How does sensor.goe_xxx_pvopt_averagepgrid get calculated? #3
Replies: 11 comments 27 replies
-
Hi Walt, the simple & honest answer is - I don't have a clue at all. All data the integration provide will be read 1:1 from the Wallbox [technically this is not 100% correct - some of the values will be divided by a constant - you can find the factor attribute in the entity descriptions in the source code. So the integration provide the data from the wall box without any kind of additional functionality. When you made the observation, that sending only GRID information would lead to more/stable results, then it could be the case, that the 'sign' of the battery might be not correct? This might be a a bug in the documentation [what means "exporting" and what means "importing" (from where to what?)]... But coming back to your initial question -> 'sensor.goe_xxx_pvopt_averagepgrid' will be calculated by go-eCharger Wallbox internally and this integration just provide the data to HA... -> I just can recommend to ask your question @ https://github.com/goecharger/go-eCharger-API-v2/discussions |
Beta Was this translation helpful? Give feedback.
-
I am/was also totally surprised, that go-eCharger offers the feature to provide PV data "manually" via the API (as alternative to the installation of the hardware solution)... Actually when I saw this in the documentation @ github that was the point where I wanted to create/provide this integration - and also 'yes' you are fully right - this is a simple wrapper of the existing API into HA entities... |
Beta Was this translation helpful? Give feedback.
-
I think the truth is a bit in between. Reading the full posts here goecharger/go-eCharger-API-v2#110 suggests that pGrid is used, and to some extend also pAkku, but not pPv. this means that for more complex settings a custom controller on the HA side might be more suitable (by modifying/hacking pGrid and pAkku when in ECO mode or directly with the commands amp (Ladestrom einstellen), frc (Ladevorgang starten/beenden) und psm (Phasenumschaltung) ) |
Beta Was this translation helpful? Give feedback.
-
Maybe I'm not seeing something. I now only submit pgrid an here's a case where I plug in the car at 21:00 after sunset when the house was drawing from the battery. Before plugging it in, pgrid was ~0, as expected: Still, charging started with 4.2 kW (6A, 3 phases) even though pgrid was far away from the configured start (1400 W): I'll now try @karliwalti's approach of adjusting pgrid by the battery power, but as I understood the semantics of pgrid, this shouldn't be necessary, no? |
Beta Was this translation helpful? Give feedback.
-
This is now a bit speculative, but it is likely that the charger "probes" how much it can drain until pGrid becomes positive (or larger Zielwert Netz). When a battery is attached, it can do this up to the max battery discharge current without seeing a change in the pGrid value. The default setting is likely that a battery discharge for charging a car is OK (since it is temporary stored PV power). |
Beta Was this translation helpful? Give feedback.
-
I don't have a battery and had the same problem. I have now set this target grid value (Zielwert Netz) to -1500 and it now leaves me with 1500 watts surplus. Is it working so far, am I right? |
Beta Was this translation helpful? Give feedback.
-
does anyone know whether it is possible to change the PV surplus calculation interval? it would be great if it could be reduced slightly rather than at 300 (stopped charging process) or 30 (active, paused charging process) second intervals. |
Beta Was this translation helpful? Give feedback.
-
The main problem is actually the calculation interval. can this be changed anywhere? |
Beta Was this translation helpful? Give feedback.
-
I am not sure whether I can follow your question. Are you asking whether it is possible to override the pause? |
Beta Was this translation helpful? Give feedback.
-
If you check the API documentation link, you will find for 'ids': Nevertheless, your logic above suggests the opposite: You need to implement a larger filter (stronger smoothing) to avoid stopping the car's charging (but you will pull power from the grid). You do not need to update the calculation frequency; you need to update the value you feed to 'pGrid'. |
Beta Was this translation helpful? Give feedback.
-
ok the solution with the deactivation of the stop automation was good. Now I still have the problem that the recalculation of the surplus (the 30 seconds) is simply too long. Example: If my car is charging with 22kW and then suddenly a cloud appears, it can happen that it continues to charge with 22kW for 25 seconds even though the value from the PV has long been less. I ask again: Is it not possible to change this calculation time???? I get the data from my inverter every 10 seconds. |
Beta Was this translation helpful? Give feedback.
-
My understanding is that this is the value that is being used to drive the controller of the PV optimized charging. The target would be a sensor.goe_xxx_pvopt_averagepgrid between -200 and 0
The reason I ask is that I observe a strong oscillatory behavior in control if I transmit PV, Battery, and Grid information. When I send only grid information, the system runs much more stable but depletes the battery before stopping.
Beta Was this translation helpful? Give feedback.
All reactions