Earth failure and then EVSE oddities

Integrations OpenEVSE
1

Last night, both our EVs tried to charge at the same time with the result that both OpenEVSEs logged an “Earth Ground check failed”. Since these have been operational for over a year now, I’m not sure what caused the error. However, that is peripheral to this problem.

The above fault than caused the main RCD to trip and both EVSEs were powered off in the early hours of the morning.

When this was noticed later that afternoon, the RCD was reset and both EVSE’s powered up, however one of them would not work.

First of all, it seemed to have lost some basic settings, so the max. current was set to I think 0 and other settings on that EVSE page of the settings were blank. Even after correcting them though it would not charge the car. Multiple reconnections of the cable, unlocking, starting the car, re-locking etc made no difference. No errors reported by the EVSE but when manually enabled the car just reported “INITIALISING” and it would never start to charge.

A quick check with yet another EVSE showed that the car was working perfectly. It was the OpenEVSE simply not starting the charge when connected to the car.

Looking further at the web interface I could see the SETPOINT showing as 6A, despite the max. current in Settings set to 32A, as was showing in the CHARGE RATE slider on the CHARGE SESSION page of the GUI.

So 2 different places to set the max and a third display of that setting. Not confusing at all - except when for no explicable reason they are different.

Having stared at that CHARGE SESSION page for a long while trying to puzzle why the SETPOINT remained steadfastly at 6A, while the CHARGE RATE slider showed 32A I decided to move that slider. At which point it immediately started to charge and ‘sliding’ the control back to 32A it charged exactly as it should and SETPOINT displayed accurately.

So, grateful that it is now working as it should, but WTF? Why are the 2 max current controls and the SETPOINT display so hopelessly out of sync, to the extent that the device will not work at all.

There is also the issue of why this EVSE was unable to maintain some basic settings when powered off then on again? Is this some hardware failure of permanent memory. Will it lose its marbles again if powered off?

Answers on a postcard please. Ok, here on the forum would be preferable. :grinning_face:

Finally for anyone reading this in the next few days - Happy Christmas. :partying_face: :santa_claus:

2

A me too post. I saw exactly the same behaviour and same settings lost (schedules and max current ) This was on a new esp32, so don’t think age related.

It also manifested itself by partial loading of the web interface, taking many reloads to get it to load fully

I’ve seen the same behaviour with emonevse, but I use the wired esp32 board, replacing the existing esp32 module

After a lot of investigations I traced it to the supply to the esp32 board being a tad under 5v. Increasing it to 5.1v cured the problem.

I’d check that out.

As for the charge max - one is the max it can ever reach and the slider is variable during charging - on mine it is configured to modulate with solar generation for example.

3

Not yet had a chance to investigate, but both EVSEs have subsequently worked without error, albeit not simultaneously.

It is puzzling why the error was indicating a ‘lack of earth’ problem (a problem that apparently no longer exists) yet the RCD tripped indicating current was leaking to earth.

The fact that both EVSEs reported the same fault, almost simultaneously would imply the fault was not in an EVSE, but to do with the supply and yet that has not changed. Nothing was ever touched on the supply itself. Earth is still connected as it was (same connection for both EVSEs) and the RCD is allowing power to each EVSE without tripping.

It’s almost as if there was a momentary problem with earthing at the substation that supplies our property and which is in fact only about 40m away. Was there a glitch that caused our EVSE to report the error, then as the substation corrected itself, a brief leak to earth caused the RCD to trip.

Just speculating, but both EVSE reporting the same error would indicate that neither were actually at fault.

4

What sort of earthing system do you have? (see https://electrical.theiet.org/media/1698/earthing-your-questions-answered.pdf)

Is it possible that a fault outside your property but close put enough voltage between your earth and the substation earth to create an earth loop current, which tripped your two RCDs?

5

As far as I know it’s TN-S as our earth connects to (I presume lead) sheath of the supply cable, but the power company have not been able to confirm that from their records. Despite that, I’m fairly certain it is TN-S.

Yes I agree there could have been a brief fault not related to my installation, but the error indicating poor earth seems inconsistent with the RCD tripping which would have been due to an apparent leak to earth.

I’ll have to try both cars charging simultaneously and see if that does cause the problem, but currently I’m thinking it was more likely a problem with the supply which although from a very new substation (actually on our property), its supply from the grid is flaky and we get frequent power cuts. So a good reason to get solar panelled and batteried up. :grinning_face:

There is still however the issue of why one of the OpenEVSEs lost some of its settings when powered off. When I was struggling to make it work I again powered it off then on and it again lost those settings.

Now it’s been working fine for a few days I probably should power it off and on to see if the settings still get lost or whether it has now recovered and is retaining everything as it should.

6

That won’t help if you have a grid-tied inverter – it will shut down as soon as the supply fails to prevent “islanding”.
:sob:

Sorry I cannot help with the lost settings problem.

7

Hence why I will be utilising a ‘backup gateway’ that will switch supply from grid to battery as soon as the grid fails. :slightly_smiling_face:

1 Like
8

Sounds like he’s talking about a hybrid inverter. (or the equivalent)

Something similar to this:

https://www.sma.de/en/products/hybrid-inverters/sunny-boy-smart-energy