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.

TIDA-010042: Not desired switching to boost converter - but worse. 19th January version

Part Number: TIDA-010042

Hi!

I downloaded new software and uploaded to PCB. All what I changed was to uncomment #define SYS12 and commented #define SYS48, because I use 12V system. When I connected the power supply(15V, 100mA) as solar panel, I noticed high voltage spikes on PV input and after few tries the MOSFET in buck stage went short.

After replacing mosfet and checking if everything is fine using old software ( 2019) with my changes, I also uncommented function "Protection with instantaneous values of variables" and uploaded new January version. This time MOSFET shorted immediately.

Then i messed with buck_lower_treshold cause in new wersion is significantly higher (520 vs 30), but no succes here.

Old software with my changes is working well at high loads (>0.05A charging) but sometimes fails as the current falls to 0.

Can you advice how to configure code correctly?

Also, I noticed some suspicious fragments of code:
 I did not notice any resetting of uint16_t maxFlips. It is responsible for speed of duty adjustion. After overflow it will behave quite randomly, but with no bigger effect on MPPT operation.

if(Panel_Power < Prev_Panel_Power)
    {
        MPPT_Direction = MPPT_Direction * -1;
        //Check for steady state
        maxFlips+=1;

        if(maxFlips==50)
        {
            dutyChange =1;
        }

Greetings

  • I have no load connected to Load terminal. I would like to use this MPPT only as charger. Does it influence behavior of charging? I will be powering low-power device with the lead-acid battery, say 20mA constant current and 7A 2sec peak roughly 20 times a day. I need large capacity of battery and high power PV to ensure proper autonomous operation in winter in most EU countries.

    Best regards

  • Hi,

    There is no requirement to connect a load. The new version of software addresses earlier issues with lower power charging. Since I am not an expert on software, I will get someone to look at this problem. Please allow us a couple more days to find out the issue.

    Regards,

    Salil

  • Hi,

    The following changes were done in the software from the older version to begin with.

    First, the MPPT loop was running in watchdog timer which was slower in execution and this was changed to a timer based execution. Since we are running the loop faster, we had the bandwidth for implementing protection against overvoltage and overcurrent. This is section can be commented and should not be interfering with the problem. Also the duty cycle limit should be ok.

    Check for the minimum battery current setting in the code. In the current code it is set to 3.

    Also ensure the deadband is configured properly.

    Please see the waveform above captured for 12V output. I had tested it with a PV input voltage of 20V and CC limit of 250mA. I had a load resistance bank connected (not a battery) of around 20 ohms connected. With the old software we had FETs blow up due to reverse power flow, but that has been corrected by introduction of feature called diode emulation mode in addition to synchronous buck mode of operation.

    Please let me know if these help.

    Regards,

    Harish

  • No succes yet. Mosfets now arent burning, maybe the last one failed due to fatigue.

    Behavior:

    1. I connect 20V 250mA regulated linear power supply. Device starts operationm and it pumps energy to battery

    2. I lower the settings of power supply. Voltage 10V, current close to 0 mA. MPPT switches off.

    3. I set voltage to 20 V and current limit to about 200mA. Voltage on PV input oscillates at 22 V. I did not measure currents yet.

    I can provide some measurments, including oscilloscope ones. But I do not know what to trigger - P1.7 as interrupt from timer A?

    Is there a guide how to safely debug high frequency power devices? Im always worried if i will stop code execution and cause some short.

    Best regards

  • Hi,

    The code is intended to work this way: as soon as you set the input and output test conditions, the software does MPPT and once it reaches this state, it does regulation of load voltage. Can you confirm if you are using the same hardware as TIDA-010042 design files?

    Thank you

  • Hi,

    I'm using TIDA-010042 hardware. I have two working boards.

    I do not get what do you mean by : "as soon as you set the input and output test conditions". 

    This software works as mentioned above - it finds MPP and regulates load voltage. 

    The problem is that when the power from PV is less than self-consumption, then It still reverses power flow. It is now very gentle compared to previous software, but still occurs. I think software senses it and terminates it, but cant return to proper operation afterwards.

    Regards

  • No succes yet.
    I am using old software with diode in terminal output to prevent backward operation. I routed battery voltage to 196K resistor directly for proper operation and shorted PV and battery mosfets. It's working fine, but diode introduces power losses.

    I will probably abandon this project due to lack of belief in safety of operation.