Upgrade to 3 phase

[ban-all-sheds]

As I said, the 'worst case' of (b) (the worst possible phase imbalance) is to have current flowing in only one of the three phase conductors, and no neutral.

And I'm struggling to understand how one could have that, how one could have current flowing in only one of the line conductors, unless it were flowing to earth.

 

[EFLImpudence]

Even without a neutral the 3ph cable will be balanced, or else the RCD would trip.

Which I'm afraid I did (and still do) struggle to understand.

Two different kinds of balanced are being confused.

 

[SUNRAY]

Indeed, but when I wrote that the cable would only be balanced if it does contain "all of the current carrying conductors in a circuit" (i.e. including a neutral - I wrote "if it had a neutral"), you responded with ...

Even without a neutral the 3ph cable will be balanced, or else the RCD would trip.

Which I'm afraid I did (and still do) struggle to understand.

Kind Regards, John

OK like so many of these threads I think we are at crossed purposes.
Hopefully this will clear the confusion:

No.1 is a multicore cable with a neutral. The CABLE and the RCD will be balanced regardless of the balance of the PHASES

No.2 is a multicore cable without a neutral. The CABLE and the RCD will be balanced regardless of the balance of the PHASES

No.3 is a multicore cable with a neutral. The CABLE will be balanced regardless of the balance of the PHASES, but the RCD will trip if there is any sort of imbalance of the PHASES .

So far the magnetic fields around the CABLE are minimal.

No.4 where the neutral is run separately for some reason, or the whole thing is in singles [very typical with temp genny feeds], the RCD is balanced and happy but the magnetic fields are potentially a problem.

Hopefully we are now on the same song sheet.

 

[SUNRAY]

Two different kinds of balanced are being confused.

Indeed, I'm hoping that post #63 [which I started 3 hours ago] has clarified the confusion.

 

[EFLImpudence]

It does.

 

[JohnW2]

OK like so many of these threads I think we are at crossed purposes. .....
No.1 is a multicore cable with a neutral. The CABLE and the RCD will be balanced regardless of the balance of the PHASES...

Agreed.

No.2 is a multicore cable without a neutral. The CABLE and the RCD will be balanced regardless of the balance of the PHASES

I don't really get that one. Unless the currents in the three phases are identical, there will surely be a finite vector sum of the currents, won't there (i.e. the cable is 'not balanced'). Think of fairly extreme situations, in which there iis little or no current in one or two of the phases. In such a situation, if only the three phases ( but not neutral) went through the RCD (per your diagram), the RCD would trip, wouldn't it?

In any event, an RCD which had only phase conductors going through it would surely serve no useful purpose - since it would not detect phase-neutral imbalanceds due to faults to earth?

Kind Regards, John

 

[davelx]

Agreed.
I don't really get that one. Unless the currents in the three phases are identical, there will surely be a finite vector sum of the currents, won't there (i.e. the cable is 'not balanced').

Doesn't the magnitude of the vector sum of the currents in the 3 phases always equal zero in this case?

Edit - 'cos, referring to EFLI's post below - if there is no neutral then there is nowhere for any unbalanced current to flow. You can't connect a load between phase and neutral if there is no neutral, so all loads must be phase-phase.

 

[EFLImpudence]

Re: #2

I don't really get that one. Unless the currents in the three phases are identical, there will surely be a finite vector sum of the currents, won't there (i.e. the cable is 'not balanced'). Think of fairly extreme situations, in which there iis little or no current in one or two of the phases. In such a situation, if only the three phases ( but not neutral) went through the RCD (per your diagram), the RCD would trip, wouldn't it?

If there is a neutral not running through the RCD, it would, but not if there is no neutral.

In any event, an RCD which had only phase conductors going through it would surely serve no useful purpose - since it would not detect phase-neutral imbalanceds due to faults to earth?

Surely it will still detect residual current.

The test button on three phase RCDs does not use the neutral.

 

[JohnW2]

Doesn't the magnitude of the vector sum of the currents in the 3 phases always equal zero in this case? ... Edit - 'cos, referring to EFLI's post below - if there is no neutral then there is nowhere for any unbalanced current to flow. You can't connect a load between phase and neutral if there is no neutral, so all loads must be phase-phase.

I was not talking about a situation in which there "is no neutral" (at all) - but, per what SUNRAY appears to have been talking about, a situation in which there is not a neutral in the same multicore cable as the 3 phase conductors (i.e. as per case 4 in his post #63).

Kind Regards, John

 

[EFLImpudence]

I was using this for information:

On further investigation, I find some which do (or show they do) connect the test button circuit to the neutral, so ...

 

[JohnW2]

If there is a neutral not running through the RCD, it would, but not if there is no neutral.

Indeed, but we seem to have 'cross-purposes' again, since I've been talking about a situation in which there was a neutral, but not in the same cable as the phase conductors.

Surely it will still detect residual current.

Again, seemingly cross-purposes. If the circuit has a neutral, it must pass through the RCD. Apart from anything else, if it didn't, the RCD would trip unless there were perfect phase balance.

The test button on three phase RCDs does not use the neutral.

Dunno. I've never used them, let alone taken one apart (I use separate L+N RCDs for each phase in my house). What does the test button do, then?

Kind Regards, John

 

[ban-all-sheds]

I don't really get that one. Unless the currents in the three phases are identical, there will surely be a finite vector sum of the currents, won't there (i.e. the cable is 'not balanced'). Think of fairly extreme situations, in which there iis little or no current in one or two of the phases. In such a situation, if only the three phases ( but not neutral) went through the RCD (per your diagram), the RCD would trip, wouldn't it?

If you have no neutral, and no current in two phases, where is the circuit which you need for there to be current in the third phase?

 

[JohnW2]

Indeed, but we seem to have 'cross-purposes' again, since I've been talking about a situation in which there was a neutral, but not in the same cable as the phase conductors.

 

[ban-all-sheds]

But you were not talking about that situation in your reply to Sunray.

No.2 is a multicore cable without a neutral. The CABLE and the RCD will be balanced regardless of the balance of the PHASES

I don't really get that one. Unless the currents in the three phases are identical, there will surely be a finite vector sum of the currents, won't there (i.e. the cable is 'not balanced'). Think of fairly extreme situations, in which there iis little or no current in one or two of the phases. In such a situation, if only the three phases ( but not neutral) went through the RCD (per your diagram), the RCD would trip, wouldn't it?

Look at the drawing, and compare it to #3 & #4.

#2 is not "a situation in which there was a neutral, but not in the same cable as the phase conductors", that's #s 3 & 4.

#2 is the situation where there is no neutral anywhere.

 

[mwatsonxx]

hahaha, once again none of them know what the other is talking about, the waffle has got too deep.

And drawing no3 is nonsense, because it is a ficticious circuit that would never be used, any sp load will trip the rcd

EDit - Didn't realise the drawing was ficticious