In Reply to: I'm not sure I missed the point... posted by Commuteman on August 28, 2003 at 17:25:28:
If you look at the instantaneous power being delivered to the load..P(l) = V^2 / R
And any peltier losses (using a factor K to describe the peltier coeff's)
P(p) = K * I (I being the instantaneous current)
I = V(l) / R (current through the loop is dependent on the voltage in the load. (I neglect the peltier voltage term for this, using the source voltage, as it is at least 10e6 times greater than the error component)
P(p) = ( K * V(l) ) /R
The total power making it to the load
P(t) = P(l) - P(p)
P(t) = V^2 /R - (K * V(l) ) / R
P(t) = (1/R) * (V^2 - K*V(l))
Notice the load power has a V^2 term and a V term.The V term being the peltier losses.
So, the voltage at the load is proportional (use M to rep 1/R) to the the sqr of the total power at the load...
V(total) = M * sqr (V^2 - K*V)
The first term is the normal undistorted term, the second is the peltier term.
So what in the name of sam hill is the spectra of sqr(V) when v is a sine?
Cheers, John
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- calcs. - jneutron 04:09:38 08/29/03 (0)