springfall2008
GivEnergy AIO dropping into Idle overnight
Describe the bug Predbat is generally operating successfully, although increasingly the AIO is sitting in Idle mode following an overnight charge. This causes the battery to sit full, and power to be drawn from grid.
A physical power cycle on the AIO kicks it out of Idle and back into normal operation.
Expected behaviour Predbat's plans look accurate and should be expecting the battery to discharge.
Predbat version v8.27.18
Environment details
- GivEnergy AIO using GivTCP
- Standard HAOS installer
Screenshots Attached log from Giv AIO
Log file Attached from suspected time of event.
Predbat debug yaml file Attached
could you give more details of what your tariff rates are, what predbat planned, and at what time do you think the 'stuck in idle' started happening?
its not clear from the information you have provided what the problem is
Sure thing. Tariff rates are 7p 11:30pm-5:30am. Predbat planned to Charge until 5:30am. I think the Stuck in Idle starts from 5am. In line with the 'Enable DC Discharge = False' on the Giv AIO.
Sure thing. Tariff rates are 7p 11:30pm-5:30am. Predbat planned to Charge until 5:30am. I think the Stuck in Idle starts from 5am. In line with the 'Enable DC Discharge = False' on the Giv AIO.
thanks
so if the cheap rate was to 5:30am then predbat would keep your AIO charging to 5:30, which would result in grid import until then, as that's cheaper than using your battery (and incurring conversion losses from the stored charge).
I can see the AC charge enable was turned off at 5:30, is there any subsequent entries in the GivEnergy portal, in particular changing Enable DC Discharge?
Also, can you look at the GivTCP log to see what it shows around the 5-6am period. We should see what commands predbat sent to GivTCP and whether there were any issues sending those to the AIO.
+what version GivTCP are you running?
No - Enable DC Discharge is not moved to True.
GivTCP is v3.5, and the log from last night is as follows;
2025-11-14 01:51:50,466 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 03:00 was a success 2025-11-14 01:52:00,667 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 02:11:36,318 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success 2025-11-14 02:11:46,828 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 03:35 was a success 2025-11-14 02:11:56,492 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 03:01:39,879 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success 2025-11-14 03:01:52,847 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 03:40 was a success 2025-11-14 03:01:59,847 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 03:21:43,164 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success 2025-11-14 03:21:53,174 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 03:50 was a success 2025-11-14 03:22:03,717 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 03:41:35,708 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success 2025-11-14 03:41:55,657 - GivTCP2 - write - [INFO ] - Setting Discharge Slot start 1 Start to: 03:30 was a success 2025-11-14 03:42:05,810 - GivTCP2 - write - [INFO ] - Setting Discharge Slot End 1 to: 04:01 was a success 2025-11-14 03:42:56,129 - GivTCP2 - write - [INFO ] - Setting Timed Export mode was a success 2025-11-14 04:01:04,357 - GivTCP2 - write - [INFO ] - Setting Charge Slot 1 Start to: 04:00 was a success 2025-11-14 04:01:14,478 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 04:40 was a success 2025-11-14 04:01:26,722 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 04:01:35,266 - GivTCP2 - write - [INFO ] - Setting Eco mode was a success 2025-11-14 04:40:11,322 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success 2025-11-14 04:40:31,429 - GivTCP2 - write - [INFO ] - Setting Discharge Slot start 1 Start to: 04:40 was a success 2025-11-14 04:40:41,088 - GivTCP2 - write - [INFO ] - Setting Discharge Slot End 1 to: 05:01 was a success 2025-11-14 04:41:31,646 - GivTCP2 - write - [INFO ] - Setting Timed Export mode was a success 2025-11-14 05:00:21,816 - GivTCP2 - write - [INFO ] - Setting Charge Slot 1 Start to: 05:00 was a success 2025-11-14 05:00:31,820 - GivTCP2 - write - [INFO ] - Setting Charge Slot End 1 to: 05:30 was a success 2025-11-14 05:00:41,506 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to enable was a success 2025-11-14 05:00:52,106 - GivTCP2 - write - [INFO ] - Setting Eco mode was a success 2025-11-14 05:30:11,158 - GivTCP2 - write - [INFO ] - Setting Charge Schedule to disable was a success
so no errors in the givtcp log
I just looked at mine and it seems that givtcp doesn't log when the enable DC discharge is turned on or off (charging slot from 4-7am):
2025-11-14 00:00:32,154 - H - write - [INFO ] - Setting Charge Schedule to disable was a success
2025-11-14 00:00:32,673 - H - write - [INFO ] - Setting Battery Pause Mode to Disabled was a success
2025-11-14 04:00:06,432 - G - write - [INFO ] - Setting Battery Pause Mode to PauseDischarge was a success
2025-11-14 04:00:07,110 - H - write - [INFO ] - Setting Battery Pause Mode to PauseDischarge was a success
2025-11-14 05:30:09,777 - G - read - [INFO ] - Saving current energy stats at start of day rate tariff (Dynamic)
2025-11-14 05:30:16,703 - H - read - [INFO ] - Saving current energy stats at start of day rate tariff (Dynamic)
2025-11-14 06:02:34,003 - G - write - [INFO ] - Setting Charge Slot 1 to: 06:00 - 07:00 was a success
2025-11-14 06:02:34,501 - G - write - [INFO ] - Setting Charge Schedule to enable was a success
2025-11-14 06:02:45,210 - H - write - [INFO ] - Setting Charge Slot 1 to: 06:00 - 07:00 was a success
2025-11-14 06:02:45,689 - H - write - [INFO ] - Setting Charge Schedule to enable was a success
2025-11-14 06:05:05,755 - H - write - [INFO ] - Setting battery charge rate 1300 was a success
2025-11-14 06:10:06,259 - H - write - [INFO ] - Setting battery charge rate 900 was a success
2025-11-14 06:15:05,501 - H - write - [INFO ] - Setting battery charge rate 2600 was a success
2025-11-14 07:00:06,825 - G - write - [INFO ] - Setting Charge Schedule to disable was a success
2025-11-14 07:00:07,306 - G - write - [INFO ] - Setting Battery Pause Mode to Disabled was a success
2025-11-14 07:00:07,797 - H - write - [INFO ] - Setting Charge Schedule to disable was a success
2025-11-14 07:00:08,293 - H - write - [INFO ] - Setting Battery Pause Mode to Disabled was a success
2025-11-14 12:30:16,431 - G - write - [INFO ] - Setting Charge Slot 1 to: 13:00 - 16:00 was a success
and looking further, its not even a control that is exposed in givtcp, so can't be something that predbat is manipulating
Looking at my own portal log for the the 4-7am charging slot, enable DC discharge is never changed, only battery pause mode and AC charge enable at the start and end of charging, and charge start/end time:
enable DC discharge is turned off on my inverter and from searching the portal log, I can't see that it was ever turned on
I don't think this is a predbat issue as enable DC discharge isn't exposed by GivTCP in HA. Am not sure that that was actually the problem as it's not turned on on my inverter. Maybe its a givtcp problem, not clear
It sounds related to the way GivTCP exposes the timed discharge, for GivTCP the inverter control is set to 'has_discharge_enable_time: False'
This means that changing the mode selector to Export does this for you and you don't need to adjust the time.
If that is not the case for this inverter then maybe changing this to True and configuring the timed discharge enable could work.
Or you can use GECloud direct which will basically do this for you without the need for GivTCP
ge_cloud_data: True ge_cloud_serial: '{geserial}' ge_cloud_key: 'xxx' ge_cloud_direct: True ge_cloud_automatic: True
