Active power import jumping to 100 000 000 W

[joernnilsson]

I'm struggling with what looks like a parsing error. ams_active_power_import is jumping from the actual value of around 1 kW to a fixed value of 100M.

Other measurements look ok, although I haven't checked the history of all of them.

This started yesterday when I upgraded to 1.9.7, but stayed when I tried downgrading to 1.9.6.

I'm quite capable of gathering debug data and testing changes, but wanted to check with the dev(s) first if you have any pointers.

Meter: Kaifa MA304H3E (Norwegian)

[turbokongen]

Can you enable debug, and attach the log. I may have an Idea of what is wrong, but I need the logs to confirm. 👍

[joernnilsson]

Attached log, from boot to having seen the issue at least once

ams.log

[joernnilsson]

I printed the parsed states as well, and it seems to happen when a larger chunk of data is read:

2022-11-07 09:44:38.838 DEBUG (Thread-3 (connect)) [custom_components.ams] data read from port=[126, 160, 39, 1, 2, 1, 16, 90, 135, 230, 231, 0, 15, 64, 0, 0, 0, 9, 12, 7, 230, 11, 7, 1, 9, 44, 38, 255, 128, 0, 0, 2, 1, 6, 0, 0, 2, 37, 242, 181, 126]
2022-11-07 09:44:38.839 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_import (state: 549)
2022-11-07 09:44:38.839 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_export (state: 0.0)
2022-11-07 09:44:38.840 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_import (state: 0)
2022-11-07 09:44:38.840 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_export (state: 81)
2022-11-07 09:44:38.840 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l1 (state: 0.787)
2022-11-07 09:44:38.840 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l2 (state: 1.874)
2022-11-07 09:44:38.841 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l3 (state: 2.108)
2022-11-07 09:44:38.841 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l1 (state: 231.7)
2022-11-07 09:44:38.841 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l2 (state: 0.0)
2022-11-07 09:44:38.841 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l3 (state: 234.8)
2022-11-07 09:44:41.291 DEBUG (Thread-3 (connect)) [custom_components.ams] data read from port=[126, 160, 121, 1, 2, 1, 16, 128, 147, 230, 231, 0, 15, 64, 0, 0, 0, 9, 12, 7, 230, 11, 7, 1, 9, 44, 40, 255, 128, 0, 0, 2, 13, 9, 7, 75, 70, 77, 95, 48, 48, 49, 9, 16, 54, 57, 55, 48, 54, 51, 49, 52, 48, 52, 49, 50, 57, 57, 53, 52, 9, 8, 77, 65, 51, 48, 52, 72, 51, 69, 6, 0, 0, 2, 34, 6, 0, 0, 0, 0, 6, 0, 0, 0, 0, 6, 0, 0, 0, 81, 6, 0, 0, 3, 13, 6, 0, 0, 7, 37, 6, 0, 0, 8, 14, 6, 0, 0, 9, 16, 6, 0, 0, 0, 0, 6, 0, 0, 9, 44, 111, 222, 126]
2022-11-07 09:44:41.291 DEBUG (Thread-3 (connect)) [custom_components.ams.parsers.kaifa] kaifa meter_type length is 8
2022-11-07 09:44:41.292 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_import (state: 100663298)
2022-11-07 09:44:41.292 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_export (state: 0.0)
2022-11-07 09:44:41.293 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_import (state: 0)
2022-11-07 09:44:41.293 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_export (state: 81)
2022-11-07 09:44:41.294 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l1 (state: 0.781)
2022-11-07 09:44:41.294 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l2 (state: 1.829)
2022-11-07 09:44:41.295 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l3 (state: 2.062)
2022-11-07 09:44:41.295 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l1 (state: 232.0)
2022-11-07 09:44:41.296 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l2 (state: 0.0)
2022-11-07 09:44:41.296 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l3 (state: 234.8)
2022-11-07 09:44:42.899 DEBUG (Thread-3 (connect)) [custom_components.ams] data read from port=[126, 160, 39, 1, 2, 1, 16, 90, 135, 230, 231, 0, 15, 64, 0, 0, 0, 9, 12, 7, 230, 11, 7, 1, 9, 44, 42, 255, 128, 0, 0, 2, 1, 6, 0, 0, 2, 48, 68, 253, 126]
2022-11-07 09:44:42.900 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_import (state: 560)
2022-11-07 09:44:42.901 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_active_power_export (state: 0.0)
2022-11-07 09:44:42.901 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_import (state: 0)
2022-11-07 09:44:42.901 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_reactive_power_export (state: 81)
2022-11-07 09:44:42.901 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l1 (state: 0.781)
2022-11-07 09:44:42.902 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l2 (state: 1.829)
2022-11-07 09:44:42.902 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_current_l3 (state: 2.062)
2022-11-07 09:44:42.902 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l1 (state: 232.0)
2022-11-07 09:44:42.902 DEBUG (Thread-3 (connect)) [custom_components.ams.sensor] Updating sensor ams_voltage_l2 (state: 0.0)

[joernnilsson]

https://github.com/turbokongen/hass-AMS/blob/1738118e8a727cd924ab255994b2b026795583a9/custom_components/ams/parsers/kaifa.py#L92-L112

Changing the call on line 100 to byte_decode(fields=pkt[71:75]) seems to give correct results on my setup. I wouldn't presume to know the history of these parsers well enough to say it's a solution, but perhaps you have an opinion @turbokongen.

If this is merely an index issue, I can create a PR with the change. If this change brakes something else, we'll have to go a bit deeper.

[turbokongen]

Fixed in #91 v1.9.8 Thank you for reporting