The power output curve is (should be?) a parabola. If you draw too much current the voltage drops and the output power decreases, likewise, if you draw too little current the voltage increases but again the output power is reduced. With a MCU you could measure output voltage and current and calculate the effective output power. Then you could vary your PWM to seek the highest power output. If illumination or load changes, your output power will then decrease. In this case you vary the PWM to seek for the highest output again. If at first you change the PWM in the wrong direction then the output power will reduce even further, then you know that PWM must be adjusted in the other direction. Perhaps you could keep sweeping the PWM in a narrow range around the current set-point to keep adjusting it all the time. That would keep your system always close to the MPP, something similar to what car ECUs do with the lambda probe. Isaac On 23/02/2015 11:13, Gordon Williams wrote: > What Harold said is the maximum power point occurs at roughly a constant= =20 > *voltage* from the solar panel. That is somewhere between zero and max= =20 > voltage output. > > You are talking about highest *current*, which is wrong. The highest=20 > current will be when the solar panel is short circuited and produces no=20 > power. > > What you can do is take you panel and measure voltage and current from=20 > it under various loads to characterize the panel output. Knowing these=20 > two values you can graph the power output and determine the peak power=20 > output voltage under sun and shade conditions. You can then confirm=20 > what Harold is saying is true. You can then use this info to design your= =20 > MPP. > > Gordon Williams > > > > > On 15-02-23 08:34 AM, Tobias Gogolin wrote: >> Thanks Harold! >> MPP is definitely is around highest output current, however I see no >> simple way to adjust feedback based on something that must be >> dynamically sought up and down? >> Constant input is just not a given when thinking about the moving >> shades and reflections a moving vehicle is exposed to all the time! >> However you do touch an important issue, as the normal circuit would >> in the effort to achieve output target, possibly drag the input down, >> way out of the MPP range! >> >> Let me see what you guys think about this type of pseudo code for the >> MPPT algorithm: >> >> D=3DFeedbackScanVoltageStep (Positive or Negative) >> F=3DFeedbackVoltage >> loop: >> P=3DPreviousOutputCurrent(A) >> F+=3DD >> A=3DMeassureOutputCurrent() >> if(A>P) Direction might be good, continue loop >> else D=3D-D, continue loop >> >> >> >> On 2/23/15, Harold Hallikainen wrote: >>> One clever thing I learned from Linear Technology is that the maximum >>> power point is very close to a constant voltage out of a solar panel. S= o, >>> a very simple MPPT converter is a constant voltage converter where you >>> take the feedback from the INPUT instead of the output. The switcher >>> adjusts the duty cycle as required to keep the input voltage constant. >>> >>> Harold >>> >>> >>> >>> >>> -- >>> FCC Rules Updated Daily at http://www.hallikainen.com - Advertising >>> opportunities available! >>> Not sent from an iPhone. >>> -- >>> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive >>> View/change your membership options at >>> http://mailman.mit.edu/mailman/listinfo/piclist >>> --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .