Linearity. Cd cells can be adjusted to match at one brightness, but they will not track at a different brightness. Adjust in full sun and the first time a cloud is in the way, the tracker will run to one side or the other.

I do not believe that this is what will happen in practice as the differences in linearity will be very small and the necessarily wide deadband and hysterisis in the control code will take care of any small differences. During daylight hours, the panel will only move in one direction (reversing mid-day would not be allowed) and does not need to constantly track the sun; it might be more efficient to move the panel at fixed intervals. But, I think to track within 1 degree will require the panel to reposition itself within intervals no greater than 4 minutes.

 

Dear Sir,

Have you any better option to solve the problem???

Ecaits: You need to learn how to follow a link. See this blue text. Click on it and read.

 

put the led's in aprox 2 inch long tubes, the tubes placed in a V with the leds in the bottom facing out. the angle of the "V" determines the step size of the tracking. the tubes should shade the led's when the sun isnt aligned with the openings.

 

I do not believe that this is what will happen in practice as the differences in linearity will be very small.

You must be working with better cells than I have worked with.

 

Dear Sir,

Have you any better option to solve the problem???

While browsing through a collection of old Elektor magazines, I spotted a design for a solar tracker - it was basically just a pair of LDRs in series across the rails, the tap between them is connected to a window comparator, the window is set by 3 resistors across the rails - in the design; the middle and top resistors were presets to set the "no action" range and one of the action levels.

The whole circuit was pretty simple, the 2 outputs from the window comparator simply drive a full-bridge output with a brush type DC motor strung between them.

 

I remember it well....