How Does An Earth Loop Tester Work?

[davetheglitz]

Sounds a stupid question but even forgetting about RCD's I can't get my head around this.

I assume the measurement is derived from a sequence of tests where voltage is measured on no load (so v = line voltage) and then voltage is measured under load (L-E load) so it is like a potential divider. A few equations later bish bash bosh you can get Ze out as everything else is known. As there is only a very small L-E resistance/impedance in a system there is only a small chance of any error.

My problem is that the same meter can measure Ipf which is most probably I l-n. If there is a  stonking great l-n parallel path then there are now 2 unknowns Ze and the parallel resistance/impedance. I can't think of a way of separating out the fault path resistance and the parallel resistance to get a meaningful measurement.

Sorry if this is a bit esoteric - but I sometimes think we take these sort of readings or granted and we accept what the meters say without any understanding any methodology.

Has anyone got any links or insight into this.

 

[binky]

you worry too much

perhaps a question to megger or fluke may be in order???

 

[davetheglitz]

From the lack of replies I obviously do!

Been watching too much of The Big Bang Theory!

 

[kerching]

When I was an apprentice...loooooooooooong time ago. 40 years probably! We HAD to buy a loop tester . All that anybody knew ( foreman kept everything a secret ). Was that " you plug it in and if the needle only moves a bit then it is ok "

Must have been years before we found that it did not actually work, probably been faulty for years

Makes you realise why you need to check them periodically

 

[sdjl]

A little search online reveals this:

In simple terms, the impedance of the phase-to-earth loop is measured by connecting a resistor (typically 10 Ohms) from the phase to the protective conductor. A fault current, usually something over 20 A, circulates in the fault loop, and the impedance of the loop is calculated within the instrument by dividing supply voltage by the value of this current. The resistance of the added resistor must be subtracted from this calculated value before the result is displayed. An alternative method is to measure the supply voltage both before and whilst the loop current is flowing. The difference is the volt drop in the loop due to the current, and loop impedance is calculated from voltage difference divided by current.

Located here -  http://www.tlc-direct.co.uk/Book/8.6.2.htm

 

[ProDave]

As to how they work, they draw a known load and measure the volt drop, and from that calculate the loop impedance.

For PEFC the measurement is done L-E for PSSC the measurement is done L-N

What more do you need to know?

 

[Evans Electric]

Sometimes there is a good case for just pressing the button and taking the reading  TBH  , Glitzy .

If you keep thonking about all these things ...your head will explode.

 

[ProDave]

Sometimes there is a good case for just pressing the button and taking the reading  TBH  , Glitzy .

If you keep thonking about all these things ...your head will explode.

I think yours already has 

 

 

[Evans Electric]

I thonked it had !!

Fancy putting the i next to the o .

 

[davetheglitz]

As to how they work, they draw a known load and measure the volt drop, and from that calculate the loop impedance.

For PEFC the measurement is done L-E for PSSC the measurement is done L-N

What more do you need to know?

L-E using that method is fine - with everything in the system the earth current should be negligible so the reading will be accurate.

L-N - anything in the system could be drawing current - so now you have 2 unknowns - Zs and the parallel impedance. In other words the applied load is now not known. There is something missing in the method.

 

[binky]

not suppossed to do test with board enegised for Ze and PFC PSCC, so no load and in theory no parellel paths. Main earth is also suppossed to be out of board. Hope your teaching the lad how to test properly ;

 

[steptoe]

I thonked it had !!

Fancy putting the i next to the o .

who'd'a thunk that?

 

[kerching]

I thonked it had !!

Fancy putting the i next to the o .

An Old Farmer around these here parts could never work his typewriter. Being a clever sod he reworked the keys so that the more often used ones were adjacent and repeated. He had. E-I...........E-I.............O

Not seen Old Man McDonald for ages now

.......just thonking

 

[ProDave]

There's plenty of McDonald's around here.  I've even met Donald McDonald, who would call their son that?

 

[steptoe]

funnily enough we had our son out of wedlock!  SHOCK HORROR !!!!!!

and our sons name was very close to my wifes maiden name, although he has always had my name, still, got her a few looks when she was asked her name, then our sons first name, and she then had to explain that his last name was actually xxxxdifferent.

 

[steptoe]

d'ya think she'd go halves?

 

[davetheglitz]

not suppossed to do test with board enegised for Ze and PFC PSCC, so no load and in theory no parellel paths. Main earth is also suppossed to be out of board. Hope your teaching the lad how to test properly ;

Of course - but Zs at a light fitting? I've never seen a recommend that all other circuits should be off. Must try a Zs test at a cooker panel with the cooker on full and off to see if there is a difference.

 

[pewter]

not suppossed to do test with board enegised for Ze and PFC PSCC, so no load and in theory no parellel paths. Main earth is also suppossed to be out of board. Hope your teaching the lad how to test properly ;

i thought PEFC and PSCC are ment to be tested with the all earths/bonding connected so you will get the highest result that might be possible under fault conditions. If the main earth was disconnected from the installation to get rid of parallel paths then you would get a lower result and not a true representation of what would happen under fault conditions.