Maker Pro
Maker Pro

Supercooling of Peltier cooler using a current pulse

G

Guy Macon

Jan 1, 1970
0
Content-Transfer-Encoding: 8Bit


Supercooling of Peltier cooler using a current pulse

J. Appl. Phys.
August 1, 2002
Volume 92, Issue 3, pp. 1564-1569
(C) 2002 American Institute of Physics.
http://link.aip.org/link/?JAPIAU/92/1564/1

G. Jeffrey Snyder,
Jean-Pierre Fleurial,
Thierry Caillat
Jet Propulsion Laboratory,
California Institute of Technology,
Pasadena, California 91109

Ronggui Yang,
Gang Chen
Department of Mechanical Engineering,
Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139

The operation of a Peltier cooler can be temporarily enhanced by utilizing
the transient response of a current pulse. The performance of such a device,
using (Bi,Sb)2Te3-based thermoelectric elements, was examined from –70 to 55
degrees C. We establish both theoretically and experimentally the essential
parameters that describe the pulse cooling effect, such as the minimum
temperature achieved, maximum temperature overshoot, time to reach minimum
temperature, time while cooled, and time between pulses. Using simple
theoretical and semiempirical relationships the dependence of these
parameters on the current pulse amplitude, temperature, thermoelectric
element length, thermoelectric figure of merit and thermal diffusivity
is established. At large pulse amplitudes the amount of pulse supercooling
is proportional to the maximum steady-state difference in temperature.
This proportionality factor is about half that expected theoretically.
This suggests that the thermoelectric figure of merit is the key materials
parameter for pulse cooling. For this cooler, the practical optimum pulse
amplitude was found to be about three times the optimum steady-state
current. A pulse cooler was integrated into a small commercial
thermoelectric three-stage cooler and it provided several degrees of
additional cooling for a period long enough to operate a laser sensor.
The improvement due to pulse cooling is about the equivalent of two
additional stages in a multistage thermoelectric cooler.
 
Content-Transfer-Encoding: 8Bit

Supercooling of Peltier cooler using a current pulse

J. Appl. Phys.
August 1, 2002
Volume 92, Issue 3, pp. 1564-1569
(C) 2002 American Institute of Physics.http://link.aip.org/link/?JAPIAU/92/1564/1

G. Jeffrey Snyder,
Jean-Pierre Fleurial,
Thierry Caillat
Jet Propulsion Laboratory,
California Institute of Technology,
Pasadena, California 91109

Ronggui Yang,
Gang Chen
Department of Mechanical Engineering,
Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139

The operation of a Peltier cooler can be temporarily enhanced by utilizing
the transient response of a current pulse. The performance of such a device,
using (Bi,Sb)2Te3-based thermoelectric elements, was examined from -70 to 55
degrees C. We establish both theoretically and experimentally the essential
parameters that describe the pulse cooling effect, such as the minimum
temperature achieved, maximum temperature overshoot, time to reach minimum
temperature, time while cooled, and time between pulses. Using simple
theoretical and semiempirical relationships the dependence of these
parameters on the current pulse amplitude, temperature, thermoelectric
element length, thermoelectric figure of merit and thermal diffusivity
is established. At large pulse amplitudes the amount of pulse supercooling
is proportional to the maximum steady-state difference in temperature.
This proportionality factor is about half that expected theoretically.
This suggests that the thermoelectric figure of merit is the key materials
parameter for pulse cooling. For this cooler, the practical optimum pulse
amplitude was found to be about three times the optimum steady-state
current. A pulse cooler was integrated into a small commercial
thermoelectric three-stage cooler and it provided several degrees of
additional cooling for a period long enough to operate a laser sensor.
The improvement due to pulse cooling is about the equivalent of two
additional stages in a multistage thermoelectric cooler.

Looks odd. Under normal conditions, heat transfer through a Peltier
cooler is linearly proportional to current, but the current produces
resistive heating in the cooler proportional to the square of the
current.

So there is an upper limit to the amount of current that you can
usefully drive through a Peltier cooler - a sufficiently high current
will produce no cooling effect at all while dumping lots of heat into
the exhaust heat sink, and half that current (the figure normally
listed on the data sheet) will produce the maximum net cooling heat
transfer - any more current will produce more heat in the junction
than it transfers through the junction.

Generally, pulsing the current through a Peltier junction is a bad
idea - obviously, if you only want cooling for a brief period anyway,
you aren't going to run the cooler continuously. One would need to see
the original paper to see exactly what the authors think they have
discovered, but I'm rather sceptical.
 
W

Winfield Hill

Jan 1, 1970
0
Looks odd. Under normal conditions, heat transfer through a Peltier
cooler is linearly proportional to current, but the current produces
resistive heating in the cooler proportional to the square of the
current.

So there is an upper limit to the amount of current that you can
usefully drive through a Peltier cooler - a sufficiently high current
will produce no cooling effect at all while dumping lots of heat into
the exhaust heat sink, and half that current (the figure normally
listed on the data sheet) will produce the maximum net cooling heat
transfer - any more current will produce more heat in the junction
than it transfers through the junction.

Generally, pulsing the current through a Peltier junction is a bad
idea - obviously, if you only want cooling for a brief period anyway,
you aren't going to run the cooler continuously. One would need to see
the original paper to see exactly what the authors think they have
discovered, but I'm rather sceptical.

I imagine they're relying on the thermal mass.
 
P

Phil Hobbs

Jan 1, 1970
0
Winfield said:
I imagine they're relying on the thermal mass.

Peltiers absorb heat right at the junctions, which are soldered to the
cold plate. Their I**2R heating is rather uniformly generated
throughout the length of the bismuth telluride bars. One of the most
important properties of a Peltier material is poor heat conduction, so I
gather that by goosing the current, you can get several seconds' worth
of lower temperature on the cold plate, before the extra heat generated
further down the bars can warm it up.

That might be quite helpful in marginal situations, e.g. ones involving
a phase change or at the ends of the temperature-tuning range of a diode
laser.

Cheers,

Phil Hobbs
 
heat in the junction



Peltiers absorb heat right at the junctions, which are soldered to the
cold plate. Their I**2R heating is rather uniformly generated
throughout the length of the bismuth telluride bars. One of the most
important properties of a Peltier material is poor heat conduction, so I
gather that by goosing the current, you can get several seconds' worth
of lower temperature on the cold plate, before the extra heat generated
further down the bars can warm it up.

That might be quite helpful in marginal situations, e.g. ones involving
a phase change or at the ends of the temperature-tuning range of a diode
laser.

I've now seen the original paper, and that is exactly what the authors
are talking about.

They are interested in cooling off a mid-IR laser gas sensor for a few
milliseconds, and it seems that pulsing the cooling current gives them
a small but useful advantage. I've only scanned through the paper,
which is heavy on numerical analysis, but it does seem to make sense.
 
G

Guy Macon

Jan 1, 1970
0
Looks odd. Under normal conditions, heat transfer through a Peltier
cooler is linearly proportional to current, but the current produces
resistive heating in the cooler proportional to the square of the
current.

So there is an upper limit to the amount of current that you can
usefully drive through a Peltier cooler - a sufficiently high current
will produce no cooling effect at all while dumping lots of heat into
the exhaust heat sink, and half that current (the figure normally
listed on the data sheet) will produce the maximum net cooling heat
transfer - any more current will produce more heat in the junction
than it transfers through the junction.

Generally, pulsing the current through a Peltier junction is a bad
idea - obviously, if you only want cooling for a brief period anyway,
you aren't going to run the cooler continuously. One would need to see
the original paper to see exactly what the authors think they have
discovered, but I'm rather sceptical.

It would be easy enough to test -- and I will certainly give it a go
the next time I am working with Peltiers if someone here doesn't post
something about it first. One would think that this is the sort of
thing that Peltier cooler manufacturers would put in their datasheets.

The bit about "The improvement due to pulse cooling is about the
equivalent of two additional stages in a multistage thermoelectric
cooler." is interesting. are they implying that I can replace a
six-stage stack with a two-stage stack? Normally I have to make
the plates bigger and bigger on the hot side[1] so as to pump out the
heat from resistive losses. If I really can replace six stages with
two, would the coldest plate be the size of the hottest plate in the
stack of six, or the size of the plate right bellow the coldest plate?
(Not that I have ever mad a stack six high... Last time I worked on
such a system we stacked three on top of a standard freon chiller)

Note [1:] Unless I made another typo, a typical 10W Peltier pumps
10W of heat into the cool side, adds another 20W from I2R losses,
and thus needs to shed 30W from the hot side. The next-warmer
stage needs to pump that 30W, but adds another 60W of heat doing so.
The next-warmer stage needs to pump 90W, but adds another 180W
of heat doing so. Then it's 270W in 810W out, 810W in 2430W out,
2430W in 7290W out, and so forth. That's why staged Peltier
coolers look like pyramids.
 
R

Rene Tschaggelar

Jan 1, 1970
0
Guy said:
Content-Transfer-Encoding: 8Bit


Supercooling of Peltier cooler using a current pulse

J. Appl. Phys.
August 1, 2002
Volume 92, Issue 3, pp. 1564-1569
(C) 2002 American Institute of Physics.
http://link.aip.org/link/?JAPIAU/92/1564/1

G. Jeffrey Snyder,
Jean-Pierre Fleurial,
Thierry Caillat
Jet Propulsion Laboratory,
California Institute of Technology,
Pasadena, California 91109

Ronggui Yang,
Gang Chen
Department of Mechanical Engineering,
Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139

The operation of a Peltier cooler can be temporarily enhanced by utilizing
the transient response of a current pulse. The performance of such a device,
using (Bi,Sb)2Te3-based thermoelectric elements, was examined from –70 to 55
degrees C.

For the uninitiated ... The heat pumping capacity is
maximum at deltaT equal Zero and the heat pumping
capacity approaches zero at peak difference
temperature which is in the order of 67 degrees
per stage. These guys operate close to the peak
difference temperature where the pumped heat is
almost zero. Meaning the application is rather
academic and hardly useable in practise.

Rene
 
P

Phil Hobbs

Jan 1, 1970
0
Rene said:
For the uninitiated ... The heat pumping capacity is
maximum at deltaT equal Zero and the heat pumping
capacity approaches zero at peak difference
temperature which is in the order of 67 degrees
per stage. These guys operate close to the peak
difference temperature where the pumped heat is
almost zero. Meaning the application is rather
academic and hardly useable in practise.

Depends what your idea of practice is. For IR spectroscopy, you're
almost always stuck between wishing you could use a Joule-Kelvin (open
cycle) refrigerator to get lower temperature, and not wanting to deal
with the gas handling, clogged tubes, and vacuum problems. Multistage
TECs are one answer, but it's really, really hard to get below, say, -50
C. The reason is (as Guy pointed out) that only the top stage is really
working near its maximum delta-T--all the other stages are having to
handle the waste heat from all previous stages. Even with spreader
plates and spreading out the TECs in the upper stages, it's hard. Since
the temperature tuning range is what it is, each degree gets you a wider
range, which is often very helpful.

IOW anyone who thinks a multistage TEC is practical is stupid enough to
think this idea is practical too. ;)

Cheers,

Phil Hobbs
 
Top