This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Implementing Arduino MPPT solar battery charger with TI chip

Other Parts Discussed in Thread: BQ25570, LM5118

I know that TI has some very robust DC-DC converter ICs.
My question is: is there any IC that can have its PWM duty cycle (NOT frequency) reliably controlled by an outside source (e.g. an Arduino) in order to achieve maximum power point tracking from a solar panel / wind turbine into a battery?

References:
http://www.timnolan.com/index.php?page=arduino-ppt-solar-charger
http://www.timnolan.com/uploads/Arduino Solar/ArduinoSolar.pdf

  • Hello Anand,

     

    Looking at the schematic in the second document it looks like you are interested in finding a synchronous gate driver.

     

    You could look at something like the LM510x or TPS2822x families combined with NexFETs to create the MPPT buck converter that yo ucan control via external PWM.

  • Hi Ryan,
    Thanks for the MOSFET driver suggestion, I'll look into that.
    I should have been more specific, though - I'd like to take advantage of the circuit protection features (under/overvoltage, sensing, etc) already present in the DC-DC converter ICs while externally controlling the duty cycle. I figure it's a long shot, but can't hurt to ask ...

  • Hello Anand,

    I would think that you would be monitoring the input voltage any for the MPPT you would just have to retain control of an enable pin of some other way of engaging or disengaging the driver.

    What sort of charge current are you anticipating? Would something like the BQ25570 work? I would think that you could actively control the MPPT set point pin with an analog voltage to accomplish MPPT.

  • Hi Ryan,
    I'm trying to work out a module that could take up to a few amps at least, perhaps more in a future version. The BQ25570 seems a bit underpowered for that.
    The LM5118 would fit the bill I think, if it'll accept a DAC input. Would it be compatible with Arduino PWM?

    -- Anand

  • Hello Anand,

    You could utilize the FB pin of the LM5118 by having it connected to the DAC to externally control the duty cycle on the device to implement MPPT instead of having it connected to a resistive divider from the output as seen in the datasheet.

  • Waiting to try this out when a sample arrives.

    Further thought: could the HO output be used with a half-bridge MOSFET driver? Replacing the buck diode with a MOSFET ...
    I'd imagine it might fuzz up the IC's function though since it's set up for a diode. Just wondering if the efficiency increase would be worth it.

  • Hello Anand,

    Let me know how the testing goes when the samples come in.

    I would not recommend configuring your system in that manner for the potential timing problems it could produce with the additional dead time from the half-bridge driver and how that could influence the boost gate control when HO & LO are driven at the same time.


    What sort of current are you expecting this converter to process? Depending on the current, FET Rdson, and diode Vfd your saving might not be that great especially factoring in the increased component cost and complexity.

  • Hi Ryan,
    Yeah, I figured the half-bridge might interfere since the LM5118 is set up for a diode. It was just a thought.

    I'm planning to try it with a 20W 1A solar panel. If it works I might add more panels later, anywhere up to 3-5A total. I'm hoping that my design may also see use in wind turbine MPPT if it pans out.

  • Samples haven't arrived yet, but here's the schematic I put together using DesignSpark PCB 5.1 and the LM(2)5118 spreadsheet calculator.

    I calculated it using its full input voltage range, switching at 300kHz and allowing for up to 3A of current in case I choose to add more panels. Could you advise if it looks acceptable?
    (The schematic should be the same as shown in the datasheet with the resistor divider replaced with the transistor / resistor / cap on the FB pin)

  • Hello Anad,

    The Transistor. resistor, capacitor is that intended to be a PWM integrator to allow you to actively control the feedback pin voltage? Depending of the gate current and PWM frequency would a PFET possibly work better?

    Also when you do the layout you want CSG placed as close to the GND side of the CS resistor to create the most accurate feedback signal.

    Otherwise when it comes to values I would trust the calculator.

  • Hi Ryan,

    Yes, the transistor (probably 2N2222 NPN) will be switched at about 500Hz by Arduino PWM, with the res/cap to stabilize the voltage. I don't know whether a PFET would be better ... it wouldn't be switching more than ~14V (maximum battery voltage) from the output pin. Or twice that, if I change the battery capacity.
    Thanks for the advice!