Go Down

Topic: Thermoelectric Cooler & battery selection (Read 2131 times) previous topic - next topic

Drew Davis

Aug 14, 2013, 04:54 am Last Edit: Aug 14, 2013, 09:23 pm by Drew Davis Reason: 1
Hello, I'm looking into thermoelectric cooling and I found one from spark fun. Sadly it does not have a data sheet. When spark fun says "lmax - 7A" is that the average amount of power it eats up?


Thanks!




https://www.sparkfun.com/products/10080

LarryD

"lmax - 7A"
Is the current your driver has to supply/source to get this to work.

Use a Power MOS FET to power this device example IRL540.
http://www.vishay.com/docs/91300/91300.pdf
The way you have it in your schematic isn't the same as how you have it wired up!

Chagrin

#2
Aug 14, 2013, 07:36 am Last Edit: Aug 14, 2013, 03:20 pm by Chagrin Reason: 1
Imax is the amount of current that the peltier needs to transfer heat at its highest level; you can supply it more current but you'll just overheat the peltier and you can supply it with less current and generate less heat transfer. Umax (or Vmax) is the same concept but with respect to voltage and Qcmax with respect to wattage.

You can treat the values as absolute limits but technically that's not the case. It's an odd way of specifying a part but you have to remember this is a collection of (in this case) 127 imperfect semiconductors in parallel and the numbers don't work as you might expect.

MarkT


"lmax - 7A"
Is the current your driver has to supply/source to get this to work.

Use a Power MOS FET to power this device example IRL540.
http://www.vishay.com/docs/91300/91300.pdf


Or choose a device with a smaller Rds(on) to run cooler.  A modern logic level device rated at 30V would
should have a far lower Rds(on) - 10mOhm would be typical.
[ I won't respond to messages, use the forum please ]

Drew Davis

I don't think I understand. When I use a dc motor it just draws as much current as it needs. Would a Thermoelectric be more like a led that has its current limited by a resister?

drjiohnsmith

ah multiple numbers:

the module is rated at moving 62.2 watts of heat from one side to the other.

To do this you need to apply 107 watts of power ( 15.4 v * 7 amps )

and in this case, your hot side needs to dissipate 169.2 watts of heat, else you will be heating the cool side..

if you have no heat sink, then the thing will heat up with the 107 watts of power till both sides of the device are 'hot'.

lower the voltage / current u put in, and the amount of heat your transferring will decrease.
  ( the cooler looks like a bad resistor , so u lower the volts and the current goes down )




Drew Davis

#6
Aug 14, 2013, 03:10 pm Last Edit: Aug 14, 2013, 09:22 pm by Drew Davis Reason: 1
So how many amps would it draw at 12v? From, the looks of this is does not appear to be very effecting at cooling. I was planning to use this to cool down a battery, however, It seems that the amount of current it pulls is not practical. Are there better cooling devices?

drjiohnsmith

at 12 volts it will draw 5 to 6 amps ( V = I * R ).

efficiency of a peltia is by definition very negative, 
   you have to put in 2 to three times the heat that you are moving

but they are one of the few ways to get a lower than ambient temperature.

would a heat sink on the battery be sufficient ?

Drew Davis

I have never thought about doing that before. I will be using two batteries and one motor. I will combine the batteries so that I get 22.2V 20Ah. I will be be pulling about 90a for just a few minutes.  Would a heat sink help cool the battery down?




http://www.superdroidrobots.com/shop/item.aspx/high-power-polymer-li-ion-module-22-2v-10ah/969/

http://www.robotcombat.com/products/0-A28-150.html

drjiohnsmith

taking your number,
   your discharging Li Ion polymer at the 5C times rate,
     
I'd hope the internal protection on the packs would kick in and stop that sort of discharge.
   even if the protection did not kick in, I'd imagine your going to seriously shorten the life time of the packs, to maybe 20 % of their life time.

the 'problem' with the cells is they are not that good at conducting heat themselves.
     as they self heat, the temperature in the centre is a LOT higher than the outside.
        in the few minutes your discharging, very little heat will actually get out.

a heat sink will lower the time to cool,
   or you could try putting a fan on them.

Another solution I have seen for such applications is putting the cells in a silicon liquid 'tank', which is then effectively a large thermal mass.

For reference,

   if a LiIon Poly cell is short circuited, if the protection doe snot go off,
         the cell goes into thermal run away,  and if your lucky just catch fire,

  If your unlucky, they cell turns to liquid at a thousand degrees and explodes,
         the resultant liquid metal spray sticks like mud to anything,
             spontaneously catches fire on contact with the air,
                 and can not be put out with CO2, water , foam or powder extinguishers.

Even boeing have had fun in their dream liner aircraft trying to do this.
   

Drew Davis

According to the battery they can supply up to 50amps discharge If I put them in parallel can't they give off up to a 100amp discharge? Also, They were going to be in a fire retardant bag… just in case. 

drjiohnsmith

how confident are you with Li Ion cells ?

if you parallel a number of cells externally, then you have to be VERY careful,

imagine one cell is charged 0.1 volt above the other,
   when they are paralleled, the higher voltage cell is going to short ALL its current to charge up the other cell.

but they are charged together you say. Ahh no.

if you charge two cells identically, then the terminal voltage of both cells is NOT the same.
   its a chemical process, that is very none linear.

A fire proof bag. Not seen one that allows the heat of the cell to get out, and can withstand exploding and metal.

Big suggestion, u always wear as a minimum a face shield when experimenting with these things.

You normally see cells in series to give a high voltage, as in the tesla car.
   so look for a cell at half the voltage you want at the same Ah, and put them in series.

these are the sort of thing we've used on such applications,

http://www.e-transportation.eu/catalog/product.php?id_product=15

Drew Davis

That is a very nice pack however at $11,797 that is way past my budget for the entire project. I have never used a Li Ion cell. Typically I just use NiMH batteries. So please excuse my limited knowledge of their dangers.  I was planning to put them in parallel as soon as I revived them.

What if I use these products...


http://www.tenergy.com/31417-02
http://www.tenergy.com/01221


Also the battery will be in a thick metal enclosure (my robot) so i'm not afraid of getting hurt. My biggest fear is having to re-buy batteries!  The reason I want to go with Li Ion cells is that they are light weight, and powerful.




drjiohnsmith

#13
Aug 14, 2013, 07:47 pm Last Edit: Aug 14, 2013, 07:55 pm by drjiohnsmith Reason: 1
dont know either of those two products, so cant comment.

if you miss treat NiMH, they might just catch fire, and smoulder.
   but thats unlikely.

Li Ion, if you hit a naked cell with a hammer they can explode.
 not the worst technology, Na cells are far more fun, but .

Dont put Li Ion packs in parallel, and you "SHOULD" be safe,
 
re metal case, dont you take that off to tinker with the robot ?
  wear a face mask.

BTW: those cells are 89 euro, which I think is about 90 dollars US,
    10 of them will give you 40 Ah at 34 volts.
       


Drew Davis

Sorry for the miscalculation. Google missed the "," and translated it wrong. I would use them but they are just to heavy. I need to keep this around 6 pounds. 

I just called the company and they said that their batteries are 100% safe. They also said that it would not be an issue to put two of those packs in parallel as long as they were charged separately.

Go Up