EU 3 phase installation setup

[marcogtt]

Hello all, i’m installing emporia in a standard EU 400v 3 phases + neutral main  + 3pahses inverter from solar.

 

I will have 3 phase users for bigger load , such as heat pump, and elevator motor but i will also have single phase (220v) users as well.

 

I have the following questions:

 

-how to power the unit for the voltage sensing?

-how to install the transformers for the 3 phase users? 1 core each phase ? will i be able to see them as a single user in the app ?

 

thank you.

 

[Emporia Support]

Hello @marcogtt,  for a 3-phase system you will want to place the Black, Red, and Blue wires on each individual phase(MCBs) then the white wire to neutral.

[AC_DC]

Hi,

Does it matter which color (black, red, blue) is connected to which phase?

I installed my Vue in a 3-phase system in Germany. Some devices (e.g. cooktop) show an idle power of 70 W, although according to the manufacturer the standby power is < 5 W. And 70 W real power in idle mode is simply impossible.
I have now made another attempt and have connected the Vue to only one phase (the phase to which my cooktop is also connected).
I have only connected one 200A sensor on the main line and one 50A sensor to the cooktop supply line, and also only one power supply for the Vue (black wire from wiring harness). In this configuration, the Vue showed 0 W idle power of the cooktop and when I turned the cooktop on the Vue showed 2000W – so everything was installed correctly. When I then reconnected all three phases (black, blue and red wire from wiring harness), the Vue again showed 70 W idle power for the cooktop and I didn’t change anything at cooktop. I guess that the 70 W is essentially apparent power and the Vue does not assign the phases correctly to the 50A sensors. Thus, the problem only occurs when connected to a 3-phase grid with 120 degrees phase shift between the phases. Is there a solution for this problem. Is it possible to manually assign the 50A sensors to the corresponding phases?

[marcogtt]

Many thanks. What about reading consumption on a 3 phase device ? Should I put the resistor on each phase ? Can I then tell the app it’s a three phase user ?

[scrambler]

You need a CT on each Phase wire for accurate reading

[djwakelee]

Wire order does not matter.  That said, multiple people have reported 3 phase problems with the Vue 2.  Their claim is that it is falsely measuring VA (apparent power) and not W (real power) when in 3 phase mode.  This sounds exactly like what you are seeing with the high idle draw of your range.

Is it a bug or a feature, that is unknown.  In many cases 3 phase is used for commercial power, and many commercial utilities bill for VA and not W.

[AC_DC]

I think that it is a bug and not a feature, because the power measurement of the 3 main phases (200 A sensors) coincide exactly (+- 2%) with the measurement of the built-in electricity meter, so the real power is measured there and not the apparent power as with the 50A sensors. If it were a feature for 3 phase systems, then the apparent power should be measured everywhere – right? I wish Emporia would take care of this problem, because from my point of view it is a pure software problem, because hardware-wise all necessary information is available (current, voltage, phase shift) to determine the real power. Evidence of this is the correct determination of the real power with the 200 A sensors. There is no reason why this should not also work for the 50 A sensors.

[djwakelee]

Yes, if you are seeing correct real power on the mains, but not on the circuits, I’d agree that is a bug.

Sadly, Emporia has not been forthcoming with any bug fixes or feature improvements to the Vue 2.  We’ve been waiting for combining of multiple Vue 2’s, as well as better handling of 120/240V split phase circuits for the US.  Going on a couple years now when these things were promised to be coming, but never happened.

[Juhe67]

Hello,

I have the same problems also.

When high load is needed, the measurements are nearly ok, but in “idle” Mode everything is worse.

I’ve had a chat with the support yesterday, but no results for the problem.

Is there a way for a factory reset? So I can try to setup everything new?

What I have also seen, ist that Main B has got less Amps. A and C are O.K. with the offical Powermeter, but B is less from 0.4A at 300W and up to 3 Amps at 2kW…

I’m very sad, but I think I will return the paket and try something different…

HomeAssistant is working well with the Vue…

 

[ArubaUser]

Hi,

So I just received my EV car charger, it’s a 32A unit, I connected it to 2 of the phases (we have 127/220 here on Aruba) to get to 220 and the reading is accurate. Also when I stop charging it shows 0.0000 correctly. I imagine that this is because the car charger won’t use any apparent power?

Either way, still REALLY hoping for a fix for this situation. Maybe in a new the gen3 this will be solved? Really hoping it because in theory this device is awesome but with such false readings on my airconditioner’s I just can’t really use it.

[emporiacs]

Our current Level 2 EV Charger model will only work on split-phase 120V/240V electrical systems at the moment. Split-phase power will have two lines @ ~120V with ~180-degree phase offset. If your system does not have these specifications it will not work and not recommended for installations outside of North America for the time being.

[ArubaUser]

Ah, I was talking about the VUE gen 2. It actually works correctly for the EV charger it seems (my EV charger is not an emporium one, sorry!). Well it shows 4.4% higher consumption than the charger shows but I’m very ok with that. Important thing is that it shows 0 consumption when I turn the charger off. As you probably know, the VUE gen2 doesn’t work correctly in 3 phase setups out of the USA, it measures apparent power instead of real power, thus shows consumption for all of my airco’s when they’re powered down, between 50 watt and 150 watt, depending the model. Just like everybody else outside the USA with 3 phase setup has been reporting on this forum. Hopefully you guys can solve this in your next version because I really love your system, but right now it’s hard to use with this bug.

I think the reason why it works fine for my car charger, because it doesn’t contain anything that uses reactive power. Just like it measures fine consumption of the lights in my house for example, all that has nothing that uses reactive power.

• This reply was modified 1 year ago by ArubaUser.
• This reply was modified 1 year ago by ArubaUser.

[emporiacs]

Hi,

Thank you for the additional details.

There is no bug and the Gen 2 Vue works in 3-phase systems without Delta. We have customers in  Europe, Asia, Middle East, and the USA using the Gen 2 Vue in 3 phases with plenty of accuracy, and our monitor measures both real and apparent power, but we display only real power within the app itself.

Hope that helps clarify further.

[emporiacs]

Also, it is possible that there is a calibration issue or similar on the Gen 2 Vue that would cause it to be a bit inaccurate in kW/kWh, but requires diagnostic tools available with Customer Support.

[ArubaUser]

Ok, so I chatted to emporia chat, the first guy thought that maybe there was a ‘channel problem’ (emporia listenening to the wrong phase). So he made some changes. My airco in standby would, according to the VUE, consume 180 watt. So he made some changes to the channel and brought it back to 90 watt. But still, this made no sense (since this is a very high SEER, energy efficient unit, it’s an expensive airco). So I decided to test with a different system from a competitor. It said that in standby it only used 3 watt and gave a VA of 168 Watt.

So then I chatted with Emporia again, told the (different) rep this time about the problem that the VUE2 seems to measure VA instead of Watts on my (3 phase) system and she said that they were aware of the problem and researching into it!!! Really happy to hear that!!

So then I decided to do another test, I bought a passthrough meter, this seemed to be one of the higher quality ones: https://www.amazon.com/dp/B0C619YRRM?psc=1&ref=ppx_yo2ov_dt_b_product_details

And guess what, it showed again the same, standby usage of only a few watts and a very low power factor (about 0.04) , showing a VA in standby of 167 watt!! So this now 100% confirms that the emporia doesn’t work correctly on some 3 phase systems out of the USA, measuring VA instead of real power.

BTW, so it turned out that the first rep, who changed the channel, actually changed it into the wrong channel, since the emporia now measures 93 watts in standby. The original measurement of 180 watts was actually pretty close to the 167 watts. So it was set to the correct channel originally, it just measures VA instead of real power.

BTW, although I do have a 3 phase system, my airco is just a two fase one. We have 127V between neutral and line here and 220V between two lines, so we connect 220V airconditioners to 2 phase lines here.

Anyway, so the rep said they were aware of this problem and researching into it. I really HOPE they can find a solution, I don’t even have care if I need to buy a gen V3, I just want this to work. It’s an awesome device but right now it’s a bit hard to use. BTW, so when I turn the airco on, the power factor goes to about 0.98, so VA is almost the same as real power so it does work nicely with the airco turned on. It’s just kind of useless for devices that have a bad power factor (like airconditioner’s in standby mode).

• This reply was modified 1 year ago by ArubaUser.
• This reply was modified 1 year ago by ArubaUser.
• This reply was modified 1 year ago by ArubaUser.

[ArubaUser]

BTW, this device DOES work correctly using clamps: https://www.amazon.com/dp/B07S6SFKTK?psc=1&ref=ppx_yo2ov_dt_b_product_details

This one does register the power factor correctly (well it says 0.02-0.03 where it should be 0.04) so it does show my airco in standby 3 watts consumption and 168W VA. However since you can’t log anything at all it’s really not what I’m looking for. I want best of both worlds, an Emporia device that actually measures real power.

So I’m REALLY REALLY hoping Emporia is working on a fix. I need this device to work.

[Dirk Grote]

Hello, I have the same problem with the reactive power. My induction stove draws 90 W when it is switched off. This cannot be right and of course distorts the power consumption throughout the day.

[emporiacs]

For your case, best thing would be to run a WiFi reset (don’t forget to create a heavy load event so the sensors can get calibrated) and see it if fixed the issue. If it does not, please give us a few more days, we are investigating a few devices that are showing real power higher than apparent.

[mako]

I have the same issue, UK 3-Phase supply and I have two air source heat pumps which are each registering ~200W  plus an oven which is registering 46W when in standby.

When I turn these loads off at the breaker the total usage remains constant with these ‘ghost loads’ removed.

[Dirk Grote]

Hello, even switching on a powerful consumer, sauna with 9 kw on 3 phase does not bring any success. The reactive power of the induction hob is still displayed as 70 VA. I have now added a PV system and when setting up the solar settings (reset WIFI/solar setup) I get the following error messages. This means that the PV inverter is not correctly integrated.

First message

Three phase system only mapped to 2 unique voltage channels
A maps to black
B maps to black
C maps to red

Second message

Duplicate Channels Detected

You have two (or more) wire harness less connected to the same Voltage line, Move a wire harness lead up or down vertically adjacent (stacked) breaker and try again.

[emporiacs]

Hi,

If you haven’t contacted our support team directly yet, please do and we’d be happy to help!  A quick check from our support team can help to ensure the wire harness is correctly hooked up.

If you’re in a split phase, you’ll want to terminate the black wire into one phase of power (L1), the red wire into the second phase – L2 (usually in an adjacent breaker), and then have the blue and white wires going into the neutral bus bar.
 

[ArubaUser]

BTW I have found out that this is NOT specific to EU, nor specific to the amount of phases. The VUE2 (not sure about VUE3) only calculates real power for the 3 mains, NOT for the 50A breakers. It’s very easy to see once you take a look at the raw data that the SAMD09 sends to the ESP. It only sends phase angle information for the 3 mains!!!! So it’s 100% impossible that it calculates real power for the 16 smaller CT’s. Not in the USA, not in the EU, not on the moon, LOL.

[Epelbaum]

<p style=”text-align: left;”>Hi, I have the same problem. Vue2 measures about 300 watt on my AC in standby mode in 3 phase residential AC system, 127/220 V.</p>
I have 3 clamps connected to three phases of AC, each of them measures 100 watt.

Talked to AC company, they tested it with special hardware and found, that Emporia measures apparent and not real power in this case.

Please advise.

 

[ArubaUser]

That’s correct, the emporia VUE only measures apparent power, no real power. I’ve completely reverse engineered the emporia firmware and there’s no real power calculation going on AT ALL. BTW We’re working on a replacement firmware that WILL calculate real power. Right now I’ve just finished the first open source replacement firmware for the SAMD09, it works now! Right now all it does is exactly the same thing as the emporia VUE original firmware, but since it’s now open source, anyone can now adjust it to include real power calculations. I will soon start looking into that myself

 

https://github.com/gekkehenkie11/emporia-SAMD09

 

[ArubaUser]

Not sure why I can’t react to the other thread (‘why things dont add up), so I’m posting it here now. I’m by no means an expert at electrical engineering, in fact I’m a beginner. I do know how to reverse engineer code though that’s how I could create the open source firmware. But honestly I’m REALLY confused at this point. I’ve been reading up and in theory, what emporia does SHOULD result in real power: if you take a voltage sample and a current sample at the same time, then multiply them and then take the average, that SHOULD equal real power. However Emporia does things slighly different, they don’t simply multiply current and voltages, but the DIFFERENCES with the average current and average power from the last half second. But I guess those averages should be 0 anyway? So why does it not result in real power for the 50A CT’s !?!? I’m still hoping someone with a major in electrical engineering can jump in.. Because I’m 100% sure that at least on the 50A CT’s the result is apparent power, I’ve tested it extensively and always come to that same result, as the others in this thread …

Ah something else that might be important is that Emporia does use the latest voltage samples but the sample from T-2 for the main currents and the sample from T-4 for the smaller CT’s …

Anyway now that the firmware is open source, hopefully we will be able to figure it out. It’s such a shame, this is such a nice device but rather useless when we can’t get it to show real power for the smaller CT’s

• This reply was modified 5 months, 3 weeks ago by ArubaUser.
• This reply was modified 5 months, 3 weeks ago by ArubaUser.

[ArubaUser]

Ok I talked to my brother who has a major in electrical engineering and he explained it, why it’s not correct what Emporia is doing. The problem is that if the current and voltage is completely out of fase, then V*I will result in apparent power. So we’ll have to adjust for that.

[ArubaUser]

So in short, emporia only measures real power when voltage and amps are in phase. When the power factor is low, it indeed measures apparent power. So, hopefully Emporia can solve this themselves, if not we’ll do it for our open source firmware. But the idea is to:

 

1. calculate apparent power via RMS
2. Measure phase difference
3. use that to calculate real power

[ArubaUser]

After quite a bit of more research I must say my earlier comments are wrong. Emporia DOES measure real power! In the world of discrete samples, real power is the average of the product of each voltage sample and a current sample taken at the same instant.  That’s what Emporia does, or I should say TRIES to do since this seems mostly a calibration issue. Within the VUE2 the voltage and amps are calibrated to be almost 180 degrees out of phase. However they’re not exactly 180 degrees. When I shift the voltage 4 samples for the smaller CT’s, my air conditioner standby power drops from 120 Watt to 16 watt (should be 8 watt). However some other measurements which were a bit too high already, get even a bit higher.

Anyway I’m surprised Emporia shipped the device like this. I’m sure it would be possible to create a better, more calibrated device. I’d definitely pay for a premium device that works better.

Anyway, going to see if I can calibrate things myself a bit better, but it’s a lot of work …

• This reply was modified 5 months, 1 week ago by ArubaUser.

[ArubaUser]

Here’s the explanation, it’s finally 100% solved:

https://github.com/emporia-vue-local/esphome/discussions/287

[Author]

Posts