ReneNulschDE / mbapi2020

Custom Component to integrate MercedesME devices into Home-Assistant

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Values jump between actual Value and old Value?

Ravenbs opened this issue · comments

The problem

During loading, the Values of our "Smart EQ" jump between the "Start to load" Value of 42% and the actual loading value.
This is how it looks in Home Assistant - makes ist more clear:
image

Thisis how it looks in Influx (&Grafana):
image

This is how the Data looks like in Influx (dont know how to output raw data from HA):
Fields:
time car car_str device_class_str domain entity_id friendly_name friendly_name_str retrievalstatus_str state_class_str timestamp timestamp_str unit_str value vin vin_str
Values:
...
1712655447798482000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409021052 2024-04-09 02:10:52 PERCENT 42 1453391151234 W1A4533911K51234
1712655455979917000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409113631 2024-04-09 11:36:31 PERCENT 47 1453391151234 W1A4533911K51234
1712655462800080000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409021052 2024-04-09 02:10:52 PERCENT 42 1453391151234 W1A4533911K51234
1712655470976120000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409113631 2024-04-09 11:36:31 PERCENT 47 1453391151234 W1A4533911K51234
1712655477800164000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409021052 2024-04-09 02:10:52 PERCENT 42 1453391151234 W1A4533911K51234
1712655485977244000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409113631 2024-04-09 11:36:31 PERCENT 47 1453391151234 W1A4533911K51234
1712655492801273000 1453391151234 W1A4533911K51234 battery sensor w1a4533911k51234_state_of_charge 1453391151234 W1A4533911K51234 State of Charge VALID measurement 20240409021052 2024-04-09 02:10:52 PERCENT 42 1453391151234 W1A4533911K51234
...

As you can see, the Database entrys for "42%" and "47%" look nearly the same - exept one difference:
Good values have the correct time in "timestamp_str"
Bad values have all "2024-04-09 02:10:52" in "timestamp_str", even if it is now 12:00. And its always this string in time, not ticking forward...

It is the same for all other Values - not only SOC.
For example Charging Power also jumps between "0" and "6.9kW" even if it is constantly charging:
image
Same with the timestamp_str here, expect that bad values have: "2024-04-06 06:53:00" and good values the actual time.

What version of MBAPI2020 do use?

v0.13.0

What was the last working version of MBAPI2020?

No response

What type of installation are you running?

Home Assistant OS

Diagnostics information

No response

Example YAML snippet

No response

Anything in the logs that might be useful for us?

No response

Additional information

No response

Interesting. Looks like the MB Servers sending different datasets. Could you go to the options of the component and activate "Debug only: Save messages...". Then let it run until you see some of these data jumps. Switch off the debug only:.... option. Go to the folder HA-config/custom_components/mbapi2020/messages, compress all files into an zip or tar. Send the archive to mbdev.issue221@nulsch.de and I can take a look. And please share the diagnostic log too...

I will do so.
Need to wait a Day until the car is empty to charge again and then send it :).
Tnx for taking a look, I love your plugin.

Short update from today:
Today driving and loading had perfect results - no longer jumping.
Only difference is, that now the debug logging is active.
Will keep the logging on for one more day - then try to disable logging if this has an impact. Will report if anything changes.

Do you have any other input? Otherwise I'll close the issue.

Thank you for tracking this. The issue has never ocurred again since then. Your Plugin is working great now, thank you for this!