Battery reporting 200%

[andromeda2537]

Hi @mike.maxwell and Hubitat support team,

I ran into this post ->

Battery reporting 200% (Linkind devices)

Replaced the CR2032 battery on a contact sensor after it failed to report for a few days, and it now shows 200% battery. The battery reads at 3.25V on my voltmeter, a bit on the high side, but not 100% higher. Could someone on staff kindly check and or adjust the battery percent calculation for the Generic Zigbee Contact Sensor (no temp). Adding the actual volts reported by the device as part of the Events Description Message would be ideal. Hub: C-4 2.3.3.140 Using system driver "Generic Z…

while searching for a solution for my ORVIBO Zigbee contact and motion sensors reporting battery level at double of the real value.

Based on that thread It appears that there is a workaround fix for it by adding the fingerprint to an exception list in the « Generic Zigbee (no temp) » driver.

If so, could you please add these two device as well?

Here's the fingerprint for the Zigbee contact sensor

• endpointId: 01
• application: 02
• inClusters: 0000,0003,0500,FFFF,0001
• manufacturer: ORVIBO
• model: e70f96b3773a4c9283c6862dbafb6a99
• outClusters: 0000,0004,0003,0005,0001
• softwareBuild: v0.1.3

And the fingerprint for the Zigbee motion sensor;

• endpointId: 01
• application: 02
• inClusters: 0000,0003,0500,FFFF,0001
• manufacturer: ORVIBO
• model: 895a2d80097f4ae2b2d40500d5e03dcc
• outClusters: 0000,0004,0003,0005,0001
• softwareBuild: v0.1.3

Advance thanks,
Sylvain.

[rwclements228]

Accurate prediction of battery life on a lithium battery such as the CR2032 is nearly impossible. The voltage discharge curve of an alkaline battery occurs gradually. With a lithium battery, the voltage remains nearly constant for about 95% of the battery life and then plunges off a cliff at the end. If your lithium battery was overcharged as evidenced by the higher than normal voltage, then the predictive curve is just estimating that it could take twice as long as normal to reach the cliff. That probably is not the case, but who knows?

I have had devices continue to function for weeks after the battery life indicator reads 0%. However, if it is a critical sensor, I will replace the battery long before it reaches 0.

[mike.maxwell]

Updated!
These will be in an upcoming 2.3.8 build.

[andromeda2537]

Excellent news… looking forward to try it out… thanks @mike.maxwell

[andromeda2537]

Latest update did the trick… all of my devices are now reporting proper battery percentage… Thanks again @mike.maxwell …

[chad.andrews]

This is also an issue on the Third Reality Contact Sensors, @mike.maxwell , could the same fix be applied?

• endpointId: 01
• firmwareMT: 1233-D3A2-00000030
• inClusters: 0000,0001,0500
• manufacturer: Third Reality, Inc
• model: 3RDS17BZ
• outClusters: 0019
• softwareBuild: 00000030

[mike.maxwell]

Have you tried updating the firmware on these?

[chad.andrews]

Yep, that's the new firmware. I have one on the new and the rest on the old and the all act the same.

This is the old firmware...

[mike.maxwell]

and all of these are using the Generic Zigbee Contact Sensor (no temp) driver correct?

[chad.andrews]

Yes sir!

[mike.maxwell]

I'm stumped then, by all means the driver should be applying logic to provide the 2x divisor to the battery report, this has been in place for this driver since platform 2.3.4
I'll have to dig one out and do some more testing on it...

[j715]

For what it’s worth, once I changed mine to the generic driver, without the no temp, I have not had a 200 reading.

[andromeda2537]

Hi @chad.andrews & @mike.maxwell

Fwiw, I have two of those sensor as well which I recently added in early March and both are reporting proper battery levels using the « Generic Zigbee contact sensor ( no temp) » driver… the only difference I noticed between Chad’s devices and mine is the f/w is at 23 while his is at 21

• endpointId: 01
• application: 23
• firmwareMT: 1233-D3A2-00000023
• manufacturer: Third Reality, Inc
• model: 3RDS17BZ
• softwareBuild: 00000023

Regards,
Sylvain

[chad.andrews]

@andromeda2537, 21 or 23 is the old firmware, the new one is on 30, scroll up and you can see the new one I screened shotted.

@mike.maxwell , See this is when it flip flops... When ever those circled values comes into the mix.
Not sure if it will provide a clue.

[mike.maxwell]

attrId 0020 is battery voltage, in the Generic Zigbee Contact Sensor (no temp) this is skipped altogether if the manufacturer is Third Reality, Inc, yet in your case it's being evaluated?
I'm not sure how this is possible if your version of the driver is using the current version code.

[chad.andrews]

To make it even more strange, some are right and some are wrong...I updated all the rest of the sensors firmware today. They are all like this now...

Have a look, some are right and some are double. Same device, same firmware, same driver.

[mike.maxwell]

@chad.andrews

To make it even more strange

Dumb founded...
When was the last hub reboot?

[chad.andrews]

3 days ago

[chad.andrews]

Maybe something else that could be helpful. Since I upgraded the firmware I haven't had a battery report in 4 days. This is the same for all of them. These devices have great battery life so maybe it only reports when there is a value change? Maybe that's why it's stuck at double the value, nothing to compute yet?
Or should they be checking in every xx hrs? They all work normally regarding the contact open / close.

[BorrisTheCat]

I had the same on a 3.0 device @mike.maxwell and it was because the standard says 200 is 100% due to lithium batteries but it's being reported as 200.

this was on the contact no temp driver