Solar Sense Rev 2.0 Validation

Owned by Aashmika Mali
Jul 09, 2021
4 min read

This page will go over the validation conducted for the rev 2.0 solar sense

The board has been almost fully assembled with the initial setup shown in the following image:

The fw on the contoller board is solar from master on July 9th 2021. When first turning on power, it can be noted that all LEDs with the exception of LED1 are on - this is good because we are testing with the “6 mppt” variant for now (since we do not want to build two solar senses anytime soon, we have just put the 6 mppts and fan control all on one board, even though normally the fan control would be on the 5 mppt variant). Note at this point, the MPPT is not providing any power, as we are just doing hw checks.

 

When I turn on the board, the relay makes random noise, which tells me that the the hold current is not sufficient. I measure 13.48V on in 12V rail and 3.25V on the 3V3 rail which is sufficient. One suggestion is to make the power rail testpoints farther from the leds and each other for easier/quicker probing.

To deal with the noise issues, we did the following:

  • put a short on D2 → no change

  • replace R21 with a 100kohm resistor instead of 160kohm → the noise became more consistent and at a slower pulse

  • take off the short → the random noise happened again, but at less of a frequency (originally it was somewhat rapid clicking)

  • I think the short has nothing to do with the clicking because when i put it back on, it made random noise again

We then started to measure the relay coil voltage using the scope to see what was happening:

ok so for the longest time the relay would just turn on and off randomly like in the first image. That is me probing either side of D2 and D1. The next image is what started happening after a while where instead of doing random clicking, the frequency of the clicking started to decrease after turning on the power supply. The video is the noise generated from the same issue.

 

Also note, the status pin was always gnd on the oscope except initially when the powersupply was first turned on (seemingly an unimportant transient).


Liam suggested doing the following, one at a time:

  • Decrease resistance of R26 (currently 255k, but it should be > 170k)

    • I changed it to 200k ohm

  • Increase resistance of R25 (I think about 200k should be fine)

    • n/a

  • Increase C21 (2 uF should be fine)

    • n/a

After changing R26 to 200kohm the relay turned on and off fine whenever the power was supplied, so we moved on in the testing for now.

I madea. harness for one MPPT to connect to the lowest referenced MPPT as seen in the below diagram (I made a harness that attaches to the battery pack as input to the mppt, and harnessed the spi from the mppt to the solar sense as well)

connecting the battery pack (11.4V) and turning the pot on the mppt, I got a constant 26V output from the mppt an into the board. I was trying to read minicom from solar sense but no values were updating for some reason, and so i tried with flashign spv1020 code, but the relay kept being turned on and off and I wasnt sure if the code was actually reading from the correct relay because this was the output

Called it a day → will continue the attempt to communicate with the SPV1020 and MCP3427 tomorrow with some fw help