Solax X3-Hybrid Gen 4 - type 14 ver 3.006.04 - Grid 1/2/3 Power values off - seem not to include actual grid draw

[darmach]

I recently installed Solax X3-Hybrid G3, api inverter type 14, firmware 3.006.04.

The values identified in here as Grid A/B/C Power seem to be weird :)

All is fine as long as PV produces all of the required current - the moment PV production is lower and we draw current from the (actual) grid there are discrepancies, the more grid draw the more off they are.

Those three values sum seem to equal PV1 Power - I'd wage that they are not actually the power being drawn from the grid, but the power delivered from PV system to local load specific phases.

For example (situation in my system right now):
PV1 production (only PV, one chain): 271W
Load power: 1140W
Grid power: (this is correct -> ~ load - prod) 857W

Grid A Power: 97W
Grid B Power: 92W
Grid C Power: 94W

Sum 283W (slightly off but usually equal to PV production, nowhere close to over 1kW pulled from the grid)

Screenshot and api output attached.

curl -sd "?optType=ReadRealTimeData&pwd=SRXXXXXXX" -X POST http://XXXXXX
{"sn":"SRXXXXXX","ver":"3.006.04","type":14,"Data":[2345,2345,2381,12,11,12,92,87,90,269,6177,0,4,0,257,0,5000,5000,4999,2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,64640,65535,0,0,0,23760,0,0,2363,0,0,1,45,1165,256,1568,2573,6155,100,0,31,0,0,0,0,0,0,0,0,0,0,0,0,0,83,0,15,41,0,0,12,0,41,0,0,0,62,0,11,0,0,0,20,0,7851,0,0,0,1000,0,0,0,0,0,0,0,0,0,1,60,1,26,68,256,2628,1800,310,350,212,197,33,33,2,1115,258,9766,9766,0,0,0,0,3292,3288,15430,0,21302,14389,19013,12594,16690,13132,13881,21302,14389,19013,12594,16690,13132,13881,21302,14389,19014,12339,16692,13132,13879,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2049,260,258,1537,0,23760,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"Information":[10.000,14,"H34A10J1536060",8,1.27,0.00,1.26,1.09,0.00,1]}

I'm trying to find the correct values - so the actual draw from specific phase from the grid, which is known to the inverter, as it displays the correct values on the status screen > meter.

Having the correct values for Grid A/B/C power, we could rename current as for ex. Inverter A/B/C power, and add them to get current total power consumption per phase. Could be nice, as current values create confusion - they are named as Grid A/B/C power, but actually they don't include power drawn from the grid for each phase, mind that the inverter 'knows' the correct values, as the total power drawn from the grid is correct, and they are show correct in system status > meter.

No luck so far, will look into it further, unless someone already sorted that out?

[darmach]

Using this oneliner to grab data fast, excluding values with 4-5 digits (depending on the load I'm using to debug):

ln=0; for i in $(curl -sd "?optType=ReadRealTimeData&pwd=SRXXXXXXXX" -X POST http://XXXXXXXXXX | jq '.Data[]'); do echo -en $ln" "; echo $i; ln=$(( $ln+1  done | grep -v " 0$" | grep -v " [0-9][0-9][0-9][0-9]"

[darmach]

Quick update, as I initially thought that Grid A/B/C Power values are showing PV -> Load power added to Grid -> Load power, it looks like when I started charging the battery from the grid those 3 went into negative, showing -2kW with the battery pulling ~2,5kW - 400W being delivered by PV

[darmach]

I think it would be good to at least rename them - as these for sure are not Grid power, with the grid being the electrical grid as external entity. (Mind that "Grid Power" is fine, it is the actual grid power pulled - I'm talking only about Grid A/B/C Power)

[darmach]

Just an update, in the evening I switched load draw from grid to battery. Grid A/B/C Power values are consistent with actual local power draw from each phase. Grid (the actual, external grid) power draw is zero.

[nazar-pc]

These names are coming from the app itself, I didn't come up with those names myself. While they can be renamed, updating this repo with new versions of the app will become more difficult.

And yes, as you discovered, power can be both positive and negative depending on the direction.

[darmach]

Hi @nazar-pc indeed, then it might be better to leave them as they are, maybe adding the comment explaining what they actually are, I'll take a stab at this.