[Bug]: Using Aeotec ZWA024, luminance is reported with wrong value

[paifre]

Checklist

• [X] I am not using Home Assistant. Or: a developer has told me to come here.
• [X] I have checked the troubleshooting section and my problem is not described there.
• [X] I have read the changelog and my problem is not mentioned there.

Deploy method

Docker

Zwavejs2Mqtt version

6.4.0

ZwaveJS version

8.11.0

Describe the bug

Aeotec ZWA024-C (EU) firmware 1.1 + Zstick7 + Docker zwavejs2mqtt image on Pi0w The differents thresholds parameters are configured to reports high/low on temperature, humidity, luminance not UV To qualify the problem, multisensor installed in a 100% dark room with a remote controlled light. Data are also captured using MQTT Explorer. While displaying a luminance of 0 Lux, at some point of time, Lux value jumps to 748 Lux. Each time this problem has been seen, same value: 748 or 749 Lux, mostly 748. Turning the light, correct 45 Lux was reported initially then keeping the light on, value bumped from 45 to 789 (748+45=793 ??) see screenshot from MQTT explorer Clicking on refresh, value to reported as expected either 0 if in dark or 45-48 if light's on Motion detection is reported correctly. Values from other sensors (Temp/Hum) are corrects compairing those values with other devices installed nearby.

Few minutes ago, doing a manual refresh of multisensor values: luminance suddendly dropped to 105. (see screenshot 2022-01-16_13-16-17.jpg) So I will deploy an Gen5 sensor to compare values, but anyway right now I'm not able to know who's wrong : silabs zwave api layer, zwavejs api layer, frontend layer or hardware layer. Need some guidance here please.

Another small detail : Having done three re-interviews, the second interview reported value for Configuration to "vundefined" instead of "v4". Other two, reported correctly v4

zwavejs_2022-01-16.log zwavejs2mqtt_2022-01-16.log

_I've tried to run in debug mode, but strange message in debug window ? Error: Not Found at /usr/src/app/server/app.js:934:17 at Layer.handle [as handle_request] (/usr/src/app/node_modules/express/lib/router/layer.js:95:5) at trim_prefix (/usr/src/app/node_modules/express/lib/router/index.js:317:13) at /usr/src/app/node_modules/express/lib/router/index.js:284:7 at Function.process_params (/usr/src/app/node_modules/express/lib/router/index.js:335:12) at next (/usr/src/app/node_modules/express/lib/router/index.js:275:10) at session (/usr/src/app/node_modules/express-session/index.js:479:7) at Layer.handle [as handle_request] (/usr/src/app/node_modules/express/lib/router/layer.js:95:5) at trim_prefix (/usr/src/app/node_modules/express/lib/router/index.js:317:13) at /usr/src/app/nodemodules/express/lib/router/index.js:284:7

I'll keep you posted once deploying a second luminance sensor to baseling comparison.

To Reproduce

See Bug description

Expected behavior

Correct value reported

Additional context

No response

[paifre]

Luminance reported by Multisensor 6 is close to value reported by Gen7. Gen7 sensors seems ok.

[robertsLando]

@paifre That error could just be a 404 so nothing related to the sensor, about your error I think it could be a device issue. cc @AlCalzone

[AlCalzone]

Yep, zwave-js (or any part of the stack really) doesn't do any correction for values. They are passed through 1:1 from the device. Some allow calibration, but if that doesn't help, it's a case for the Aeotec support.

[paifre]

Tomorrow, I'm working remotely. So I will connect with Simplicity studio as I was able to install and configure it yesterday evening. Il will let things run. I should know what is pushed by the multilevel sensor.

[paifre]

Hello, as you've correctly highlighted, I confirm that PC Controller reports also 750Lux I can see a nice 02 EE at the end...

[paifre]

Hello, some update.... for resolution record Resetting (hw reset) the sensor, problem disappeared. This was a software stack issue on Aeotec side, could bad configuration being pushed generating such weird behavior (remember value was correct using a refresh, wrong when pushed at device initiative...). Also on the other hand, I've not done exactly same setup. Outlimit related parameters are not used this time, only report on threshold. Could also be something related to the fact that the device file was not existing initially for the EU version and has been deployed on-the-fly without excluding then including. Potentially secure what is supposed to be default values entering them manually Rule of thumbs: remove, reset, include again !