CT Clamp calibration ideas

Power Sensor
1

Hi team,
I’m currently using the Evnex charger to send excess solar power to the car during the day, but I’ve found that the CT clamp that it uses doesn’t seem to be calibrated quite right, as it seems to always use approx 200-300 watts more power than is being generated, so resulting in a net import from the grid.

While this doesn’t seem like a big deal - my daytime grid power price is triple the price I get for export to the grid, and also the grid power during the day is quite ‘dirty’ power which I would prefer not to be using.

Perhaps two options would be:

  • Add a user level setting to the Evnex system to allow a calibration value to be added / subtracted to the CT clamp values, so I could then compare the CT clamp data to my actual smart meter, and tell it to subtract 0.3kw from the measured value

OR

  • Add a setting to the solar diversion area in the app, telling the system to use percentage of the excess power (at the moment it tries to use all 100% of it) so perhaps I could set it to use only 90% of the excess.
1 Like
2

Hmm… you have an intriguing idea, but it would complicate the user-interface. If it’s not a widespread issue then maybe the best thing for Evnex to do is to file-away the idea? (Disclaimer: I don’t provide professional advice to Evnex or to anyone else these days – I’m fully retired, thank you very much, so I just give “armchair advice” :wink:

It’s certainly possible that the Evnex chargers overestimate the solar excess in most installations. I think that very unlikely, as it’d be easy to add a “fudge factor” of 100W or 200W in their next firmware rev. But hmmm… maybe your installation is quite different to the usual one. In particular:

  • Is there a much longer-than-usual run of cable from the CT clamp to your Evnex charger? The resistance of the cabling matters, if the CT clamp is of the TVS variety rather than one which has an internal burden resistor. I really don’t know – I have never sighted an installation manual for an Evnex charger, nor have I ever inspected the CT clamp on a household which has an Evnex charger! However I’d guess that its a TVS. See a geeky discussion of whether an existing and disused 40m run of CAT6 cabling could be used to connect a CT clamp to an Evnex E2

  • Does the cabling to the CT clamp run for a significant distance, near the cabling to power-hungry appliances, or near the AC cabling for your PV inverter? I’m wondering about the noise injection at 50Hz – which would be especially problematic when there’s a small “signal” from the CT clamp to the Evnex charger. I note that you’re looking at cases when there’s only 200W or 300W of supply to your household. I see a bloggish discussion of this issue at Noise on CT lines?

It’s surprisingly difficult to measure kW on an AC circuit accurately! I think getting within +/- 300W is pretty good, for estimating household draw using an appliance that’s located at some distance from its main switchboard. To reliably do better than this, I think you’d need to be measuring both voltage and current at the switchboard at pretty high sampling rates. Your power meter is doing this already, but (sad to say) it does not reveal its power measurements to your household devices. In my dreams!

3

Hi there.

If the offset is fairly consistent then we do offer the capability in our app to change the “target” during a solar charging session. So if the CT is consistently reading 200W over then you could set a target of 200W instead of 0W which is the default (and then shift the start and stop conditions by 200W as well).

It’s not as elegant as an offset parameter, but it achieves broadly the same thing.

Kind regards,

Tom

1 Like