3.0.1.130 Stable now ?

[sooty]

Unfortunately these devices use multichannel and there is information that is saved in the device when you add it to the network. Since I am not aware of any class changes (class changes require the device to be removed and re-added), you should be able to delete all the devices for the node then just do an import. I know that still messes up your events but the node ID does not change.

I have the latest dimmer 2 on a system here and from all my testing it looks ok.

I would remove association 2 and 4 if they are added as that is for older controllers that only accept BASIC commands for control. HS can handle the multilevel command class ok. The basic command could cause the wrong devices to be modified. That is another reason I normally suggest to remove and re-add the device as HS will set the proper associations for you. But in your case you may have older associations set and removing the devices will not remove existing associations.

OK After some playing around I have one of these dimmers sort of working.
Can you confirm that the below 2 items are not an issue on your system?

Issue 1:
It won't let me set an association from Group 3 to HS Endpoint 1.
The status message says that it has completed successfully but repeated updating and audits show that it hasn't stuck.
It will let me set Groups 2, 4 and 5 to either Endpoint.

Issue 2:
With this association config:
G1 > HS (No EP)
G2 > HS (EP 1)
G3 > HS (No EP) because it won't let me set it to EP 1
G4 > HS (EP 2)
G5 > HS (EP 2)

everything appears to work with the exception that if the directly connected light (S1) is operated in this order:

Set level to 50% from HS
Switch of at unit with single click.
Switch back on at unit with single click.

In this scenario HS should show the status of Switch_Multilevel 1 as 50% but it shows it as ON. If it is polled it will update to the correct value.

I'm thinking this behaviour is caused because it won't let me set an association from G3 to EP1 which would sent the multilevel status of S1 however if the light is dimmed from say 50% to 60% at the switch it does update the status correctly in HS. The issue is that basically a single click on the switch from an off state will set the unit to it's last level but this is always shown in HS as ON unless its either polled or adjusted to another level at the switch.

Note:
With Z-Wave plug-in 3.0.1.143 issue 2 does not exist and the HS device correctly updates to the "last level" value.

Paul..

[sooty]

Update...

After several hours of pi**ing about with 3.0.1.150 and a Fibaro FGD-212 I found that the only association configuration that works properly for all child devices is:

Group 1 > HS (No Endpoint)
Group 1 > HS (Endpoint 1)

All other groups removed.

The default way that HS sets it up causes multiple processing of values, so for example if groups 1, 2 and 3 are all associated to HS, then it will process 3 lots of commands (one for each group) and I found that this was causing devices to appear as though they were not getting the correct value.

e.g.
Group 1 will send a SWITCH_MULTILEVEL command to set the level to say 60% but group 2 will then send a BASIC command that sets the device to ON giving the impression that the incorrect value was being set.

So, the thing to watch if updating to 3.0.1.150 is the way in which the plug-in incorrectly sets the associations.

Paul..

[johnny_holden]

I've at to revert back to 3.0.1.130 as thermostat and Stella Z actuator setpoint values didnt pass to the devices for heating. Bug issued.

[budejake]

I've at to revert back to 3.0.1.130 as thermostat and Stella Z actuator setpoint values didnt pass to the devices for heating. Bug issued.

I've just come on to write something very similar!

My z-wave Popp thermostat (similar to Danfloss LC13) haven't changed setpoint since installing .150. They read temperature from devices fine and if you manually temp on device it reflects in homeseer but not setpoint changes are passed to the device.

The log says it's been added to the wakeup queue but it never updates. (I've tried removing and re-adding)

[stefxx]

I have Qubino sending us some samples. This will be ideal as we can test going forward to avoid this type of issue. It will be easy to fix once we have the actual device. I suspect some changes that we made for the Fibaro micro switches have something to with this.

Hi Rich, I know you are a busy man, but can you at least confirm if you ever received the samples from Qubino? If not, I would like to chase them again so you could get some... it has been almost two months ago that Goap promised me to send it to you.

We have the following open issues with Qubino devices:
- On many devices, Watts are not reported anymore
- On/Off sets either Watts or kWh to 0/255 value
- Rescan or include device reports "Could not get the METER TYPE from node 12, so devices for the METER class will not be created or rebuilt."
- Reset of accumulated values do not hold (https://forums.homeseer.com/showpost...&postcount=139)

Thanks!

[whippersnapper]

rjh, any update from Aeotec regarding their Nano Dimmers and scene activations?

Odd timing, we were testing this today. And yes, the dimmers do not work and the issue is with the dimmer. We have sent network traces to Aoetec and are waiting to hear back. We are sending the config commands to the dimmer, but the dimmer ignores the scene activation.

[stefxx]

All recent betas output "NodeChangeEvent_Real CC" debug information, which is quite annoying on Linux. Especially because of the rate this information is shown. Can that be removed again, or made optional? Thanks!!

NodeChangeEvent_Real CC: COMMAND_CLASS_BATTERY
Instance: 0
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 0
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 0
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 4
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 4
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 3
NodeChangeEvent_Real CC: COMMAND_CLASS_SENSOR_MULTILEVEL_V2
Instance: 3
NodeChangeEvent_Real CC: COMMAND_CLASS_METER_V3
Instance: 2
NodeChangeEvent_Real CC: COMMAND_CLASS_METER_V3
Instance: 2

[stefxx]

Hi Rich, I know you are a busy man, but can you at least confirm if you ever received the samples from Qubino? If not, I would like to chase them again so you could get some... it has been almost two months ago that Goap promised me to send it to you.

We have the following open issues with Qubino devices:
- On many devices, Watts are not reported anymore
- On/Off sets either Watts or kWh to 0/255 value
- Rescan or include device reports "Could not get the METER TYPE from node 12, so devices for the METER class will not be created or rebuilt."
- Reset of accumulated values do not hold (https://forums.homeseer.com/showpost...&postcount=139)

Thanks!

Come on Rich, this is getting annoying. I asked several times over the last few months if you received the samples (for which I tried to help with Goap support), but a simple yes or no is too much?

And we are now at beta 162, still no change in the behavior of the Qubino devices

[rjh]

Sorry, I never saw those requests, you can always email me, or better, yet send a request to support@homeseer.com and I will have the support guys keep on the issue as that opens a ticket.

To answer your question, no, I never received them. I assume you tested with the very latest plugin? There have been a few changes that may help with the issues.

Come on Rich, this is getting annoying. I asked several times over the last few months if you received the samples (for which I tried to help with Goap support), but a simple yes or no is too much?

And we are now at beta 162, still no change in the behavior of the Qubino devices

[stefxx]

Sorry, I never saw those requests, you can always email me, or better, yet send a request to support@homeseer.com and I will have the support guys keep on the issue as that opens a ticket.

To answer your question, no, I never received them. I assume you tested with the very latest plugin? There have been a few changes that may help with the issues.

Thanks. I will reach out to Qubino again to see if they are still willing to send you the samples.

And yes, I tested with build 162 and all the issues are still there. I will open a ticket for this.

[stefxx]

All recent betas output "NodeChangeEvent_Real CC" debug information, which is quite annoying on Linux. Especially because of the rate this information is shown. Can that be removed again, or made optional? Thanks!!

Thanks! Fixed in 162

[rjh]

I have not, I will ping them again.

rjh, any update from Aeotec regarding their Nano Dimmers and scene activations?