Solar Freaks

Hi Lee,

There are a few designs floating around on these forums. Tony (the board owner) has the FatMAX using an atmega48. He's doing some improvements to it.

I haven't seen anyone using a PIC but my first suggestion is make a tiny test case that just runs the PWM in some very set ways. Do you have access to a scope to see what the output looks like? You mention duty cycle, so you're probably using a 50/50 duty cycle, not a fixed-frequency PWM mode?

Good luck. -dch

Hi,
Thanks for your quick reply,
Well I have the PWM output of a PIC which changes according to the measured Voltage / Current of the cell,
However as I tend to have a more 'Software Brain' about these types of things, some of the calculations for the Inductor value and PWM frequencies go over my head!

I have a string of small PV cells which output about 4 Volts in strong sunlight, and I'd like to keep them at their measured maximum power point when charging a battery. (The actual charging circuit is not really important at this stage)

Are their any tutorials around which explain how to select suitable components for the converter, and ascertain the correct PWM frequency? To change the frequency seems to be a simple case of writing to a register in the PIC.

Many Thanks for your help,
Lee

Hi Lee,

Sorry, no tutorials that I know of. I am a software guy too, well but I have "branched out" into hardware, and I'm studying it now at University.

I could guess at what you want to do with a PWM output - control the duty cycle of a switching regulator?

Your comment about getting 4V was interesting, but I still don't get it. What is your design? Are you aware of the MPPT chips in some of the topics here?

You've got the right idea. Most MPPT designs use a high precision high current resistor (1 Ohm is typical) to directly measure current. You're making the assumption that your load is equivalent to a 1K resistor. That's not wrong: it sounds like you have hooked it up to a 1K resistor.

However, you said you wanted to hook it up to a boost converter. That's going to present varying resistances (and reactance, like an inductor). That's when you'll need another way of calculating current.

Hi,

Thanks for your help, it looks like I'm going to be using a buck converter instead due to finding some useful details on how to design one!

How would you suggest that I calculate the current flowing in the device for this configuration? I was simply going to use a 1K load resistor on the output and measure the voltage across this, however I realise this is not ideal.

Thanks in advance for your help!
Lee

For current sensing have a read of the attachments in this thread.
There is a good document from Linear Technology.

The circuit im using on the FATMAX uses a very Low R current shunt on the input, then i use a Zero drift Op-amp to amplify the voltage to levels the ADC can read with reasonable accuracy.

Make sure the Op-amp your using has a common mode input range starting from ground.
I recommend the LTC2050.

[quote][/quote]

Try an LT1013