Ultrasonic2's TEC liquid chilled pc

[Ultrasonic2]

no, i meant electronically - not stacked (i know that has a set of problems all by itself). if you connect TECs in series (electrically) will you still get the desired dT increase? or do they have to be stacked to get that effect? ie: if they were side-by-side would you have a greater dT with half the voltage each vs a single at full voltage? what effect on QMax does this have? (still working exactly what QMax does for a TEC)

so far i understand UMax (max voltage), IMax (max current), QMax (max heat transfer), dT (temp differential between hot and cold sides at IMax), and COP (heat transference curve relative to current (amperage) at specific hotside temp) and how they all interrelate (mostly). i understand why more TECs is better than a single with similar total UMax/IMax. i assume the massive QMax you guys recommend is due to the undervolting of the TECs, whereas the actual power draw from those undervolted TECs is considerably less than total UMax/IMax for all the TECs combined?? still trying to figure out the relationship between UMax/Imax and QMax - i thought QMax was simply the product of volts and amps, but then i saw high QMax TEC's that were well under what i thought they would be. (35V/20A@403QMax) note: CoP will always be highest as dT approaches zero. but, since our chiller loops depend on a greater dT, how do we compensate for the lack of efficiency (ie: higher power requirement/current draw) of an ever increasing dT - at what point are we simply wasting power for diminishing returns?........anyway - still reading/learning/trying to figure this all out. thanks for all the input and help - and thanks for putting up with my MANY questions!!

idea: i was thinking about a barely sub-ambient (no condensation) water loop. since TEC's are most efficient at dT=0 then (ideally) you would want your hot side close to your low side. now, since it's impossible to get your hotside below ambient.......why not put a radiator in teh cold loop? (hear me out)...by doing so you would increase the cold side of the TEC to ambient - and if you can keep your hotside to within a few degrees of ambient you would then have a super efficient cooling method that would ensure ambient even under high load. you would have more cooling **potential** than you would at higher a dT and would (should) be able to remove more heat from your cold side loop (note: and with your radiator taking out some of the heat - as well as the TEC keeping the temps at room temp) you would then have a C/W of ZERO!! right? granted, you would not have a negative C/W but the cooling potential would be the same or greater than a subambient loop.

if you connect TECs in series (electrically) will you still get the desired dT increase?

no as you decrease the input voltage the delta will also decrease


interestingly applying 50% of Imax while applying Umax like a pwm controller at 50% is not the same thing as applying 50% of umax

or do they have to be stacked to get that effect?
stacking is the only way to increase their delta

if they were side-by-side would you have a greater dT with half the voltage each vs a single at full voltage?
assuming no load is applied then the one at full voltage would have the greatest delta


what effect on QMax does this have? (still working exactly what QMax does for a TEC)
placing TEC next to each other increases the Qmax while stacking decreases it. (relative to a single TEC)

i assume the massive QMax you guys recommend is due to the undervolting of the TECs, whereas the actual power draw from those undervolted TECs is considerably less than total UMax/IMax for all the TECs combined??
YES

i thought QMax was simply the product of volts and amps, but then i saw high QMax TEC's that were well under what i thought they would be. (35V/20A@403QMax)
PMax is 35V*20A Pmax is the power required to achieve dtmax (no load applied )

CoP will always be highest as dT approaches zero
YES

since our chiller loops depend on a greater dT, how do we compensate for the lack of efficiency (ie: higher power requirement/current draw) of an ever increasing dT
Add more radiators

at what point are we simply wasting power for diminishing returns
That depends on a lot of things

thanks for all the input and help - and thanks for putting up with my MANY questions!!
All good i enjoy it

.by doing so you would increase the cold side of the TEC to ambient - and if you can keep your hotside to within a few degrees of ambient you would then have a super efficient cooling method
Adding a rad to the cold side would then add heat to the cold side that the TEC's would then have to move back . On paper the TEC's would be operating at a very high COP. but you wouldn't actually see any benefits from it. not having the cold side rad and simply deceasing the input voltage further would be a better way to increase COP


but that was the whole point - to warm the cold side (to get closer to dT=0). i know - i have seen it said here many times not to do this because it defeats the purpose. and if you are trying to go subambient than yes, it certainly does, but if you only want to maintain ambient (better than normal LC, and without the increase in temps under load) then why wouldnt this make an efficient heat pump. it is still doing what it is intended for - it is removing heat generated by your heat sources.

you've just created a loop . tec's move heat to the hot side which gets put into the air via the rads. then on the cold side this air and heat is then put back into the cold side via the rads. The TEc's then move it back to the hot side. it's a ever worsening situation that never ends.

[bds71]

OK - some things are becoming clearer, but some things just aren't quite clicking - hope i don't sound like a broken record here (especially on the last part) - eventually, i'll just click the "i-believe" button and that will be that....but for now, i want to take another stab at it....

interestingly applying 50% of Imax while applying Umax like a pwm controller at 50% is not the same thing as applying 50% of umax

you lost me - current follows voltage: by decreasing voltage, you will decrease current. whether you reduce UMax with a PWM, or by putting 2 TEC's in series, or simply by putting a resistor/potentiometer in front of the TEC the current should decrease proportionately with the voltage??

placing TEC next to each other increases the Qmax while stacking decreases it. (relative to a single TEC)

increasing QMax is a good thing, yes??

PMax is 35V*20A Pmax is the power required to achieve dtmax (no load applied )

this does not compute: 35*20=700; UMax=403 w/dT=67. dT is reached at UMax, not PMax?? question: if i applied 24V@17A (ie:~400W - UMax) would i hit QMax? <confused>

not having the cold side rad and simply decreasing the input voltage further would be a better way to increase COP

decreasing input voltage decreases dT. increasing cold side decreases dT. all things being equal, i don't think the TEC can distinguish between the two methods so CoP should be the same regardless how its done?

you've just created a loop . tec's move heat to the hot side which gets put into the air via the rads. then on the cold side this air and heat is then put back into the cold side via the rads. The TEc's then move it back to the hot side. it's a ever worsening situation that never ends.

at idle, the situation would be as you state - the radiator would be adding heat to the system (assuming TECs would normally operate below ambient). but at load you have more heat coming from the sources than you do being input from ambient air cooling of the rads (ie: the amount of heat being generated by all the sources - in my case 2x OC GPUs and an OC CPU) would bring a normal water loop above ambient. at **that** point the TECs are adding to the efficiency of the cooling loop to maintain ambient - ie: the radiator is no longer adding heat to the system, but hepling to take heat out of the system. once you reach equalibrium (ambient) the radiator no longer acts as a heat source. at that point it creates a base-line for the TEC to hit. and with the low dT you can easily maintain the hotside within 2-5C of ambient.

i think your scenario assumes that the TECs would be able to **hold** subambient even under load (which is what we all strive to accomplish). in that case, yes - the rad would be adding heat to the loop. (sorry, i'm just working through this as i type this lol)....let me get back on this...more to follow - feel free to comment on previous remarks!!

note: if anyone else wants to chime in on the discussion (keep bashing to a minimum please) please do!!

[Ultrasonic2]

OK - some things are becoming clearer, but some things just aren't quite clicking - hope i don't sound like a broken record here (especially on the last part) - eventually, i'll just click the "i-believe" button and that will be that....but for now, i want to take another stab at it....




you lost me - current follows voltage: by decreasing voltage, you will decrease current. whether you reduce UMax with a PWM, or by putting 2 TEC's in series, or simply by putting a resistor/potentiometer in front of the TEC the current should decrease proportionately with the voltage??



increasing QMax is a good thing, yes??



this does not compute: 35*20=700; UMax=403 w/dT=67. dT is reached at UMax, not PMax?? question: if i applied 24V@17A (ie:~400W - UMax) would i hit QMax? <confused>



decreasing input voltage decreases dT. increasing cold side decreases dT. all things being equal, i don't think the TEC can distinguish between the two methods so CoP should be the same regardless how its done?



at idle, the situation would be as you state - the radiator would be adding heat to the system (assuming TECs would normally operate below ambient). but at load you have more heat coming from the sources than you do being input from ambient air cooling of the rads (ie: the amount of heat being generated by all the sources - in my case 2x OC GPUs and an OC CPU) would bring a normal water loop above ambient. at **that** point the TECs are adding to the efficiency of the cooling loop to maintain ambient - ie: the radiator is no longer adding heat to the system, but hepling to take heat out of the system. once you reach equalibrium (ambient) the radiator no longer acts as a heat source. at that point it creates a base-line for the TEC to hit. and with the low dT you can easily maintain the hotside within 2-5C of ambient.

i think your scenario assumes that the TECs would be able to **hold** subambient even under load (which is what we all strive to accomplish). in that case, yes - the rad would be adding heat to the loop. (sorry, i'm just working through this as i type this lol)....let me get back on this...more to follow - feel free to comment on previous remarks!!

note: if anyone else wants to chime in on the discussion (keep bashing to a minimum please) please do!!

you lost me - current follows voltage: by decreasing voltage, you will decrease current. whether you reduce UMax with a PWM, or by putting 2 TEC's in series, or simply by putting a resistor/potentiometer in front of the TEC the current should decrease proportionately with the voltage??

right lets use this tec as an example
243 Qmax
15.2 Umax
26 Imax
Pmax is therefore 395watts

if you apply Umax(15.2) then amps will Be Imax (26) and Pmax will be consumed (395)

50% PWM Current while applying Umax will be 15.2v * 13amps so consumed power is 197.5watt

If you apply 50% of Umax or 7.6v then amps is 11.19 so watt used is 85watts



If you bin the extra complexities of a TEC right now and just use ohms law for a given resistance then the same is true

Resistance of 5 ohms with 12 volts = 2.4 amps and therefore wattage is 28.8 watts

if we then 50%pwm (only allow it to draw 1/2 the current it wants to draw) the current or apply 1/2 of it we get 5 ohms 12v and 1/2 of 2.4amps (1.2) so wattage is 14.4watts

However if we apply 1/2 the volts 6 to a 5ohms resistance we get 1.2amps for a wattage of 7.2watts


SOO no matter what you do you should always apply the lowest input voltage as possible for maximum COPs




increasing QMax is a good thing, yes??

YES

[Ultrasonic2]

i think your scenario assumes that the TECs would be able to **hold** subambient even under load (which is what we all strive to accomplish). in that case, yes - the rad would be adding heat to the loop. (sorry, i'm just working through this as i type this lol)....let me get back on this...more to follow - feel free to comment on previous remarks!!

note: if anyone else wants to chime in on the discussion (keep bashing to a minimum please) please do!!

Right this comment of yours from the previous reply through me why not put a radiator in teh cold loop? by doing so you would increase the cold side of the TEC to ambient

but at load you have more heat coming from the sources than you do being input from ambient air cooling of the rads (ie: the amount of heat being generated by all the sources - in my case 2x OC GPUs and an OC CPU) would bring a normal water loop above ambient. at **that** point the TECs are adding to the efficiency of the cooling loop to maintain ambient - ie: the radiator is no longer adding heat to the system, but hepling to take heat out of the system. once you reach equalibrium (ambient) the radiator no longer acts as a heat source. at that point it creates a base-line for the TEC to hit. and with the low dT you can easily maintain the hotside within 2-5C of ambient.

Yes you are right. and that would make scene if you didn't have enough Qmax to cool the total load via the TEC's

thats the idea my hybrid block works on .. TEC can't move all the wattage, but can say move 1/2 of it so the load applied to the cold side rad is 1/2 what it would be without the TEC's on and therefore the rise above ambient for the cold side is only 1/2


2 things work against you though

1 if you have one rad in the cold side loop and one in the hot side TEC loop, then you could 1/2 your delta by simply having both rads in the cold side loop and not bother with the TEC's

2 the TEC's are using power and therefore heating up your room resulting in a increase in ambient that you wouldn't see without the TEC's on

[bds71]

1 if you have one rad in the cold side loop and one in the hot side TEC loop, then you could 1/2 your delta by simply having both rads in the cold side loop and not bother with the TEC's

i thought about that, but then i wouldn't be able to have fun tinkering with TEC's!! seriously though, the truth is i'm not sure if my boreas can cool my entire loop by itself. i am inclined to believe it can, but i have not tested it as i don't have the proper (OK, i don't have ANY) insulation and don't want to mess anything up. i put the rad on the cold side to keep temps at ambient - on purpose. when i run benchmarks, or game, or encode my DVDs, or run my 3 instances of F@H - basically 24/7 my PC is under full load - the TECs kick in when/if i hit my set point and cool things down quick fast and in a hurry. if i want to do benchmarks or whatever i adjust the setpoint to zero and the TECs/fans ramp to 100%. with this method i can run at ambient all day long under load but it's loud (TEC's on the boreas are air cooled) and draws a considerable amount of juice. normally, i put the set point about 5C over ambient and the Boreas runs about 50-60%.

that is all current stuff. the project i was working on was to replace the Boreas with a couple regular TECs operating at double the amperage at 12V and i thought i would have "double"(ish) the cooling power. i have since realised my mistake, and now i'm not sure what i'm gonna do with all the stuff i bought lol. i'll probably still be able to use some of it. but the TEC's, as i understand now that they are crap, will not be used. i might still do it - just to do it. but my performance expectations have seriously diminished.

2 the TEC's are using power and therefore heating up your room resulting in a increase in ambient that you wouldn't see without the TEC's on

the way i had planned to set up the new rig would have the heat from the TECs exhausting out of the room (blowing towards the door). i was basically going to (probably still will) build a wooden shroud for the 2x GTX240 rads and the accompanying 8 fans with an extra few feet of tubing and set it angling towards the door to exhaust the heat out of the room - then i was going to mount the pair of TECs/4x water blocks setup to the wooden shroud using a perforated steel custom enclosure/"wrap." as i said before though, not sure what i'll do now regarding the TECs.

[bds71]

if we then 50%pwm (only allow it to draw 1/2 the current it wants to draw) the current or apply 1/2 of it we get 5 ohms 12v and 1/2 of 2.4amps (1.2) so wattage is 14.4watts

i could have this wrong, but wouldn't it actually be:

P=I*I*R (I^2*R) so P at PWM 1/2 I (I=2.4) = 1.2*1.2*5=7.2W (not 14.4W)

and likewise (as you stated i the next step) P=(V/R)*V so P at PWM 1/2 V (V=12V) = 6/5=1.2*6=7.2W

Current follows Voltage and Voltage follows Current - if you 1/2 one, you have to 1/2 both.

but i get what your saying: for best COP you want the lowest Voltage (probably between 20-30% UMax). in this regard 3-4 TEC's wired in series would give the appropriate COP - then wire another 3-4 "sets" in parallel to boost dT and keep COP's high (yea, 12-16 TEC's) to get the best performance!! <light bulb comes on> that's why my Boreas (using 12 TECs) only consumes about 10-12A @ 12V (120W) (MUCH higher COP) and will WAY outperform the two 10A/12V (240W) TEC's i bought. i get that now - lesson learned now i just have to buy 14 more TECs and a whole lot of copper lol

[Ultrasonic2]

Explaining how PWM might help you

It turns the current no and off so full power or no power or 28watt or 0w

So at 50% PWM the power is on for ½ the time and during this period it is drawing the 28 watts. The other 50% of the time the now power is being applied

On,off,on,off,on,off

28w,0,28w,0,28w,0

The average is 14watts or 12v * ½ the potential current


By the way you can test a true PWM controller and this is exactly what you get





If you have a TEC setup cooling “A” load while achieving B delta why consuming C power

If you then double the load AA and you want to keep the delta the same B and the consumed power C the same then you have to add more Qmax and under volt them more to keep C the same and get B around the same

However it maybe impossible to achieve the above because you have lowered the in put power and therefor max achievable delta has been reduced so it maybe impossible to achieve B anymore while using C power

[bds71]

yea, i guess PWM's don't fit so nicely into Ohms law because they use more of a duty cycle than a true measure of constant voltage/current. i do understand how they work, just not sure on the math hehe - we didn't do that kind of math when i went to school (ITT) :P as for the output of a PWM controlled TEC i think it's just time for me to press that "i believe" button. i get the concepts (much better now - after "talking" about it with you for the last couple weeks or so) and i think that's enough for now. i can at least go back and re-read some of the projects around this thread and "understand" why folks are doing what they are - and more about the suggestions being made.

with that, i'll let this thread get back on track - you were talking about your new TEC project, but i guess it's finished, eh? did you ever get a chance to install an OS and test it properly? i would be very interested in the results. oh, and i do like like them blocks you made (a little flashy for my taste on the covers, but very nice base and exquisit craftsmanship - good on ye!!) once i figure out how i want to proceed with my own project, may i inquire about getting some blocks made? or do you even make custom blocks for others?

[Ultrasonic2]

I've been on holiday

FFS !!!! another TEC has died ! ARRR

[thePoison.]

I've been on holiday

FFS !!!! another TEC has died ! ARRR

oouch those can add up quickly, where do you order yours from?

[Ultrasonic2]

well these are old ones that have been in a thousand blocks so im not surprised just disappointed

[Morphling]

This is a real master work There is too much technical knowlege here. Congratulations

[NEOAethyr]

Ultrasonic2, have you put any thought into making ram blocks?
At the moment it's the only thing I can think of to put into real use with tec's.
I'de love to sub ambient my ram 24/7.

If anything, at least for my nf2 which loves ram in the lower teens.
Uh that don't look right lol, it's funny though, I mean, lower teen c's in temps lol...

These thigns sure get darned hot though regardless of what voltage you put into them.
I fried my finger a month ago showing my cousin just how much they heated up with low power.
Didn't expect it to be that hot, I knew it got hot before but d@mn..., melted my skin, no blister, just plain fried my finger.
I've got a 12709, it's a 9amp 40mm with 127 thermal couples in the array block.
Supposed as a teg it can produce around 13-15 watts but I'm doubtful.

Anyways I'm really interested in some sort of halfway cheap/reasonable ram tec block.
The prob I'm thinking of is how to get such a thing mounted up to the sticks and the gaps needed, etc.
Thinking around 1/4 in copper plates on both sides, some sort of screw setup to tighted them together, and then another block on top with the tec's and so on.
Probably not feasable at all.

[chaotic]

It is very easy to make a ram/tec waterblock...this is my way- http://tecpeltier.com/viewtopic.php?f=4&t=58 you can make the same with this products- http://www.aquatuning.de/shopping_ca...f8ff24db4f2d4c but this is only for naked ram not dominator