Peltier modules????

[Glenn Ashmore]

Can someone explain the idiosincracies I have observed in a peltier module?

In scrounging parts for the watermaker I cam across an interesting
surplus assembly. I was really after the 1 gpm flowmeter and stainless
needle valve but it came with a really strange device. One of those
things you just sense is to good to throw away. This thing is a pair of
hollow black anodized aluminum blocks about 4.25"x1.75"x1" with pipe
fittings on each end. They are bolted together with what appears to be
a couple of 1.5" square Peltier modules separating them.

Being a curious sort, I did the natural thing and took it apart to play
with the Peltier modules. First I put them on the VOM to check
resistance. They seems to float around between 19 and 22 ohms. Then I
wired it to a DC power supply and slowly cranked up the voltage. While
one side slowly got warm and the other cooler I could not get it higher
than 3 volts and the current never got above .25 amps. Changing
polarity did not make any difference.

This does not seem to fit any of my perception of the rules of
electricity. What is happening?

--
Glenn Ashmore

I'm building a 45' cutter in strip/composite. Watch my progress (or lack
there of) at: http://www.rutuonline.com
Shameless Commercial Division: http://www.spade-anchor-us.com

 

[Peter Gottlieb]

Peltier modules are interesting devices. I think of them as like permanent
magnet DC motors in a way. If you apply voltage to a motor, it will
accelerate until the back emf minus the input voltage leaves just enough
current to overcome friction. In a Peltier device, the temperatures will
make a difference from side to side until there is an equilibrium, with just
enough current to overcome thermal losses. And, like the motor when you
turn it and it acts like a generator, the Peltier device, when fed a
temperature difference, will produce a voltage and can be used to convert
heat to power. Reversing polarity simply swaps sides.

A 1.5 inch square Peltier may very well run on up to 5 or 6 amps and around
12 volts for full operation. Search the web for data sheets, many of these
devices are very similar so if you find specs for something of that size you
will be right in the ballpark.

 

[Spehro Pefhany]

Can someone explain the idiosincracies I have observed in a peltier module?

In scrounging parts for the watermaker I cam across an interesting
surplus assembly. I was really after the 1 gpm flowmeter and stainless
needle valve but it came with a really strange device. One of those
things you just sense is to good to throw away. This thing is a pair of
hollow black anodized aluminum blocks about 4.25"x1.75"x1" with pipe
fittings on each end. They are bolted together with what appears to be
a couple of 1.5" square Peltier modules separating them.

Being a curious sort, I did the natural thing and took it apart to play
with the Peltier modules. First I put them on the VOM to check
resistance. They seems to float around between 19 and 22 ohms.

Your measurement below indicates an effective resistance more like 12
ohms at realistic voltages. There are usually more than 100
semiconductor junctions in series in these devices. They probably
don't behave exactly like a resistor at low voltages/currents.

Then I
wired it to a DC power supply and slowly cranked up the voltage. While
one side slowly got warm and the other cooler I could not get it higher
than 3 volts and the current never got above .25 amps.

Chances are your DC power supply is limiting at 250mA.

Changing polarity did not make any difference.

It should change which side gets hot and which gets cold.

;-)

Best regards,
Spehro Pefhany

 

[Ian Stirling]

Can someone explain the idiosincracies I have observed in a peltier module?

In scrounging parts for the watermaker I cam across an interesting
surplus assembly. I was really after the 1 gpm flowmeter and stainless
needle valve but it came with a really strange device. One of those
things you just sense is to good to throw away. This thing is a pair of
hollow black anodized aluminum blocks about 4.25"x1.75"x1" with pipe
fittings on each end. They are bolted together with what appears to be
a couple of 1.5" square Peltier modules separating them.

Being a curious sort, I did the natural thing and took it apart to play
with the Peltier modules. First I put them on the VOM to check
resistance. They seems to float around between 19 and 22 ohms. Then I
wired it to a DC power supply and slowly cranked up the voltage. While
one side slowly got warm and the other cooler I could not get it higher
than 3 volts and the current never got above .25 amps. Changing
polarity did not make any difference.

This does not seem to fit any of my perception of the rules of
electricity. What is happening?

Whatt happened if you tried to increase the voltage over 3?
Was the supply current limited?

 

[Glenn Ashmore]

Problem solved. Looks like the old Lambda PS has given up. Hooked the
modules up to a heavy regulated 13.8 volt power supply. It pulled 6
amps and froze my fingers.

These things are fascinating. I put the assembly back together, filled
one side with ice water and the other with hot coffee. Darned thing
cranked out almost 9V!

Whatt happened if you tried to increase the voltage over 3?
Was the supply current limited?

--
Glenn Ashmore

I'm building a 45' cutter in strip/composite. Watch my progress (or lack
there of) at: http://www.rutuonline.com
Shameless Commercial Division: http://www.spade-anchor-us.com

 

[Ian Stirling]

Problem solved. Looks like the old Lambda PS has given up. Hooked the
modules up to a heavy regulated 13.8 volt power supply. It pulled 6
amps and froze my fingers.

These things are fascinating. I put the assembly back together, filled
one side with ice water and the other with hot coffee. Darned thing
cranked out almost 9V!

Would be really handy if the efficiancy wan't so low.
In theory it can hit almost 25%, with 0C and 100C.
In practice it's a whole lot less.
When I actually did some tests, I got .1% efficiancy.

 

[Vladimir Vassilevsky]

Typical Peltier can generate maximal temp. difference about 50C. At this
difference, the heat conductivity losses are equal to the amount of cold
generated, so the efficiency is 0%. That is quite poor in comparison
with conventional freon refrigerator.

Vladimir Vassilevsky, Ph.D.

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

 

[richard mullens]

Problem solved. Looks like the old Lambda PS has given up. Hooked the
modules up to a heavy regulated 13.8 volt power supply. It pulled 6
amps and froze my fingers.

make sure that you incorporate some means to take away the heat from the hot
side. If you don't, you will destroy the device when the temperature rises
too high.