fix #452 reevaluate claims if amps > pilot

[KipK]

Until the bug is corrected on evse fw, this solves the issue and prevent the charge_current going to max when Evse goes from Disabled to Active state ignoring claims & pilot values
fix #452 and should also fix #513 fix #388 fix #481

Edit: It seems those bugs are related to PP_AUTO_AMPACITY beeing set. Posted an issue there: lincomatic/open_evse#156

So this is not really an OpenEvse bug, but more that OpenEvse and Wifi module don't share the non tethered chargers checking cable ampacity state and we have confusions here.

[jeremypoulter]

Actually I don't think getAmps is what you should be checking against, that is the live amp reading. What may be an idea is calling openevse.getCurrentCapacity after a state change/current change then doing the check to see if the pilot on the EVSE is the same as what the Monitor thinks it should be, so somewhere here, https://github.com/OpenEVSE/ESP32_WiFi_V4.x/blob/master/src/evse_monitor.cpp#L478, and here, https://github.com/OpenEVSE/ESP32_WiFi_V4.x/blob/master/src/evse_monitor.cpp#L280

[KipK]

Do you think then we should reevaluate claim or just set the pilot to evse ?

[jeremypoulter]

I think just setting the pilot should be sufficient

[KipK]

I have rewrote it on getChargeCurrentAndVoltageFromEvse()

Can't make the change using openevse.getCurrentCapacity , i'm lost with those callbacks.
But it works good as it, ( still the 2-3 sec glitch )
Can you merge for now as it solves important issues, and remove it later when you have time for a proper way of doing it?

[KipK]

Ok I got it thanks.
I've made the change, gotta test first.

[KipK]

Jiust tested but it doesn't work now. Amps stays at 32

[KipK]

I have added some debug:

pilot = 32
getPilot() = 6
####  Pilot differents set again

So the check works ok, but setCurrentCapacity seems to not have effect here.

[KipK]

Ok it works now. I've rebased the commits in one.
Ready to merge here

[KipK]

After few tests, I can say it reacts faster than the claim approach I did first. I don't see spikes over the set limit and it stabilise quiet fast.

Until we got a proper integration of auto_ampacity between openevse fw and ESP32 one, this should fit.