Ah, great. I removed the entity ID from the calc node and entered it into the call service node, and it works.
One thing unrelated to this method is that when I tap the physical switch, the notification goes away. I don’t know whether there’s a switch config parameter that I could set to allow the notification to persist.
These switches have both notification and normal LED colors. The notifications can be cleared after a set amount of time or after you press a button. Setting the LED color and brightness are permanent until you set it to something else.
I believe the script is for notifications only.
See parameters 13, 14, and 15 for LED settings.
Parameter 16 is for notifications only.
https://devices.zwave-js.io/?jumpTo=0x031e:0x0001:0x0001:0.0
Anyone else having an issue with removing the effect? All other commands effects work.
I’m not able to run the script to perform an effect: “off”. It appears it is doing the math wrong and sending a Parameter 16 value of 0 whenever I set effect as “off”.
My “off” works fine, on dimmers and switches. I haven’t tried the fan/light combo switch yet.
My script is working again (and so is my UNRAID server after my old SSD failed and took Home Assistant offline for several days). I’m guessing a Z-Wave JS update fixed something in the meantime. Here are the changes:
March 26, 2021
- Added fields to help users experimenting in the Services Developer Tool.
April 3, 2021
Incorporated @firstof9’s changes:
- Set execution mode to “parallel” to all this script to potentially run on more than one devices simultaneously.
- Implement ZWave JS’s new zwave_js.bulk_set_partial_config_parameters command.
Works awesome, thanks Brian.
Just need to think of how to add multiple conditions for the LED indicator based on the state of various locks. Node-Red is great for stuff like this but I’m trying to stay away.
I too was unable to get “Off” to work when using this with ZWave JS. I have the same problem mentioned above - I can set new effects, but never remove an effect. I am using a LZW31-SN dimmer with the latest version of the script from April.
I enabled the debug logging that’s in the script and this is what appears:
zwave_integration: zwave_js
model: dimmer
color: '0'
level: '4'
duration: '0'
effect: '0'
node_id: 'None'
parameter: '16'
value: '0'
This was working for me on the old zwave integration with an earlier version of the script. 
After some more testing, I’ve learned that calling this service does not clear the effect:
service: zwave_js.bulk_set_partial_config_parameters
target:
entity_id: light.test_light
data:
parameter: '16'
value: 0
… however calling this service does turn the effect off:
service: zwave_js.set_config_parameter
target:
entity_id: light.test_light
data:
parameter: LED Effect Type
value: '0'
Using that information, I’ve worked-around this issue by updating my copy of the script to call set_config_parameter in the spot that is intended to clear the previous effect for zwave_js.
Here it is in case it helps others: Worked around issue preventing effect from clearing in zwave_js when setting effect type to 'Off'. · GitHub
Yeah, I am not sure why it sets it as value: ‘0’
What I am doing is just just using a value obtained from the Inovelli Switch Toolbox to set the manual command for off:
service: zwave_js.bulk_set_partial_config_parameters
target:
entity_id: light.red_series_dimmer
data:
parameter: '16'
value: '16714410'
I don’t recall how we started using the zero, but I can tell you that it worked with previous Zwave integrations that I have used.
Thank you for resolving this! I noticed my effect wasn’t clearing, but I hadn’t had a chance to troubleshoot it yet.
Updated code below:
So i found this script the other day and its really useful, thank you for that! I havent completely reverse engineered it, so this is a lazy question, but what are the possibilities for creating a list of effects to ‘loop’ through?
I imagine its outside of the scope of this script, but that this script could be integral in creating that effect. for example, lets say i turn all leds to pulse yellow when my garage is open, or pulse orange when the front door is unlocked for longer than 5 minutes, or to pulse red if the front or back door was left open. I think it would be neat to have a script that continually rotates through existing ‘alerts’ every 10 seconds or so.
Anyone have any tips on how that might be done?
There’s probably a better way, but I would use one automation to set an input_boolean ‘on’ or ‘off’ for each of the alerts you want to cycle through, and then use a second automation to call the effects script and set an input_select to track where you’re at in the cycle.
The triggers for the second script would need to be:
A condition that at least one of them is ‘on’.
The action would be a series of nested if / elif statements or a choose. I’d start with looking at the trigger to see if an input_boolean changed so you can jump to its position in the loop then, if the time pattern triggered it, looking at where you’re at in the cycle and testing what the next active boolean is, setting the input_select to that option and finally calling the script with the “color” of that alert.
It seems like a lot of flashing lights for me, personally, but I think it’s possible to do.
I got around to implementing the device_id() feature released in 2021.8.0. This allows us to do the following exciting things now.
No longer should you have to manually set the zwave_integration variable or pass the model type variable in the script call!
According to the current Z-Wave integration counts (from Home Assistant Analytics) there are 11,539 total Z-Wave integrations in use. Of those 11.5% of those integrations are using one of the two deprecated integrations.
At some point Home Assistant is likely to remove support for both the OpenZWave and Z-Wave integrations. Which leads me to consider removing all of the extra code that deals with those two integrations.
What are your thoughts?
| Integration | Installations | Percent |
|---|---|---|
| Z-Wave JS | 10,212 | 88.5% |
| Z-Wave (deprecated) | 1,025 | 8.9% |
| OpenZWave (deprecated) | 302 | 2.6% |
| Total | 11,539 |
Oops my bad, that’s weird, I deleted my post to avoid confusion!
Not sure why I keep getting this when I copy your link:
Someone shared an Automation Blueprint they made in this discussion. Apparently Home Assistant likes that. That reminds me Script Blueprints were added recently, I am going to see if I can migrate this script over.
Okay, I’m sure I’m missing something obvious, but how do I add this to my HA?
An up to date version of the inovelli_led script can always be found at this link in my GitHub repository.
I’ve got that, but when I try to copy it into my scripts file or into the script editor, it generates errors and I can’t save.
You can paste the entirety of the YAML code into the Home Assistant script editor, as long as you remove the first line ---.
However, it looks like in doing so the YAML code errors occur due to the formatting gets scrambled. For example:
variables:
debug: false
zwave_integration: "{%- if state_attr(entity_id, 'product_name') is string %}\n\
\ zwave\n{%- elif device_attr(device_id(entity_id), \"model\") is string %}\n\
\ zwave_js\n{%- else %}\n ozw\n{%- endif %}\n"
The solution is to store the script store the script manually in a YAML file. For example I can run both the UI generated script code and YAML stored script code side-by-side by using the following trickery.
homeassistant:
script ui-script: !include scripts.yaml
script: !include_dir_named scripts
Storing the above code in your configuration.yaml file would allow you to save my file as is inside of the /scripts directory at the root of your Home Assistant config. You could literally save my file to /scripts/inovelli_led.yaml and it should work.