Printable View
I have started the work on a waterchiller.
The idea is to use 10 of each of these:
70W Qmax peltiers, $50
Scythe Heatlane Zen NCU-1000 heatsinks, $100
Koolance CPU-100 waterblocks, $40
So far $190 spent.
I have an old 12V 300W PSU to I plan to use to power the TECs, passivly cooled.
I will update as I make progress.
Some pics of the heatsinks and waterblocks.
I plan to stick a peltier between each heatsink and waterblock. I will connect to waterblocks in parallell. The 3/16" connections on the Koolance WB makes this a must.
The whole thing will be mounted on top of my case. As the pictures above demontrates. Some kind of frame will hold the heatsinks above the top of the case so that air can flow up through the fins of the HSs. If necessary, which is likely, some quiete fans blowing air from beneath will help convection.
I am not sure how I will control the power to the peltiers.
I have been thinking of using the CPU-fan control output voltage as a control voltage for a mosfet powertransistor that would control the voltage fed by the extra PSU to the peltiers. To bad I dont know enough electronics to know if this would work.
This method should allow easy control to run condensation free, but still keep the system just a few degrees above condensation temperature.
I am open to suggestions..
Wahooo
I love TEC Projects, keep us posted, lots of Pic's..
cool:D
Glued together silicon tubing, two 12mm/18mm tubes with ten 5mm/9mm each. The 10 Koolance blocks will go onto the small tubing.
(camera phone picture)
The option to control cpu fan speed via pulse width modulation is available via the motherboard cpu-fan connector. And I was planning on using it to control current to my peltiers.
According to one TEC manufacturer switching frequency need to be in the order of 1khz or higher, or the TECs efficiency will suffer noticably.
http://www.marlow.com/technicalinfo/...tions_faqs.htm
So I am asking if anyone know what frequency that cpu-fan speed PWM works at?
I don't think a 300W old psu is going to hold you.
I smell doom.
I wish I could smell over the internet :(
No wait, no I don't.
I will undervolt. I could get another powersupply if i wanted to go past 300W of input power. But I think thats wasteful enough as it is. And as you can see from my post about temperature control I dont plant to get water temps down much below ambient. Atleast not until I feel like undertaking a big insulation/isolation project.
I am not sure what you are trying to calculate there Hollo. If you compare cooling power with max 70W you need to stay at 0 deg temp delta.
And when temp is constant cooling power is linear to voltage in the working range of the peltier element.
So a better calculation would be
Qc@6V (6/15)*70 = 28W (280W for all 10)
P@15V = 121W
I@6V = (121W/15V)*(6V/15V) = 3.2
Pi@6V = 3.2*6 = 19.2W
COP = 28W/19.2W = 1.458
Althoug I feel this is wrong too. I will look up correct numbers later today.
COP should be higher.
With 300W of input power a working point COP could be as low as 0.66.
And peltiers should be able to reach between 20-30 degrees delta at this temperature. That would load my HS with 500W, 50W each.
If voltage goes down current must go up!Quote:
Originally Posted by Jorlen
What you really should be doing is wiring them in series so they can take 140V, then rectify AC to DC with a filter then feed it to the pelts through a veriac. This will keep current within sane levels and still provide full power.
If you wanted to get really fancy you could use a chopper circuite to regulate current and inject a PWM to the chopper reset circuit to control over all cooling capacity.
@Jorlen
sorry you can pretty much ignore my post, i thought they were 70W power consumed, not 70W Qmax
i'll delete the post since it's useless
did you get the 121.6W ones?
Dimensions: 40mm x 40mm x 3.2mm
Wire length: 34cm (13.5")
Model number: TEC1-12708
Voltage: 12V
Couples: 127
Umax (V): 15.2
Imax (A): 8
Resistance ( £[): 1.66 ~ 2.0
COP = heat-load over power-consumed
= Pload / (19.2 x 10)
since Pload is probably about 100W for a dual-core, while theoretical cooling power is about 280W (cooling power is linear with voltage while power consumption is square with voltage? :eek: that's great), it's looking good
if you don't want condensation problems and want to start straight away (before building a voltage controller) you could do 9 peltiers @ 4V (76W total power draw ((4/15.2)^2*121.6*9), 167W capacity at dT=0 (Qmax conditions, 4/15.2*70*9)) :D
there's no way it would overload... in theory....
what CPU do you have?
No. That is only if power consumption stays the same ;)Quote:
If voltage goes down current must go up!
Thanks for the tip Xeon. My plan is to use the PWM of the motherboard cpu-fan output to a drive chopper circuit that controls the current from my PSU. Is chopper circuit the correct term? :)
Yes Hollo, the ones on e-bay. Though I havent got them yet, but I have paid for them. We shall see if and when the show up. Its a long way from Hongkok to Stockholm.
@NOL I can always get more PSUs if I want more cooling power. But at this point I am not aiming for maximum cooling, just really good watercooling. Think of it as watercooling with a slight performance boost. Not waterchilling to compete with direct phasechange systems.
And here are some pics. Scythe packaged a nice copperplate that mounted according to instructions. I drilled 2 extra holes in the copperplate and threaded them. And cut up a few of the motherboard back-plates that came with the HS to make the small beam that goes across the waterblock. The beam is stainless steal 1.5mm thick sheet metal.
Neither do I. Also, I think if you just ran hose the normal way, a lot more water would move through the blocks and therefore more cooling would get done. Also, those heatsinks don't look like they can take 70 watts, especially passively.Quote:
Originally Posted by n00b 0f l337
As ben said.... I would have run the hose normally. the way youve done it wont the water just rin along the bottom? as its not being forced into the blocks. just a thought.
300W if input power is just fine for my purposes. So the PSU will hold me. What is the normal way to run a hose? Are telling me to run 10 pcs of 5mm innerdiameter waterblocks in serial?
Thats just crazy.
And the water wont run at the bottom. Water will travel in parallell through all the waterblocks. Do I need to mention that i wont short-circuit the waterloop by allowing water to circulate through the system without passing through the waterblocks?
what is your CPU?
and will you do a 4V x 9 peltier run for me :D?
i've never seen a peltier setup with a COP greater than 1 before
Sure, I will can run them at a variaty of voltages. I dont remember the range of my PSU exactly. But i think its 10.5V to 13V. Its tunable via a potentiometer. Using it and different ways of serial/parallell circuits I should be able to test a lot different voltages. I can measure the PSU range later today when I get home to see how far it ranges.
My system is a E4600 on a Gigabyte DS3 P35, Ati HD2600 pro. I havent taken my time to figure out highest stable overclock yet. And its possible I wont overclock at all. And I dont run folding or other stuff that consume my idle cycles.
I got the tracking ID from the ebay peltier purchase, and the package is being shipped out from hong kong today.
i just had a great idea - gotta share it :rolleyes:
set it up with 9 peltiers
eg. Pin from your 10.5-13V power supply:
min (((10.5 / 3) / 15.2)^2) * 121.6 * 9 = 58.0263158W
max (((13 / 3) / 15.2)^2) * 121.6 * 9 = 88.9473684W - high enough??
set up the 10th peltier between a waterblock and air-cooler, like the others, but connect it to a small amplifier circuit that is connected to a relay so that when it generates a certain voltage due to the water (water-block) temperature being lower than the ambient (heat-sink) temperature it will deactivate the entire peltier loop.
(i'm not sure what the circuit would look like, but it'd be pretty simple)
it'd be perfect because condensation is dependant on the difference between the loop temperature and the ambient temperature rather than an absolute temperature. it'd just need a circuit that was quite sensitive to (or good at amplifying the) current from the peltier so a water temperature 1-2 degrees below ambient would be enough to shut the circuit down.
it'd be a cool 'feature' for day to day usage :hrhr:
The idea would work. I am trying to stay away from solutions where thermal cycling of the peltier elements occur. It reduces the life span of the TECs.
PWM is the way to go. Since it can control the power much more fluently, even though in the microscale it switches on and off at kHz speed.
Either I use the motherboard, or I create my own potentiometer+thermistor controlled PWM circuit. Or rather I copy/modify someone elses.
I have a psu I pulled from a device that used an NTC thermister controling the inverter chip, I'll see if I can't get some revers engineering don on it today, I won't make any gurentees though.
ahh i see
perhaps you could create a circuit so that voltage generated by a peltier when the water-block temperature is higher than the heat-sink temperature would feed the "base" of a "transistor amplifier circuit" which would have a gradual effect on the amount of current that could flow through the "collector" and "emitter" - as water temperature drops to ambient the current shuts off in a nice analogue curve.
(i've drawn the "transistor amplifier circuit" with "base, collector and emitter" as symbols of transistor function - i dunno what the real circuit would look like)
then you'd still have the 'luxury' of a water-to-ambient-dT based power control - but not a digital on/off one.
you might also want a variable resistor that you could use to set a minimum amount of power that'd flow into the peltiers regardless of water temperature - this'd allow you to set a minimum temperature near or under ambient. without it the minimum temperature would be slightly over ambient. if full load temperature is over ambient i guess you'd leave it out.
though if the ambient temperature in your computer room is fairly steady this'd be a bit overkill
Please post what you learn from that Xeon.
Hollo, its completly possible to do as you describe. The drawbacks are that a power transistor works a variable resistor in the way that it increases its own resistance to control the current/voltage. Thats why you need massive heatsinks on these suckers. So although it would work, its my second choice. Dont wanna generate more heat then I absolutely have to, you know. ;) Its a cooling system after all.
A peltier device is one way to measure temperature difference. But its a rather bulky operation considering you need to dedicate a whole HS waterblock peltier assembly just for it. I imagine a much more space and cost efficient solution can be made from using two thermistors that are for measuring temperature in air and water. And via some resistors and transistors you can transform the difference in resistance into difference in voltage. And use that voltage differential to drive a power transistor. Maybe you need one inverted ranged thermistor if you wanna make it easier. As you wanna decrease voltage differential the cooler it gets.
I dont know.
My first priority is to measure the output of my motherboards cpu-fan socket. And see what I can make of that. There are opto-transistors that I can use to keep galvanic separation between the TECs and their high currents and the motherboard.
a few people have said it before and they're right....you're not going to get much flow through those waterblocks with the hosing in parrellel. Unless you have something putting direct flow into those tubes the water is going to go the path of least resistence; the larger diameter tube. You'l have some really cold stagnant water though.
Ok here is a picture. Look carefully, water HAS to pass through the waterblocks.
ahhh. ok... sorry,i misinterpreted your first pic. looks like it should work nicely.Quote:
Originally Posted by Jorlen
I like this:
http://www.overclockers.com/tips1240/
Another way to do it would be to make a box out of copper/aluminium instead of acrylics and positions the fan to create best possible circulation and let the mineral oil cool against the walls of the box. Then place it outside the chassi.
Quote:
Originally Posted by hollo
You could do this by feeding the pelts out put to a Voltage controlled Oscilatore then filter it through a schmit trigger for a more streamlined aproach.
When that thing goes it's going to go good, not worth it IMO, fare more effective and safer ways to achieve the same effect.Quote:
Originally Posted by Jorlen
Ok, I dug out the old PSU. Its called HiTRON model HSH250C-11. Its actually only 20A@12.5V, 250W.
The unloaded voltage range goes from 9.35V to 15.5V. I hope I have that range available under load too. I dont have any suitable resistance available to test it under load atm.
The coils (I suppose) started making some faint noise when voltage was tuned below 11.5V. Nothing to write home about, but I prefer it makes no noise in the final system. :)
Its completly silent above 11.5V which is nice. Hope it stays like that under load too.
you may have to put some kind of load on the 5v rail to get a hgher/more stable 12v rail. some PSUs like a little on the 5v to get em going too.Quote:
Originally Posted by Jorlen
Thanks for the advice Sam. But this PSU is not from a PC. It only has one output.
This is a picture of the HiTRON next a regular PC PSU.
ah now I see the tubing more clearly. I didn't realize you had it configured more like a HX. Sorry for my mistake. Otherwise, it looks good so far. You may have said this, but i think you will need an enclosure with fans for the Heatsinks to be effective.
with a 'coarse' heatsink (wide fin spacing) on the water-block with the internal fan pointing at it it'd be excellentQuote:
Originally Posted by Jorlen
well in a way the power consumption of a transistor in such a circuit would be a moot point, because whatever voltage controlling mechanism you use it'll boil down to a resistor connected in series with the peltiers, and the resistor will need a heatsink and the voltage moderating circuit will be inefficient (unless its a binary on/off circuit, which you don't want).Quote:
Originally Posted by Jorlen
however using 3 sets of 3 peltiers would be extremely efficient in its own right, because it would require the least damping of the voltage going into the peltiers during normal operation to avoid condensation. 5 sets of 2 peltiers would require a bit more resistance in series with it to moderate the peltier voltage enough to avoid condensation, and 10 peltiers in parallel would require the most resistance.
since 3 sets of 3 peltiers would be the most efficient use of the peltiers and PSU (so long as it wasn't underpowered) it would make sense to set it up like that .... the possible use of the extra water-block/peltier/heatsink for dT based temperature control was kindof a bonus
what were you planning - 5 sets of 2 peltiers each?
Hollo switching on and off at high speed (PWM) is very energy efficient. Losses in TEC performance is 1% roughly, and losses in the switching transistor I do not know exactly, but they are very small in comparison.
Yes I plan to run them 5x2. 2 in serial 5 in parallell.
Resistance of one peltier
15.4/8 = 1,925
Total array resistance = 1.925 * 2/5 = 0.77
at 3.5A per peltier (17.5A total) this requires 13.475V, gives 6.7V per TEC.
Total power draw 236W.
So thats pretty much my max without changing PSU.
I plan to cool GPU and NB too. So total cooling heat output might be in the range of 100W-150w.
38W per HS, at a passive C/W of around 0.8, means a temperature increase of, eeek, 30C degrees. but its easy to improve upon it, a quite fan will take me down to 15C degrees above ambient.
ahh, PWM
http://en.wikipedia.org/wiki/Pulse-width_modulation
i get it now :slap:
Got the peltier elements now. Whats missing now is the pump, more hose, insulation, structure that holds the HS and waterblock strucure a few inches or more above the top of my case.
The pump is a DDC 10W, which can be modded to 18 if I am lucky with the version number. I might build a box around it if I find it to loud for the rest of the system. Would be nice if someone could confirm that the DDC will be fine just being cooled by the coolant.
Pics:
Pump arrived. I tried to solder it to get 18W. But I screwed up and tore the targeted solder face from the PCB by mistake. I will need a finer soldering Iron if I wanna repair it. For now 10W will have to do.
Bought the OCLabs top cause it was the cheapest and capacity wasnt that far behind the XSPC top.
I managed to borrow an oscilloscope. And the PWM working frequency of my Gigabyte DS3 cpu-fan output is close to 23kHz. I changed to speed of the fan with Gigabytes Easytune, and the oscilloscope showed how the width of the pulse changed as one would expect.
To complete the TEC controller I now need a set of transistors that can pass a combined current of 20A and switch efficiently at 23kHz.
Didnt bother to take any picture of the measuring procedure. For those who want to know. I used an unused PWM fan from an Intel standard cooler and connected the oscilloscope to the black and blue wire. Then I counted on the screen the nr of cycles completed over the course of 1ms, which were 23.
Hope it will work fine for you, cause I'm thinking of doing something similar just with one 62mm 545W TEC ...
Do you think it will be cold enough and ?
Awaiting your results so get your build done asap ;)
Update. I managed to break one of the heatsinks when mounting the TECs and peltiers. I am being a bit careless, I stress to get more work done. As a result, things go wrong. I dont recommend this way of working. ;)
But things progress.
This is how far I have gotten. 9 out of 10 heatsinks+TECs+Waterblocks mounted in a frame.
looks sweet
@Hollo
Since I broke one heatsink I can only run 9 TECs. I am running the system outside the case atm running 9 TECs. I have no load on the system its just circulating water. The 9 TECs together draw only 5.1-5.4A from the supply.
Each TECs running at 4.1V and 1.7A. 63W of total energy draw to the TECs.
The water is getting a bit cooler. And the Heatsinks are getting a little hotter. Not much to speak of.I might run it like this on my CPU to see what the cooling capacity is. But first I need to fix a few problems.
I got two problems atm, reservoir is crap. Cant bleed worth a damn with it. Need to rebuild it. And the waterblock for the CPU is leaking. This probably wont be to hard to fix.
That looks sweet, nice job so far :up:
I wouldn't worry about this. In a normal setup, a more powerful pump will put more heat into the water and hopefully this will be counteracted by the increased flow.Quote:
Originally Posted by Jorlen
In a chiller setup, because of the peltiers, the increased heat in the water can become more or a negative factor than the increased flow is positive (I think...).
So, in my air cooled chiller project I am aiming to use a DB-1 pump as its only 6w despite the fact I own a more powerful 18w DDC. Then again my cold side loop is very unrestrictive which is a major factor. When I have finally finished the chiller I will test both pumps so try and prove this.
I can change voltage to the pump and see how the performance of the system changes with pump power. If the effect is noticable and favorable I might attempt the modification. But noise level is even more important for me. I havent heard the pump at 18W.
I have modified the reservoir and repaired the cpu-wb now. And I bought a new heatsink to replace the one I punctured.
Do you have any thermal probes? Even those cheap ones like this.
Good, I think you will find more or less the same temps with it at 18w but with more noise and wasted electricity. Personally I found my 18w pump too be too noisy as a fan of quiet computing.Quote:
Originally Posted by Jorlen
I dont have any thermal probes. I will probably get one.
System is up and test running atm. Using the configuration described above with 9 TECs. No fans are running anywhere near the heatsinks. E4600 @ 3ghz, 1.4V core voltage setting. Ambient around 23C
Idle temps from Everest:
NASA -we have lift-off
Sweet, fun stuff
First screenshot is during Prime95 that has been running for two hours and second screenshot is just a few seconds after i stopped Prime95.
Ambient temperature stayed the same, about 23 degrees, during the whole time. The CPU temp shoots up about 10-13 degrees C in a second when Prime95 starts. Then water temperature rises about 5-6 degrees over 10-15 minutes and then reaches equilibrium. When Prime95 stops cpu temps drop the same amount just as fast. And when I am writing this the water temperature is falling and I assume the cpu will reach same idle temps as before within a few minutes. As a comparison I had Arctic cooler Freezer 7 pro mounted prior to this setup.
http://www.arctic-cooling.com/cpu2.php?idx=79&disc=
And under same load it maxed out at about 45 degrees C, using a 7W 92mm fan at full blast.
I am very happy with the results. I am only using 25% of the what the 250W PSU can dish out. I not using any fans to cool the heatsinks. And at current humidity I am not condensing any water.
More pictures to come.
Nice results
So how much did you invest in total, 190usd?
What is the water temperature at load and idle?
Water temps I havent measured yet. But I would guess the Idle temps is about where its at. So 7-9C after long time Idle. And 16C at max load.
The cost has run away for sure. If you wanna count everything including the things I had lying around (ie. the PSU) it cost a lot more then $190 US.
Pump was $100, so was PSU 6 years ago.
Hoses will be almost $100 when include the NB and GPU. Silicone hose is freaking expensive, and I didnt get a good price. Only fortune is that I could build the reservoir and cpu waterblock for free at a friends workshop.
So more like $500 in total.
Normal watercooling is expensive. And when you change the radiator for 10 x (TECs + Waterblocks + Heatsinks) it doesnt get any cheaper.
:shocked:Quote:
10 x (TECs + Waterblocks + Heatsinks)
now that is a monster TEC build, :up:
I won't even go near that price thing, I know how the
cost just keeps going up, there is
always one thing that was forgotten or
had to be added, always,,,,:rolleyes:
I think the only reason your getting this is your only adding about 100W to the loop.
I wonder if the temperature difference of 5 degrees C between my two cores is because of uneven mounting of my waterblock.
I dont have any springs to even the load of the 4 bolts that hold the waterblock. And the hoses are thick and elastic, pushing the waterblock sideways as they are bent.
What do you mean NOL, only reason. Reason I get what I get is that I have what I have and load what I load.
Other load other temps. More fans other temps. More wattage to the Peltiers other temps. Gotta sleep now...
nice job. have you tried upping and lowering the voltage to the TECs?
Ya, I wanna see something like this under a quad
Assuming my CPU puts out about 100W, I have a total load on 9 heatsinks of 163W + heatdump from pump, 170W.
With one more TEC and going 4V -> 5V, power will 109W to the TECs. System total will be 46W higher, 216W. Thats a 38% increase in cooling power and a 27% increase in heat output. With 11% more heatsinks.
I have just taken a regular inside/outside digital thermometer and plunged into the reservoir. And the temperature does not match what Everest shows. Water temperature lowest is 15.7C, 23C ambient. And it seem to stagnate at 21.5 under load. This is a cheap thermometer. But I doubt its scaling is off by much. So temp rise 5-6 degrees C is about right.
But the CPU temps are probably way off.
You can always turn up your vcore to simulate hotter cpus and do your tests again. It would give you an idea of how it performs with various heat loads.Quote:
Originally Posted by Jorlen
At 3GHz,
1.7v =150W
1.6v=133W
1.5v=117W
1.4v=102W
1.3=88W
I am currently building a little calculator program for this kind of application. So far its very rough, but I just ran some numbers through it and was pleased to see it seems to match what you are saying.Quote:
Originally Posted by Jorlen
with 100w heat load, 9 heatsinks with peltiers at 4v I get power consumption of just of 63W.
with the same load and 10 heatisnks/peltiers at 5v I get power consumption of 109w (you said 63 + 46 = 109).
In the first case, to get these figures and water temp to match your 21.5c at ambient 23c I had to make assumptions about the c/w of the heatsinks. To get this to work I had to put in a value of 0.9C/W which is frankly, terrible. But then they were cheap and are passive.
However, this also showed that going to 10 units at 5v with heatsinks as they are, would give a water temperature more like 17.5c, does this come anywhere near your expectations?
EDIT: a quick look on frostytech suggests that the C/W of the heatsink with basically no airflow with a 27mm 50w heatsource is 0.94 C/W. This will improve with a lerger heatsource, modern thermal paste and get worse due to the lower heat load you are applying and less air speed, so 0.9 probably Isn't miles of. Also I add the thermal resistance of the thermal paste separately and they obviously include it. This probably brings the C/W of your hot side cooling dangerously close to 1 C/W.
Have you considered some low speed fans to give the system a boost? I think it will make the world of difference.
I have considered fans. But I need to point out the keeping noise down is more important.
There is a rather large copper heatplate between the heatlanes on the heatsink and the hotside of the TEC. The plate is warmer then the heatsink by a noticable margin. Which to me suggests that thermalresistance between heatsink and TECs is far from optimal.
I knew this when I decided to include the hotplate. It comes with the heatsink. It made things a lot easier when coming up with a way to assemble everything. And the peltier is 40mm while the heatsink is about 60mm, which mean I would use only 2/3 of the small heatpipes in the heatlane without at heatplate in between.
Thermal resistance of heatsinks using passive convection goes down as they get warmer. I dont know if its obvious, but at constant airflow, energy flow between air and heatsink is in linear relation to temperature difference of the air and heatsink. But since the airflow increases with the higher temperature of the heatsink the C/W goes down. With this in mind I am still hoping to cool well without fans.
There are so many materials for the heat to transfer through in my setup:
CPU, thermalcompound, heatspreader, TC, Waterblock, water, waterblock, TC, peltier, TC, heatplate, TC, heatsink(which is complex aswell, since it heatpipe-based), air.
Something about this is not optimal. ;)
I think you are right to use the plate for mounting. Its very very important to cover the whole peltier and finding heat sinks with a large enough base area isn't easy.
As I'm sure you know; things to check are that the plate top and heatsink base are both flat, that you have a thin but even layer of good quality thermal paste and that you have the two clamped together as evenly and tightly as you can manage.
It might also be possible to permanently bond the plate to the heat sink for better thermal conductivity.
I know this very well. On frostytechs review then got 0.94 C/W with a 50w load and 0.64 C/W with a 100w load. But you have less than 20w load, so you are hardly pushing your heatsinks. Hmm, I need to think.Quote:
Originally Posted by Jorlen
Great work on this Jorlen -
thoughtful design and a cool idea. Not exactly a price performer, but hey, when is anything extreme also affordable?! :)
Ok, the system is running with 10 TECs now at 6V per TEC. I have connected the GPU and northbridge to the loop. It was a pain to make a mounting system for the koolance waterblocks, to mount them on the GPU and NB. But it turned out ok in the end. Since I didnt know if the wiring would be final I didnt bother to make it pretty. Have that in mind when I post pictures later. ;)
If I get my hands on a high precision thermometer I can find out if all the TECs get the same amount of hot water. They feel about the same temperature, but that tells me close to nothing. ;)
Since I hooked up the waterblocks in parallell and GPU CPU NB in parallell I can choke the flow to each part independantly. I doesnt seem necessary with my current temps and ambitions. But I have that option atleast.
Regarding the temperatures I am pushing the limits of the passive cooling capability of my 10 Scythe Heatlane. Its barely worth giving more voltage to the TECs the C/W is so bad that the increase in performance is barely measurable.
So with double the energy expended in running the TECs all i get is a water temperature that is 1C lower at max load.
My system is down to 3 moving parts now. A reasonably quiet 500gb Samsung HD, the 10W DDC pump, Corsair 450W, 120mm PSU-fan.
The pump is hanging from the hosing outside out in the open. Its fairly quiet. The HD is suspended in rubber. And the PSU is about as noisy as the pump.
I have an old PSU that will come under the knife next, I plan to lengthen the legs of the power transistors in the PSU. And move them outside the PSU box, possibly outside the computer case aswell. And with bigger heatsinks on them, their temperature should stay within operational range, using passive convection only.
I dont know at which frequency the transistors work. And I dont know the inductance the extra length of the legs will add. Or how that might have an effect on the operation of the PSU. If the PSU mod works and it doesnt get too warm under these new conditions, I will perform the same modifications on the Corsair.
The final task will be to build am sound proofed box to house the pump and HD.
E4600 @ 3350mhz and 1.55V setting (shows as 1.5v in CPUID)
10 TECs @ 6v
Prime95 for 1h 30min
Ambient 23.5C
Watertemp 22.3C
Well done, those are good water temperatures for a passive cooler (obviously).
Whats the power consumption of the CPU at those settings? Around 150w?
Did you measure the current draw of the TECs running like this?
Do you know how this compares to running the TECs at 5v?
I have not connected the Ampere-meter, although I have an old instrument that will sit on top showing current draw of the TECs.
I have not run the tecs at 5V under this load. Will get back to you on that.
I dont know the exact load, the GPU and NB add a few watts. I suppose its between 120W and 145W all together.
Comparing external thermometer in the waterloop and CPU temp in Everest. I find it strange that the water temperature ranges about 7 deg C from idle to max out temp, and CPU temp ranges 33 deg C during the same test.
One explanation is that the thermal diod on my CPU is somewhat broken. I already know its off measuring the individual cores. Since the measure below water temps. What else is wrong I dont know.
I wouldn't use Everest at all for reading CPU temperatures. Try something like Coretemp or TAT.
Either way, I think the water temperatures are the most important measure of a chillers performance, assuming you have the CPU well enough connected so that most of its heat actually reaches the water in the first place.
everest uses the same readings as core temp
Oh, ok. I'm a bit behind with temperature reading software. Have you tried Real-Temp?
Very cool idea. I really like your concept and setup. Great job!
The reliable way would be to place a thermalprobe under the heatspreader of the cpu. I think heatsink reviewers out there do that.
I can try to do a idle calibration when I get home with realtemp. But the would only shift my temps so they are more correct at idle. I need to add probably 10 deg C on core one and 5 deg C on core two.
That takes care of idle temps. But that just puts my load temps more 30 degrees above water temps. Which would indicate my waterblock is the worst in history (it might be), but I seriously doubt that. I will measure the temperature of waterblocks copperplate. I bet its very close to watertemperature.
And if it turns out as I expect that leaves me with the conclusion that the integrated heatspreader(30C temp delta, eeeek) is crap or internal temperature measurements are useless.
Jorlen -Quote:
Originally Posted by Jorlen
Brilliant work - I especially like the idea of using the fan controllers on the MB to control the TEC. If you need any help on that, glad to provide it - I have designed a lot of big power chopper control circuits and can help with device selection etc. With high speed PCM it's a good idea to add a low pass filter on the output - any old burned up PSU has all the parts you need for any power PCM designs at these levels.
I know the Heatlane Zen is designed for passive, but for others reading the thread, heatsinks designed for forced air don't do well passive - the fins are usually too close to each other to promote air flow from the convection transfer. Forced air sinks often have spacing of 1/8 inch or less, while passive usually have 1/4 or 3/8 spacing with longer fins.
I'm not a fan of the Heatlane Zen, despite its cool looks. I played around with one a few years ago and was disappointed with performance. My testing was done with a controlled heat source at 40W, 50W and 60W in 25C ambient. Even with a 40W load, the base was at almost 70C. In comparison, a big typhoon with the same load held the base under 30C, and a Melcor HX5-301 holds it to 28C. The Heatlane Zen is big and hard to mount, and passive performance is mediocre - about 1 C/W. But I was able to get better cooling with a little airflow - I mounted a Vantec Stealth 120 to blow through it, with the fan at 9V turning under 1000 RPM, which was virtually silent. That airflow took the performance to .55 C/W.
I'm pretty sure you are using 12708 TECs. At 5V, you should see about 2.1A per TEC, with a 25C differential. That puts you at a CoP of 1.5, about as good as you can do with a 25C differential, which is all those sinks will do without active airflow. That means you are pumping about 15W per TEC, and a total heat load of about 25W.
Going to 6V probably takes you to the 2.45A level, and now you are burning almost 15W per TEC. You will move 20W per TEC, but efficiency has fallen to 1.36 CoP. Your hot side is probably around 60C - I'm surprised you got any improvement in cold side at all.
Just adding minimal airflow to those sinks should take them to under .6 C/W, which with 5V on the TEC takes your hot side from probably around 45C to more like 35C. At 5V you will move to CoP of 1.9, and you will move 20W per TEC or better, with no increase in power, still at 5V. If you can get the Vantec Stealth fans, give them a try - even at 12V they are only 28db, and at 9V I really can't hear any sound at all.
One other point - I had a hard time getting good contact with the Heatlane Zen. Are you sure that the heatsink makes contact all across the TEC? Also, correct and even pressure is important - there are several references to the right way to do it on this forum, including mine at http://www.xtremesystems.org/forums/...=184686&page=2
About the choice of heatsinks, they might not be the best heatsinks but they were about 1/6 of the price of the heatsinks that could condend. And when you need 10 of them price is an issue, I would not have started this project if the first thing I needed was to spend $500 on heatsinks. Same with the waterblocks and TECs. Even though the resulting pricetag did run away a bit, I think it would be hard to do it much cheaper. I will be on the lookout for cheap quiete fans. I think it would be smarter for me to cycle fans on and off and run TECs at low power. The idea of driving TECs harder when there is a load falls short when the thermal resistance is as high as it is without fans. So I will need fans to justify running TECs higher then 5V.
I am not completly satisfied with the way the Heatlane is clamped to the hotplate. The plate is to thin that clamps the heatlane to the hotplate. So the pressure on the heatlane is at the edges closest to the bolts. And not in the center. The pressure over the TECs is probably fine. Cant improve that without reworking the whole thing. Right now I have 2 M4 bolts and a very sturdy U-profile of 2mm stainless steel that secures the waterblock and TEC to the 4mm hotplate.
I understand about the sinks - any price times 10 is significant. Those have a lot of area, and with a little airflow should do fine as long as you are getting good contact everywhere. That baby looks about right. Leaving the voltage at about 5V per TEC and getting the cooling right is probably the best way to get efficiency up.Quote:
Originally Posted by Jorlen
I saw your comment about modding the PSU. But as you suspect, the dimensions are pretty important - switching speeds are over 50KHz and even a little parasitic transfer can cause big problems which can wipe out the other components in your rig. There is a much easier way if you have the space. Get a PSU with much more capacity than you need, put it in a bigger box, and bolt on bigger sinks to the main cooling inside the PSU. That is usually easy to do and the sinks that need cooling are obvious.
An alternative is oversize the PSU, but instead of increasing the sinks, use a bigger fan running slower. I have seen some 'silent' PSUs that do this using a 140mm fan running 800 RPM - not totally silent but very quiet.
By the way, you had an earlier question about frequency of PWM power for TECs. You need to decouple the PWM with a low pass filter including a decent sized capacitor ahead of the TECs but it's very easy to do. Cycling TEC power using essentially a thermostat is OK too, as long as the hot side and cold side don't change temp much - as in a design with WC on both sides. But with air cooling, the hot side will have a bigger spread between powered and unpowered. The issue is that thermal cycling causes expansion and contraction of elements inside the TEC, which causes stress fractures, which cause the TEC to fail. You want to keep that cycling to a few degrees at most to avoid problems.
I have an alternative solution to the noise issue. Since the pump, HDD and PSU create about the same amount of noise it only makes sense to take care of all three. So my solution is to make a soundproof enclosure of the right compartment of my servercase. Where HDDs PSU anb pump will be. The picture below explains it pretty good. When running the sidepanel will seal the remaining side of the box and the hole where the hose and wires enter the left compartment will be tight. The exhaust of the powersupply will push air through the radiator into the chamber on the other side of the radiator. And since that chamber is separate the air will leave by passing through the radiator in the other direction above the PSU. This will limit the need of fans to just the one in the PSU. I do not know how much extra energy will leak through the sound isolation material and into the water via the radiator. I dont know if the water will stay below ambient in this setup. I feel confident that I can run my system without fans as long as I dont overclock to much. And if I add fans to the Heatlanes I should have plenty of headroom.
I have upgraded the CPU MB and memory, to a more overclocking friendly setup. But I have not had time to play yet.
DFI blood iron, E7200 and Cellshock PC9200.
Discovered a poorly placed capacitor stopped me from using the same waterblock on the NB chipset. And the memories are running a bit warm without any airflow. So aside from soundproofing, a ram watercooler and a new chipset WB is on the the todo list. After that I will look at adding fans.
The BI northbridge runs very cool. I'd recommend using a fanless HS like this: http://www.newegg.com/Product/Produc...82E16835109019. The only problem with that HS is that once installed it is very close to the video card -- almost touching it. In my rig I just placed a plastic cover behind the cooler and I have no fan on mine. Temps at 517 FSB never exceed 45C with that cooler. Although I do have a lot of airflow in my case.Quote:
Originally Posted by Jorlen
Nice but much more Watt gg ?
Hello again. Been running my system since early may. Performance is still the same as when first assembled.
I have not taken the time to finish the complete plan that included soundproofing the case and cooling the inside of it with the cooled water passing a radiator inside. And minimizing airflow through the case to keep the noise and cool air inside.
What I have learned so far. Silicon sucks. I have lots of silicon tubing to get all that water around. Silicon is to pourus, it lets air in and water out. My reservoir makes this issue more of a pain. I need to keep it completly filled with water otherwise turbulence in the reservoir will create airbubbles that make noise as the rush through the pump and the rest of the system.
The volume that can vary is that inside the tube coming off the top of the reservoir. So I need to refill that tube every week or so. To keep everything running with minimum noise.
Other then that I like the system very much. Silent and effective.
silicone... i think maybe its the way ur applying it, maybe the surface isnt grease, oil, and dirt free when u made the res, silicone shouldnt leak atol when applied properly
if silicone is pourus and lets air in and water out, why would it be used in every glass aquarium?
Update again. It wasnt the silicone it was the waterblock that leaked. Its leaking very slowly and drips down on my graphics card. It broke. I fixed it now. Life is still to busy to give me time to continue on this project.
Edit: Its not that purous. And when you seal with it like in an aquarium. The water is exposed to a small surface of silicone, and the seal is quite thick.
And even for me with all the hoses it wasnt the issue.
My system is still up and running. It works exactly as I expected it to work when I first planned it. Very quiete when the pump is suspended properly.
Changes made since last post:
New CPU block XPSPC Delta v3
New Reservoir. DDC top+reservoir integrated.
There are 3 blocks now cooling CPU, MB and Gfx. Added the MB last.
Since I have an Ati HD3600 I dont need a fancy waterblock, so I use the same Koolance blocks as on the MB and TECs (they were almost free when I bought them).
Aside from the pump, moving parts still in use:
PSU fan
2 x HDs
This project cost way to much if you just look at the value it produces at thist stage.
What you can do with this, as I have said earlier, is to build the system inside a soundproofed box and transport the heat out with the water and have the large passive sinks cool the whole thing. Without TECs the passive cooling outside the box would be huge in order to keep the system within operational limits. So there is potential.
To bad I dont have time for this anymore. Maybe next year who knows.
Very Cool. :up: Glad it's worked out for you. :up: Thanx for updating this....:up:
Awesome project. These are clever ideas but still nothing too complicated...why no one has built a sealed TEC chiller with feedback, I don't know...it would outperform everything else and be worth paying for.
What the TECs do is to allow smaller convection based cooling. A combination of the Zalman TNN-500A and TECs should be very effective. Using heatpipes instead of water to expand the cooling surface enough for passivly cooled TECs to come into play.