Fitting a 3-phase meter

[Corazones]

My electrician has just fitted a 3-phase meter to my 3-phase supply, as I need to separately meter the consumption from a heavy-duty well pump. The meter has the usual connections - terminals 1/3 for L1, 4/6 for L2, 7/9 for L3, and 11 for neutral.

He connected each of the 3-phase cables in turn; when he connected the third one there was a short circuit and a lot of smoke, so he removed it. The pump had been isolated at the time, so this wasn't responsible.

The meter was bought in Germany by a neighbour, who swears the equipment is OK and the electrician must be at fault. I'm stuck in the middle and trying to sort out what might be the problem.

How important is it that Phase 1 (supposedly brown) is connected to the L1 terminals, phase 2 (black) to L2, and phase 3 (grey) to L3? The electrician is trying to convince me that it doesn't matter, but I can only imagine that maybe they were connected differently - maybe even due to incorrect cable colouring.

How can I find out who or what is at fault?

 

[Lectrician]

Standard polyphase meters usually (never seen otherwise) have the terminals presented at the bottom, with left to right, L1 Supply L1 Load, L2 Supply, L2 Load, L3 Supply, L3 Load, N and N.

Are we talking a 80/100amp meter here, or a panel or din rail mount sort of meter?

 

[RF Lighting]

They don't half sound like CT meter connections

 

[Spark123]

It won't matter which particular phase is connected to L1, which one is connected to L2 and which is connected to L3. As long as the ins and the outs are correct, which for what you describe sounds like what may have happened or there is some sort of an internal fault. The phase sequence will matter at the pump though (otherwise it will run backwards.)

 

[Corazones]

Thanks for your advice on this.

Lectrician, the terminals are exactly as you describe, with a removable link between supply and load in each case. It's an independent box unit, a German make (Heliowatt EM10) which looks identical to the Siemens type, 10/40A. Any idea where I can download installation instructions?

Spark123, this is just what the electrician told me. It's good to have this confirmed. As the smoke came from inside the meter, I suppose I should cut the seals and see where the problem lies - unless that will invalidate the guarantee, maybe...

 

[plugwash]

what is the make and model of this meter?

 

[Spark123]

Opening it up will probably invalidate the guarentee as you say. Is it a digital meter? If it is the old analogue clockwork ones then there shouldn't be anything stopping your electrician insulation testing (aka meggering) it to see if there is an insulation problem.
Can you post up a piccy of the offending meter?

 

[Corazones]

It's not a digital type. It would take a while for me to organise a pic but if you search on Google under "Heliowatt EM10" you'' find a clear picture of one offered on German ebay, if that helps. It's a heavy plastic box, connections along the bottom, number display and revolving dial in the window above.

I could run a resistance check if that would pinpoint the problem, but between which terminals?

 

[Corazones]

I've now tested the resistance between all terminals, and I think I've located the fault. Between supply and load of each pair (phase 1-1, 2-2, 3-3 and N-N) there's a normal 6 ohms, just the coils of the meter. Also between pair 1 and pair 2 is completely open circuit, similarly between pair 3 and N is open. BUT... between 2 and 3 is a short of 18 ohms. And the smoke occurred when the phase 3 cable was connected!

Can someone please confirm this definitely seems to be a fault - and if so, what on earth could be causing it?

 

[plugwash]

6 ohms also sounds extremely high for a meter designed for direct mains use.

 

[Corazones]

Oops, you're right - that should be 0.6, I overlooked the decimal point on the digital reading. It also means that the short is only 1.8 ohms! Could it be more than coincidence that's the total resistance of the three meter coils, and indicates incorrect internal wiring? I would have thought that most of these connections were copper strips, which couldn't be wrong, but then I've never taken one of these things apart.

 

[Spark123]

Without opening it up it isn't possible to tell as there will be internal wiring for it to detect the voltage of the phases, so I suppose even IR testing it may not show up much. Are there any markings on the meter i.e. rated voltage etc?
Opening it up will probably ruin any warranty, could it have been damaged in transit etc? You can buy new meters from the likes of TLC for about £100 http://www.tlc-direct.co.uk/Products/RDCRED3slash100.html

 

[plugwash]

Oops, you're right - that should be 0.6, I overlooked the decimal point on the digital reading. It also means that the short is only 1.8 ohms! Could it be more than coincidence that's the total resistance of the three meter coils

What are you using to make the resisance measurement, ordinary multimeters cannot measure accurately at such low resisances.

 

[Corazones]

Plugwash - I'm sure you are right, as it was indeed a multimeter, but quite a good one and used at its lowest ohm setting (200). But the readings were at least consistent, and the whole point is that surely there should be no inter-phase connection at all?

 

[Spark123]

A wattmeter needs to measure voltage as well as current hence I'd expect there to be measurable connections between the phases and neutral within the device.