Author Topic: A look at Peltier cooling  (Read 81166 times)

0 Members and 1 Guest are viewing this topic.

Nautilus1

  • Newbie
  • *
  • Posts: 17
  • Country: ro
Re: A look at Peltier cooling
« Reply #180 on: May 15, 2015, 03:48:13 PM »
Acquired rather cheaply in the meantime an adjustable power brick, able to give 4.5 amps while the voltage is set to 12, 15, 16, 19, 20 or 24 volts.

Tested each component with a K-probe. The interior fan created a small hot spot at the motor, it heated itself to >20C and it could hit 32C if the Peltier and fans were fed the nominal Vmax of 15.4 volts. Dropped the inner fan altogether.

Outer fan was in a poor shape and rather noisy. Removed it and fit a Sunon 80mm fan on an elastic mount to avoid vibration. New fan is visibly more powerful, air flow blown through the heatsink can be felt 2 ft away. Now the heatsink is much cooler.

On 12 volt setting (about 36 watts counting the fan motor as well), the fan is less noisy and it can be left overnight to cool. The Peltier can run in 15 volt (15.4 actually) or 16 volt (real 16,2 volts) to cool relatively quickly the content, in a few hours. This raises the power consumption in the 50-watt range. It may run as well in higher voltages, but the amps raise just as quickly, to the point the power brick heats itself and can no longer supply the current.

Now the load can be cooled in a few hours for the cost of energy and noise, and left in a maintenance mode around 12 volts for most of the time, fan turning slower and less energy spent.

micropv

  • Newbie
  • *
  • Posts: 27
  • Country: us
Re: A look at Peltier cooling
« Reply #181 on: June 14, 2015, 07:58:30 PM »
Been playing with this old (circuit board bad) 12V cooler that I purchased at a thrift store for a couple $.  TEC and fan are still good.  Managed to hit 47.7F inside the "cooler" feeding it 5V/1A directly to the TEC and using a 6V battery pack to run the fan.  It took about 2 1/2 hours to reach that temp (ambient in the basement is 69.4F to 69.8F on the floor).




Today, tried to cool 3 room temp cokes, from 8 AM until 5:30 PM or so, lowest I could get them was 52.4F.  Might try again placing cold items in it and see how it does. 

Just something to do more than anything really..


Nautilus1

  • Newbie
  • *
  • Posts: 17
  • Country: ro
Re: A look at Peltier cooling
« Reply #182 on: June 15, 2015, 09:30:34 AM »
This is quite a problem: cooling from room temps to 7-8C (44-46F) takes a lot of time in a Peltier cooler, maybe even 2-3 days, which is quite wasteful.

Taking the beer or juice bottle already cooled to a reasonable temp from the store and using the Peltier to just keep away heat from creeping in (active insulation) works much better.

Feeding the Peltier higher voltages works up to a point, where the heat generated is above the heatsink's ability to dissipate it and the cooling process is much hampered.

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4581
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #183 on: June 15, 2015, 09:40:57 AM »
When trying to do active cooling , it's better to have the TEC to it's near maximum rating to get the cooling cycle working, then use intelligent programing to step the power down to keep contents at or near set temps.
 Those little fridges are cute, but I think it's TEC is a 12V unit, but it would be better to check it's rating with the marking on the chip (IF you can get to it).

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

micropv

  • Newbie
  • *
  • Posts: 27
  • Country: us
Re: A look at Peltier cooling
« Reply #184 on: June 21, 2015, 02:06:41 PM »
Hey Bruce,

Yes, it is a 12v 5a TEC.  Interesting to experiment with, I have it hooked directly to "50W" of solar at the moment, I am running the fan from a 12v battery.

As you view the photos, keep in mind nothing is "final", just playing more than anything. 












Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4581
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #185 on: June 22, 2015, 08:51:52 AM »
While running the fridge, put a temp probe over the exhaust fan to see what it is has, I'll bet there's too much pretty coverage around that back fan to let the heatsink get rid of excess heat.
I took the back off mine, found a 120mm fan to replace the smaller one ( I also tried to get a lower current rated one) and see if it keeps the contents cooler.
Of the contents I see inside, the metal ones will cool much better/quicker and I'm thinking the gel ice units are taking up too much space.
Also, metal lined on the inside should help wick away to heat from the other contents.

NICE use of foam board! I like the idea of using a portion as a door, better than opening the entire door.

A kind word often goes unsaid BUT never goes unheard

micropv

  • Newbie
  • *
  • Posts: 27
  • Country: us
Re: A look at Peltier cooling
« Reply #186 on: June 25, 2015, 10:16:55 AM »
I agree, a metal lined interior would really help!  I wonder if storing some items in those old aluminum "Coffee, Sugar, Salt, etc." containers would help some?  They were all the rage in the 70's and 80's.  I see them at thrift stores sometimes.  The gel packs do take up some room, kind of a in progress test, idea from the no longer made Sundanzer bfr105.

So far it is holding well, I use a small inverter upstairs, to feed the juice downstairs...  full throttle during the day (when sunny), and trickle at night.





It was aprox. 1.3CF before the addition of 1" foam.  4.8v to run the fan when on trickle.  An AA pack would make it 24hrs, just switched over to a D pack this morning.



I seriously need to start cleaning up the wires and connections!

MattM

  • Sr. Member
  • ****
  • Posts: 440
Re: A look at Peltier cooling
« Reply #187 on: June 25, 2015, 12:13:46 PM »
So line the interior insulation with tin foil.  The use of foil on insulation is purely related to stopping infrared radiation and does really almost nothing as far as spreading out the cool.  Entropy does it already.  Blocking infrared light is very important as all heat-excited objects tend to radiate in the infrared band.  Not all, but most.

dewzenol

  • Newbie
  • *
  • Posts: 2
  • Country: us
Re: A look at Peltier cooling
« Reply #188 on: January 21, 2016, 12:30:02 AM »
I just built a peltier-based cooler using two peltiers, two heat sink / fan combos, a 1'x1'x2' Styrofoam cooler, and some 2"x1/8" aluminum plate.
Initially I made a 'T' using two (inverted) 'L'-shaped sections of plate (about 10" long each) and placed the cold side of one peltier on each of the two 'L's bases.  I then stuck another length of flat on top of the 'T', bridging the two hot sides, and placed my monster CPU cooler/fan (from an old Dell XPS) in the center. 
The vertical of the 'T' (or the legs of the 'L's depending on your perspective) were shoved through a snug slot into the Styrofoam cooler.  A smaller CPU heat sink / fan was placed on the aluminum bar inside the foam cooler.  Thermal grease was used on every junction.

The net result of this was a whole lot of current being used and a whole lot of heat being generated and no noticeable cooling effect.

Just like any bad hobbyist would do, I changed not one, but several variables in an attempt to correct this.  I ditched one of the 'L's and the flat bridge on top, and stacked the two peltiers directly between the big heat sink and the base of the remaining inverted 'L'.  In the process, I noticed that my thermal grease had squeezed out and onto the sides of the peltiers.  I think this was a big factor in my failure.  I believe the goo was conducting the heat right back to the cold side.  I thoroughly cleaned the excess paste and assembled as I described. 
Another variable I changed was to switch from 12v in parallel to 12v in series, giving each peltier only 6v.

The net result was that with the device running, the temperature inside the cooler dropped about 2 degrees centigrade per minute.  I'm curing meat so I didn't try to get to freezing -- I'm sure the rate would have dropped the closer I got to that.  I only went from 70F ambient to 55F.

I don't know if the extra aluminum plate on the hot side (and its surface area fractions of an inch away from the cold side) was a big factor in my initial failure.  I think the excess thermal paste was a/the huge factor.

I am also not sure whether switching to 12v in parallel would greatly speed the cooling.  I suspect it would pump heat faster than my sink/fan could dissipate it and would therefore be much less than twice as fast.  So I'm keeping it as is.

Hopefully my story will have some value to you.

Good luck! 

dewzenol

  • Newbie
  • *
  • Posts: 2
  • Country: us
Re: A look at Peltier cooling
« Reply #189 on: January 21, 2016, 12:46:18 AM »
Just wanted to add before and after photos..



Mary B

  • Hero Member
  • *****
  • Posts: 825
Re: A look at Peltier cooling
« Reply #190 on: January 21, 2016, 03:25:09 AM »
What model Dell is that from? I have a 350 watt peltier I am building into a water chiller for the laser cutter and need a monster heatsink to remove heat

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4581
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #191 on: January 21, 2016, 01:51:30 PM »
Mary B;
I have some of those very same heat sinks, mine came from a Optiplex 620 and Optiplex 760.
Hope that helps
Bruce S
A kind word often goes unsaid BUT never goes unheard

Mary B

  • Hero Member
  • *****
  • Posts: 825
Re: A look at Peltier cooling
« Reply #192 on: January 21, 2016, 08:25:46 PM »
Thanks! Found one on ebay cheap. Need to measure the peltier and see if it is a big enough contact patch.

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4581
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #193 on: January 21, 2016, 08:35:38 PM »
Mary B;

Let me know what they go for on ebay, I could send you one FREE!! just pay shipping.
OR let me know what size you're looking for I'll dig into my grab bag of stuff .
At the FD we MUST trash older computers , being in their IT dept  8) I get first pick. I think I have about 6 of these!

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

Mary B

  • Hero Member
  • *****
  • Posts: 825
Re: A look at Peltier cooling
« Reply #194 on: January 22, 2016, 04:45:41 AM »
Think it was $15 shipped... but my Peltier is huge... 62mm square! The one I was looking at had a ~ 57mm square contact patch... so back to the drawing board. I may go with a water cooled unit but they are not cheap!

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4581
  • Country: us
  • USA
Re: A look at Peltier cooling
« Reply #195 on: January 22, 2016, 08:50:43 AM »
Mary B;
That's not huge! IF google is correct, 62mm is only 2.5 inches. Let me dig into my stash and see what I've got.

I know for sure my wife would love to have some of these out of the way  ::).
They don't weigh much at all, I'm certain the fans that keep them cool are 3x3inches.

Bruce S



 
A kind word often goes unsaid BUT never goes unheard

Mary B

  • Hero Member
  • *****
  • Posts: 825
Re: A look at Peltier cooling
« Reply #196 on: January 22, 2016, 08:56:28 PM »
the aluminum fin area is huge, the copper heat spreader is smaller!