Possible issue with Sengled Element Classic Bulb driver

corerootedxb · October 28, 2019, 12:16pm

My Sengled Element Classic BR30 bulbs seem to have an issue when using the Sengled Element Classic Driver. When issuing a dim command, they don't turn on and set the level (if off). I have to issue a command to turn them on first and then set the dim command. If I switch them to the Generic Zigbee CT Bulb (dev)/Generic Zigbee Bulb, they turn on by just issuing a dim command. This seems to only affect the BR30s.

Logs (Generic Zigbee CT Bulb (dev)):
dev:12182019-10-28 08:11:35.519 am infoOffice Desk Light 3 was turned on

dev:12182019-10-28 08:11:35.516 am debugevt- rawValue:1, value: on, descT: Office Desk Light 3 was turned on

dev:12182019-10-28 08:11:35.512 am infoOffice Desk Light 3 is 25%

dev:12182019-10-28 08:11:35.509 am debugevt- rawValue:63, value: 25, descT: Office Desk Light 3 is 25%

dev:12182019-10-28 08:11:35.495 am debugParse: read attr - raw: 78010100060A00001001, dni: 7801, endpoint: 01, cluster: 0006, size: 0A, attrId: 0000, encoding: 10, command: 01, value: 01

dev:12182019-10-28 08:11:35.488 am debugParse: read attr - raw: 78010100080A0000203F, dni: 7801, endpoint: 01, cluster: 0008, size: 0A, attrId: 0000, encoding: 20, command: 01, value: 3F

dev:12182019-10-28 08:11:34.173 am debugParse: catchall: 0104 0008 01 01 0040 00 7801 00 00 0000 0B 01 0400

dev:12182019-10-28 08:11:33.655 am infoOffice Desk Light 3 was turned off

dev:12182019-10-28 08:11:33.652 am debugevt- rawValue:0, value: off, descT: Office Desk Light 3 was turned off

dev:12182019-10-28 08:11:33.641 am debugParse: read attr - raw: 78010100060A00001000, dni: 7801, endpoint: 01, cluster: 0006, size: 0A, attrId: 0000, encoding: 10, command: 01, value: 00

dev:12182019-10-28 08:11:33.638 am debugParse: catchall: 0104 0006 01 01 0040 00 7801 00 00 0000 0B 01 0000

dev:12182019-10-28 08:11:30.753 am infoOffice Desk Light 3 was set to 25%

dev:12182019-10-28 08:11:30.750 am debugevt- rawValue:63, value: 25, descT: Office Desk Light 3 was set to 25%

dev:12182019-10-28 08:11:30.739 am debugParse: read attr - raw: 78010100080A0000203F, dni: 7801, endpoint: 01, cluster: 0008, size: 0A, attrId: 0000, encoding: 20, command: 01, value: 3F

dev:12182019-10-28 08:11:30.456 am debugParse: catchall: 0104 0008 01 01 0040 00 7801 00 00 0000 0B 01 0400

dev:12182019-10-28 08:11:21.712 am infoOffice Desk Light 3 was turned on

dev:12182019-10-28 08:11:21.709 am debugevt- rawValue:1, value: on, descT: Office Desk Light 3 was turned on

dev:12182019-10-28 08:11:21.689 am debugParse: read attr - raw: 78010100060A00001001, dni: 7801, endpoint: 01, cluster: 0006, size: 0A, attrId: 0000, encoding: 10, command: 01, value: 01

dev:12182019-10-28 08:11:21.567 am debugParse: catchall: 0104 0006 01 01 0040 00 7801 00 00 0000 0B 01 0100

dev:12182019-10-28 08:11:21.070 am infoOffice Desk Light 3 was turned off

dev:12182019-10-28 08:11:21.065 am debugevt- rawValue:0, value: off, descT: Office Desk Light 3 was turned off

dev:12182019-10-28 08:11:21.045 am debugParse: read attr - raw: 78010100060A00001000, dni: 7801, endpoint: 01, cluster: 0006, size: 0A, attrId: 0000, encoding: 10, command: 01, value: 00

dev:12182019-10-28 08:11:19.789 am debugParse: catchall: 0104 0006 01 01 0040 00 7801 00 00 0000 0B 01 0000

Logs (Sengled Classic Bulb driver):
dev:12182019-10-28 08:13:43.352 am debugdescMap:[raw:78010100080800002000, dni:7801, endpoint:01, cluster:0008, size:08, attrId:0000, encoding:20, command:0A, value:00, clusterInt:8, attrInt:0]

dev:12182019-10-28 08:13:42.339 am infoOffice Desk Light 3 was turned off

dev:12182019-10-28 08:13:42.330 am debugdescMap:[raw:78010100060800001000, dni:7801, endpoint:01, cluster:0006, size:08, attrId:0000, encoding:10, command:0A, value:00, clusterInt:6, attrInt:0]

dev:12182019-10-28 08:13:42.090 am debugdescMap:[raw:catchall: 0104 0008 01 01 0040 00 7801 00 00 0000 0B 01 0400, profileId:0104, clusterId:0008, clusterInt:8, sourceEndpoint:01, destinationEndpoint:01, options:0040, messageType:00, dni:7801, isClusterSpecific:false, isManufacturerSpecific:false, manufacturerId:0000, command:0B, direction:01, data:[04, 00]]

dev:12182019-10-28 08:13:42.006 am tracesetLevel(0, 0) called...

dev:12182019-10-28 08:13:42.002 am traceoff() called...

dev:12182019-10-28 08:13:38.413 am infoOffice Desk Light 3 level was set to 50%

dev:12182019-10-28 08:13:38.391 am debugdescMap:[raw:7801010008080000207F, dni:7801, endpoint:01, cluster:0008, size:08, attrId:0000, encoding:20, command:0A, value:7F, clusterInt:8, attrInt:0]

dev:12182019-10-28 08:13:37.281 am tracesetLevel(50, 0) called...

dev:12182019-10-28 08:13:32.108 am infoOffice Desk Light 3 is on

dev:12182019-10-28 08:13:32.095 am debugdescMap:[raw:78010100060800001001, dni:7801, endpoint:01, cluster:0006, size:08, attrId:0000, encoding:10, command:0A, value:01, clusterInt:6, attrInt:0]

dev:12182019-10-28 08:13:32.087 am debugdescMap:[raw:catchall: 0104 0008 01 01 0040 00 7801 00 00 0000 0B 01 0400, profileId:0104, clusterId:0008, clusterInt:8, sourceEndpoint:01, destinationEndpoint:01, options:0040, messageType:00, dni:7801, isClusterSpecific:false, isManufacturerSpecific:false, manufacturerId:0000, command:0B, direction:01, data:[04, 00]]

dev:12182019-10-28 08:13:31.924 am traceon() called...

dev:12182019-10-28 08:13:25.914 am infoOffice Desk Light 3 was turned on

dev:12182019-10-28 08:13:25.895 am debugdescMap:[raw:7801010008080000203F, dni:7801, endpoint:01, cluster:0008, size:08, attrId:0000, encoding:20, command:0A, value:3F, clusterInt:8, attrInt:0]

dev:12182019-10-28 08:13:24.779 am tracesetLevel(25, 0) called...

dev:12182019-10-28 08:13:20.915 am debugdescMap:[raw:78010100080800002000, dni:7801, endpoint:01, cluster:0008, size:08, attrId:0000, encoding:20, command:0A, value:00, clusterInt:8, attrInt:0]

dev:12182019-10-28 08:13:19.893 am infoOffice Desk Light 3 was turned off

dev:12182019-10-28 08:13:19.884 am debugdescMap:[raw:78010100060800001000, dni:7801, endpoint:01, cluster:0006, size:08, attrId:0000, encoding:10, command:0A, value:00, clusterInt:6, attrInt:0]

dev:12182019-10-28 08:13:19.785 am debugdescMap:[raw:catchall: 0104 0008 01 01 0040 00 7801 00 00 0000 0B 01 0400, profileId:0104, clusterId:0008, clusterInt:8, sourceEndpoint:01, destinationEndpoint:01, options:0040, messageType:00, dni:7801, isClusterSpecific:false, isManufacturerSpecific:false, manufacturerId:0000, command:0B, direction:01, data:[04, 00]]

dev:12182019-10-28 08:13:19.678 am tracesetLevel(0, 0) called...

dev:12182019-10-28 08:13:19.674 am traceoff() called...

Ryan780 · October 28, 2019, 2:03pm

Do you have Level Prestaging Turned on in the driver?

corerootedxb · October 28, 2019, 2:46pm

Nope. I thought about that as well.

Ryan780 · October 28, 2019, 2:52pm

I have a bunch of Sengled Element Classic bulbs using that driver and I'm not seeing the same thing,
Are you putting a 0 in for fade time?

I just issued a setLevel of 100 to a Sengled that was currently off. The bulb turned on to 100%.

Ryan780 · October 28, 2019, 2:56pm

What happens with the Generic Zigbee Bulb driver? These are only dimmable so i'm confused why you'd even try the CT bulb driver.

corerootedxb · October 28, 2019, 2:58pm

Nope. Just a simple rule. It's weird as it only happens to my BR30s. My A19s are fine with the Sengled driver.

Ryan780 · October 28, 2019, 2:58pm

Why are you setting CT and level. That won't work for a dimmable bulb. The command isn't going to be recognized by the Sengled Driver because it doesn't support CT. Only level.

corerootedxb · October 28, 2019, 2:59pm

The Generic drivers work as expected. I tried the CT driver just as a test to see if it was something specific to the Sengled driver or if either the Generic Zigbee or Zigbee CT driver would show the same results.

Ryan780 · October 28, 2019, 2:59pm

But the command isn't support by the Sengled driver.

corerootedxb · October 28, 2019, 3:00pm

Read the rule again. The device is Office Desk Light 3 and it's only dim. My others (3 Sengled color bulbs and 1 Sengled strip) all support CT.

Ryan780 · October 28, 2019, 3:02pm

What happens from the edit device page though? Can you issue a setLevel there and get it to change to the correct level without issuing the On command?

corerootedxb · October 28, 2019, 3:02pm

For CT? This I know. I tried and showed results for both the Generic Zigbee CT and Generic Zigbee to highlight where the Sengled driver was failing to turn on and set level on a BR30 compared to the generic drivers.

corerootedxb · October 28, 2019, 3:03pm

Nope. It fails on the Sengled driver and works with the generic drivers. But only on the BR30s. I have 4 of them and they all exhibit the same quirk.

Ryan780 · October 28, 2019, 3:06pm

That's really strange indeed! Have you tried changing the reporting interval to see if that has any affect? It's possible the firmware is different and doesn't support the same reporting as the E26 but I can't imagine why it would.

corerootedxb · October 28, 2019, 3:10pm

I haven't gone that far yet. I might play with it a bit more later just to see if that makes a difference. Keeping them on the generic driver doesn't bother me that much and I rarely use the color or level prestaging.

I thought about the firmware being different, which why wouldn't it be, right? LOL

Ryan780 · October 28, 2019, 3:18pm

Have you tried with the Generic Zigbee Bulb driver (not the CT one)? I see the Sengled and the CT driver but I'm wondering why you wouldn't try the regular bulb driver next.

corerootedxb · October 28, 2019, 3:34pm

I did. I just didn't include the results in the log as they should be the same for the generic zigbee and generic zigbee ct driver.

Ryan780 · October 28, 2019, 3:38pm

Oh...ok. Well, using the Generic regular instead of the CT will at least prevent you from accidentally issuing a CT command to them.

The other thing I find odd about what you posted...only in one section of your rule is the current mode bolded (that's why my eyes centered on it. I honestly thought those were all different modes. You are, after all, Mr. Mode. )

See what i mean?

Wonder if that's a UI bug? I wonder if anyone's ever used CT by mode and dimmer by mode in the same rule before.

corerootedxb · October 28, 2019, 3:43pm

I hadn't even noticed that! The dim by mode should be bolded as well. Hmmmm, two bugs in one day? Yay me? What's funny is that I actually do CT, Color and dim commands in the same rule all the time and always by mode and I've never noticed that before. Now I have to go see the rest of my rules to see if it's an actual bug! lol

[Edit]: Yup, looks like a UI bug:

Ryan780 · October 28, 2019, 3:48pm

I used to do testing for medical diagnostics lab software. So, this kinda stuff is in my blood. I always won our internal "bug hunts".