Wallbox Pulsar Plus (FW 6.7.4.1): SetChargingProfile accepted but charging never starts (Current.Offered remains 16A)

[cabvol]

Hello,

I have spent several days troubleshooting this issue and would appreciate your advice before I continue looking in the wrong direction.

Environment

• Home Assistant Add-on
• evcc 0.313.1 (same behaviour with 0.313.0)
• Database-only configuration
• Wallbox Pulsar Plus
• Firmware 6.7.4.1
• Tesla Model Y (2023)
• OCPP

Problem

The Wallbox connects successfully to evcc.

Everything looks normal:

• BootNotification succeeds.
• The vehicle is detected correctly.
• Tesla API communication works (SOC, odometer, range, status).
• MeterValues are received.
• The Wallbox reports SmartCharging support.

However, charging never starts.

evcc reports:

• charger logic error: current mismatch (got 16A, expected 6A)
• later current mismatch (got 16A, expected 10A)
• sometimes charger out of sync: expected disabled, got enabled

OCPP trace

The OCPP trace shows that every SetChargingProfile request is acknowledged by the Wallbox with Accepted.

For example, evcc sends a charging profile with a limit of 6 A or 10 A and the Wallbox responds with Accepted.

However, the following MeterValues still report:

• Current.Offered = 16

The charging session remains in Preparing and no charging starts.

What I already checked

• Tesla tokens are valid.
• Tesla API communication works.
• Only one charger exists in evcc.
• No duplicate loadpoints.
• Wallbox local load management disabled.
• Wallbox local PV surplus mode disabled.
• MeterValueSampleInterval tested with 30 seconds.
• Same behaviour with evcc 0.313.0 and 0.313.1.

Without OCPP (OCPP disabled in the Wallbox app), the Wallbox starts charging the Tesla immediately.

Question

Is this a known compatibility issue with the Wallbox Pulsar Plus firmware 6.7.4.1?

Or is there anything in my setup that you would recommend checking before I conclude that the problem is related to the Wallbox OCPP implementation?

I have a complete OCPP trace available and can upload it if it would help.

Thank you very much for your support.

[mfuchs1984]

Enabled remotestart in evcc?

[cabvol]

Thanks for your reply.

I enabled remotestart, saved the configuration, restarted evcc and tested again.

Unfortunately the behaviour is unchanged.

The Tesla still does not start charging and evcc still reports:

• charger out of sync: expected disabled, got enabled
• charger logic error: current mismatch (got 16A, expected 6A)

One additional observation:

The error

cannot create charger type 'template:ocpp-wallbox': cannot create charger type 'ocpp': connector already registered: 1

only appears when clicking Check status in the charger configuration. It is returned by the /api/config/test/charger/... endpoint.

During normal operation the Wallbox connects successfully via OCPP.

[mfuchs1984]

The pulsarplus is know for its problematic ocpp implementation. Is there an option to downgrade the software on the charger? Could you plase export your full evcc configuration?

[mfuchs1984]

When Current.offered stays on 16 A even when not charging, you can add -Current.Offered to the meter values field in the charger config.

[cabvol]

Thanks.
Currently the MeterValues field is empty, so evcc uses the default values.
Should I enter only

-Current.Offered

or the full list with -Current.Offered replacing Current.Offered?

[mfuchs1984]

Just -Current.Offered

[cabvol]

Thank you very much. The workaround with -Current.Offered solved the original problem 👍🏼. The Tesla now starts charging correctly via evcc.

However, I still have one remaining issue with PV charging.

EVCC correctly detects the available PV surplus and calculates a higher charging current. The debug log shows for example:

pv charge current: 11.7A

followed by

set charge current limit: 11A

So evcc is clearly increasing the requested charging current.

However, the actual charging current does not increase accordingly. The Tesla continues charging at only about 5 A per phase (around 3.7–3.8 kW), although there is still about 3.4 kW being exported to the grid. The Tesla app also shows 5/9 A, so the allowed current changes but the actual charging current remains at 5 A.

In addition, I still occasionally get the message:

charger out of sync: expected disabled, got enabled

even though charging continues.

Is this another known issue with the Pulsar Plus OCPP implementation, or is there another setting or workaround that I should try?

[cabvol]

Thank you very very much again.

After applying -Current.Offered, restarting evcc and disabling the Wallbox’s own PV surplus charging, the system now appears to work correctly.

EVCC adjusts the charging current dynamically and the Tesla follows the current limit as expected.

I’ll continue observing it over the next few days, but it looks like the issue is resolved.

[lostOzone]

Restore the charger.
Or connect the EVCC OCPP forward to HASS. And send ocpp.clear_profile from HASS dev tools.

[mfuchs1984]

Power.Offered is probably also wrong, you can use -Current.Offered,Power.Offered to disable both. Do you really have 253 V on L2 and L3?

[cabvol]

Thanks. I'll keep -Current.Offered for now since everything is working correctly again. If you think Power.Offered is also unreliable with this firmware, I can test -Current.Offered,Power.Offered as well.

My Sungrow inverter / Home Assistant reports:

L1: 233.0 V
L2: 232.8 V
L3: 233.8 V

So the 253 V reported on L2 and L3 in the OCPP trace are definitely incorrect.

This seems to be another issue with the values reported by the charger via OCPP.

[lostOzone]

I have investigated tons of wallbox.com charger ocpp logs. With each release, they manage to introduce new bugs or reintroduce previously fixed ones... But auto detection worked reliable in evcc since ever.

With v5.x, Current.Offered and Power.Offered were faulty, that's why the template https://github.com/evcc-io/evcc/blob/master/templates/definition/charger/ocpp-wallbox-fw5.yaml removes them by default, in addition, the charger did not correctly handle charging profile stacking. Later, these things started to work, but then, the charger suddenly needed relative charging profiles. Some 6.x versions again failed to report correct Current.Offered. Some even report only values for two phases, for them, even more measurands need to be removed. The history is long, these chargers are from hell...

But you are right, the truth lies in the trace log.

Thanks for this comprehensive firmware history. This looks like a mess.
When I upgraded to a Pulsar Pro I had issues with the factory 6.xx firmware luckily an upgrade was available and fixed it.

[mfuchs1984]

So the 253 V reported on L2 and L3 in the OCPP trace are definitely incorrect.

Yeah, looks like. The only other option is really bad contacts if the L2 and L3 cables. But not car would start charging at 253 V, I think. The good thing is, these values are not relevant for evcc..

[cabvol]

Thanks to both of you for your help.

The workaround with -Current.Offered solved the issue. EVCC now starts charging correctly and the charging current is adjusted dynamically as expected.

I'll keep observing the system over the next few days, but so far everything is working reliably.

Thanks again for your time and the detailed explanations!