sun tracker - make a project ?
Archiver
Posts: 46,084
>
> If you have a decent clock and calander you could point at the Sun
without seeing it at all, which would be good when there's clouds.
It's also good not to be making any moves that are not necessary
since that uses extra electricity.
>
> Note that the COS function is broad at the peak. For example you
only have a 10% loss if the panel is misaligned by 25.8 degrees, or
5% at 18.2 degrees, so tracking to sub degree accuracy is not
needed. Since it takes more power to start moving than to move
continously it will take less energy to make fewer moves, the ideal
case is continous movement.
>
> Just some ideas,
>
> Brooke Clarke, N6GCE
> http://www.RPC68.com
I have been following this thread and since it will not die, I
figured why not add some life to it ?
OK, we can agree that following *CAN* be easy with a constant speed
motor, but that does not use the Stamp, nor does it allow for errors
in power, controls or mechanics.
A sun tracker can be as simple as a couple LED's into a comparator or
other sensors into a Stamp.
Following the sun can be agumented by monitoring the solar radiation
and the PV output and ossasionally, testing the posistion to verify
the optimum setting. ie: move foreward a little and measure the
effect.
Then monitor the changes in output as the sun crosses the path. Then
verify if your set point is the optimum or off a few percent.
Yes, a RTC can be added and specific numbers called out, and even
servo feedback for location, but that goes more the open loop control.
Feedback is too easy not to take advantage of it.
One could even calculate the total current per day/month/year.
Maybe one could even data log clear days or clear hours, where the
sky is not overcast based on some baseline.
And, don't forget the dirty glass sensor.
Dave
> If you have a decent clock and calander you could point at the Sun
without seeing it at all, which would be good when there's clouds.
It's also good not to be making any moves that are not necessary
since that uses extra electricity.
>
> Note that the COS function is broad at the peak. For example you
only have a 10% loss if the panel is misaligned by 25.8 degrees, or
5% at 18.2 degrees, so tracking to sub degree accuracy is not
needed. Since it takes more power to start moving than to move
continously it will take less energy to make fewer moves, the ideal
case is continous movement.
>
> Just some ideas,
>
> Brooke Clarke, N6GCE
> http://www.RPC68.com
I have been following this thread and since it will not die, I
figured why not add some life to it ?
OK, we can agree that following *CAN* be easy with a constant speed
motor, but that does not use the Stamp, nor does it allow for errors
in power, controls or mechanics.
A sun tracker can be as simple as a couple LED's into a comparator or
other sensors into a Stamp.
Following the sun can be agumented by monitoring the solar radiation
and the PV output and ossasionally, testing the posistion to verify
the optimum setting. ie: move foreward a little and measure the
effect.
Then monitor the changes in output as the sun crosses the path. Then
verify if your set point is the optimum or off a few percent.
Yes, a RTC can be added and specific numbers called out, and even
servo feedback for location, but that goes more the open loop control.
Feedback is too easy not to take advantage of it.
One could even calculate the total current per day/month/year.
Maybe one could even data log clear days or clear hours, where the
sky is not overcast based on some baseline.
And, don't forget the dirty glass sensor.
Dave