3 Phase System with MultiPlus: Breaker Types and Sizes

Q&A and troubleshooting
, , , ,
1

I’m in a weird situation, I have some knowledge and a bit of education in this field, but didn’t complete electric engineering college and I’m not a certified electrician so I’m not allowed to do 230V stuff myself. So, I’m aware I might be wrong and missing something, but I need a third or possible 4th, 5th, 6th person to confirm or disprove my suspicion to be sure.

I have a 3 Phase Setup with 3x MPII 10kva. Inverters are only connected via AC-In, no AC-Out. And AC coupled PV, but the PV doesn’t matter now.

Currently each phase and a combined neutral are going through a 40A 3P+N MCB.

My electrician just said he realized that was a mistake and now wants to swap that out and instead use a 1P+N 20A breaker for each Multiplus. He argues that, in the current setup, the MCB will trip earlier, because N will have higher current than 40A when multiple phases see unequal load. Also, the wiring can only take 60A so he can’t go higher than 3x20A. He still says that I’ll be able to draw more power that way than I’m currently able to do.

However, I’m pretty sure that the phase shift between those phases causes the current on N to balance eachother out to some degree, maybe not completely, due to uneven load, but at the very least it will never go above 40A(rms). I also did a LTspice simulation that confirmed that.

He says this is not a true 3 phase system so they don’t balance eachother out, but rather add up on N and that the spice simulation is wrong, but he doesn’t know how to simulate it correctly.

The easiest solution would be to test that, but we currently don’t have enough consumers in the house and are not allowed to export anything. Plus my whole house is behind a 40A breaker so I can’t even use the forced charging from grid to test it up to those 40A without risking to trip the main breaker. (I might be wrong here? can I test this somehow?)

Thus, I have currently no way to test those claims in practice, but the installation is mostly finished and they want to replace the breakers now, which I want to preven,t because I think that would be worse for me and I don’t want to realize that after we signed off on the install.

This is the current setup:

2026-02-04 20_51_13-LTspice - Draft2.asc
2026-02-04 20_51_13-LTspice - Draft2.asc1320×602 10.8 KB

LS stands for breaker (LTspice doesn’t have the symbols for them) And here is a picture of it just to be sure:

20260204_1828411
20260204_18284111920×2560 459 KB

So, which is correct?

3x Multiplus 10kva in 3 phase setup.

3x 1P+1N 20A breakers and break each neutral separately. (Neutral can only take 60A at most)

or 1x 3P+1N 40A MCB with shared Neutral going through the MCB.

Remember that it’s a 3 phase setup and they are shifted 120° each, which I believe to be the most important point here.

Thanks in advance!

2

Wrong - it won’t. The Neutral will only carry the current equal to the out of balance load. Leaving the input breaker as a 3P OR 3P+N (depending on code requirements) is fine, as if one inverter of a 3 phase group loses AC in, the whole group will typically switch to invert mode. However this behavior can be changed.

Yes, this is true.

Again wrong. IF the source is 3 phase, that is all that matters. The loads can be 3 phase, single phase or mixed.

sounds like your sparky is trying to rip you off for un-necessary work.

2 Likes
3

Thank you very much for confirming!

I personally feel much more confident in my understanding now, but is there maybe some Victron Guide that confirms all this? Would probably be easier to convince them if it comes from Victron officially rather than pointing at a community forum.

I looked through the ultimate wiring guide and 3 phase examples and even though I found their schematic that shows a 3P+N breaker I couldn’t find any text that explains it.

I’m not paying for the time it takes them to fix their own mistakes and I doubt that’s the case. He really seems like he wants to make it right/perfect and seems to think that this would be better for me.

However, if I sign off and then realize its wrong, it might be hard to get them to fix it, especially, if I can’t convince them that their theory is wrong, but now I already have the 40A MCB, so I can at least prevent them from removing it and, worst case, just tell them I accept the current situation even if they don’t believe its okay…

4

@realdemayo

Du scheinst deutsch zu sprechen (LS).

Wenn Du ganz sicher gehen willst, suchst Du einen 4P LS mit nacheilendem N.
Der 4P LS überwacht auch den N. Der 3P+N überwacht die Phasen, schaltet aber auch den N.

Wenn der N nicht nacheilend ist, also nicht nach den Phasen öffnet, kann es zur Sternpunktverschiebung kommen. Dann liegt an einer Phase 400V statt 240V an.

Leider habe ich bei den Hager MBN440 keine Information gefunden, ob der Neutralleiter nacheilend ist.

Das Problem der Sternpunktverschiebung gilt auch beim 3P+N ohne nacheilendem N.

Wie es sich bei 3x 1P+N oder 3x 2P ist, weiß ich nicht.

Minute 15:00:
Der Strom auf N übersteigt nie den Strom der Phase mit größter Leistung

Mehr Theorie:

1 Like
5

Bei rein ohmschen Lasten, bzw. wenn der cosPhi in etwa gleich ist über alle Phasen. Braucht schon eher extreme Werte dazu, alltäglich ist es nicht, “nie” ist jedoch nicht korrekt

2 Likes
6

I have BSc(Hons) in electrical & electronic engineering. Don’t normally push that, but Tell him a qualified engineer says he’s wrong.
However, individual breakers also should not be an issue. But the breakers should all be 40A. Still limits you to ~9.4kVA per phase.
RE: Neutral current : Yes, in theory, if you had different power factors on each phase such that for e.g, red phase =1, blue phase = +120 degrees (blue phase V lags red by 120) and white phase pf -120 degrees , then the currents from all phases are ‘in phase’ with each other, and then the neutral current is tippled. Chance of this really happening: 0.
For true 3 phase loads, the neutral ends up just carrying harmonic currents, and any phase mismatch current. Normally less than 5% of total power on all 3 phases. In these instances, it is permitted to install a neutral conductor of 50% phase conductor size, BUT if you do that, and there is a possibility of an unbalanced load, then the there should be a neutral current sensor that trips the phase breaker (not the neutral) if the 50% current level is exceeded.

2 Likes
7

Danke für die Bestätigung!

hager ist der links daneben, der LS um den es geht ist Siemens, hier das Datenblatt:
2323248B-1-Siemens-5SY6640-7.pdf

ich konnte auf google aber praktisch garnichts über “nacheilenden” Neutralleiter finden, kannst du mir dazu einen Link geben?

Auch danke für den Hinweis, dass die meisten MCBs den N garnicht messen, das war ja eigentlich der Hauptgrund für den Tausch.

translation:

clarified that the MCB in question is siemens (see datasheet above)

couldn’t find any info on MCB with “lagging/delayed” neutral which was supposed to be better

realized my MCB doesn’t actually protect neutral so it can’t trip due to overcurrent on neutral…

8

Stick with the 3 phase +n 40A breaker.

1 Like
9

I am not a professional, but if I read the symbols on the hi-res picture correctly, all 4 conductors are measures and switched.
You see a coil and a switch per phase and N and the four are mechanically linked.

I also did not found any information a latching neutrals. For my transfer switch, I got confirmation by the technical specs of the manufacturer, but for MCB I did not found any.

Maybe that’s the reason why 3p+N exist for real 3p loads and 4p MCB exist for unbalanced 3x 1p loads.

But please double check with the manufacturer.

11

Not really. 3p breakers exist for regions where the wiring code insists that the Neutral conductor remain unbroken, this is normally TC-N grounding scheme (If I remember correctly), other regions (mainly European) insist on breaking the neutral conductor too. So the use of 4p as opposed to 3p breakers is not dependent on the load, but the wiring code.

12

@MikeD
as I understood… but please correct me if I am wrong.

A 3P+N breaker measures only the three phases but disconnects all three phases and the neutral. Though the neutral does get opened.
The 4P breaker measures the three phases and the netural and also opens all three phases and the neutral.

The arrow indicates measurment. The disconnected line is the switch.

3P+N Hager MBN640

image
image540×301 15.5 KB

4P Hager MBN440

image
image540×329 16 KB

13

Even so, you would not use 4p breakers for single phase loads, here you would use 1 or 2p breakers according to code. In such cases, the main switch / breaker would be a 4p device with current sensing on each of the 3 phases. Neutral current sensing is fairly rare, as explained above - in virtually all 3 phase systems, the phase currents exceed the neutral current.

14

I agree that both a 3p+N and 4p will work in most situations. I and the referenced videos did not cover cos phi != 1 in conjunction with taking from grid on one phase and feedback to grid on other phase.

In my system I haven’t seen any annoing N-currents yet.
On the other hand my 3p MP2 5k are connectd 10qmm and have 3P+N MCB 50A. The grid connectin is 63A (NH00) wiht 16qmm and protected by 35A SLS (MCB). I ssume my MP2 will die before any of my calbles get even warm :wink: