Peter - Sounds OK to me.

One way of tracking the maximum - If that was what you wanted to do  -
which it isn't - would be to increase the mark-space ratio until the output
starts to drop and then increase it again until it drops etc etc. This way
the duty cycle wanders backward & forward across the peak - but is never
very far from the ideal point.
May be advantageous to restart at longish intervals just to make sure there
is no lockout but  I have a sample system that works like this running & it
appears to track OK.
Fairly slow response & crap efficiency but it is tracking.

Richard P



                    Peter Mcalpine
                    <petemca@BIGPO        To:     PICLIST@MITVMA.MIT.EDU
                    ND.NET.AU>            cc:
                    Sent by: pic          Subject:     Re: [EE]: Solar charging of batteries
                    microcontrolle
                    r discussion
                    list
                    <PICLIST@MITVM
                    A.MIT.EDU>


                    28/01/03 16:50
                    Please respond
                    to pic
                    microcontrolle
                    r discussion
                    list






Ok, so modify the software flow for a simple MPPT, the uP
scans through the PWM range looking for either the target current
supply, or if it can't reach that, goes for the peak current level.

Also, I would guess that if I can meet the target current supply (ie
the cell can supply more) then it would probably meet these at 2 points
on the PWM range, so I would guess that the one that had the highest
solar voltage would be the better one to pick. So perhaps that can also
set the PWM scan to start from a low on duty cycle.

does that come closer to a better solution?

Given that this routine will track the maximum current output via
varying a PWM signal, I can't see the need to open cct the solar
cell and check the voltage there. Am I missing something?
(If I was using a fixed look up table then I can see the need).

Cheers
Peter Mcalpine

>
> > Basic software flow...
> > FET off - check solar voltage
> > If solar volt enough (ie daytime), start PWM at low level.
> > Measure shunt V drop
> > if lower than target V drop increase PWM
> > repeat till target is met.
>
> This is not MPPT, and will not work where the illumination is too low to
> sustain the target current.  Instead, "hunt" around the current operating
> point to adjust for maximum output current.  This is what a lot of the
> discussion has been about the last few days, so I won't repeat it.
>

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads