We have started publishing OTA updates to the Homey Test App:
Inovelli | Homey
For clarity, the white series is handled by the CSA and Homey has supported updates through that process for a while. They recently released an update for OTA via Zigbee and Z-Wave which need to be added to the app on an individual basis. So far, in the test app, we have the VZM32-SN and VZW32-SN beta firmware updates. I will be adding more down the road, but if anyone has any feedback, please let me know.
All current gen devices with OTAs are now available in the Homey test app.
@EricM_Inovelli — following up on this thread now that current-gen OTAs are live: could the VZM31-SN beta (3.07, or at least 3.05+) be added to the Homey test app manifest? Details below.
Setup
- VZM31-SN on firmware 3.04 (Homey shows it as 0.1b2 / 0.3b4), neutral install
- Switch Type (#22) = 3-Way AUX, with an AUX01 on the traveler (traveler + neutral wiring)
- Homey Pro (2026), firmware 13.4.0-rc.4, mobile app 10.0.0.2028
- Inovelli Homey app: test channel, v1.0.28
Issue I’m chasing
With #22 = 3-Way AUX, every off — app or paddle — fades the load out, the LED bar flashes full-bar blue for a moment, then the switch turns itself back on within a few seconds. Setting #22 = Single Pole eliminates it completely (off stays off, normal operation), so parameter writes are landing and the load side is fine. Wiring is meter-verified (line/load/neutral), and the re-on also persists with the traveler disconnected at the aux end. This looks like the aux false-detection behavior that P124 / Aux Detection Level (3.05+) was added to address.
OTA status on Homey
“Check for device updates” reports up to date at 0.1b2 / 0.3b4 — consistent with app 1.0.27 (Jul 7) bundling 3.04 as the VZM31-SN image. No 3.0x beta in the manifest yet.
Ask
Publish the VZM31-SN 3.07 beta to the Homey test app, same as the VZW31-SN 2.01 beta in 1.0.28. Happy to beta test — I have 11 more Blues and 4 AUX01s going in over the next few weeks on this Homey Pro, so I can give the Homey OTA path and P124 a real workout in 3-way aux configs and report back.
Thanks for the fast iteration on the Homey app lately — three test builds in a month hasn’t gone unnoticed.
@EricM_Inovelli — found what looks like an enum off-by-one in the Homey app’s Switch Type (#22) mapping, confirmed by behavior on a VZM31-SN (FW 3.04, shown as 0.1b2 / 0.3b4). Homey test app v1.0.28, Homey Pro (2026) FW 13.4.0-rc.4.
Setup: VZM31-SN, neutral install, AUX01 on traveler+neutral, wiring meter-verified. Output mode set to on/off.
The bug: The app’s Installation Settings shows three Switch Type options — Single Pole / 3-Way AUX / Full Sine Mode — but firmware #22 has four values (0 Single Pole, 1 3-Way Dumb, 2 3-Way Aux, 3 Full Sine). The driver appears to write by list index, so:
- “3-Way AUX” → writes 1 (Dumb)
- “Full Sine Mode” → writes 2 (Aux)
Evidence: With “3-Way AUX” selected, the master exhibited classic dumb-mode behavior with an aux attached: (a) turning off from app or paddle caused the switch to turn itself back on within seconds (dumb false-detection — P165 territory), and (b) AUX01 presses toggled the master’s reported state in Homey while the load stayed dark, i.e. output routed to the traveler as if driving a dumb 3-way leg. Selecting “Full Sine Mode” instantly produced correct aux behavior — taps on the AUX01 latch the load on/off, states track in Homey, no phantom re-ons.
So the workaround for Homey users with an aux is to select “Full Sine Mode,” which is obviously not discoverable. Looks like a one-line mapping fix in the driver (and ideally add the missing 3-Way Dumb option as a fourth entry).
Secondary ask from my earlier note: publishing the VZM31-SN 3.0x beta (3.07 with P124/P165) to the Homey test app manifest would still be appreciated — happy to beta test. I have 11 more Blues + 4 AUX01s going in on this Homey Pro over the next few weeks and can regression-test the driver fix and OTA path.