Hello everyone,
I’m looking for a way to implement additional button functionality on my AQARA Shutter Switch H2 (EU version, model DS-K02D) connected to Samsung SmartThings. Currently the switch is connected via Matter to Aeotec V4 hub and shutter control works.
If the EU version doesn’t differ too much from the US version, it’s a matter of minutes to add support to my driver.
I think it is just a different hardware, mainly the shape and voltage differs, function should be the same, shutters open, close, stop, go to specific position 0-100% and two additional buttons, non visible with standard, matter window covering, driver.
But I did not find the shutter switch or any xx-K02D switch.
If I replace the current one with standard 4 button I will lose the shutter control function, I guess?
Yes. They are completely different devices.
Give me a minute to do more research.
Alright.
Install the Aqara H2 Shutter Switch driver from my channel. Then delete the device from SmartThings and re-add it.
It’s been a while since I wrote the driver and the plan was to merge it with my Aqara H2 Wall Switch driver, but if I remember correctly, there was an issue with the two buttons on the right-hand side.
I had to use aqara app to switch the device from matter to zigbee mode, than delete and add worked.
Your driver looks great, lots of nice features, I am curious about how the power meter works, is it just an estimation? Unfortunately the only thing that does not work are the additional buttons.
Thanks! The energy meter is not just an estimation. The H2 exposes its accumulated energy consumption through the standard Zigbee Simple Metering cluster, and the driver reads that value directly from the device. I did experiment with instantaneous power in earlier versions, but couldn’t verify a reliable live power value, so I deliberately removed that from the production version rather than showing something misleading.
The additional buttons are unfortunately a much more complicated story. We’ve actually been working on those for quite a long time with another tester, who provided literally thousands of lines of logs and tested quite a few specially prepared driver versions.
The interesting part is that we know the buttons themselves work. When pressed, the H2 changes the corresponding internal Zigbee attribute exactly as expected. We also successfully tested the necessary bindings, reporting configuration and Aqara’s multi-click setting. I even tried the slightly different approaches used by ZHA and Zigbee2MQTT.
But the actual button press messages never arrive at the SmartThings Edge driver.
That is why additional Edge driver logs unfortunately won’t tell us anything new at this point. The problem appears to be below the driver layer: either the H2 firmware is not transmitting those unsolicited button reports when it is connected to a SmartThings coordinator, or the SmartThings Zigbee stack is receiving them but not passing them on to Edge.
To determine which of those two possibilities it is, we would need either Zigbee radio/packet captures or lower-level hub/Zigbee stack logs. If the H2 sends the packet over the air but it never reaches the Edge driver, that would be very strong evidence of a SmartThings Zigbee stack issue. If it doesn’t transmit the packet at all, we’d be looking more toward an Aqara firmware/compatibility issue.
So I haven’t given up on the buttons at all. We have just reached the point where changing the driver blindly isn’t useful anymore - the evidence says the remaining problem is one layer deeper.
Unfortunately, there’s no developer support here anymore, and therefore we can’t get someone to look at the hub logs anymore…
That is unfortunate… Thank you very much for all the effort and help. If there is anything else I can do to help you with this please let me know.
There is actually a new SmartThings hub firmware on the way - the 0.62.x release is currently in beta testing.
There is nothing in the release notes that specifically mentions this Zigbee issue, so I don’t want to raise expectations too much, but considering where the problem seems to be, a hub firmware update is one of the few things that could potentially change the situation.
So for now, fingers crossed. And thanks again for offering to help - if the new firmware doesn’t make any difference, I may come back to you for another test.