After changing my OpenEVSE setup to use the feed ‘emon/emonpi/power2’ plus changing the Mode to Production my solar generation is now very accurately tracked by the car and no energy is wasted. However, when other house loads are switched on the extra power only comes from the grid, there is no corresponding reduction in ev use to balance the load even if I enable the ‘No Import’ filter.
I would like the ev power to reduce as the house load increases to maintain a balance with the available solar power. Is this possible?
I am including some information images below in the hope that someone will spot my setup error.
At the risk of steering you wrong again, I recognise what you describe and it is the reason why in my installation the value I published to MQTT for the EVSE was equivalent to “Solar production minus household consumption minus a margin for error”
I don’t know how you can achieve that MQTT feed on the current software, but maybe someone else here can
You will need to change the MQTT feed to emon/emonpi/power1 and the Mode to Export.
According to your diagram CT1 is your grid import, the CT must be orientated, so the reading is positive import/negative export.
But as I mentioned previously, from my experience using Production mode and the PV generation often works better. You can adjust the PV power ratio to allow for household loads. It depends on your use case, but in my experience the household loads are small compared the EV.
Please look at the diagram again. The Emon pi grid monitor is CT2, it’s located between Henley block 1 and Henley block 2. There is another CT monitoring the grid but that is for the diverter.
Are you saying that the way my system is working is correct for Production mode?
I would like to try using export mode so I moved CT2 as you suggested and switched the mode to Export but the OpenEVSE was not enabled even though kW’s were being exported. I noticed that the Grid +Import/-Export is constantly stuck showing -12W for some reason. See attached.
There must be an issue with the reading coming from the emonPi if it’s always reporting -12W. Can you login to the emonPi Emoncms and check to see what the Input from Power2 is reading?
Are you able to (temporarily to test) put CT2 on the same wire as CT1 and compare readings - they should be more or less the same (± 10% or better). If not, again temporarily, swap CT1 & CT2 at the input of your emonPi. If the bad reading transfers to Power1 input, the CT is faulty, if it doesn’t, it’s either CT input 2 of your emonPi, or your emonhub.conf file is wrong - check the line scales=... for Node 5 (the emonPi).
I shall do the test you have asked for but bear in mind the system has worked fine in export mode for the last 7 years using the feed from the diverter (SolarDiverter) rather than the Emonpi.
Briefly swaping the feed back to SolarDiverter restores normal export mode operation.
Well, that was interesting. Removing Emon pi CT2 from the cable between Henley Block 2 and Henley block 1 I then waited for the power indication to drop to zero. It did not change! I then removed the Grid connection CT2 for the diverter (eMC) and the power reading dropped to zero.
I have clearly got the inputs confused in the software somehow.
If the data in the emonPi “is stuck showing -12W.” (your words) – and bearing in mind the Inputs page above shows +10.22 W – something looks wrong, and the best place to start is with a check on the incoming data – because if that is wrong (maybe something has failed), you stand no chance.
[Edit]
Obviously, we were both typing at the same time.
Looking at your diagram in the top post, emonPi CT1 is reading the PV Infeed, emonPi CT2 is reading the TOTAL infeed (grid + solar, if the c.t’s are orientated according to our convention, which is power into the house is positive). Is this your intention? If it is, that’s OK but you’ll need to adjust most of the things in our documentation because I think everything is written with the expectation that CT1 is the Grid infeed and CT2 is the Solar infeed (and this is why Power1plusPower2 is available as the total). Is this the source of a mixup?
