Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier/TEC Hybrid Water Cooling vs 2.0
New Posts  All Forums:Forum Nav:

Peltier/TEC Hybrid Water Cooling vs 2.0 - Page 22

post #211 of 399
Thread Starter 
Quote:
Originally Posted by toolmaker03 View Post

I would like to take the time right now to make a couple of suggestions.

your using EK supremacy EVO water blocks on the cold side of the TEC's a all metal top for the cold side TEC's water blocks are only $16.00 each.
I would suggest getting two metal tops for the cold side water blocks as the TEC's are strong enough to chill the entire water block.

http://www.performance-pcs.com/ek-top-ms-supremacy-evo-full-copper-ni.html

I know that this may sound strange but I would also suggest getting 2 pounds of clay, if you pack the clay right around the TEC instead on using eraser putty it will provide better results.
you don't need a lot 1/2" wide so that the clay covers the TEC and both the hot side and the cold side base plates, and the clay will need to be a 1/2" thick. this will prevent any kind of stalling of the TEC and it will help to keep the entire cold side of the TEC cool.

I am using a lot more clay than that, but the 1/2" X 1/2" would be the bare minimum needed to get the benefits that the clay will provide, for the TEC itself. than you just add the foam insulation as you would normally.

EDIT (molding clay does not ever get hard) that is what I am using it's only $4.00 a pound

Totally unnecessary!
post #212 of 399
Quote:
Originally Posted by OCAddict View Post

Toe, Do not listen to tools regarding the metal block top and the clay you don't need to do that, Sir!

He's all about sub zero and hasn't even reached it yet, I thought you were after a below ambient solution like mine but water cooling the hot side, or have you changed your goal?

Tools is telling you to keep the cold in the water block with the metal top and insulating it but that is not even actually proven to work.

My setup stores the cold in the insulated reservoir not in the water block, if you want any kind of success close to mine you do the same.

Getting the cold produced by the peltier from the water block as fast as possible keeps the peltier producing the cold and also keeps the hot/cold balance of the peltier constantly producing cold and not stalling.

My setup has been running over 4 years and Tools setup isn't even fully operational yet, Tools has a lot of ideas in his own mind which are not proven reliable, be careful from whom you take your advice or an actually simple cooling solution can turn into a nightmare.

Okay, let's take a second and parse through this. I think LiamG6 may have been on the money with his comment here:
Quote:
Originally Posted by LiamG6 View Post

If you already have metal blocks, by all means, use them, but I don't think it is worth the expense to get metal top's, money would be better spent getting better blocks in general.

There's nothing inherently wrong with the suggestion to use metal tops as a way to increase coldplate surface area with the coolant. I am interested in how much it helps, but suggesting to have the entire cold side block conducting heat out of the coolant vs just the base conducting heat out of the coolant seems like a legitimate suggestion.

I think you're misinterpreting what toolmaker said if you think the waterblock is acting as a reservoir.

Also, I'm not on board with this stalling idea yet. I don't know of any physical or theoretical reason this would happen, though I'm open to ideas. To me, it's sounded like the results you would expect if you put too much load on the TEC.
post #213 of 399
Thread Starter 
Quote:
Originally Posted by Skyl3r View Post

I think you're misinterpreting what toolmaker said if you think the waterblock is acting as a reservoir.

Also, I'm not on board with this stalling idea yet. I don't know of any physical or theoretical reason this would happen, though I'm open to ideas. To me, it's sounded like the results you would expect if you put too much load on the TEC.

No the water block is not acting as a reservoir and it never successfully could, but insulating the water block and using a metal top is an attempt to hold cold in the water block that may be necessary for sub zero, but not for below ambient cooling that this thread covers.

This cooling works simply because the cold is picked up from the water block as fast as it is being produced and carries it to the insulated reservoir, eventually the cold that is constantly being added increases in the reservoir and drops the coolant mass temperature to the below ambient 10c that I run with.

Stalling is not my idea or concept it has happened in my setup and can happen in any setup that the hot/cold is not balanced on the peltier or TEC assembly, too much of either hot or cold out of balance can cause stalling.
post #214 of 399
Quote:
Originally Posted by OCAddict View Post

No the water block is not acting as a reservoir and it never successfully could, but insulating the water block and using a metal top is an attempt to hold cold in the water block that may be necessary for sub zero, but not for below ambient cooling that this thread covers.

This cooling works simply because the cold is picked up from the water block as fast as it is being produced and carries it to the insulated reservoir, eventually the cold that is constantly being added increases in the reservoir and drops the coolant mass temperature to the below ambient 10c that I run with.

Stalling is not my idea or concept it has happened in my setup and can happen in any setup that the hot/cold is not balanced on the peltier or TEC assembly, too much of either hot or cold out of balance can cause stalling.

The reason toolmaker suggested a metal top wasn't "to hold cold in", it is to provide more conductive surface area to conduct heat out.
As far as insulating the block goes, you should do that anyways tongue.gif I don't think it matters if you're subzero or not, the same concepts will apply.
I don't mean to say there's anything wrong with what your doing. I think your setup is fine.

Can you provide steps to reproduce stalling?
Edited by Skyl3r - 5/2/17 at 5:48am
post #215 of 399
Thread Starter 
Quote:
Originally Posted by Skyl3r View Post

The reason toolmaker suggested a metal top wasn't "to hold cold in", it is to provide more conductive surface area to conduct heat out.
As far as insulating the block goes, you should do that anyways tongue.gif I don't think it matters if you're subzero or not, the same concepts will apply.
I don't mean to say there's anything wrong with what your doing. I think your setup is fine.

Can you provide steps to reproduce stalling?

Since you have not experienced any type of peltier stalling it is a foreign concept to you, eweast does touch on it in his "Ultimate Guide To TECs" linked below.

http://www.xtremesystems.org/forums/showthread.php?38367-The-Ultimate-Guide-to-TECs
post #216 of 399
you should always insulate when below ambient ryan. when there is a temperature differential, energy will move to equalize that difference, meaning if your cold block is below ambient, you need to insulate, otherwise energy will attempt to equalize with the ambient air, ie warming the cold block up (or cooling the ambient air temp down, much harder to do) this is important even if it isn't below dew point otherwise your losses will be bigger than they could be.

tools idea to increase the conductive area for heat exchange is sound, but the problem is the top of the block has bad contact with the base, ie an o-ring, no TIM and just generally bad contact, and then there is nothing on the inside of the top that is very effective and transferring that energy to the coolant, so in theory it's a good idea, but in reality the thermal conductance isn't really good enough to do anything with metal tops and is basically a waste of time, add a big heat load in and it will just store the energy from the heat load and be counteractive to the TEC's cold side. You could design a custom block that had good contact with the base and used TIM outside the o-ring and hell even had a big air cooler on the top of the "top", it would all work but it's also completely crazy and not very effective compared to a well designed custom block.

this stalling that ryan and tools speak of is not stalling. stalling is when the TEC cannot handle the heat load, simple as that. ie it's ability to pump heat is overwhelmed and it stalls out and gets hotter and hotter. this can also happen if your hot side cooling cannot dissipate the total heat load and you get thermal runaway
Edited by LiamG6 - 5/2/17 at 6:15am
post #217 of 399
Quote:
Originally Posted by OCAddict View Post

Since you have not experienced any type of peltier stalling it is a foreign concept to you, eweast does touch on it in his "Ultimate Guide To TECs" linked below.

http://www.xtremesystems.org/forums/showthread.php?38367-The-Ultimate-Guide-to-TECs

What he mentions here:
Quote:
The biggest danger is attaching a TEC that is underpowered; many have cooked their brand new processors doing this. :cry Of course, this is easily avoided; just don't attach a TEC will less cooling power than your CPU generates. Duh.

The second best method to frying your CPU is to have an insufficient method of cooling the hot side of the TEC. The hot-side can become so hot that it begins to warm the cold side which adds to the heat load of the cold-side which adds to the hot-side's heat and the cycle begins. Not a pretty thing. I’ll go into more detail about TEC cooling further down.
Is what I am imagining you are seeing when the "stalling" occurs. It's not related to a balance between the cold side temperature and hot side temperature though; it's just too much heat load on the cold side. So the peltier can move 200w, but the heat load is 210w, so every second you're getting 10 joules left over on the cold side which starts the cold side warming up.
Edited by Skyl3r - 5/2/17 at 6:28am
post #218 of 399
Thread Starter 
Quote:
Originally Posted by LiamG6 View Post

you should always insulate when below ambient ryan. when there is a temperature differential, energy will move to equalize that difference, meaning if your cold block is below ambient, you need to insulate, otherwise energy will attempt to equalize with the ambient air, ie warming the cold block up (or cooling the ambient air temp down, much harder to do) this is important even if it isn't below dew point otherwise your losses will be bigger than they could be.

tools idea to increase the conductive area for heat exchange is sound, but the problem is the top of the block has bad contact with the base, ie an o-ring, no TIM and just generally bad contact, and then there is nothing on the inside of the top that is very effective and transferring that energy to the coolant, so in theory it's a good idea, but in reality the thermal conductance isn't really good enough to do anything with metal tops and is basically a waste of time, add a big heat load in and it will just store the energy from the heat load and be counteractive to the TEC's cold side. You could design a custom block that had good contact with the base and used TIM outside the o-ring and hell even had a big air cooler on the top of the "top", it would all work but it's also completely crazy and not very effective compared to a well designed custom block.

this stalling that ryan and tools speak of is not stalling. stalling is when the TEC cannot handle the heat load, simple as that. ie it's ability to pump heat is overwhelmed and it stalls out and gets hotter and hotter. this can also happen if your hot side cooling cannot dissipate the total heat load and you get thermal runaway

My system is insulated, my water blocks simply don't need it as the coolant flows through the water block so fast it is just not necessary.

As far as stalling maybe that was the wrong word to use for what happens!
post #219 of 399
Quote:
Originally Posted by Skyl3r View Post

What he mentions here:
Is what I am imagining you are seeing when the "stalling" occurs. It's not related to a balance between the cold side temperature and hot side temperature though; it's just too much heat load on the cold side. So the peltier can move 200w, but the heat load is 210w, so every second you're getting 10 joules left over on the cold side which starts the cold side warming up.

precisely
post #220 of 399
Thread Starter 
Quote:
Originally Posted by LiamG6 View Post

precisely

Whatever rolleyes.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier/TEC Hybrid Water Cooling vs 2.0