Hello everyone, I'm currently working on a project where I am trying to design an MPPT (Maximum Power Point Tracking) circuit by myself, and I’m looking for some guidance. I would really appreciate it if someone could explain how MPPT circuits work, the key components involved. If anyone has experience with MPPT circuit design, could you please share your insights or resources that could help me in building and understanding these circuits? Thank you in advance for any help!

 

They are quite simple in theory - the controller adjusts the load the solar panel sees by varying the operation of the SMPS circuit, to keep the panel at the VI maximum operating point. As the maximum operating point will change with variations in insolation, the controller must track this change to keep the load at the maximum. Some very sophisticated algorithms can be implemented. Most MPPT controllers are microcontrollers and the algorithm is implemented in software, but there are designs that are linear.

https://aurorasolar.com/blog/the-importance-of-modeling-global-maximum-power-point-tracking/

https://ww1.microchip.com/downloads...plicationNotes/ApplicationNotes/00001521A.pdf

here's an interesting paper on a linear mppt design:
https://ieeexplore.ieee.org/document/954019

 

Look at the Power v. Output Voltage for your solar panel. You will note a few things:
1. The peak is quite broad, so small inaccuracies are not going to result in much power loss
2. The peak hardly varies with light level
3. The peak varies with temperature.
Because the battery impedance is very low, but the solar panel impedance is high, a buck regulator will regulate the panel output to Vbatt/D (where D is the duty cycle), rather than regulating the battery voltage.
So what you need to do is adjust D so the panel voltage hits the peak.

You can do it very simply, by setting D to a fixed value that corresponds to an average temperature and average battery voltage. The disadvantage is that the lowest battery voltages are usually first thing in the morning when the panels are cold and giving the highest output voltage, so the two variables go in opposite directions.
You can make it a bit more sophisticated by adjusting D for the temperature, and the battery voltage.
Or you can use one of the many algorithms, the most common being Perturb and Observe, where you adjust D a little bit alternatively up and down, a note whether the power increases or decreases, then change D in the direction which makes the power increase.

 

here's an interesting paper on a linear mppt design:
https://ieeexplore.ieee.org/document/954019

Any chance of a copy which isn't password protected?

 

Any chance of a copy which isn't password protected?

It wasn't when I originally sourced it (from the author's website if I think back)

 

It wasn't when I originally sourced it (from the author's website if I think back)

Thanks. That’s a really interesting paper.
UnfortunatEly, you can buy a lot of processing power for the cost of a AD633!
However, they use the AD633 to produce a signal which represents the panel power (Vpanel x Ipanel), which must be the same as the power going into the battery (Vbattery x I battery). As the battery voltage changes very little on that timescale, and we’re onlyinterested in dP/dt, then the battery current should be proportional to the power, and eliminate the need for the multiplier.

 

Unfortunately, you can buy a lot of processing power for the cost of a AD633!

With a digital solution, it's never been the silicon that costs the money. Good code is a scarce and expensive resource.

...the battery current should be proportional to the power, and eliminate the need for the multiplier.

Good thinking - I need to digest that one.

 

The simplest MPPT algorithm is called perturb and observe.
There is a ton of information, please google it, no need to attempt to condense the information here.