Propeller Head Plaza Technical and scientific discussion of amps, cables and other topics.

---

[ Follow Ups ] Thread: [ Display  All  Email ] [ Propeller Head Plaza ]

---

[ Alert Moderator ]

---

Calculation of small effects..

Posted by jneutron (A) on August 26, 2003 at 09:26:58

Hi peter..got tired of the right hand side of my screen.

Been thinking about how to calculate what's going on..

First, peltier..

Assume at time t=0, instantly pushing current through a junction.

Because delta t is zero, the peltier efficiency is at max.
As time progresses, the thermal gradient builds up, with the heat (cold) front moving away from the junction at a characteristic speed, defined by diffusivity of the metals.
As the gradient moves out and builds, the delta t across the actual junction increases, causing a decrease of the efficiency.
Eventually, the gradient at the junction will assume steady state, that being minutes, seconds, defined by the overall structure.

But, at some arbitrary time, say 1 millisecond, turn off the current.
For that one milli, thermal gradient has been established..Yah, it hasn't travelled far into the materials, perhaps one to ten mils..but, energy has been stored.
Now, with the current off, the gradient across the junction is producing electricity, returning it to the current source(go with me on that, I make the assumption the drive is bipolar and able to support four quadrant operation.)

Now, the efficiency of seebeck? It starts high with high gradients, and diminishes with time here, becoming zero when the gradient is gone.

But, in addition to the seebeck generation, is the material attempting to go back to uniform temperature. That would be dissipating the energy within the metals, not to be returned to the source.

Both the drive part, and the return part have non linearities.

Now, for a sine drive: during the rise of the current, peltier occurs, putting energy in. After the peak, seebeck occurs, trying to return the energy. Then, the opposite polarity does the same, albeit with the gradient reversed.

Note that energy is stored first, peak energy at peak waveform, like a cap. Then, it returns during fall, part of it anyway, again like a cap. But, while all this is going on, the materials are trying to short circuit the thermal gradient by conducting heat back across it.

As a first start, I'd use the following:

Square wave input, say 1mS..into an already established gradient across a 10 mil thick junction. That locks the efficiency and gets rid of time constants, while allowing heat capacity to be averaged, to get rise and energy delivered.
Then, Back the process out, this time using average seebeck coefficient.

Cheers, John

This post is made possible by the generous support of people like you and our sponsors:

Sonic Craft

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2025, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---