Incorrect consumption readings for L1, L2 and L3 on the Cerbo GX with an EM540 grid meter

My Cerbo GX, connected to an EM540 used as a grid meter, is showing incorrect consumption values for the individual phases.

My system consists of 3x MultiPlus-II 8000s, a Cerbo GX, an EM540 as the grid meter, and 3 additional EM540s as meters for the PV system, household consumption and heat pump consumption.

First of all, the values for the PV system, household consumption and heat pump are exactly right.

Only the values on the grid meter are completely wrong. The total values are correct; I compared them with the grid operator’s meter and they are exactly the same. It’s only the sum of the L1, L2 and L3 values that’s completely wrong.

Why is the EM540 counting and displaying such incorrect values here? With the other uses, the values are absolutely identical; in other words, the sum of L1, L2 and L3 matches the total value exactly.

The grid meter EM540 displays these values.

The PV meter, house meter and heat pump meter are all exactly right.

On the grid meter display, I noticed that it constantly registers both positive and negative values on the individual phases, while the total remains roughly 0.

Of course, in principle, I can ignore the incorrect values. But if they aren’t right anyway, why are they displayed? Or why can’t they be hidden?

Is anyone else seeing these discrepancies in the display?

Could I just need to adjust the settings to get it working properly?

One more thing I noticed: at the start of the year, I reset the values. Everything went to 0, except the grid meter’s “Reversed Energy” values, which remained. Victron’s German support team hasn’t been able to help. I reported the issue months ago, but nothing has happened.

I hope you can help.

Hi,

the values are not wrong at all. They’re the result of the net-metering method used by a meter that nets readings:

Feed-in and consumption are offset against each other BEFORE the total meter reading changes:

If you have readings like -300/200/100 on the grid, for example, that nets to 0. Over one hour, that would mean:

  • L1 registers 300 Wh of feed-in
  • L2 registers 200 Wh of consumption
  • L3 registers 100 Wh of consumption
  • The total meter registers neither feed-in nor consumption. No change of 0 Wh.

(You can’t work out the total meter readings from the individual values afterwards, because the netting happens in real time, as the readings occur)

Hello,

Thanks for your reply.

That’s what I thought, but it does make me wonder why these completely pointless values are being displayed. They’re wrong for both the import and the feed-in, and on top of that I can only reset the import to zero; the feed-in figure stays as it is.

They’re correct on the individual PV, house and heat pump meters.

They do match, and provide valuable insight into the average phase imbalance your loads cause on the grid.

Not relevant for private users, but phase imbalance can cost businesses a lot of money.

OK. I see. So I can use this to see how much is actually being drawn from and fed into the grid on each phase. I could also see whether the load is unevenly distributed across the individual phases.

However, because of the netting, the total amount of imported or exported electricity is lower.

Do you have an explanation for why resetting the grid meter zeroes the import values for L1, L2 and L3, but not the export values?

Exactly.

Nah, unfortunately I can’t answer that. I’ve never reset an EM540 or paid any attention to this function. It definitely sounds illogical, though. Maybe you have to reset it separately, as a feature for users who want to reset their import reading (when switching supplier) but not their feed-in reading?

That didn’t make sense to me either. There’s only one menu item on the meter to reset everything. As far as I remember, it’s the same in the software. But nothing has solved the problem.

Well, I’ll just ignore it now too.

I’d be grateful if it could be hidden as well.

Ah, it’s just clicked: the EM540 is available as a MID-certified version — which you apparently have. That means resetting the total counters isn’t permitted; they have to be tamper-proof.

Yes and no.

I was also surprised that I could reset almost all the values.

If that were the case, it wouldn’t have worked at all. But Energy was set to 0. Reversed Energy wasn’t.

The totals were also set to 0.

So it has nothing to do with MID.