Hybrid Car – More Fun with Less Gas

Thermal couples - Page 4

register ::  Login Password  :: Lost Password?
Posted by Peterthinking on November 30, 2003, 6:47 pm
My head hurts...GAhhhhh!

why not just electroplate an entire pot bellied wood stove.
Or a 55 gallon drum.

there has to be a way to make it work...surface area seems to be the
key...plating would solve that.



address is fake.)

Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com ).
Version: 6.0.545 / Virus Database: 339 - Release Date: 11/27/03

Posted by Anthony Matonak on November 30, 2003, 11:33 pm
Peterthinking wrote:

Electroplating a pipe would place the layers the wrong way around.
Also, I'm no expert in thermocouples but I don't think it's dependent
on surface area so much as the difference in the electrical and
thermal conductivity of the metals.

scholar.lib.vt.edu/theses/available/etd-8497-205315/ unrestricted/chap2.pdf


Posted by daestrom on December 3, 2003, 12:06 am



Two dissimilar metals will only produce a voltage potential in the millivolt
range.  To get higher voltage, multiple 'connections' are needed.  Say you
use copper and iron.

When you place the iron against the copper, you get a millivolt range
voltage.  To add another 'connection, you need to connect another
iron-copper connection in series with the first.  If you connect the iron
from the second junction to the copper of the first, you've created a third
junction with the *opposite* polarity from the first and second.  Net
result, still just one junction's potential available.  Multiple layers, all
at the same temperature, will not develop any higher voltage than the one
junction.  Even if you mix iron, copper, constantin, tin, and other metals
in various combinations.  The various junctions formed will just add and
subtract from each other leaving you with the same potential as combining
the first and last metal together directly.

In real-world systems, they get around this by making the connections
between the first and second junction 'someplace else'.  The would run the
copper wire from the first junction to the a colder place.  And run the iron
wire from the second junction to a colder place where it connects with the
first.  You still have three junctions, but two are 'hot' but the one
in-between and opposite polarity is 'cold'.  So the voltage you gain is the
difference between the potential of the second junction, minus the potential
formed in the parasitic junction that is colder.

Increasing the surface area will reduce the resistance of the circuit.  This
can be important in determining the overall current that can be developed,
but not the voltage.


Posted by Anthony Matonak on December 3, 2003, 7:21 am
 daestrom wrote:

A device that uses multiple connections is called a thermopile.

A thermocouple generates it's voltage NOT at the junction but through
the entire length of the wire that extends from the hot to cold zones.
The junction simply provides a path for the electricity to flow. The
electricity is generated from the flowing of heat from one end of the
wire to the other and the fact that the two wires, being made from
different metals, conduct heat at different rates. The total power
generated isn't a question so much of surface area but electrical
and thermal conductivity of the wires (or plates).

A thermopile around a pipe might be composed of two different layers
arranged in an alternating zig-zag fashion. Something like this...


Where the / would be one metal, the \ would be the other and the
===== would be a pipe (isolated electrically from the thermocouples)
containing something very hot. Assuming the outside of the entire
device was in something cooler.


Posted by daestrom on December 4, 2003, 1:22 am


Not true.  Look up 'Seebeck effect'.  It (the voltage potential) is created
by dissimilar metals in contact.  The wire can be as long, or as short as
you want.  The voltage is the same.  The connection of the wire to *any*
other metal creates yet another junction.  The temperature of the junction
point and the particular metals used determine the voltage developed.  Same
materials at different temperatures develop different voltage.

Typically, the junctions end up going from copper wire from the measuring
instrument on one end, to a metal forming one wire (first junction), then
the junction of 'interest' where a different metal is used for the second
wire, and finally from this different metal wire to another copper wire back
to the instrument.  All in all, there are three junctions.  The first and
third (from thermocouple wire to instrument copper wire) are at ambient
temperature, the second junction 'of interest' is at some other temperature.

The voltage potential measured simply shows the overall difference between
the 'cold' junction and the middle junction.  For high-accuracy
applications, you measure the temperature of the terminal strip where the
'cold' junctions are and adjust the reading accordingly.  If the terminal
strip where the thermal-couple wires are 'landed' is hotter than the cal.
conditions, the difference in temperature there must be added to the
temperature reading obtained from the instrument.

The amount of 'heat' conducted along the wire, or the total length has
*nothing* to do with the voltage developed.  The diameter and length *do*
affect the resistance of the circuit of course.  In instrumentation, this is
not critical since the instruments have high input impedance and draw very
little current.  In 'power production' this resistance can be the major loss
of total power developed and so is often minimized by large area conductors
and short as can be adequately insulated (thermally) between hot and cold

 The total power

Thermal conductivity has nothing to do with it.  The voltage is a function
of temperature difference and the particular metals used.  The current flow
is a function of the voltage developed and the internal and external circuit

A 'side' effect is the 'Peltier Effect'.  The direction of the current flow
caused by a hot junction will be such that it cools the hot junction, while
the current flow in the cold junctions will tend to heat them.  So, heat is
removed from the hot and 'moved' to the cold by electron current flow.  But
the wire in the middle can be in an ice bath and have no affect on this.
The energy isn't conducted from the hot to the cold by thermal conduction,
it is transferred by the 'Peltier effect'.  Of course, if you *have* an ice
bath, you probably want that for the 'cold' junction instead of whatever you
were using before ;-)

Yes, *this* arrangement would work.  But you would have to keep the cold
cold, and the hot hot.  Any heat that is *conducted* from the hot to the
cold would be a waste.  So a tradeoff between longer / and \ to lessen the
energy that is conducted thermally, and  increased resistance of the longer
wires has to be found.

The idea another posted had of plating one metal over the other and
repeating the plating would *not* work as every other junction would be the
'wrong polarity' yet still at the same temperature.


This Thread
Bookmark this thread:
  • Subject
  • Author
  • Date
please rate this thread