Electrical measurement: Power Factor value

[dduransseau]

Hi,

By discussing issue #19705 I figure out that power factor might be wrongly interpreted by Z2M (front-end).
Following Zigbee Cluster Library p348 the power factor is represented as

Contains the single phase or PhaseA, Power Factor ratio in 1/100ths.

So it's divided by 100 by zigbee-herdsman-converter to have a power factor between 0 and 1, which is the correct behavior. However the front-end display it as a percentage which is wrong but a common usage.
I see 3 solution to display the correct value:

• Avoid to display a percentage in front end and just display the value between 0 and 1 with two decimal
• Multiply the value by 100 and let the percentage as it is
• Avoid to divide by 100 the power factor in zigbee-herdsman-converter and let the value as it is

I'm not really convinced by last solution since it broke a bit the zigbee interpretation but that would work.

[Koenkk]

I think the correct fix is to remove the % unit

[dduransseau]

Agreed that would better match the zigbee definition. However that might imply a regression for tuya device where power_factor is provided as raw value so between -100 and +100, at least for one I have (_TZE200_bkkmqmyo).

[dduransseau]

But how we can identify if the exposed power_factor metric come from tuya device or standard zigbee device ?

[Koenkk]

In the exposes of the tuya device, use e.power_factor().withUnit('%') and remove withUnit('%') from exposes.ts

[dduransseau]

Ok, so yes by updating tuja file. I'll have a look at this and PR.

[dduransseau]

PR for this #6847

[Patricen1]

Hello,

Just updated to 1.39.1 and issue looks to be still there.
On the attached screenshot, we should have apparent power @392VA, power @392*0.84 = 329W and reactive power = 392-329 =63W

Maybe I'm wrong as this concern is about BITUO TECHNIK SPM01-U01