I’m running openHABian on a RPi4 with a Aeon Z-Stick Gen5+.
There seems to be a channel mapping issue with the DSC11 Smart Strip. I understand that there are over 1000 things supported by the Z-Wave binding and it requires community participation to help resolve issues. I would like to better learn how I can help when I discover issues.
The strip has six outlets. Four are switchable and two are always on. Looking at the strip (left to right), the first two outlets are the always on. They are followed by four outlets are labeled “Switchable 1” through “Switchable 4.”
In Paper UI, the Power Strip thing shows up with 21 channels. The first three channels coorespond to the entire strip and work as expected:
switch_binary
meter_kwh
meter_watts
The next 18 channels are for the outlets and that is where things seem to be a little mixed up. Below is how I believe how the functioning channels currently map to each outlet. Switched Outlet 1
switch_binary1
meter_kwh3
meter_watts3
From what I can tell, these channels do not seem active:
meter_kwh5
meter_watts5
meter_kwh6
meter_watts6
switch_binary5
switch_binary6
I attached the XML and z-wave Debug Logs. I did the following to generate data for the logs:
switch_binary on
switch_binary off
switch_binary1-4 on
switch_binary1-4 off
switch_binary5-6 on
switch_binary5-6 off
What other resources should I be using when trying to troubleshoot something like this?
Thank you for the response. I should have indicated that I did use that viewer, but I wasn’t quite sure what I was seeing. I can confirm that the logs showed no information for the channels that I assumed were missing data.
There does appear to be some things to cleanup in the database, but it would be very odd if the Channels mapped to the physical device that way. If so, it sounds to me like a hardware defect, where they assembled the device incorrectly. Aeon support is great, so they might send you another one. If not, you may be able to take it apart to map the physical outlet and Channels properly.
How have you setup any of the configuration parameters? My guess is that you have not set outlet 3, 4, 5 or 6 to report power and energy.
From the config params, it looks like a temperature Channel might be missing, but I have not read through the XML you provided or the manual.
Based on the other Aeon energy meters, it looks like there should be a parameter 103 for group 3. The manual or engineering doc should provide this info. There is also reference to Item1 and Item2, but this probably should be Group1 and Group2, so parameters 101 and 102 could probably use some cleanup.
If the last two outlets are always on, then it sure seems like switch_binary5 and switch_binary6 should be removed. I haven’t read through the manual to confirm.
The strip is probably five years old. I don’t think that Aeon would replace…and it really isn’t an issue I “need” to resolve. TBH: I was more so looking to learn so I may contribute to the openHAB project. I really don’t need power monitoring at the outlet level.
Here is my uneducated guess: The unswitched outlets should be linked to the channels ending in 1 or 2. They are for the power metering but switching should be not applicable for those. My guess is that the four switched outlets should be 3-6 for everything. That is the way that the outlets are physically laid out and the power metering for the first two switched outlets follows that. With no knowledge, I’m guessing that the power metering for channels for the second two switched outlets just isn’t set up correctly. I saw an old post that seems to confirm that.
Users are key to getting the database populated, but there is also a small group of us helping @chris to maintain it. If you want to update the Z-Wave device database, start here…
Setting the log level for the Z-Wave binding to debug is one of the most useful tools there is. You’ll find instructions for this at the very end of the Z-Wave binding documentation…
I suggest linking an Item to each Channel and confirming all of the switch_binary Channels turn the outlets ON/OFF properly. Only 4 of these should do something… well, maybe 5, if there is a master switch on endpoint0. Then set the configuration parameters so that the energy/power monitoring is enabled for each outlet and confirm those Channels are working properly too. From your results, we can cleanup the device in the database. You’ll need help though, since removing Channels is not something that average users have permissions to do.
@5iver Thank you for the tips on where to start. I previously did what you said regarding the debug logs and adding all the channels. The database manual was very informative. I still haven’t absorbed all of it but I will reread a few more times. Also, to answer an earlier question, I am running 2.5.10-1 (Release Build).
I have made some observations which seem to be pointing me in the direction of a solution.
The device XML has endpoints 0-4 but the database has endpoints 0-6. I found an old mention from @Chris that endpoints 5-6 needed to be added the database.
The database shows a COMMAND_CLASS_METER_V2 for endpoints 0-4 but COMMAND_CLASS_METER_V0 for endpoints 5-6. Shouldn’t they all have the same version? If so, that would explain why two of the outlets are not showing power metering.
My testing shows that SWITCH_BINARY does nothing in endpoints 5-6. My feeling is that they should not be there since there are only five items (strip and 4 outlets) which would need COMMAND_CLASS_SWITCH_BINARY_V1.
If my assumptions from my analysis are correct, then endpoints 5-6 should be have COMMAND_CLASS_METER_V0 changed to COMMAND_CLASS_METER_V2 and COMMAND_CLASS_SWITCH_BINARY_V1 removed. If you agree, how do I submit this change? I am logged into opensmarthouse but I don’t see the edit option like in the documentation. Also, how long does it take for the change to apply?
If the above solves the problem of the missing metering for switched outlets 3-4, then the only remaining issue will be how the channels are named. My assumption is that the always on outlets should be named 1-2 and the four switched outlets should be named 3-6. This would match the order of the outlets on the strip. As of today:
meter_kwh1-2 and meter_watts1-2 report for the always on outlets
meter_kwh3-6 and meter_watts3-6 report for the switched outlets (once 5-6 are updated)
switch_binary1-4 controls the four switched outlets.
I appreciate the help and direction. I look forward to trying help maintain the database. My next task will be to diagnose the issue with a door sensor that is not triggering the alarm channel.
It would be helpful to see the xml file in the zwave directory of userdata. That contains information from your device. It would be interesting to see if that device only recognizes 4 Endpoints. firmware updates sometimes add new features to devices.
It was in my original post. Here it is again. I have tried looking to see if there is a firmware update but I have yet to find it. network_f67d3453__node_27.xml (24.1 KB)
Here is where a newbie like me needs to be careful about suggesting changes to channel names. Page 9 of the manual attached to the database entry labels the channels as they are now. (assuming channels 5-6 get updated). Maybe just the labels could be updated because to make it clear that one shouldn’t assume that channel 1 for measurement does not correspond to channel 1 for switch.
Based on your testing, I have removed these Channels.
The Command Classes are not manually entered but read directly from the device, so these should be correct.
This is how it is described in the manual, where the switches are 1-4 and the meters are 1-6. This is confusing though, since switch_binary1 is not the same outlet as meter_kwh1 and meter_watts1.
Outlet
Switch
Meter
1
N/A
meter_kwh1 and meter_watts1
2
N/A
meter_kwh2 and meter_watts2
3
switch_binary1
meter_kwh3 and meter_watts3
4
switch_binary2
meter_kwh4 and meter_watts4
5
switch_binary3
meter_kwh5 and meter_watts5
6
switch_binary4
meter_kwh6 and meter_watts6
It could also be confusing for some people if the Channels were renamed though, especially if they have read/understood the manual or have been using the device with the current Channels names. This device has been setup like this for about 3 years, so changing things now is probably more disruptive. I suggest adding an explanation to the Device Overview and/or Usage Information.
After reading the manual and considering how long the device has been active, I agree with keeping the numbering the same.
Regarding the command classes, I have some more info which might change your thoughts. This device only reports endpoints 0-4. Endpoints 5-6 were manually added so maybe they were mistakenly entered as COMMAND_CLASS_METER_V0 instead of COMMAND_CLASS_METER_V2. I’m having a hard time understanding why two of the same type of outlets of the same device would have different command classes.
I really appreciate your attention to this matter. It has been a great way for me to learn more.
As a side topic, do your ZD2102 report alarm? Mine do not. We should probably move that to a private conversation.
Ah… usually, this information is added though importing the XML that the binding generates after interrogating the device. If these were added manually, then absolutely they could have been entered with an incorrect or missing version of the Command Class. However, some of the information in the device database is just informational and is not used by the binding. The versions of the Command Classes is one example. Your XML shows 5 instances of COMMAND__CLASS__METER and all are version 2. The devices are interrogated for this information during initialization. If you were to update the database with your XML, it would update the version for at least one of them, but there is no need for that. For completeness, I’ve updated them to version 2.
I’m glad I can help and that you are learning! When you get more proficient, subscribe to the “zwave” and “z-wave” tags to see what people are reporting as they come in. Hopefully, you can help solve them!
