so far is working flawlessly except for one of the Thermalright 140mm cooling fans froze up.

It was an internal bearing problem, but I replaced the fan the same day, luckily I had an exact specification Thermalright fan on hand, it's a different color, but the RPM speed, power draw, and sound level, are the same.

So far version 3.0 has surpassed all my expectations, but it was a challenge getting all I did inside one case, worth it in the end though!

Take care, Ryan

maxfly · Nov 21, 2017

wow! thats a ton of work! great job man, love the old dd block for your cpu. very clever! i think i have one in my old wcing box...somewhere? btw the link to your tec supplier is 404ing.

Silver Surfer · Nov 21, 2017

maxfly said:
wow! thats a ton of work! great job man, love the old dd block for your cpu. very clever! i think i have one in my old wcing box...somewhere? btw the link to your tec supplier is 404ing.

Thanks I fixed the link.

Silver Surfer · Jan 19, 2018

The setup is now cooling a 6 core i7-8700K Coffee Lake CPU off the shelf from Newegg, that so far has not been delidded and is overclocked to 5ghz stable on an ASUS ROG Maximus X Hero motherboard with hyper threading enabled. I have corrected the listing in my sig but for some reason it has not shown the corrections yet.

Deous · Jan 28, 2018

What exactly is used for cooling? Thread is bit long and shows some variants of build.
You using water + tec or just tec elements with fans?

maxfly · Jan 28, 2018

the title of the thread is
Chilled Water Cooling Version 3.0

Silver Surfer · Jan 28, 2018

Deous said:
What exactly is used for cooling? Thread is bit long and shows some variants of build.
You using water + tec or just tec elements with fans?

Well I thought it was 100% obvious, water is the cooling medium that transfers the cold produced by the peltiers to the insulated reservoir and then to the pump which pumps the coolant through the CPU water block cooling the CPU, and then flowing back through the 3 TEC assemblies to complete the water loop. Each TEC assembly consists of a water block on the cold side of the peltier and a Thermalright 140 True Spirit Power heat pipe air cooler to cool the hot side of the peltier and it works, and works very well.

I have been running this type of chilled water cooling, with various improvements through the years, for over 5 years now!

Deous · Jan 28, 2018

Silver Surfer said:
Well I thought it was 100% obvious, water is the cooling medium that transfers the cold produced by the peltiers to the insulated reservoir and then to the pump which pumps the coolant through the CPU water block cooling the CPU, and then flowing back through the 3 TEC assemblies to complete the water loop. Each TEC assembly consists of a water block on the cold side of the peltier and a Thermalright 140 True Spirit Power heat pipe air cooler to cool the hot side of the peltier and it works, and works very well.

I have been running this type of chilled water cooling, with various improvements through the years, for over 5 years now!

Yes that cools nice. Have you tried putting tec on top of cpu and remove heat from it with water block?

Silver Surfer · Jan 29, 2018

Deous said:
Yes that cools nice. Have you tried putting tec on top of cpu and remove heat from it with water block?

No absolutely not!

That method is completely different from the way I am using it and, requires fully insulating the motherboard socket area, and it comes with it's own problems. I am running below ambient and above condensation forming point so I don't have any problems with condensation at all. There are some that use that method at overclock.net I'm sure they'll be glad to help you.

Deous · Jan 29, 2018

Well, I will try cooling with such set. For one reason - water absorbs heat very very slow, retains heat for very long time and when cooled it gives out heat same slow as absorbing it.
Unlike solid materials - e.g. metal or ceramics that bounce heat back and forth quite fast. That means if you cool hot surface of tec with waterblock it will be cooled much longer before water heats up. Example is when you cool tec with fan and fan turns off - the heat immediately increases - with water block that is not the case.
The only problem is to have larger amount of chilled water and do not let it go above ambient. My tec freezes below 0F in 20sec with water temps 1~8C

EarthDog · Jan 29, 2018

Example is when you cool tec with fan and fan turns off - the heat immediately increases - with water block that is not the case.

sorry...what does that prove? We know water isnt as good as solids....but is a lot better than air.

Deous · Jan 29, 2018

EarthDog said:
sorry...what does that prove? We know water isnt as good as solids....but is a lot better than air.

I disagree. Water is no better than air. Eventually all goes to air anyway. Besides water is harder to cool - needs powerful fan and large radiator.
I believe tecs could be improved in future by adding materials that absorb heat to conduct electricity. Like weird stuffs with Vanadium Dioxide.
Like I said I am waiting for compressor coolers time.

Silver Surfer · Jan 30, 2018

Deous said:
Well, I will try cooling with such set. For one reason - water absorbs heat very very slow, retains heat for very long time and when cooled it gives out heat same slow as absorbing it.
Unlike solid materials - e.g. metal or ceramics that bounce heat back and forth quite fast. That means if you cool hot surface of tec with waterblock it will be cooled much longer before water heats up. Example is when you cool tec with fan and fan turns off - the heat immediately increases - with water block that is not the case.
The only problem is to have larger amount of chilled water and do not let it go above ambient. My tec freezes below 0F in 20sec with water temps 1~8C

I see you have a lot to learn the hard way, I actually started with 10 gallons of water in a Coleman cooler chilled with ice thinking thermal mass was the way to go and that volume has dropped to the 150ml reservoir I run now. I am not going to share with you what I learned the hard way as I have shared all you need to know to successfully incorporate this type of cooling, if you so desire. It would not be fair to you to cheat you out of your learn the hard way adventure. If your thoughts are aiming towards using water to cool the hot side of the peltier get ready to spend a lot more money in cooling hardware, than you see depicted in my build. You may think from the outside looking in you have a better way of doing it, and if you do go for it, don't talk about it, Do It, and then supply a link to your 100% operational build, OK.

Silver Surfer · Jan 30, 2018

Deous said:
I disagree. Water is no better than air. Eventually all goes to air anyway. Besides water is harder to cool - needs powerful fan and large radiator.
I believe tecs could be improved in future by adding materials that absorb heat to conduct electricity. Like weird stuffs with Vanadium Dioxide.
Like I said I am waiting for compressor coolers time.

Like Bruce Lee said, "It's like a finger pointing away to the moon, don't concentrate on the finger, or you'll miss all the heavenly glory!"

Translation: Don't overthink it, just do it!

One other thing comes to mind: For every action there is an equal and opposite reaction.

There are those always learning but never doing, what good is unused knowledge to the world. (That's not a question)

For any onlookers to this thread, Deous did not just stumble across this thread, he has come from another thread to share his wisdom.

Deous · Jan 30, 2018

Listen, I am not here to philosophy or argue. All I am trying to say is that it is more efficient to cool peltier with chilled water
than cool water with peltier connected to monstrous fan. We are not talking here about standard water cooling but water chilling
with other means than air fan. Good ice-maker would do or small business water chiller.
I understood it through properties of water that cannot cool solid objects as fast as tecs but it removes enough superheat from tec hot side
to create extreme cooling/freezing.

EarthDog · Jan 30, 2018

Then stop posting because that is all you are doing. I asked you keep out of this thread and keep this bantor to yours... but showing proof. Please... enough.

Dolk · Jan 30, 2018

Just utterly ****ing ridiculous. If I could hand out awards for the most insane cooling solution of the year, I'd give it to you. Just utterly redic.

But all in the awesome-I-need-it-because-OC-potential way

Chilled Water Cooling Version 3.0

Member

DRAM Guru Senior

Premium Member

Member

Member

Member

Member

Registered

Member

Member

Registered

Member

Registered

Gulper Nozzle Co-Owner

Registered

Member

Member

Registered

Gulper Nozzle Co-Owner

I once overclocked an Intel

Similar threads