Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › TEC Chill Box Chamber Build log
New Posts  All Forums:Forum Nav:

TEC Chill Box Chamber Build log - Page 78

post #771 of 1654
Thread Starter 
ok, so after really looking through CTE stock of TEC's, I decided to get 3 X 26311-5M31-17CW's, and a 24V power supply for them, to replace the 12726 TEC's that are in the main TEC assembly now. if my understanding is correct the 12726 TEC's are 245W, and the 26317 TEC's are 331W so they should be able to handle the heat load from my PC components much better than 12726 TEC's.
I think that I am starting to get how these TEC's work, anyway LiamG6 thanks for all of your the help, and let me know what your thoughts are on this idea, thanks. I will send the 12726's back to CTE once they get here.
post #772 of 1654
Thread Starter 
yea, I was able to cancel the order for the 12726's through CTE. so, the TEC's that will be getting from CTE, are the 11911-5L31-15CQ's, and the 26311-5M31-17CW's. these are the TEC's that will be on the build, in the Main TEC assembly, I will have the 26311-5M31-17CW's, powered by a 24V power supply. in the secondary TEC assembly, I will have the 19911-5L31-15CQ's powered by a 14V power supply. I hope this will be enough TEC to cool my system properly, and allow me to clock my hardware to higher levels.
post #773 of 1654
Thread Starter 
Quote:
Originally Posted by OCAddict View Post

Wow!

At this point I would like to exercise my right as a true a$$hole, and say I told you so!

When I suggested about 40 pages ago to pretest everything before applying all that clay and just assuming it was going to work as you had planned.

Peltiers do not always cooperate, you have to discover the sweet spot they can give you, it is not a set in stone calculation when it comes to chilled water cooling.

And peltier wise you get what you pay for, you buy cheap Ebay, or China crap don't be surprised at the end results.

That said:

I am going to say once more at this point that you need to insulate those reservoirs!

You are after as much cold as you can possibly achieve and insulating the reservoirs will tremendously affect that.

yea, your right, I got what I paid for. sad-smiley-002.gif well, I am getting good TEC's this time, so they should work as described. I will never get cheap china TEC's again, I have learned my lesion.

I know that I will have radiated loss of coldness by not insulating the reservoirs, but I still do need some kind of a radiated loss of coldness inside the ice chest. if I don't have some coldness in the ice chest, the inside of the ice chest will get really hot, with all the components radiated heat and no way for that to escape.

it's a thermal dynamic nightmare, but if the radiated loss of coldness through the reservoirs is to great for the TEC's to handle, I will insolate the reservoirs and see what my temps are like inside the ice chest with the reservoirs insolated. it's all trial and error at this point, I don't really know how much radiated coldness I need, to keep the components from heating the inside of the ice chest.
Edited by toolmaker03 - 7/15/16 at 8:22am
post #774 of 1654
Good start with getting some quality TEC's, and running them at better Umax percentages. I would change your TEC block insulation. We use rubber art eraser as a protective layer on our motherboards, but its not actually a very good insulator, you would be better off getting some sticky sided neoprene that you can attach all around the TEC blocks, this stops thermal transfer between the hot and cold side of the TEC, currently your clay is actually transferring heat to the cold side. You only need a few layers of sticky sided neoprene to stop condensation and it doesn't transfer the heat like the clay does.

I would insulate your reservoirs as recommended, this will keep the chilled water buffer as cold as possible, I would ditch the vacuum idea, it's unnecessary complexity, albeit a cool idea, its hard to implement well, I would then add in a 240-360mm radiator plumbed into your cold loop inside your ice chest with some low speed fans circulating the air inside the chest and cooling the air inside the chest to close to the chilled water temp, this radiator should be the first component that the chilled water passes through inside the chest after the reservoirs, but understandably that may not be possible with your limited space, just make sure it is in there. This will also remove the days of drying/vacuuming etc that you have to put up with currently. Turn you chiller on, let the air temp inside chest drop with the coolant temp, when it gets to the right temp inside chest, start your PC up.

EDIT, actually if you insulate your reservoirs extremely well, you can remove them from the ice chest, to free up a lot of space inside there, and then you can place a radiator inside the ice chest easily, and there will be plenty of space to make sure air is moving around well inside the chest keeping the air temp consistent so there are no hot spots where condensation could occur.
Edited by LiamG6 - 7/15/16 at 5:58pm
post #775 of 1654
Thread Starter 
Quote:
Originally Posted by LiamG6 View Post

Good start with getting some quality TEC's, and running them at better Umax percentages. I would change your TEC block insulation. We use rubber art eraser as a protective layer on our motherboards, but its not actually a very good insulator, you would be better off getting some sticky sided neoprene that you can attach all around the TEC blocks, this stops thermal transfer between the hot and cold side of the TEC, currently your clay is actually transferring heat to the cold side. You only need a few layers of sticky sided neoprene to stop condensation and it doesn't transfer the heat like the clay does.

I would insulate your reservoirs as recommended, this will keep the chilled water buffer as cold as possible, I would ditch the vacuum idea, it's unnecessary complexity, albeit a cool idea, its hard to implement well, I would then add in a 240-360mm radiator plumbed into your cold loop inside your ice chest with some low speed fans circulating the air inside the chest and cooling the air inside the chest to close to the chilled water temp, this radiator should be the first component that the chilled water passes through inside the chest after the reservoirs, but understandably that may not be possible with your limited space, just make sure it is in there. This will also remove the days of drying/vacuuming etc that you have to put up with currently. Turn you chiller on, let the air temp inside chest drop with the coolant temp, when it gets to the right temp inside chest, start your PC up.

EDIT, actually if you insulate your reservoirs extremely well, you can remove them from the ice chest, to free up a lot of space inside there, and then you can place a radiator inside the ice chest easily, and there will be plenty of space to make sure air is moving around well inside the chest keeping the air temp consistent so there are no hot spots where condensation could occur.

I know that this might sound off, but wouldn't the reservoirs without being insolated inside the ice chest, be basically doing the same thing that a radiator inside the ice chest would be doing?

to me, it seems like they would both be radiating coldness into the ice chest.

but I will seriously consider this idea of removing the reservoirs, and adding a radiator to the ice chest, as that also is the way everyone else has done there chill boxes.
post #776 of 1654
yes and no, not insulating the reservoirs and keeping them inside the chest will cause the reservoirs to absorb the heat from the air in the chest, but it won't circulate, I think it is a primitive solution and would lead to hot spots and a generally higher coolant temperature due to absorbed heat while the coolant is stagnant in the res.

If it is in a vacuum it's beyond me what will happen. In any case, I think keeping them external and insulated and using a radiator to chill and circulate the air inside the chest would be a better solution, I think the vacuum/nitrogen charge is pointless complexity without a perfectly sealed case. I think insulating the res and keeping the chilled fluid buffer as cold as possible and only using rad to circulate chilled air is a more effective solution.
post #777 of 1654
Thread Starter 
well at this point the ice chest can be perfectly sealed when I seal the ice chest lid. the rest of the build like the ice chest, the area where the wire and water lines go in and out of the ice chest has no leaks. it did take awhile to do that and I had to redo it 3 times to get that perfect seal on the ice chest, but at this point I am there. I already have the nitrogen, and the fittings to make filling the ice chest with nitrogen after vacuuming the ice chest easy it can now be done one right after the other. so, I would like to go ahead and do that part.

as for removing the reservoirs, damn it, that means I have to remove them again, but I am starting to get real good at doing that, because I have done it so many times already. so, I will do that as well, but it will take some time for me to get the radiator, fans, and fittings for the radiator.

question how am I going to power those fans, going on the radiator, inside the ice chest?
I do not have enough fan headers on the motherboard, I am using two now, so I only have one left.
I do have a open 4 pin molex connection, I know that fan controllers have a molex connection to power them, should I put a fan controller inside the ice chest?
Edited by toolmaker03 - 7/16/16 at 3:40am
post #778 of 1654
Thread Starter 
ok, so I got a alphacool UT60 240mm radiator, and 4 fans to fit it, but I still need fittings to plumb it into the system. I have started building the frame to hold the DYI reservoirs, that will be mounted on the table top. I ordered the 6 X CPU blocks for the second TEC assembly, but I still need the fittings to plumb it into the system. I have removed one of the 600W TT power supply from the build, because the 850W enermax power supply can handle the 21 fans, 4 X water pumps, and all of the displays. I am moving the TEC power supply assembly closer to the radiator bank, and the PC system 1250W power supply, will be where the TEC power supplies where on the table top. this way I now have room to mount the DYI reservoirs, in the area right beside the ice chest. I do not have the norprene foam to insolate the reservoirs, so for the time being they will remain bare. I am removing the 12V 45A TEC power supply from the build, and replacing it with a 24V 50A power supply this will power the new TEC's from CTE. I will add another two sections of aluminum angle to the fan frame on the PC case, to secure the 240mm radiator inside of the ice chest. I am also adding a small 4 fan, fan controller to the inside of the ice chest to power the radiator fans. ok, so I am going to build another 3-way DYI reservoir, combined that will give me a 12 liter cold bank. I think that will be a large enough cold bank to keep me gaming for a few hours. when I build the frame for the current DYI reservoirs I will design that frame work so that it can handle both builds of the reservoirs. the new set of 3-way reservoirs will be directly above or on top of the current set of 3-way reservoirs. ok, so I think that about does it, for all of the changes that are going to be on the build, at this point. biggrin.gif
Edited by toolmaker03 - 7/16/16 at 8:47pm
post #779 of 1654
sounds good. not sure you need 12 liter reservoir buffer though, that just means pc will take a long time to drop to temp. Also unless it's insulated extremely well it will allow the coolant to rise in temp slightly.

I've always been a fan of sizing the TEC's correctly and minimizing the reservoir buffer required to hold temps. The lower the amount of coolant there is the faster the PC will drop to temp and also the faster the TEC's can react to high load situations without having to rely on a buffer. I thought the amount of res you already had was overkill but I guess it could allow you to hold even colder temps than just the dT of the TEC's with 600w heat load, maybe you can test it if you already have all the materials, ie 1 res, 3 res, 6 res. Also think you need to try your cold side TEC blocks in series as well as in parallel to see which works better. I guess with so many water blocks now though series becomes highly restrictive, not a terrible thing in the cold loop though as high flow isn't required.

Sounds like you are on a good track, time and testing will tell if it reaches your goals and expectations.

I think I'll stop badgering you with my opinion for a while and sit back and enjoy what you are doing, which is, once again, an epic build. thumb.gif
post #780 of 1654
Thread Starter 
Quote:
Originally Posted by LiamG6 View Post

Also unless it's insulated extremely well it will allow the coolant to rise in temp slightly.

I will do my best to make sure that it is insulated extremely well, it will most likely be all over the reservoir like 3" thick. thumb.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › TEC Chill Box Chamber Build log