GivEnergy EV Charger with Predbat

[AberDino]

I have a GivEnergy setup with an AIO, gateway and EV charger, all connected with network cables (no Wi-Fi). I've recently acquired an EV, so over the past few days I've been updating my Predbat configuration to 'integrate' the charger. I have a 'basic' ToU electricity contract with cheap electricity between 12am and 6am, so my supplier does not control the charging slots.

My aim was to use Predbat to plan and control the charging slots, but ideally I also wanted it to be aware of "non-Predbat" charging events, for example a forced manual charge from the GivEnergy app. I was going to use the "car_charging_now" sensor for that, but I understand that's not recommended in combination with Predbat-led charging, so I've left that option out.

The GivEnergy EV charger is on firmware 1.13 and I can successfully communicate with it locally using GivTCP. After writing and testing the HA charging automation, I found out that GivTCP can read from the charger, but it cannot 'write' a start/stop command. I confirmed this by changing the 'charge control' drop-down on the GivTCP charger device page, and it had no effect. The command is logged, but not actioned by the charger. I found a post which confirmed local write control is unlikely to ever be implemented due to "SmartCharge regulations". Bit odd that I can't locally control equipment which I own, and to be reliant on a third-party for that, as the charger would be of no use if GivEnergy would stop trading for any reason.

As a workaround, I implemented the GivEnergy cloud API connection for Home Assistant to communicate with the charger, and this all works a treat. I now have Predbat planning and actioning the charging of my EV!

My question is around the method for preventing the AIO discharging to the car. Both Predbat and GivEnergy have an option for this, and I have both options enabled, but the implementation is different and this appears to cause a conflict. Predbat puts the AIO in "PauseDischarge" mode while charging the car, while GivEnergy apply an AIO discharge power limit of 0. As you can see from the attached screenshots, the two actions are implemented (as expected) when the car charging commences. However, GivTCP then detects the AIO 0W discharge rate and resets it to 6000W, which causes a loop, as GivEnergy's EV charger inverter control then sets it back to 0W, and so on. As a workaround, I'm going to disable the GivEnergy option "Prevent Battery Discharging to EV" and see what happens when I charge the car tonight. I suspect this will prevent the loop from occurring, but I'd rather stick with the "belt and braces" approach.

Any thoughts?

[gcoan]

However, GivTCP then detects the AIO 0W discharge rate and resets it to 6000W

I think it unlikely GivTCP is doing that as GivTCP just mirrors what the inverter settings are in HA controls and vice versa.

It is quite probable though that its Predbat changing your battery discharge rate, and finding its not what it expects it to be, is changing it back.

Unfortunately no-one has contributed their configuration of the givenergy EV charger to the predbat documentation, so new discoveries ahead ...
(be useful if you do get this working to document it for others).

Have you set switch.predbat_car_charging_from_battery to False? https://springfall2008.github.io/batpred/car-charging/#additional-car-charging-configurations

The documentation says it should set the discharge rate to zero, and I'd expect that to happen as well as setting Pause Discharge.

[AberDino]

Hi @gcoan, I can confirm that switch.predbat_car_charging_from_battery is set to False, and I'm pleased to report that after disabling the GivEnergy option "Prevent Battery Discharging to EV", the discharge power adjustment loop is gone:
Predbat:


GivTCP:

GivEnergy portal:

So it appears Predbat is not changing the AIO discharge power setting to 0, which is fine in my opinion, as pause discharge does that job nicely.

I then did a separate test by manually changing the AIO discharge power on the GivEnergy portal to a lower value. The portal indicates the value was written successfully, but when I then looked at the AIO control page within HA, the discharge rate hadn't changed. Looking at the AIO log on the GivEnergy portal, "EV Charger Inverter Control" had set it back within seconds:

Looking at my EV charger configuration on the GivEnergy portal, it's set to "solar" mode, which means the vehicle will only charge when there is >1.4kW excess solar power available. On another page there is an option "Max Battery Discharge Power to EV", which should only take effect in "hybrid" or "grid" mode. It was set to 6,000W, so I lowered it to 0W and tried changing the AIO discharge power again, just in case. It appeared to be successful initially, as the discharge rate slider on the AIO control page within HA updated to the new value, but a minute later "EV Charger Inverter Control" restored it back up to 6,000W! I will pick this up separately with GivEnergy, although I suspect it might be a while before I get a response... It might be related to the fact that my EV charger configuration is set to "Inverter Control Cloud", which I believe is correct for my setup, but there are five options to choose from.

I'm the only one in the household with the GivEnergy app so the risk of the car charging at the wrong time and/or draining the AIO is minimal, and I'm happy run with the setup as is. I'll keep testing it for a bit longer and I will then follow up with my configuration so it can be added to the documentation.

[AberDino]

I'm now on Predbat v8.26.1, and it appears the 'pause discharge' during EV charging is no longer working as expected:
Predbat:

GivTCP:

GivEnergy Portal:

As you can see, 'pause discharge' is enabled at midnight when the car is starting to charge, but is then disabled 10 minutes later because Predbat stops issuing the 'hold for car' command, which worked fine previously. The car actually continued to charge until 1:45.

Any chance somebody could verify this behaviour? Thanks

[gcoan]

I agree it looks like PauseDischarge was turned off by predbat when it changed from 'charging hold for car' to 'charging', but am not sure this actually made any practical difference

If the battery is in charging mode from 00:00 to 06:00 then the battery won't discharge regardless of the battery pause mode?

[AberDino]

That's true, but my concern is that if the AIO was full and not in charging mode, then disabling Battery Pause Mode too early would result in the AIO discharging to the car.