OpenEVSE charging rate shows wrong

OpenEVSE charging rate is always 3A more than what the car actually charges. I can tweak this via RAPI but the charging starts only when the 3A lead distance has been reached and the kWh calculation also gives too large numbers apparently for this same reason?

Does someone how to fix this?


Firmware config:openevse_wifi_v1
Firmware version:4.1.7


Hi Tonza, welcome :grin:

I had a similar issue with my Leaf not responding to the calculated pilot signal in OpenEVSE. I wrote some external control software to manage the openEVSE pilot signal and soak up excess grid power rather than rely on the internal PV divert function. You can find the thread and links to GitHub here:

I don’t actually know if this is an issue with the car or the OpenEVSE or it’s just ‘normal’ :stuck_out_tongue_winking_eye:. It doesn’t matter since the software forms a closed loop feedback path to overcome any offset issues like this. So long as changing the pilot causes the actual consumption of the vehicle to change in the same direction it should work. When the car decides it has had enough and stops charging the pilot will just float up to max and stop there. Which is fine since this is what a fixed output wallbox would be like anyway. I also included just turning charging on and off into the software e.g. for night time charging.


Hi Stu, thank you.
It’s nice to hear that I’m not alone with this problem :slight_smile: , but I’m not in my comfort zone when it comes to programming. I hope there is an easier solution for this or I’m just suffering with this…

What car do you have? I’ve noticed BYD would draw something like 8A when using the Polestar 10A EVSE, or Khons at 10 or 15A; Always a couple of amps less. It wasn’t an EVSE problem, just a quirk of the car. You might have to builld a fudge factor into how you transmit the solar PV data. How are you doing that?

This looks like the issue I might have with my i3, but the difference is only 1A, not 3.

I’m quite sure today this is the issue I have with the i3.

I always have 1.2 or 1.1 amp difference between “setpoint” and “current”.

So when “setpoint” is 6 amps (manually or PV-divert), i3 gets less than 5 amps.
It goes in error mode because “grid doesn’t send enough”.

If this an hardware or software issue?

I hope I eventually nailed it.

OpenEVSE is configured for 220V (Current scale) while our main voltage here is around 240V.
With 240V, the calculation is ok: 3802W / 240V = 15.8 A.

So I just modified the calibration to 240V using RAPI command.
However, RAPI don’t survive reboot (and should disappear in a future version).

Do I have to compile my own version of the software to get 240V as default?

Do you have any way of measuring the gird voltage e.g emonTx or emonPi? Grid voltage will often vary, the only way to get it really accurate is to provide an MQTT feed with the grid voltage.

If no volt meter is available the voltage is hardcoded here:

Hi Glyn.

I have ways of live-measuring the grid voltage (EmonTX and the Fronius Inverter).

Edited message.

I’ve tried just tried it.
Grid voltage is read in the Fronius inverter then published to mqtt.
OpenEVSE is configured with the grid voltage topic (in MQTT tab), currently 245V 8-/

However, when asking the voltage with RAPI, I still get the voltage I setup a couple of days ago (240V).

There’s no car charging right now, do I have to wait till a car is charging for OpenEVSE to read the correct value in MQTT?

Currently charging, asking for GA via RAPI keeps on answering “240”.
But GG seems OK:

> $GA
< $OK 240 0^26
> $GG
< $OK 9600 239000^27

MQTT seems to work, the inverter updates the voltage and openevse gets it (data taken a couple minutes away above RAPI request):

fronius/tensionalternative 238.7
openevse/voltage 238.6999969