Emulating solar battery function

[greengeek]

...and another web page I need to look at more closely:
https://www.homemade-circuits.com/homem ... t-maximum/

[Moose On The Loose]

Moose, if it's any help, I'm pretty sure that the output voltage of a solar panel (at its maximum power point) doesn't vary by much between when it begins to put out usable power in the morning and when the sun is directly overhead at noon.

The point of maximum power should vary very little with the amount of sun light. Basically here is how I think about it:

A photon of light comes it and hits the panel.
This creates an electron, hole pair.
These go to their respective sides of the panel.
With a short circuited panel, this will lead to just about the normal 0.65 Amps per watt of light near the peak for a single cell. (just into the infrared from red light).
A solar panel is a series connected string of cells. Each cell is a silicon diode. The forward voltage drop of a diode applies to each of these diodes. If the voltage across the diode is about 0.5V, not much of the short circuit current is lost by flowing back through the diode. If the voltages rises to about 0.7V per cell, then basically all of the photocurrent is lost to the diode current.

I have a "more complicated than it needs to be" design working in the modeling but this fact about the panel voltage is making me reconsider the need to optimize over a wide range. I also don't have any circuit in place to stop it over charging a supercap etc.

The folks I work for, have gone back to full time so until the weekend I will only spend a little more time on the idea. I am thinking that I should make something more along the lines of "here is how to think about this and understand the electronics" than a "here build this". I figure more people will get use from it if I go that way.
I think I will make it in the form of an HTML page I can give a link for. Making a photo ends up making a quite huge file.

What do you guys think?

[Flash]

You could scan (or take a high-res picture of) a schematic, then, if necessary, use mtPaint to crop the picture and/or reduce its resolution to the minimum that still shows all the details. I use mtPaint to do that all the time.

[Moose On The Loose]

You could scan (or take a high-res picture of) a schematic, then, if necessary, use mtPaint to crop the picture and/or reduce its resolution to the minimum that still shows all the details. I use mtPaint to do that all the time.

Editing I will start on the left to explain:

I1 and the big chain of diodes is my model of the panel

R8 = 0.01 Ohms assumed for wiring from panel

C9, L1 is an EMI filter you will want in the final design to keep it out of your radio etc.

C5 is a place to store up some charge

D2 protects the "LM7908" against a sudden short at the panel. They don't like sudden 0V or negative V at their inputs when the output side has voltage

Q3, Q4 is my fake LM7908 I didn't have a good model of a real one.
I expect to make this a real one. We may change to LM7812.

Remember what I said about "ground" not really being ground. The rest of the stuff runs between 17V and {17-8V}

R10,R11 and "knob" is how I model the manual adjust 10K pot

U3 look up LM741. They quit making it but opamps are well explained using that as an example. The real one will be an 8 pin DIP for ease of prototyping. This circuit is what adjusts the PWM to make it hold the panel at a given voltage.
More explanation if you ask

LT1011 AKA LM311 is a comparator. When "-" is above "+" it pulls its output down to ground. When "+" is higher it doesn't do that so the resistor R4,R4 can pull it up.
Q1 and Q2 provide the ability to move a lot of charge in and out of the gate of M1
R6. C8 are positive feedback so that when the LM311 moves its output up or down it tends to stay there for a while.
R7, C4 are negative feedback so "for a while" comes to an end after about 10uS
R3, R4, C3 as so that when the output of the LT1011 swings up, it doesn't slow down as it nears the positive voltage. This is a "bootstrap"

D1 is a "catch diode" for L3
M1 and D1 alternate as M1 switches on and off. The average voltage across an ideal inductor is always going to be zero. The other end of the inductor is at our output voltage.

R2 is my current sense for making the change from having a pot to automating the action.