#Need help debugging my energy meter installation

1 messages · Page 1 of 1 (latest)

void plinth
#

For those who followed my last thread: Yes, I did it. I've got the bloody Vue3 bought, flashed and installed. However, my troubles are just starting.

I swear I have all phases set up correctly, and I'm getting sane values off the input phases. However, some of the individual circuit clamps are reading negative values. ALL of them are oriented the same way, but of the four on phase L1, one is reading positive, the other three are reading negative - as in sane, but sign-inverted values. What could be the issue here? I could inverse the clamps or slap a filter on the sensors, but I don't want to just stupidly fix symptoms just to find out later my data is corrupted and useless.

wheat chasm
#

make sure K side go to power source, L side to your lamps/whatever

void plinth
#

I assume by that you mean the orientatio of the current clamps. As I said, they are all the same orientation and at least by the labels they have on them, it's correct.

acoustic sapphire
#

I just had a look at the Vue3 documentation - the schematics in there aren't very clear about this, but is it possible for the clamp wires to be reversed in the wiring harnesses?

jade cipher
#

It wouldn't be the first time that the printed on labels are wrong

#

If you are sure that those individual circuits aren't producing any energy and it's actually the right circuit then just turn those clamps

#

Or check wether the software allows for a reverse reading

void plinth
#

Yes, I'm bloody sure my dishwasher isn't outputting any power 😉 And there's your crux right there. All of them are wired the same, configured the same and oriented the same, yet the baking oven one has a positive reading, all others have a negative reading. So unless all but one current clamps are wired backwards internally, I don't know how that could be explained. I even tried switching the voltage tap around, but the same still applies.

void plinth
jade cipher
#

I would kinda put my money on that the QA for such clamps is basically non existent for them being both cheap to make and easy to reverse, either physically or in software

void plinth
jade cipher
#

Hmm 🤔 that's for sure worse ^^

acoustic sapphire
#

That sounds alarming - this could potentially mean you have the live and neutral reversed in some of your breakers? I'd be very careful... and maybe have a qualified electrician look at the board.

void plinth
#

There's an idea. The guys who built this back in the day weren't exactly the brightest candles in the church. On the other hand, I don't see how it could be, my breakers are european style single-phase modular breakers. One wire in, one wire out. I don't see how a neutral could have been mixed in there, except if it was done somehwere outside the main distribution box... and yes, I have outlet cups stuffed to the brim with connectors...

#

I don't think I did anything wrong, I'm painstakinly mediculous when it comes to anything involving mains wiring. I got 100% on my test in school too. The only time the teacher found an error happened because I was payin attention to the pretty gurl next station instead of my wiring.

acoustic sapphire
#

If you turn a breaker off and you still have voltage on the "live" in the circuits that'd be a dead giveaway for live and neutral being reversed. But I don't advise messing with and measuring mains voltage unless you A) know exactly what you're doing and B) have the necessary equipment!

#

It could be that your board has different parts / bus bars and only in one of them it was reversed?

#

In a messy board it can however still be hard to trace the wires and thus get the direction wrong... So it's not necessarily that L and N was reversed, but if there is any doubt, I 'd have it checked!

void plinth
#

Don't worry, I A) am trained if not formally certified and B) own certified, high-quality measuring equipment that won't explode on me even if I do make a mistake.

#

As for the setup, I have three phases, each comes out of the central RCD and goes into a block of breakers connected by a bus bar. The N and PE go to separate bus bars. Since you seem to know your stuff, it's a TT installation.

acoustic sapphire
#

Ok, then it's unlikely that L and N are reversed even for one phase I think, that would have gone bang the first time power was switched on...

void plinth
#

Now that you mention it, yes it would have. The RCD doesn't like that kind of thing at all.

acoustic sapphire
#

The other thing that could be happening is that you have large inductive loads in some circuits? That could sometimes mess up the power factor calculation. Are the negative wattages you're seeing small (i.e. more like standby) or large?

void plinth
#

I'd say -150W is no longer in the realm of "small". The largest mildly inductive load in the kitchen is the freezer. Next largest reading is the living room at -60W. There's definitely nothing even mildly inductive there.

acoustic sapphire
#

Hmmm... I was asking because the clamp of my induction hob does show consistently -6W when the hob is on standby, but goes to positive power draw when switched on. But no, that's not it then.

#

Well, what I would do (after ruling out any safety issues with the electrical installation) is to shrug my shoulders, reverse the clamps for those circuits and find the underlying issue some time when I'm bored...

void plinth
#

It so happens I work for an electricity company at the moment, albeit as a software developer. I just put the problem before a bona fide electrical engineer. He said, there is no way this could be a physical anomaly in the system. Short of a wiring problem in the device or the clamps, it would have to be a software issue.

#

So I guess I'll go over my config again with a fine comb to rule out any stupidities on my part, and if that comes up blank, I'll just reverse the wiring on the bloody clamps and be done with it.

#

I've just had the idea to measure voltages across the current transformers. Never seen a multimeter come up so consistently blank. I'll be damned if I know what's going on there.

acoustic sapphire
#

You'd need to measure current across the clamps, not voltage, they are a current source.

void plinth
#

Now that you mention it... Didn't have my head in gear, did I?

void plinth
#

@acoustic sapphire In case you are still following this, the last current clamp turned negative too. The only thing I did was move it to a different location on the same wire, to make room for the next batch. Seriously, don't ask me, so long as it works...

acoustic sapphire
#

That is very strange. Might be worth contacting Emporia and asking them about this?

void plinth
#

The problem with that is that I'm running ESPhome on it, which Emporia might take as a welcome excuse to void my warranty...

#

I could go over the code for the component in ESPhome, but it's been inactive mostly and hasn't integrated any pull requests for months.

#

It works well enough for the most part, so I guess I'll live with the abs filter in my code and move on to new projects.

#

If I had to venture a guess, there's probably some wonky logic in the data acquisition chip that tries to guess whether the clamp is reversed or not, and gets it wrong.

#

Or somebody in the factory wired a few dozen boxes of sensors backwards, it is made in India after all...

acoustic sapphire
#

I just saw this post on the esphome github: https://github.com/emporia-vue-local/esphome/discussions/264 - you probably have seen it also? Some people are reporting negative readings as well and having to use the abs or invert filters.

GitHub
Emporia Vue Gen 3 Support · emporia-vue-local esphome · Discussion ...

Back in June, a FCC entry for 2AS6P-EMV3A (https://fccid.io/2AS6P-EMV3A/Internal-Photos/Internal-Photos-6584145) went public, which shows the gen 3 device should still be ESP32 adding ethernet, and...

#

Got me thinking - if you use a clamp for which the voltage is measured on one phase around a wire that is on a different phase you WILL get weird and possibly negative readings. So even taking a clamp that worked well on one circuit and putting it on a different circuit that is on a different phase will lead to strange effects. I don't have a Vue3 so I can't tell if this is possible or likely happening.

void plinth
# acoustic sapphire Got me thinking - if you use a clamp for which the voltage is measured on one ph...

I have seen that discussion, yes. That was one of the first things I checked. All clamps are religiously assigned to the correct phases in the configuration. I've also conferred with a few electrical engineers I happen to work with, and their unanimous opinion was that it can only be the internal wiring of the device or the firmware. My guess is the latter. If I had to venture a guess, the V3 is probably a bit more differnt internally from the V2 than people thought. The fact that one clamp showed the right way round but then changed its mind might hint to there being an internal logic that tries to guess the right orientation of the clamps but gets it wrong. Anyway, the abs filter is a viable fix, so I'm not planning on digging deeper at this time.