• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Peltier (TEC) + Liquid Cooling Project

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
What would you say is "a fast speed"? Temps might have to jump 10-12 degrees C within at most 30s-1min

Also, perhaps the liquid cooling is a little excessive/too complex. Wondering if a quick and dirty heatsink can do the trick..

I'm trying to imagine how you'd change the liquid into a heatsink. Since the liquid (blood) needs to flow to cool or heat.....

Let me ask where you are going to take temperature readings? At all five finger tipes?

And lastly, you can heat exchange rather quickly I think with a strong enough tec and small amount of liquid. Say running less than 200ml would be easier to heat or cool than more than 200ml.

In your picture, you have water blocks at the finger tips. Is this the temperature reading point? This I'm the most curious about......

Are you taking a temperature of the liquid (blood) at the same time as the finger tips heat sinks? (hopefully to watch heat exchange rates)

You asked about small pumps. I've used this with TEC cooling so I don't need an expensive PC water cooling pump. Under 20 dollars with a reservoir. Probably small enough to fit inside a baby. http://www.ebay.com/itm/DC-12V-Wate...05603?pt=US_Water_Cooling&hash=item4acb501f23
 
I think what you have planned is really cool. I don't see why it wouldn't work. Haven't seen any show stopping arguments either. I think the hardest part would be the blocks and attaching small tubing without it leaking profusely.

Listen to these guys about tec size and look for some ebay tec's on a ballpark of what you think you need and test. If 30s to 1min is your goal, I think that's easily doable.

Good luck and pics or it didn't happen.
 
Im both intrigued and really confused. What is this even for?

Well, here are some problems I can see

1) peltiers are used to heat OR cool. Doing both ... you'd have to create two separate loops or find some way to alter the flow path.

2) 25c is below the temperature that sustains life, I'd love to know what this temperature is supposed to represent.

3) temperature regulation in a loop that small will be more pick a temp and go for it. There's no way you're going to get a predictable temperature transient without a lot of trial and error.

4) quite unsure what the waterblocks on the finger tips are for

5) At best, you're going to need a low ambient temperature, and let the TEC heat the water to 25c, get it stable, then figure out what kind of voltage you're going to need for 36c. Cooling it back down to 25c would be quite slow.
 
you know you can solve this with a 3-way valve, yeah?

The "loop" through the fingers is a fixed loop.
You stick a TEC between two blocks/reservoirs so you have a cold supply and a warm supply. You "flip" the 3 way valve as needed.


You know , if this does not need to be "super portable" , you can set this up anywhere there is a water tap/faucet.... By using an under sink waterheater and a ordinairy (kitchen) mixing faucet.....

Oh, ... and a bucket/drain of course :)

http://www.clage.com/en/product-range/mini-instantaneous-water-heaters/produktlist.php

ALSU__Under_Sink_Water_Heater.jpg
 
I'm trying to imagine how you'd change the liquid into a heatsink. Since the liquid (blood) needs to flow to cool or heat.....

Let me ask where you are going to take temperature readings? At all five finger tipes?

And lastly, you can heat exchange rather quickly I think with a strong enough tec and small amount of liquid. Say running less than 200ml would be easier to heat or cool than more than 200ml.

In your picture, you have water blocks at the finger tips. Is this the temperature reading point? This I'm the most curious about......

Are you taking a temperature of the liquid (blood) at the same time as the finger tips heat sinks? (hopefully to watch heat exchange rates)

You asked about small pumps. I've used this with TEC cooling so I don't need an expensive PC water cooling pump. Under 20 dollars with a reservoir. Probably small enough to fit inside a baby. http://www.ebay.com/itm/DC-12V-Wate...05603?pt=US_Water_Cooling&hash=item4acb501f23

I was considering ditching the liquid entirely and just using TECs at the fingertips, then heatsinking.

Another thought I had was using TECs at the fingertips and watercooling each TEC with the liquid cooling loop. So now the TECs aren't heating+cooling the water, but the water will be cooling (and not heating) the TECs.
hbzj7sM.jpg

Temperature readings would either be at all 5 fingertips or perhaps even just 1 finger, because ideally they'd all heat/cool to the same temps.

I haven't really thought much about measuring the liquid temp.. I'm hoping I'll be able to find a good temp that'll cool down the TECs (if I pursue the system in the above drawing).

Thanks for the pump+res. I'll definitely consider using that (it's cheap!)

I think what you have planned is really cool. I don't see why it wouldn't work. Haven't seen any show stopping arguments either. I think the hardest part would be the blocks and attaching small tubing without it leaking profusely.

Listen to these guys about tec size and look for some ebay tec's on a ballpark of what you think you need and test. If 30s to 1min is your goal, I think that's easily doable.

Good luck and pics or it didn't happen.

The only "show stopping" factor is probably size-I don't know of any water blocks that come small enough to fit in a baby's finger.. That's why I considered dropping the liquid cooling and just using make-shift heatsinks to cool the TECs. Thanks for the encouragement; means a lot! Pics and updates will definitely be coming.

1) peltiers are used to heat OR cool. Doing both ... you'd have to create two separate loops or find some way to alter the flow path.

Not with a MOSFET I don't. I can reverse the direction of current to flip the "hot" and "cold" sides of the peltiers.

2) 25c is below the temperature that sustains life, I'd love to know what this temperature is supposed to represent.

I kind of arbitrarily chose a low temperature, but ultimately it's gotta reach a low enough temperature for a thermochromic material that I'm using to change color (to blue). These thermochromic materials will be at the fingertips and will change color from skin tone to blue with dropping temps.

3) temperature regulation in a loop that small will be more pick a temp and go for it. There's no way you're going to get a predictable temperature transient without a lot of trial and error.

Yup, as stated above (in response to ShrimpBrime), now I'm thinking of putting the TECs at the fingertips and cooling them with the loop, and I'm hoping to just reach a low enough temperature to compensate for the excess heat that the TECs will be dissipating.

4) quite unsure what the waterblocks on the finger tips are for

With the 1st loop I posted, the waterblocks were to transfer the cold/hot energy to the fingertips to induce the color changes in the thermochromic materials. Now in the 2nd loop I posted above, the waterblocks will only be transferring cold (and NOT HEAT) to the TECs, and the TECs will be controlling the heating/cooling of the fingertips (Arduino controlled).

5) At best, you're going to need a low ambient temperature, and let the TEC heat the water to 25c, get it stable, then figure out what kind of voltage you're going to need for 36c. Cooling it back down to 25c would be quite slow.

Yeah I figured it's quite hard to fluctuate the water temps, so I'm considering the loop#2 I posted.

you know you can solve this with a 3-way valve, yeah?

The "loop" through the fingers is a fixed loop.
You stick a TEC between two blocks/reservoirs so you have a cold supply and a warm supply. You "flip" the 3 way valve as needed.

Wow that's a really cool idea. My only concern is that that will require 2 water blocks on 1 fingertip and there may not be space for that. Also, how would I be able to "flip" the 3-way valve electronically?

You know , if this does not need to be "super portable" , you can set this up anywhere there is a water tap/faucet.... By using an under sink waterheater and a ordinairy (kitchen) mixing faucet.....

Oh, ... and a bucket/drain of course :)

Unfortunately it'll have to be confined within a baby :( Thanks for the ideas though!
 
I think measuring temps at the finger tips with TECs may not be the way to go.

See the TEC instantly cools or heats. You'd be in shock how hot they can get a heat sink.

I think the liquid cooling / heating the sinks at the fingertips would be the best route because it will take time for the heat exchange in the water, wheres the heat exchange on the finger tip heat sinks with TECs will be very fast... guessing being able to start freezing about instantly with the right amount of current.

But an issue with the liquid loop is that you need a loop to cool the tec and a loop for the finger tips. Since 25c is room temperature, I'd think you'd want to heat the loop that would be considered blood with one TEC And cool that same loop with it by reversing polarity. You could run 2 loops, or perhaps because your using a short amount of time, just a heat sink on one side of the TEC and the other to alter liquid temps (blood)

Your going to do some math to do this right. Water Delta of 25c is sitting in my glass. My glass is 200ml full. I need to add heat to make 200ml of liquid come to 37c the temperature of blood. How big of a peltier and how much power should be used. I think that 150w of heat could have a water loop running 37c within the 60 second time period give or take a few watts.

Since you may need to cool the liquid just as fast, you just use the same peltier and same wattage, voltage and amp out put to cool the (blood) at the same rate. 12v will react slower than the 15v the TEC is rated for. Therefore you are not getting say 136w of cooling output, perhaps closer to 105w, amps can make up for the loss of voltage. Higher volts, less amps..... but don't run less than 12v, just use the TEC to heat for a period of time instead.

But a bunch of mini tecs could work as well, but I haven't experience with something so small, But I assume they can create frost just as fast as the big brother 40mm TEC I like using.
 
dont get me wrong for the fixed loop the "fingertips" are just 1/4" U shaped copper "refrigerator" tube ... a coil at 10ft length sells for about $6 at home depot
http://www.homedepot.com/p/Cerro-1-...EC-_-NavPLPHorizontal1-2-_-NA-_-203451405-_-N

Yes, 3-way valves used in CV/HVAC can be "flipped" electronically, but just by hand would work also :)


A variant:
following needs some verification, especially if a Switech Apogee Drive can handle it

Loop: Swiftech APDrive > Y-splitter > 2 Y-splitters > tube(s)/hose(s) > fingertips (U shapes) > hose(s) > Y-Splitters > waterblock (eg a normal apogee) > hose > Radiator > hose > Swiftech APD.

The loop always runs in the same direction ;
The tec is sandwiched between two waterblocks;
Hot or cold is done by flipping polarity on the tec.
The radiator is there as a balancer to get rid of excess heat;

Now, it comes down to correctly dimensioning the tec & the Rad. And to use a switch that can flip the tec polarity and OFF, because of the "perpetuum mobile" setup of the loop it can easily run out of control... so some electronic (PWM) on-off based on a temp probe or two might be needed.
Rocker-Switch.jpg



PS: actually, come to think of it... you can drop the Y-splitters completely if you bend the copper tube (carefully) in the shape of hand.
The challenge will be to a tight enough U without folding/cracking the tube (but there are guides for that - using either sand or water)

.
.
 
Last edited:
I think the liquid cooling / heating the sinks at the fingertips would be the best route because it will take time for the heat exchange in the water, wheres the heat exchange on the finger tip heat sinks with TECs will be very fast... guessing being able to start freezing about instantly with the right amount of current.

Very fast is IDEAL! The hope is to trigger the Arduino and increase temps asap (slowest I'd want is 30-60s).

You could run 2 loops, or perhaps because your using a short amount of time, just a heat sink on one side of the TEC and the other to alter liquid temps (blood)

Yup, that was the plan (heat sink for the TEC), but the temp regulation might be too slow.


RnRollie said:
dont get me wrong for the fixed loop the "fingertips" are just 1/4" U shaped copper "refrigerator" tube ... a coil at 10ft length sells for about $6 at home depot
http://www.homedepot.com/p/Cerro-1-4...-203451405-_-N

Can those conduct heat/cold well enough to not need a waterblock? These aren't hoses right..the hoses will lead to the copper tubes (which essentially act as waterblocks)?

Loop: Swiftech APDrive > Y-splitter > 2 Y-splitters > tube(s)/hose(s) > fingertips (U shapes) > hose(s) > Y-Splitters > waterblock (eg a normal apogee) > hose > Radiator > hose > Swiftech APD.

The tec is sandwiched between two waterblocks;

I'm not really understanding what the two waterblocks are for.

qIpN15T.jpg

Where does the TEC go?? Also, what's the benefit of the Swiftech APDrive versus getting a cheap waterblock + pump?
 
Last edited:
you basically sandwich the TEC between two CPU waterblocks
one block to heat up the water and cool the tec's hot side , one block to cool the water and thus unfortunately heat up the tecs cold side - which is where the rad comes in to balance the extreme difference between hot & cold side a bit (less steep/sharp response to changes).

The Swiftech apogee is because it combines block & pump in one... thus less fittings and failure points .... and is a very good block.
Of course there are small "piezo" ceramic pumps like the DC-LT2400 , but you need to ADD a pump top, fittings, a cheap waterblock, more fittings .. its gonna be more expensive as the ApogeeDrive where you only reducerfittings or barbs

Now, forget about the Y splitters :)

Each waterblock has an IN & OUT ...
waterflow using ApogeeDrive: APDrive - OUT > handshaped single tube > block2-IN > block2-OUT > radiator > APDrive IN

waterflow using any block + pump : Block1-OUT > pump > handshaped single tube > block2-IN > block2-OUT > radiator > Block1-IN

Each ">" is a fitting or barb also indicating the flow directions

TEC between the two waterblocks , polarity change to change the temp from cold to hot and back. OFF to avoid overshoot.

The handshaped copper coil is basically one tube-run... this can possible lead to a very small temp difference between thumb & pinkie... its not that the thumb will be +50°C & the pinkie -20°C , maybe 0.5°C which is human anyway i guess, no?

IF you only want the fingertips to significally change temp, wrapping some insulating tape around the rest of the tubing will work, just leave the U shaped fingertips unwrapped.

I assume the whole needs to fit into some kind of fleshcoloured glove.. looking "a bit" like a hand... unless you plan on shooting a horrormovie :)
 
you basically sandwich the TEC between two CPU waterblocks
one block to heat up the water and cool the tec's hot side , one block to cool the water and thus unfortunately heat up the tecs cold side - which is where the rad comes in to balance the extreme difference between hot & cold side a bit (less steep/sharp response to changes).

How can each block both heat and cool at the same time? Doesn't one block heat up while the other cools when the TEC is in operation? Then it'd be vice versa when reversing the current going into the TEC. Not really understanding the sandwich very well.

Each waterblock has an IN & OUT ...
waterflow using ApogeeDrive: APDrive - OUT > handshaped single tube > block2-IN > block2-OUT > radiator > APDrive IN

JnYvNqS.jpg

Is this the right idea of "sandwiching" the TEC? Not really sure where the rad is going still.. Apologies
Also, the APD has a heatsink in it, so I'm wondering if a rad is necessary.
Oh and I'll need a reservoir or T-line in the loop.

TEC between the two waterblocks , polarity change to change the temp from cold to hot and back. OFF to avoid overshoot.

The handshaped copper coil is basically one tube-run... this can possible lead to a very small temp difference between thumb & pinkie... its not that the thumb will be +50°C & the pinkie -20°C , maybe 0.5°C which is human anyway i guess, no?

IF you only want the fingertips to significally change temp, wrapping some insulating tape around the rest of the tubing will work, just leave the U shaped fingertips unwrapped.

I assume the whole needs to fit into some kind of fleshcoloured glove.. looking "a bit" like a hand... unless you plan on shooting a horrormovie :)

Thanks so much for the "copper hand" idea! It's a really great idea.. I was dreading figuring out how to fit small waterblocks at the fingertips haha.
 
Last edited:
yes, one block goes hot, while the other goes cool, but the whole setup would be like a perpetuum mobile and run out of control before you know... the TECs cold side getting to warm, resulting in the TECs hot side growing to hot and frying the TEC or the hot side get "too cold" resulting in the cold side getting so cold it freezes the water. Although in reality the whole ting running too hot is more likekly.

The thing with the blocks is that whatever side is the hot side, the block will lack the capcity to move the heat away fast enough, while the cold side is probably even slower in cooling the water. Thus the whole loop will have the tendency to run hot.

Given a body of water the energy requirement to raise or lower the temp of the body with 1°C is about the same, only the timeframe differs. :)

This is where the radiator comes in: to act as a balancer , an equaliser, to dampen the difference between hot & cold. You dont *need* a radiator.. you could achieve the same with a 5 gallon reservoir... but that is slightly bigger as a radiator :)
The radiator will probably not be enough to keep everything in check, but it will slow down heating or cooling long enough so it is less brutal.

One one end of the Block/TECBlock sandwich you have "the hand"-loop , on the other end the Radiator loop.

For the electrical part:
The most imporant thing is to put in an OFF switch , preferably a thermal OFF, something which powers OFF the TEC when the watertemp goes over say 40°C
An inline probe/sensor is easy... making a circuit/relais to switch off the TEC when a certain temp/resistance is reached might be a bit more difficult.

When you switch polarity, the previous HOT side/block will cool down (fast) while the other block goes from cold to hot. Since the water just goes around relentlessly in a circle through the hand & blocks it will start heating up or cooling down depending on the polarity you are feeding. There will be a short moment of equilibrium :)


Getting the TEC & rad right might involve some advanced math & experiments :) My guts say that you can probably get away with a relative low power TEC maybe 40W or less. But this is where the real TEC specialist come in :) You might need more, you might need less.

Of course if you have access to switching power supply like one of the better Meanwells where you can regulate volts & amps seperatly, then you can get there by experimenting :)

Also, the Radiator might be able to work with the fans turned down when cycling "hot" fingers and the fans ramped up with running the cold hand

.

With the right kind of rad -like one with a build in reservoir, you wont need a seperate reservoir.. But a T line is always helpfull in bleeding & filling
.
 
Last edited:
One one end of the Block/TECBlock sandwich you have "the hand"-loop , on the other end the Radiator loop.

So it'll be block+pump > radiator > hand loop > block > block+pump? Your other loop flow had the radiator between the 2 blocks 0_0

For the electrical part:
The most imporant thing is to put in an OFF switch , preferably a thermal OFF, something which powers OFF the TEC when the watertemp goes over say 40°C
Got the switch OFF under control using a thermistor + Arduino :)

Started to order parts! I bought the 1/8'' ID copper coil (http://www.homedepot.com/p/Cerro-1-...tility-Coil-PCLA-250U010/203451405?quantity=1), and I own 1/8'' ID plastic tubing.
I'm wondering what kind of fittings I need to secure everything together. Not really understanding the G1/4'' threading vs 1/8'' ID fitting concept..
Would I need a straight connection "Compression on All Ends" fitting to attach the plastic tubing to the copper coil? (http://www.mcmaster.com/#standard-single-sleeve-compression-tube-fittings/=sshjhq)
 
Last edited:
You only need barb fittings on the blocks/rad. To make to connection between the block/barb and the copper tube, we'll cheat using the hot water trick :)

G1/4 is the BSPT indicator of the threaded hole.
1/8 , 1/4 , 1/2 , 5/8 , 3/4, etc is the "thickness" of the hose/tube/barb

NOTE that 1/8" ID copper pipe will actually be 1/4 OD...

What does "ID", "OD", and "G 1/4 BSP" mean?
* ID means "inner diameter". It refers to the internal width of a hose or connection. 6mm (1/4"), 10mm (3/8"), and 13mm (1/2") are common ID sizes.
* OD means "outer diameter". It refers to the external width of a hose or connection. 10mm (3/8"), 13mm (1/2"), 16mm (5/8"), and 19mm (3/4") are common OD sizes.
* G 1/4 BSP (British Standard Pipe) is a pipe thread specification. Nowadays, it does not refer to the diameter of the fitting. G 1/4 BSP is the most common parallel threading type used by the computer liquid cooling industry. Unlike NPT tapered threads used in North American plumbing, parallel BSP threads seal with an o-ring collar and do NOT use sealing compound or teflon tape. In fact, sealing materials should be avoided to prevent interference with the o-ring.

Koolance used to have some nice pictures where you could clearly see the difference between a 1/4" barb and a 1/2" barb .. in both cases having the same G1/4 base that screws into the block/rad.
http://koolance.com/products?path=62_63


PS: i have been thinking, and have a few concerns... probably not show stopping ... i'll get into them later :)
need to feed now :)
 
Last edited:
You only need barb fittings on the blocks/rad. To make to connection between the block/barb and the copper tube, we'll cheat using the hot water trick :)

That totals to 6 barb fittings right? I'm assuming the blocks/rad that I'll buy will be G1/4 threads. Could you direct me to what kind of fittings I'll need then? My tubes/copper wire are 1/8'' ID. Sorry for the cluelessness :(

PS: i have been thinking, and have a few concerns... probably not show stopping ... i'll get into them later :)

Thanks for the support - much appreciated :D
 
Well....
a few questions, to see if i didn't make you stray away from the path :)

First: What is the actual size of this? Does it have to be lifesize?
Because it is not that easy to make very tight bends in small bore copper without kinking or folding it. Some very fine tablesalt in the tubing will help for bending the tube. There are numerous copper tube bending guides on the net. Either way, baby might end up with some big hands for its age :)

2nd: what is the temp range we are looking at? What is "cold", or "hot"?
A bit morbid, but in "life" simulation, "cold" can be ambient.. leave a corpse logn enough alone, and its body/extremities will be room temp. Same for the other end, leave over 40°C "fever" unchecked long enough and you end up with a corpse.
The required temp range will play in the design... i think we might be able to drop one waterblock and just use a cheap aircooler. And thus only need 4 barbs :)
 
First: What is the actual size of this? Does it have to be lifesize?
Everything will be contained in an infant mannequin. Hopefully within the arm/hand area.
Because it is not that easy to make very tight bends in small bore copper without kinking or folding it. Some very fine tablesalt in the tubing will help for bending the tube. There are numerous copper tube bending guides on the net. Either way, baby might end up with some big hands for its age :)
Yup, I looked into using salt, and it seems like the copper can bend a decent amount.

2nd: what is the temp range we are looking at? What is "cold", or "hot"?
A bit morbid, but in "life" simulation, "cold" can be ambient.. leave a corpse logn enough alone, and its body/extremities will be room temp. Same for the other end, leave over 40°C "fever" unchecked long enough and you end up with a corpse.
The required temp range will play in the design... i think we might be able to drop one waterblock and just use a cheap aircooler. And thus only need 4 barbs :)

Temp range is maybe from 20C to 30-35C, which is a little below room temperature and up to about skin temperature. Since this is only a 10-15C range, I agree, 1 waterblock may do the trick.
 
Begin condition:
FANS: OFF , PUMP OFF , TEC OFF => whole is a room temp

Init:
FANS: ON (low) , PUMP ON , TEC OFF => RAD keeps loop at room temp

Heating Phase:
FANS: ON (low) , PUMP ON , TEC ON => loop temp will rise fast ; RAD keeps it from running out of control

Steadying Phase:
FANS: ON (high) , PUMP ON , TEC ON => loop temp will probably keep rising slowly (that's what tecs do) ; RAD keeps it "steady"

Cooling Phase 1:
FANS: ON (high) , PUMP ON , TEC OFF => loop temp drop (quickly) towards room temp thanks to RAD

Cooling Phase 2:
FANS: ON (low) , PUMP ON , TEC OFF => loop temp keeps dropping (slowly) towards room temp thanks to RAD


The ordinairy air heat sink is to keep the cold side of the TEC from freezing when the TEC is ON.


Thus the loop is "simple" : Block/Pump > "hand" > Radiator > Block/Pump

The TEC is sandwiched between the waterblock and the ordinairy heatsink/cooler. The waterblock is on the "hot" side.

Needs TWO switches: One to power on/off the TEC. One to power on pump + fans. A rheo to change fan speed would be nice.


Now, for the exact dimensions/size/type of the TEC... i wouldn't know, i'll post an invitation for the TEC specialist in the extreme section :)
Given that a faily ordinairy radiator can dissipate 100Watts easy enough, and if the fingers dont have to exceed 50°C i guess the TEC can be kept a "small" enough to not exceed 100 Watts "dump" :)


Fittings: examples

http://www.plumbingsupply.com/barb.html
Looking at "king nipples" Male Barbed Adapter (barb x mips)
either 1/8" barb x 1/4" Mips or 3/16" barb x 1/4" Mips or 1/4" barb x 1/4" Mips
Best walk into your local DIY to find G1/4 x 1/8 or G1/4 x 3/16

OR a W/C Webshop
http://www.aquatuning.de/water-cool...tings/5395/6mm-fitting-g1/4-with-o-ring?c=511
http://www.frozencpu.com/products/1...etallic_Silver_HF-G14-14.html?tl=g30c407s1028

But 1/8 or 3/16 is difficult to find on W/C Webshops
 
Last edited:
Begin condition:
FANS: OFF , PUMP OFF , TEC OFF => whole is a room temp

Init:
FANS: ON (low) , PUMP ON , TEC OFF => RAD keeps loop at room temp

Heating Phase:
FANS: ON (low) , PUMP ON , TEC ON => loop temp will rise fast ; RAD keeps it from running out of control

Steadying Phase:
FANS: ON (high) , PUMP ON , TEC ON => loop temp will probably keep rising slowly (that's what tecs do) ; RAD keeps it "steady"

Cooling Phase 1:
FANS: ON (high) , PUMP ON , TEC OFF => loop temp drop (quickly) towards room temp thanks to RAD

Cooling Phase 2:
FANS: ON (low) , PUMP ON , TEC OFF => loop temp keeps dropping (slowly) towards room temp thanks to RAD

Actually, I'm wondering if it's possible to keep the body warm at "steady state," and then cool temporarily to 20-25C, then bring it back up to a warm steady state (Kind of how the body works! Usually "warm" but gets cold with bad blood perfusion). Does this require too much energy?

Also, any suggestions for a small but good radiator (trying to keep things as small as possible!)
 
Actually, I'm wondering if it's possible to keep the body warm at "steady state," and then cool temporarily to 20-25C, then bring it back up to a warm steady state (Kind of how the body works! Usually "warm" but gets cold with bad blood perfusion). Does this require too much energy?

Also, any suggestions for a small but good radiator (trying to keep things as small as possible!)

The cheapest RAD will probably a Magicool or Phobya or Coolgate 120.1 or 120.2
http://www.performance-pcs.com/catalog/index.php?main_page=index&cPath=59_457_667


that might require the waterblock to be on the "cold" side of the tec. Might be a bit more difficult to keep the TEC from burning itself up. But the whole logic/sequence doesn't change much.
But keeping the logic/sequence as it is, it might require a bigger rad. We'll need to wait for the TEC specialists to chime in on the TEC size.

Getting the exact settings for fan high/low , TEC power feed, etc is a bit experimental anyways (and depending on ambient room temp)..

A bit like flying an helicopter... you just push & move random things and remember how it reacts and what you've done in case you want the bloody thing to do that again. By all logic, an heli should not be able to fly and yet it does... and each and every one of them behaves differntly. Getting the b*tch to do what you want is a GD struggle of willpower EVERY TIME - and this is coming from somebody who held a license for decades :)
 
Last edited:
that might require the waterblock to be on the "cold" side of the tec. Might be a bit more difficult to keep the TEC from burning itself up. But the whole logic/sequence doesn't change much.
But keeping the logic/sequence as it is, it might require a bigger rad. We'll need to wait for the TEC specialists to chime in on the TEC size.

Is this setup still assuming that the TEC polarity will be reversed when signaled to do so? If not, which is the "real cold side," and how will the TEC heat up the water to "body temp" if the waterblock is on the "cold" side of the TEC?

Found a Swiftech MCP35X lying around! (That saves me a 100 bucks heh.)
Added a water block to my Parts List

-Water block: http://www.performance-pcs.com/cata...oduct_info&cPath=59_971_294&products_id=30144
-Radiator: http://www.performance-pcs.com/cata...t_info&cPath=59_457_667_670&products_id=21112
-Pump: Swiftech MCP35X
-1/8" ID Tygon Tubing
-1/8'' ID Copper pipe: http://www.homedepot.com/p/Cerro-1-...943611&cm_mmc=CJ-_-3754110-_-11210757&cj=true
-1/8'' barb x 1/4'' Male iron pipe size (4 of these)
 
Last edited:
Back