I’m trying to work out whether this behaviour is coming from the
Evnex charger or the
Kia EV3.
I have a fairly advanced electrical monitoring system that I designed myself, with high accuracy and high time resolution, so this is not just a rough app estimate.
In this graph, the residual main supply line can effectively be read as the car’s kW draw in this situation.
What I’m seeing is that when solar tracking ramps the car charge load up to around 3.4–3.6 kW — roughly 15–16 A — the EV charging draw sometimes drops away very suddenly for about a second.
Solar production does not drop. The house then briefly exports power because the car has stopped using the available solar.
So I’m trying to determine whether:
- the charger is briefly reducing/stopping the offered current, or
- the EV3 is briefly stopping/reducing AC draw around that threshold.
Has anyone seen similar behaviour with an EV3, an Evnex charger, or solar-tracking AC charging around this power level?
Image attached.
Hi PeterH,
Thanks for your query. It is really a combination of both the hardware and software layer contributing to what is happening here. That is, software can adjust in milliseconds however when it comes to actual physical hardware the timings can be more varied. Standard rules (IEC61851) mean cars can take up to 5 seconds to slow down their charging, and there’s no time limit at all when speeding it up.
Even though the solar system constantly adjusts the charging limit, the car itself decides how much power it actually takes. We don’t update our system every second, and standard rules (IEC61851) limit how fast we can change those numbers. So, if solar power or home energy use shifts, there will be a slight delay from our charger system, a bigger delay from the car, plus some minor measurement errors. The key consideration is whether these minor delays have a material impact on overall energy usage.
Thank you
Hi,
Thanks for the explanation. That makes sense in general, but I don’t think it fully explains the specific behaviour shown in my graph.
The important point is that during the period I’m referring to, the solar production is essentially steady. So the issue does not appear to be caused by rapidly changing solar generation.
I understand that the car ultimately decides how much power it draws, and that there can be delays in how quickly it responds to changes in the charger’s allowed current. However, to determine whether this is mainly the charger control logic or the car’s onboard charger response, I think we need to compare the charger’s commanded current/PWM limit with the actual draw from the car.
Can you please clarify whether the charger was changing the Control Pilot/current limit during the period shown in the graph? If so, are you able to provide or confirm the commanded current limit over time?
The useful comparison would be:
- solar generation
- household load
- grid import/export
- charger commanded current limit/PWM
- actual EV draw
If the commanded current limit was stable but the car draw moved around, that would point more towards the vehicle. If the charger was changing the current limit while solar generation was steady, that would point more towards the charger’s control algorithm, measurement filtering, or update rate.
My concern is not a few seconds of normal response delay. It is that the system appears to be reacting around the import/export threshold even when solar generation itself is not changing materially.
Regards,
Hi PeterH,
Thank you, I don’t think we have remote visibility at the time resolution you are talking about so cannot give you the information you’re looking for. The system will have inaccuracies, delays, filters, noise and so forth and a certain level of error is actually to be expected. As long as the long term average results in a reasonable utilization of your solar energy then it is considered acceptable.
Thanks again
