Battery Drain - Danfoss - Pulling every few seconds

Tags: #<Tag:0x00007f43330a9380> #<Tag:0x00007f43330a9290> #<Tag:0x00007f43330a91c8>

@chris

Here are the logs

openhab2.3.0.log and openhab2.4.0.log

Same behaviour, only with 2.4.0 - After clean install… Even bootstraping new Raspberry image

I have an idea of what is up. Give me some time to work this through (over the next week).

@Chris
Thank you for your help till now an patience. Sure IF you want me to test some custom build ping me.

Hi Chris,
Did you had time to work on this issue?

@chris maybe I can contribute to the changes too (somehow)

Does the problem still exist with the latest binding?

@chris
What is the best way to update to latest binding?

Use @5iver script.

@chris

Using the latest snapshot 20181120. LC-13 and Popp doesn’t seem to register/finish registration. and the ones that we have had issues (Old LC-12) stay on the alarm mode like they have no connectivity. (2.4.0.M5) also the same
Debug log: binding and wakeup https://pastebin.com/n1CyYbdz

Node 10

@chris After many tries it paired. Now the problem still exists!

Drain and constantly connecting, as before

Sorry - I’ve been busy with other things (still am) but I will take a look at this today.

I’ve made some further updates and updated the log viewer to better interpret the messages from the device. Please try the latest version and if it doesn’t work still, send a log and I’ll take a look.

@chris thank you for your effort.
Here is the latest log: 20:22 (log time the problem begins with node 11
https://pastebin.com/LVDAh7mE

also adding: Still having a 15% battery / day drain

I’m not sure there’s much more I can do. The updated code seems to be working as it should, so without more information I’ve probably come to the end of the road. I would suggest to contact the manufacturer to find out what the device wants to see.

What I have added is responses to two messages that the device sends periodically - you can see this below -:

These commands are acked by the device, but if it’s still not happy, then we need to find the user manual for this device that explains what is required.

@chris There is no manual explaining anything like that on these. I think it is time to through them out of the window. As danfoss is also non responsive on that.

Though it is strange 2.3.0 and other software does something to keep them happy. :\

Thank you for your effort! really appreciated.

As far as I can tell, the latest binding is doing the same thing as 2.3. If you can double check it would be useful - I’ve upgraded the log viewer in the past few days to try and make it easier to see what is happening.

I saw that. But still strange, right?

Sure, but what do you suggest?

I recognised that updating the value of current temp or something on the device, it stops this endless loop.

Similar to this issue: https://github.com/OpenZWave/open-zwave/issues/218
by sending a command class 70 or 129 (remove) in order to force cleaning up

in this particular situation the patch for open-zwave sends remove to these 2 classes. (If i understand well). this makes device to not go to Panic mode! and drain.
the device is exactly the same.

I have several Develo Thermostats MT02650 which have the Danfoss Thermostat as basis and Devolo Firmware. I think this topic applies for both, but correct me if I am wrong.

I switched from FHEM to OpenHab. Now I am missing the clock command class in OpenHab. CCS is not necessary as it is strange to handle anyway.

I know from my own experience and from Danfoss support, that you need the clock set correctly on those devices, as the device will perform a calibration procedure sometimes (at least once after initialization, but I believe it is also after battery change). During this calibration, the device will step by step open its valve in the morning ours and measure how the room is heating up (depending on room and heater size).

In FHEM I added a “rule” to automatically update the clock once a day since it was drifting quite heavily on some devices.

If you cannot set the clock, the devices will perform this calibration heating randomly during the day. In my case I have one device which is doing it in the evening at the moment when the child needs to sleep. So I have to remove the battery everyday, to abort this procedure.

I would like to add the clock command class again, to be able to manually set and read the clock. How can we do this?

Should I better create a new topic for the Devolo device?