Author Topic: A look at Peltier cooling  (Read 80500 times)

0 Members and 1 Guest are viewing this topic.

dnix71

  • SuperHero Member
  • ******
  • Posts: 2278
Re: A look at Peltier cooling
« Reply #30 on: February 17, 2011, 09:57:20 AM »
One more suggestion. Since this is solid state, it shouldn't matter which way the door is mounted. I checked my peltier fridge and it works fine with the door face up like most upright freezers. That keeps the cool air in when you open the door. That fridge makers do not already do this shows how stupid they are. My fridge has the same space either way, it's basically a cube.

My Engel is made that way.

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #31 on: February 17, 2011, 10:27:49 AM »
Quote
I wonder if an exhaust fan would help in removing the heat?

Not sure if I'm following your logic here... ?

Basically, system function would be like so -

Initial power up and cool down will be handled by the Carnot cycle (compressor)

Heavy loading (hot item inserted, door open for too long, etc) also handled by Carnot.

Brief periods of the door opening for inserting/removing items will cause minimal heat shock, and the more heavily stocked it is, the less of a factor this will be. If the temp doesn't rise above a certain level, the compressor will not kick in.

Maintenance and small temperature adjustments (drops) will be handled purely by the Peltiers. The current delivered to the modules will be proportional to the difference between the internal temperature and what the thermostat is calling for. Probably going to use a buck converter for this process since it will be easy to integrate a simple thermostat into it (thermistor + pot).

There will need to be fans involved, but they are only for moving air across the heatsinks, both inside and out. The original design was calling for a single fan on each side to move the air across one large mass of fins for each side, but with the direction this is taking, I am going to have to get creative (more than usual LOL) to keep from having to use multiple fans, which would kill efficiency and make the whole project pointless. :(

There is another factor that needs tending to; Peltier modules create more heat than they pump, so the hot side heatsinks technically need to be larger than their cold side counterparts. The original heatsinks I was looking into accounted for this. Unfortunately, however, since it looks like all I am going to be able to get are 20 of the same exact heatsink, I'm going to have to deal with the surplus heat in another way. Not sure how I'm going to go about that just yet.

Considering a blower for each side, as they are somewhat more efficient in moving air than everyday fans are, but it's easy to get carried away and go over the 'energy budget' without really even trying.

Some designs I've seen use a single motor to drive both the hot and cold side blowers/fans, and while this makes sense from a consumption standpoint, it limits the versatility of the design.

To keep everything in check as close as possible, I may even look into high performance brushless motors (like those used in RC helicopters, etc) purely for their torque to power ratio. By going this route, I could throttle the blowers at almost any speed imaginable (within reason of course), and only spend as much energy as is absolutely required to provide the necessary airflow. Problem there becomes cost.

The other tweak I am probably going to integrate is a cold-side fan kill for when the door opens. This will help reduce the heat load on the Peltiers by not pushing the warm outside air across the cold side heatsink. The hot side fan will continue to run, so that for whatever length of time the door is open, the cold side sink can 'build up' as much cold as it can for a quicker recovery when the door is subsequently closed again.

It's definitely become more complicated than I originally envisioned, but I think it will be fun at the very least, and might actually turn out to be worthwhile if close attention to detail is given.

Peltiers clearly have their limits; in my second test, I ran the Coleman full throttle with a load of 54 soda cans (not sure how I managed to do that, its technically a '44 can' cooler...?) and it took 3 days to get down to ~6C (~43F). It hit 8C (~46F) in two days. I haven't plotted out the curve yet, but it looks like the butter zone in terms of energy efficiency is about 10C (50F). This of course is above a 'comfortable' temperature for refrigeration, by 7-9C (12-15F). Ideal is just above freezing, "between 35 and 38 degrees F (1.7 to 3.3 degrees C)" according to one hit I peeked at.

This poses a problem, because my house stays at roughly 72F, so getting the cold side down much lower than what I've seen is a completely uphill battle. My goal is to completely eliminate the Carnot cycle from the picture for holding the temperature, but if the Peltiers can't do it alone, I may be at another brick wall.

I've also considered cascading the modules to gain a higher temperature difference between hot and cold sides, but this too does not come without caveats. That pesky 'Peltiers make more heat than they move' problem becomes very apparent when attempting to stack modules. Not to mention the efficiency hit; to keep the cooling power (BTU transfer), I would have to double the number of Peltiers. At that point, I believe I'd be on the other side of the efficiency line, meaning it would likely be using more energy running the Peltiers than it would if I had just left it alone as a pure Carnot system.

We will see...

Steve

EDIT -

dnix - Yep, top loader. That was one of my first criteria. :)
« Last Edit: February 17, 2011, 10:34:34 AM by Madscientist267 »
The size of the project matters not.
How much magic smoke it contains does !

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4576
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #32 on: February 17, 2011, 12:02:33 PM »
Forgot you were going with a hybrid system.
My thought was towards the pure solid state units.
 By having a small exhaust fan located at the highest part of the back interior when the door was opened a trigger could turn that fan on to evacuate the hot air coming in.
This would allow the solid state to cool the rest of the air. BUT in doing a little further thinking it would be much harder to turn on an exhaust fan than to simply turn up the current to the solid state and let the air circulate around the heat/cooling sink.

going solid state allows the fridge to be at any position and still work whereas the freon pump needs to be level for the oils to be circulated.
I still follow this just in case I clear up a few round TUITS  ;D I have a small cubic footer that has a toasted pump just screaming to have solid states to be used.

Have fun!
Bruce S
 
A kind word often goes unsaid BUT never goes unheard

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: A look at Peltier cooling
« Reply #33 on: February 17, 2011, 03:24:14 PM »
I have recently bought a 134 w chip and played with it for a few days.  As for your efficiency, I thought it was closer to 10%, which is much less than a refrigeration pump.

For my supermileage car, we are buying a ~240 w chip to make a cooled seat.  We plan on running it at half power for better efficiency.  I have read that going past 75% of the power rating is pointless for cooling things down, unless you have a massive heat sink.  Our chip will be hooked up to a water block and a aluminum heat sink + a 100cfm box fan.

Once you turn the chip off, they heat up to the hot side temp real fast.  Trying the generate power is almost impossible.  My max was about 2v at 1 amp for a few seconds using some big pieces of aluminum on both sides.


Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #34 on: February 17, 2011, 04:08:31 PM »
Quote
less than a refrigeration pump

I have my reservations about this for sure. Chicken or Egg is in play here, so to speak.

My only light in the tunnel at this point is the idea that they can hold a reasonable facsimile of temperature difference at very low power levels.

That's probably the key point in all of this; Peltiers can be throttled, compressors cannot.

It's going to come down to average power consumption. So let's play with some math real quick...

These numbers aren't really directly comparable, because they are taken from two rather different sized implements, but the concepts are there...

The compressor on the Lowes box draws just shy of 2 amps (RLA) at 120V. That's 240W.

If it runs for 6 minutes out of every hour, that's 24wH of power used just countering the leakage in the insulation.

Take the 10 Peltiers, and run them continuously, however throttled way back.

If they can maintain the temperature in the chamber for less than 24W, it's plausible to do something like this.

If they can maintain it for less than 12, its completely viable.

The Coleman can do it with 9W. It's also all of maybe 2 cubic feet.

Problem is, a bigger chamber means more surface area, and all else being equal, means more thermal leakage, which means more power to keep the heat where it needs to be.

Not sure if it scales linearly (probably not, my luck it's more likely a square...?), but even if it is linear, 5 cubic feet would translate into 2.5x the power required to keep that same balance.

Works out to 22.5W, which is still less power than the compressor.

Also, with Peltier, not contending with motor start surges.

Steve
The size of the project matters not.
How much magic smoke it contains does !

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #35 on: February 17, 2011, 05:10:52 PM »

Problem is, a bigger chamber means more surface area, and all else being equal, means more thermal leakage, which means more power to keep the heat where it needs to be.

Not sure if it scales linearly (probably not, my luck it's more likely a square...?), but even if it is linear, 5 cubic feet would translate into 2.5x the power required to keep that same balance.

I've been following this post from the start because I've always wanted to play w/a peltier, the one thing that doesn't seem to have been mentioned, much, is the insulation... orrrrr... did I miss a big chunk?  :-\

Seth7

  • Newbie
  • *
  • Posts: 30
Re: A look at Peltier cooling
« Reply #36 on: February 17, 2011, 05:14:16 PM »
Me too .. i bought 50 @ 100W peltier's 7 years ago .... and i have a 80lb heatsink from a 900Mhz CDMA amp .....
building a PicMicro to PWM dosent seem too hard ....

Mike KC7NOA

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #37 on: February 17, 2011, 06:30:37 PM »
Zap -

The concept of insulation has gone all over the place on this one. Passive, active, back to passive, and now homebrew...  ;D

This is where I'm at for the moment on this...

From the post with the heatsink dimension specs:

Quote
Also, considering using some 'aerogel' based insulation to give it that final space-age touch (the 'SpaceLoft' blanket, near the bottom):

http://www.buyaerogel.com/

Anybody have any experience with this stuff? Claims to be a couple times more insulating than styrofoam... and only a minor irritant to various body parts.

It resembles something between felt and wool...

Not using that for the very first tests however. Decided I'm going to cheap out until I get an idea of whether to continue funding the project.

Going to build a surrogate door out of cardboard, fiberglass insulation, and duct tape to see if the principles are worth chasing.

At least that way if it bombs on me and all I end up with is a bunch of Peltiers and a freezer, I still have just that. A working freezer.

If the preliminary tests work as expected, I'll move on to pseudo-destruction of the real door and go for the good insulation.



Seth -

Quote
building a PicMicro to PWM dosent seem too hard

Just don't forget that raw PWM doesn't work as one would expect with a Peltier module. The power input needs to be as smooth as possible, ideally a pure DC 'waveform'.

HEAVY filtering is in order here to maximize efficiency.

PWM taking an all-to-nothing approach causes multiple problems with Peltiers;

1 - When in 'all' state, the module is running full throttle at it's least efficient operating point

2 - When in 'nothing' state, the module is not providing any cooling whatsoever, and the residual heat from the hot side will tend to head back toward the cool side, defeating the purpose of running it to start with

3 - PWM can be stressful on thermocoupled joints such as in a Peltier module, due to the repeated heating/cooling cycles happening several thousand times a second, eventually leading to module failure.

A buck converter is likely the best solution, again, heavily filtered.

Steve
« Last Edit: February 17, 2011, 06:33:58 PM by Madscientist267 »
The size of the project matters not.
How much magic smoke it contains does !

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #38 on: February 17, 2011, 06:44:30 PM »
By the way, the Coleman is now in 'coasting' mode, running at 5V, using 9 watts.

I have the data from the full power cool down, just need to plot it on paper.

I'll put it up after I have a few days of data collected from the coasting...

Steve
The size of the project matters not.
How much magic smoke it contains does !

DanG

  • Hero Member
  • *****
  • Posts: 1018
  • Country: us
  • Twin Cities Minnesota
Re: A look at Peltier cooling
« Reply #39 on: February 17, 2011, 08:11:23 PM »
Umn - in a dual-source unit the peltiers can never be shut off, I guess you're ready for 24/7/365 DC amps.

Environment heat freon eventually will have to deal with unless energized, even (baffled, shrouded, gated) it will be shining more heat than the original insulation design.

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #40 on: February 17, 2011, 08:32:21 PM »
Quote
Umn - in a dual-source unit the peltiers can never be shut off, I guess you're ready for 24/7/365 DC amps.

Correct; as long as this continuous drain is less than the averaged consumption that the compressor would use to perform the same work by cycling, it's Peltier for the win.


Quote
Environment heat freon eventually will have to deal with unless energized, even (baffled, shrouded, gated) it will be shining more heat than the original insulation design.

On this, you lost me just a bit. Elaborate? :)

Do you mean as in heat leaking in from the outside via the freon plumbing, acting like heat pipes?

I had considered that, not entirely clear on how to get around it. Chalked it up as one of the losses I was just going to have to accept as part of the design...  ???

Steve
The size of the project matters not.
How much magic smoke it contains does !

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #41 on: February 17, 2011, 09:30:59 PM »
I don't understand why the peltier can't be shut off ???

Do I need to increase my fish oil intake?

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #42 on: February 17, 2011, 10:20:44 PM »
It's a two part problem that only really has one solution: Trickling.

First, a Peltier that is turned off isn't too much less of a heat conductor than if it were just replaced with an aluminum block of similar dimensions. The heat from the outside readily leaks back through to the cold side from the hot side.

Second, because they are so inefficient, they generate more heat than they move, so the excess heat found on the hot side will quickly find its way back to the cold side, in addition to whatever is leaking from the ambient air on the hot side.

The result is you will actually put more heat in the fridge than you were taking out.

The only way around this is to keep them powered up, but throttled back so that they are not allowing the reverse heat creep.

In this design, however, I'm taking it a step further and adding just a touch more juice to keep what is leaking through the box insulation pumped to the outside.

Steve
The size of the project matters not.
How much magic smoke it contains does !

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #43 on: February 17, 2011, 10:39:09 PM »
OK... so why not make the module movable?
Imagine a CD/DVD tray holding the module on top of the lid.  The module slides out from over it's "hole" when not working and a "plug" of insulation is pulled into it's place.  I would think a stock CD/DVD tray could handle that and it could be automated to work with a thermostat.
Since the module is outside the chest it could then be powered off without harm?

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #44 on: February 17, 2011, 10:59:09 PM »
The logistics in that would be a nightmare... :(

The biggie that comes to mind right off the top of my head is the heatsink grease.

If using mechanical means to isolate the two worlds were employed, it would be more feasible to either insulate the cold side sink from the chamber, or remove the entire assembly and plug the hole.

Either way, separating and pairing the module on demand is pretty much out of the question. It would be like taking your CPU out of your computer whenever you're not using it.

Steve
The size of the project matters not.
How much magic smoke it contains does !

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #45 on: February 17, 2011, 11:42:27 PM »
If using mechanical means to isolate the two worlds were employed, it would be more feasible to either insulate the cold side sink from the chamber, or remove the entire assembly and plug the hole.
That's exactly what I was talking about.  Forget separating the assembly.
Have the whole thing sitting up far enough so you can slide a piece of insulation under the whole thing?  A CD/DVD tray could do that more easily.

I can draw a pic if ya want.


Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #46 on: February 18, 2011, 07:38:59 AM »
Hmm... Yeah, I think I can see the basic concept you're getting at. I just don't know if it would be worth all the trouble.

Another side effect of powering down the Peltiers when off-cycle comes from the idea that they aren't exactly quick about 'setting up'.

It takes a little bit to get the temp of the cold side heatsink down after initial power application. So, for whatever length of time it takes to get the sink down below the temp of the cooled chamber, it will be depositing heat into the chamber until the point of thermal equilibrium (and subsequently reversal) is reached.

This might take a good percentage of the energy that would have otherwise just have been used in maintaining the differential.

Thermodynamics is a ^#$@&, I know. LOL

I'm piecing the big picture together as I go, and so far, it's falling into place why Peltier is not more commonplace as a heat pump.

I'm pursuing it really only for a few reasons. One, and this is big, no moving parts to wear out. Two, no noise. Third, power conversion losses would be less (thinking in terms of comparing running a compressor from an inverter vs directly driving the Peltiers).

Problem is, it's already a very close call, if it even is possible to do the job with less energy than a compressor uses. But that's what I'm here to try and find out. :)

One of the first things I intend on doing when I receive the modules is taking one of them, and temporarily mounting the hot side on a heatsink, and leaving the other side exposed.

I'll get (at least) two useful pieces of info out of this:

First, I'm going to find out what the absolute bare minimum power requirements are to get various temperature differences, then plot the results. There will be a knee somewhere in the curve, and that is roughly the butter zone for the most efficient active cooling.

The other, is to find out where the lowest possible power input begins to have an effect. I have a theory that I haven't really been able to confirm just yet; Once they begin to pump heat, they seem to resist heat flow better from the opposite direction. In other words, they may be able to provide active insulation (at least for themselves) from the outside temperature with even less power than what it takes to just maintain the temperature. If I am correct, and the relationship is non-linear, small changes of a few degrees in temperature on the outside would not readily propagate to the inside through the module. With passive insulation, there is strictly a linear relationship between temperature difference and rate of transfer.

The effect of this would be that even though they may end up being less than useful for active cooling (or even maintenance), if they can provide active insulation for the entire box for much less power than I am even seeing for maintenance, this is still very much so doable. They would need to be spread out, so we've gone full circle on that again as well...

I apologize for the 'bipolar/circular' tendency of this entire thread. Lots of variables, lots of thinking on paper, and lots of theory. I fear the zig-zagging won't stop until I can physically try some of my thoughts in real application.  :-\

Steve
The size of the project matters not.
How much magic smoke it contains does !

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4576
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #47 on: February 18, 2011, 07:42:05 AM »
Steve.
The no big surge start-up is one thing I really do like about these.
Even if they turn out not to be so efficient, being able to start one up without a large invertor is very worth while and appealing.

Cheers;
Bruce S
   
A kind word often goes unsaid BUT never goes unheard

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #48 on: February 18, 2011, 08:21:50 AM »
I apologize for the 'bipolar/circular' tendency of this entire thread. Lots of variables, lots of thinking on paper, and lots of theory. I fear the zig-zagging won't stop until I can physically try some of my thoughts in real application.  :-\

Steve

None needed.
It's an interesting post... and... I'll :-X now.

dnix71

  • SuperHero Member
  • ******
  • Posts: 2278
Re: A look at Peltier cooling
« Reply #49 on: February 18, 2011, 02:46:59 PM »
Zap maybe not moving the peltier assembly out of the way, but rather just sliding an insulated cover over the assembly would be easier. Mylar with some foam for a shield. Maybe like a roll up door.


zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #50 on: February 18, 2011, 03:37:25 PM »
Yup... that's what I was saying up there a ways.

I think the peltier should be on the top.  Seems to me the cold would naturally fall into the cooler.



You could ignore the drive mechanism if need be and make the sliding part also be the on/off switch.

dnix71

  • SuperHero Member
  • ******
  • Posts: 2278
Re: A look at Peltier cooling
« Reply #51 on: February 18, 2011, 07:02:16 PM »
The drive mechanism could be simply a spring loaded solenoid. When energized, it pulls back the insulation. Cut the power and the spring closes it. The spring doesn't have to be large and it doesn't have to snap shut, just gently pull shut.

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #52 on: February 18, 2011, 11:41:01 PM »
What effect will this have from heat shock when the Peltier warms up... You then have to drop the temps again.

I don't think they're efficient enough to ever completely cut off. "Cheaper ta keeper". :)

Cycling a heatsink down to temp can take minutes. :(

Steve
 
The size of the project matters not.
How much magic smoke it contains does !

DamonHD

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3622
  • Country: gb
    • Earth Notes
Re: A look at Peltier cooling
« Reply #53 on: February 19, 2011, 04:25:57 AM »
I think the sliding flap is Maxwell's demon!  B^>

Rgds

Damon

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #54 on: February 19, 2011, 07:52:40 PM »
Got the sinks yesterday BTW!

Was easy, too... no digging. They had been stacking them all, some with fans still on them.

I'll be returning the fans (due to a glitch in communication and understanding of what was what on my part), but the sinks are mine! All 30 of 'em!



The heatsinks laid out before cleaning and de-fanning.





Detailed shot of one of them, with an old pack of smokes for size comparison.





All cleaned up (even ran them through the dishwasher). This is the rough configuration physically that they will be in.

Main difference is that I will probably use two rows on the cold side instead of three to make up for the inherent heat generation.

Still not entirely clear how I'm going to go about holding them all together and in place, although I'm thinking something like thermal epoxy on each sink to something like a 0.25 inch slab of aluminum plate for each side. The Peltiers would then be sandwiched between the two slabs with standard heatsink grease.

Or something like that.

At least it didn't cost me a thousand bucks!  ::)

Steve
The size of the project matters not.
How much magic smoke it contains does !

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: A look at Peltier cooling
« Reply #55 on: February 19, 2011, 09:35:12 PM »
Still not entirely clear how I'm going to go about holding them all together and in place, although I'm thinking something like thermal epoxy on each sink to something like a 0.25 inch slab of aluminum plate for each side. The Peltiers would then be sandwiched between the two slabs with standard heatsink grease.

Or something like that.


Folks have been using JB Weld for many years for such things... the metal in it tends to conduct heat very well.

I think the sliding flap is Maxwell's demon!  B^>

Rgds

Damon

The design I posted is my intellectual property and mine alone... if Maxwell doesn't like it, Maxwell can sue me ;)
« Last Edit: February 19, 2011, 09:39:58 PM by zap »

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: A look at Peltier cooling
« Reply #56 on: February 20, 2011, 01:24:20 PM »
Freezer got here today as well. Had to pull some strings - they were trying to deliver it next Saturday.

I don't think so.

Here it is with the heatsinks 'mocked up' on a board sitting on top of the unit...



Looks like with a little careful spacing and substrate management, I'll be able to cover the entire top. :)

Steve
The size of the project matters not.
How much magic smoke it contains does !

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: A look at Peltier cooling
« Reply #57 on: March 07, 2011, 05:26:54 PM »
I have been messing around with these last few days and there are so many variables that I just decided to make a spreadsheet that does it for me.  If you guys want to make a fridge, the biggest bang for your buck comes in the form of a bigger fan or bigger heat sink.  Getting the lowest C / W rating will let you lower your input power to the chip. 

I'm in the same situation right now, as I am trying to make a cooled seat, but I have some serious limits to how much power I can use.  At 25w of cooling, I can achieve a temp drop of 22.4 C at 300W input and 200 CFM on my 2"x5" heat sink.  But at 300w, my battery will die in 5 min.  Most likely I will end up with a 23 min run time w/ 2 batteries and a 9.5C temp drop.  I will have to update once I get the whole system working.

DanG

  • Hero Member
  • *****
  • Posts: 1018
  • Country: us
  • Twin Cities Minnesota
Re: A look at Peltier cooling
« Reply #58 on: March 08, 2011, 07:00:34 AM »
Umnn - you have moving air already, and the auxiliary fan.. maybe you can go a little more basic..

Evaporating a gram of water at 106F in one minute absorbs 569 calories or 2.25646442627 BTU or 39.65 watts heat energy.

The same amount ice to evaporated vapor carries away 720 calories per gram or 2.8552801176 BTU or 50.17 watts heat energy.

An average sensible heat for a human standing might be 4 BTU per minute, with brisk activity that is 6 BTU.

If you can keep wicking from acting as an insulator then having just enough water holding vs. surface area might give ten minutes of coolth. The old dip seat in water, shake twice and ride trick.

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: A look at Peltier cooling
« Reply #59 on: March 08, 2011, 07:24:16 AM »
We could spray water on the heat sink, in which the water should evaporate fairly quickly.  Our idea will work, but we probably bought TEC's that were too large (Qc = 248 and 134w). 

Here is the set up right now:







Things that majorly change the performance include where we put the fan and the power input.  Our fan is only 100 CFM, so we plan on buying a 250 CFM soon. According to my calculations, that will drop the temperature at least 10C.  The water pump pushes 2 GPM and the water block is 3" x 3" for scale.  The water pipes are 3/8" I.D, 1/2" O.D.

« Last Edit: March 08, 2011, 07:40:22 AM by taylorp035 »