Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier chilling system
New Posts  All Forums:Forum Nav:

Peltier chilling system - Page 14

post #131 of 167




I will try to explain what my opinion is.

ok, so that little area where there is a indent for the bolt that holds on the water block, that little indent is why the TEC is broken in that area. there is so much outward force on the TEC's surface, that a little indent on the surface of the mounting block, is enough to cause the ceramic on the TEC to crack apart. copper is a soft material, and if made too thin, the TEC will literally imprint itself into the copper. the effect of this would be, that the TEC would be able to tear itself apart.
Edited by toolmaker03 - 8/21/16 at 4:24pm
post #132 of 167
Thread Starter 
Quote:
Originally Posted by toolmaker03 View Post





I will try to explain what my opinion is.

ok, so that little area where there is a indent for the bolt that holds on the water block, that little indent is why the TEC is broken in that area. there is so much outward force on the TEC's surface, that a little indent on the surface of the mounting block, is enough to cause the ceramic on the TEC to crack apart. copper is a soft material, and if made too thin, the TEC will literally imprint itself into the copper. the effect of this would be, that the TEC would be able to tear itself apart.


I understand what your opinion is on this matter i'm just trying to explain to you that what you experienced is not due to the hardness of the copper. Clamping force and where you apply it can cause these same issues. Also while different coppers offer different strengths you can also get a copper grade in various conditions. Do you have ANY way to verify that what you are experiencing is due to deformation of the copper? Other than a marking in the copper? Which could be a result of miss-clamping. You could also be using fully annealed copper, which is the softest form of C110 you can get.

Here's something that's lacking in this discussion... data

C110 Copper O60 (soft / annealed) temper
Physical and Mechanical Properties Ultimate Tensile Strength, psi 31,900
Yield Strength, psi 10,000
Rockwell Hardness F40
Machinability 20% of free-cutting brass rod (C360)


C110 Copper H02 (half hard) temper
Physical and Mechanical Properties Tensile Strength, psi 42,100
Yield Strength, psi 36,300
Rockwell Hardness B40
Machinability 20% of free-cutting brass rod (C360)


C110 Copper H04 (full hard) temper
Physical and Mechanical Properties Tensile Strength, psi 50,000
Yield Strength, psi 45,000
Rockwell Hardness B50
Machinability 20% of free-cutting brass rod (C360)



This is why i made mention of the particular C110 used was H02 condition. At the strengths listed above i have no concern of the copper deforming. Again i don't mean to toot my own horn here but i have a very good understanding of material grades along with their conditions.
post #133 of 167
Thread Starter 
Also now that i'm noticing are those tops to those water blocks made of Delrin or ABS? Delrin and ABS both are a very soft-impact friendly plastic. Yet stiffness properties are lacking. This alone could allow the copper to deform and stress where ever it would want to. Not an ideal choice of plastic IMO. Had an acrylic been used, you lend yourself to gaining torsional strength in the copper as it becomes supported by the stiffness of the acrylic.
Edited by mrkubanftw - 8/22/16 at 9:59am
post #134 of 167
Thread Starter 
If words don't get us there, here is an actual simulation of the full waterblock being subjected to clamping forces with a 62mm ceramic pad present. While colors are pretty, you'll notice the deformation strays on the side of; not even measurable. Roughly 1 micron, in the worst spot. IE 0.000039". Now lets take a look at the other sim i did.






Here we have your recommendation of Beryllium Copper C17000. You'll see a difference in the best case deformity on the scale from 1.0mm e-30 go to 0.0mm e-30. So from a number that is 30 decimals back to a number that doesn't exist. Okay so NO feasible changes there, lets take a look at the worst case. It's identical.

Here you have it. An actual engineering sim showing that Beryllium Copper yields NO structural integrity gains over standard fully soft C110 copper (i'm using 1/2" hard).


post #135 of 167
@ mrkubanftw

Very interesting Thank You! smile.gif
post #136 of 167
Thread Starter 
Quote:
Originally Posted by OCAddict View Post

@ mrkubanftw

Very interesting Thank You! smile.gif

Not a problem. Liam had mentioned the method of clamping can affect this very topic and he makes a point. My original sim shows how this block would react with even forces being applied on the outside of the acrylic (my original method of clamping). I re-ran the sim showing how the stresses would be affected by using the 4 bolt pattern closest to the TEC. You can see that even applying all that force in only 4 places... the copper still does not tweak. Worst case being around 0.001" and only extending slightly beyond the bolt face in the acrylic. This is to be expected and is not a valid concern either. You'll notice in the other image, the TEC area is unaffected.


post #137 of 167
Thread Starter 
I should also note that force was calculated at 75psi to the TEC surface area. The first sim shows that direct PSI force.

The second sim took a calculation to calculate force of each bolt face.

(2.44*2.44) -TEC Area

(5.9536) * 75 - Surface area * pressure = total force

446.52 / 8 = 55.815LBF -Total force divided by 8 affected bolt faces


= 55.815 lb's of force were applied to each bolt face in this calculation.
post #138 of 167
Quote:
Originally Posted by mrkubanftw View Post

I should also note that force was calculated at 75psi to the TEC surface area. The first sim shows that direct PSI force.

The second sim took a calculation to calculate force of each bolt face.

(2.44*2.44) -TEC Area

(5.9536) * 75 - Surface area * pressure = total force

446.52 / 8 = 55.815LBF -Total force divided by 8 affected bolt faces


= 55.815 lb's of force were applied to each bolt face in this calculation.

yea, I here what your saying, it should be fine,
I just was showing that there are options for this kind of work, I have worked with copper to make different types of plates, and connectors before. the first time I made a copper plate, I taped the plate for fine thread screws, and as I tightened the screws down, the threads striped out. yea, no fun there. so, I drilled for a larger screw, and this time I taped for a curse thread screws, and that solved the issue of the threads stripping out when I tightened the screws down. I just know that copper, when compared to other raw metals, it is one of the softer metals. it is probably strong enough for what your doing. I have never tried to make a water block, I have always just modified what was available to meet my needs.
Edited by toolmaker03 - 8/22/16 at 3:33pm
post #139 of 167
Thread Starter 
I had an opportunity to make up 5 blocks last night minus the slotting operation. Scrapped one out due to a model issue. So i have 4 that need the slitting operation. Still waiting on my G1/4 tap from china. Should be here any day. With that i should be able to finish these things up in a week or so and test them!
post #140 of 167
Thread Starter 
Also ordered two of these pumps today with reservoirs. One more radiator, a set of 8 fans and i can test two full TEC assemblies. Also ordered some extremely dense two part polyurethane, i plan to use it as insulation between the two water-blocks.


http://www.aquatuning.us/water-cooling/pumps/d5-series/d5-pre-installed/20543/alphacool-vpp655-pwm-g1/4-ig-inkl.-eisdecke-d5-acetal-v.3?c=6541
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Peltiers / TEC
Overclock.net › Forums › Cooling › Specialized Cooling › Peltiers / TEC › Peltier chilling system