[ade7880] phase angles feature and forward/rewerse active energy issue

[eskalator]

Component name

ade7880

Link to component documentation on our website

https://esphome.io/components/sensor/ade7880/

Describe the enhancement

I'm not an active programmer, but i manage to "read" and even edit a little something that already exist. I'm an electrical engineer though and i have some thoughts/requests.

1. I was checking the aforementioned component and it seems to me like phase angle callibration does not work.
   First of all it seems to me that the values- based on datasheet- should be in -383 : 63 range if they are written directly without any conversion into register. In the example the values are 185, 188 , 180. I've checked, using purely resistive load if changing the values does anything to power factor(as it should) but i can't see any change. So my request is if it would be possible to get phase angle values form registers ANGLE0, ANGEL1, ANGLE2 available directly as sensor entities. That would allow to check whether the change in phase calibration value affects the phase angle correctly.
2. Minor formal issue- for me at least- the way of presenting power factor. It is accepted in the industry (and mentioned in the datasheet) to present the power factor as fractions of 1 with +or - sign (depending on the phase current lagging or leading the voltage). In my opinion that way of presentation should be used instead of the percentage used now or it should be explicitly explained in the documentation how to interpret the percentage values.
3. I'm a little lost with reverse/forward active energy entities. They are both calulated in the same way:

this->update_sensor_from_s32_register16_(chan->forward_active_energy, AFWATTHR, [&chan](float val) {
  return chan->forward_active_energy_total += val / 14400.0f;
});
this->update_sensor_from_s32_register16_(chan->reverse_active_energy, AFWATTHR, [&chan](float val) {
  return chan->reverse_active_energy_total += val / 14400.0f;
});

so, if i understand correctly it's the same value only named differently. And when the forward ACTIVE energy is presented in "Wh", the reverse ACTIVE energy is presented in "varh" which is reactive energy unit.

What is the point of this differentiation?
4. Would it be a big problem to facilitate an interface to choose a phase to do harmonics analysis(ade7880 is capable of that) and present as an entity total harmonics for that phase?

Use cases

As for nr 1. it would be useful for calibration check of the current transducers phase error.
Number 2. is purely formal but there should be at least good documentation for the currently used solution.
Number 3. is just wrong or i don't understand something.
Number 4.- it would make the component a poor-mans power quality analyzer.

Anything else?

Sorry for my english- it's not my native language.
I've got one Shelly 3EM with damaged calibration data so i've had to try to calibrate it myself. I can use it for tests.
If i could be of any help please tell me :)

[github-actions[bot]]

🏷️ I've automatically added the component: ade7880 label to help categorize this feature request.