Predbat incorrectly allowing discharge to EV and remaining in Demand mode during IOG daytime EV charge

[hamiltongross]

Describe the bug
After plugging in my car today at 1305 to my Ohme charger, an IOG charge session ran from 1306 to 1330. During that time Predbat remained in Demand mode and allowed discharge to the EV from the battery.

Expected behaviour
I was expecting Predbat to change to Hold for Car mode and prevent discharge from the battery to the EV

Predbat version

v8.31.14

Environment details

• GivEnergy 9.5kWh battery. GivEnergy AC coupled inverter.
• Standard HAOS installation.
• Ohme ePod EV charger (connected to IOG)
• Skoda Enyaq (one of two EVs configured in Predbat in exclusive mode)

Screenshots

Log file

predbat-2.log

Predbat debug yaml file

predbat_debug_13_20_00.yaml.txt

[gcoan]

Looks like you are on Predbat Octopus direct

Ohme identified the car being plugged in, triggering a predbat replan:

2026-01-10 13:06:27.320059: Watched event: sensor.ohme_epod_status_id_3 = {'entity_id': 'sensor.ohme_epod_status_id_3', 'state': 'plugged_in', 'attributes': {'friendly_name': 'Ohme Epod Status iD.3'}, 'last_changed': '2026-01-10T13:06:27.312029+00:00', 'last_reported': '2026-01-10T13:06:27.312029+00:00', 'last_updated': '2026-01-10T13:06:27.312029+00:00', 'context': {'id': '01KEM07TBDJ5M6WW6AQDV9KBDW', 'parent_id': None, 'user_id': None}} will trigger re-plan

which car was plugged in, car 0 or car 1?

Looks like Predbat thought that car 0 was on IOG but there was no charging planned and car 1 charging was not for some time

2026-01-10 13:06:54.048782: Cars 2 charging from battery False planned [False, True], charging_now [False, False] smart [False, False], max_price [0.0, 0.0]p, plan_time ['07:00:00', '07:00:00'], battery size [58.0, 58.0]kWh, limit [46.4, 46.4]%, rate [7.4, 7.4]W, exclusive [True, True]
...
2026-01-10 13:06:56.040434: Car 0 on Octopus Intelligent, no active plan
2026-01-10 13:06:56.040527: Car 1 plan charging from 46.4 to 46.4, with slots [{'start': 1410, 'end': 1440, 'average': 7.0}, {'start': 1440, 'end': 1470, 'average': 7.0}, {'start': 1470, 'end': 1500, 'average': 7.0}, {'start': 1500, 'end': 1530, 'average': 7.0}, {'start': 1530, 'end': 1560, 'average': 7.0}, {'start': 1560, 'end': 1590, 'average': 7.0}, {'start': 1590, 'end': 1620, 'average': 7.0}, {'start': 1620, 'end': 1650, 'average': 7.0}, {'start': 1650, 'end': 1680, 'average': 7.0}, {'start': 1680, 'end': 1710, 'average': 7.0}, {'start': 1710, 'end': 1740, 'average': 7.0}, {'start': 1740, 'end': 1770, 'average': 7.0}, {'start': 2850, 'end': 2880, 'average': 7.0}, {'start': 2880, 'end': 2910, 'average': 7.0}, {'start': 2910, 'end': 2940, 'average': 7.0}, {'start': 2940, 'end': 2970, 'average': 7.0}, {'start': 2970, 'end': 3000, 'average': 7.0}, {'start': 3000, 'end': 3030, 'average': 7.0}, {'start': 3030, 'end': 3060, 'average': 7.0}, {'start': 3060, 'end': 3090, 'average': 7.0}, {'start': 3090, 'end': 3120, 'average': 7.0}, {'start': 3120, 'end': 3150, 'average': 7.0}, {'start': 3150, 'end': 3180, 'average': 7.0}, {'start': 3180, 'end': 3210, 'average': 7.0}] from SoC 46.4% to 46.4%, ready by 07:00:00
2026-01-10 13:06:56.040697: Car 1 charging is exclusive, will not plan other cars

the subsequent predbat run identifies the extra OIG slot, but then seems to have the wrong times for it, and ignores it in the plan:

2026-01-10 13:07:49.364940: Car is charging now - added new IO slot {'start': '2026-01-10T13:00:00+0000', 'end': '2026-01-10T13:30:00+0000'}
...
2026-01-10 13:07:49.980808: Car 0 on Octopus Intelligent, no active plan
2026-01-10 13:07:49.980891: Car 1 plan charging from 46.4 to 46.4, with slots [{'start': 1410, 'end': 1440, 'average': 7.0}, {'start': 1440, 'end': 1470, 'average': 7.0}, {'start': 1470, 'end': 1500, 'average': 7.0}, {'start': 1500, 'end': 1530, 'average': 7.0}, {'start': 1530, 'end': 1560, 'average': 7.0}, {'start': 1560, 'end': 1590, 'average': 7.0}, {'start': 1590, 'end': 1620, 'average': 7.0}, {'start': 1620, 'end': 1650, 'average': 7.0}, {'start': 1650, 'end': 1680, 'average': 7.0}, {'start': 1680, 'end': 1710, 'average': 7.0}, {'start': 1710, 'end': 1740, 'average': 7.0}, {'start': 1740, 'end': 1770, 'average': 7.0}, {'start': 2850, 'end': 2880, 'average': 7.0}, {'start': 2880, 'end': 2910, 'average': 7.0}, {'start': 2910, 'end': 2940, 'average': 7.0}, {'start': 2940, 'end': 2970, 'average': 7.0}, {'start': 2970, 'end': 3000, 'average': 7.0}, {'start': 3000, 'end': 3030, 'average': 7.0}, {'start': 3030, 'end': 3060, 'average': 7.0}, {'start': 3060, 'end': 3090, 'average': 7.0}, {'start': 3090, 'end': 3120, 'average': 7.0}, {'start': 3120, 'end': 3150, 'average': 7.0}, {'start': 3150, 'end': 3180, 'average': 7.0}, {'start': 3180, 'end': 3210, 'average': 7.0}] from SoC 46.4% to 46.4%, ready by 07:00:00
2026-01-10 13:07:49.981006: Car is charging now slot {'start': 780, 'end': 810, 'average': 29.14}
2026-01-10 13:07:49.981067: Car 1 charging is exclusive, will not plan other cars

[hamiltongross]

At 1305 I unplugged car 1 (which finished charging hours before) and plugged in car 0, which then started charging.

Yes, I'm on Octopus Predbat direct.

Thanks!

[gcoan]

At 1305 I unplugged car 1 (which finished charging hours before) and plugged in car 0, which then started charging.

Thanks for the info, that didn't seem to be detected that you had swapped cars, predbat continued thinking car 1 was plugged in.

Probably part of the issue ....

[hamiltongross]

Sorry, just to clarify.

Although I did unplug car 1 and plug in car 0 at that time, it seems that my automation to switch the Ohme vehicle didn't run properly and Ohme remained set to car 1, although car 0 was plugged in.

[gcoan]

Thanks, that's useful to know. So even though predbat thought car 1 was still plugged in, are you saying that it should have accepted the extra IOG charging slot as being for car 1 and planned accordingly?

With thinking that car 1 was plugged in and the future charging for car 1 was in the future, predbat kept the battery in demand mode.

[hamiltongross]

Also, just to mention that I use two helpers, one for each car, which contain the charging status for that car. Depending which car the Ohme vehicle is set to, that car's status is either set to unplugged or to the value of ohme_epod_status.

Those two helpers are used for the car_charging_planned and car_charging_now for each car.

So in this case car 0's car_charging_planned and car_charging_now would have indicated unplugged, and car 1's car_charging_now and car_charging_planned would have indicated charging.