Testing Question

[Sidewinder]

OK,

I was doing a job a couple of weeks ago in a very controlled premises.

Once complete, no certification was required, however, I needed to prove that there was still an acceptable earth fault loop impedance at the point of works.

No permit to open up the DB to do any works within the DB, no requirement for IR, or R1+R2, due to the nature of the works.

No neutral present at the point of test, it was a balanced 3ph load, so, only 3 phases & cpc present.

Incomer to the DB 30mA instantaneous 4 pole RCD.

Not supposed to trip off the power.

However, I need to do a circuit Zs @ the point of work.

Meter is a Fluke 1653B, not a cheap tester, but, it does a 3 wire non-tripping loop test.

I can't think of any way that I can configure the meter to do a 2 wire non-tripping loop test, conventionally, or unconventionally.

Now I CAN'T open up the DB to bypass anything, plus that would not be a true Zs.

I HAVE to do the Zs measurement before I leave site and record this.

I could not think of any way to do it, checked the meter instructions, could not see anything obvious, I was in a rush, I admit.

So I just used my Megger LTW425 2 wire non tripping loop tester to get a value.

Not before I hasten to ad I had tripped the RCD twice!

Client not happy, but, managed to talk them round, still no chance of opening the supply board though.

I can't see how to do what is basically a 2 wire non tripping earth loop test on a Fluke 165xx series tester.

Anyone have any ideas?

Before I go to Fluke with this?

I have thought about this a little, not too much because I used the LTW to get a reading, and have the reading needed now, but, it would be nice to know!

 

[Blue Duck]

Can you not change the parameters of the zs test on the fluke?

Been a while since I done any 3 phase testing

Not much in the manual about it either ?



Your client must be one serious geezer to not let you fettle with the db

 

[NozSpark]

Extension lead off a socket circuit fed from same incommer db and use that neutral?

You have to have a neutral so that the tester can saturate the RCD

 

[Andy™]

short L-E. if it trips, all good. if it doesnt trip, not so good

 

[Sidewinder]

Extension lead off a socket circuit fed from same incommer db and use that neutral?

You have to have a neutral so that the tester can saturate the RCD

OK, possible, but no 1ph loads on the DB, only the 4 3ph machines.

Can't open the DB to check what is in there.

However, no sockets fed from the 4pole RCD that I could connect to.

Good idea mate, but... ; (

short L-E. if it trips, all good. if it doesnt trip, not so good

So what is the EFLI measured by that test for me to record it Andy?

Also, IF it does not trip, I'm relying on the B32, may not be so good, if, the fault current "flash" is big enough to cause any sort of arc, as that might have set all the wood dust & wood flour on fire about the place.

Now that would have been a bit of a claim, plus, I don't thing Her Majesty would be too pleased about me mucking up one of her hotels!

 

[Andy™]

Can you not do a 3 wire test on a 2 wire (i.e. N+E) connected to E

Pretty sure the fluke does this

too much current L-E, it would trip the RCD

So what is the EFLI measured by that test for me to record it Andy?

if its trips,< max Zs for MCB. if it doesnt trip, > max Zs for MCB

 

[Sidewinder]

Can you not do a 3 wire test on a 2 wire (i.e. N+E) connected to E

Pretty sure the fluke does this

I can't find a way of making the Fluke do a 2 wire non-tripping test mate.

If you know better please tell.

I tried it without the N, with the N on E & another way I can't recall now, as I was in a rush, I had to get out of the premises by a certain time.

N&E from the meter together on E, tripped RCD.

No N from the meter connected, would not test.

The "other" way I dreamt up tripped RCD, though I can't remember what I did!

 

[Andy™]

no idea if it would work, but could you get 230v by putting a transformer between 2 of the lines?

 

[Sidewinder]

too much current L-E, it would trip the RCD

if its trips,< max Zs for MCB. if it doesnt trip, > max Zs for MCB

Like that but, a bot dodgy for me and, not really accurate enough.

As I said, I don't think Her Majesty would like me possibly destroying one of, or part of one of her luxury hotels, IYKWIM!

no idea if it would work, but could you get 230v by putting a transformer between 2 of the lines?

It would generate an N point, but, there was not one there at the time, & I needed to get the test done so I could leave!

Else, I could have been asked to take one of the "Hotel" rooms, and, if it went wrong the same!

 

[NozSpark]

The only other way would be to connect the neutral to the star point of the "machine" you're testing at

 

[Andy™]

The only other way would be to connect the neutral to the star point of the "machine" you're testing at

what if there is no star point?

 

[NozSpark]

Wouldn't work anyhow

 

[Andy™]

I am pretty sure you do this

Fluke - Test

L        -     L

N       -     E

E        -    E

3 Wires in Use on Fluke but connected to 2 wires at Test point

I am pretty sure you do this

Fluke - Test

L        -     L

N       -     E

E        -    E

3 Wires in Use on Fluke but connected to 2 wires at Test point

but L-N has a much higher current than will trip an RCD, so connect the N to E and it will trip

 

[NozSpark]

You just have to have a neutral

 

[Blue Duck]

I am pretty sure you do this

Fluke - Test

L - L

N - E

E - E

3 Wires in Use on Fluke but connected to 2 wires at Test point

That is what I meant ...not that it's any help :|

 

[Sidewinder]

Fluke - Test (I take it you mean circuit under test?)
L - L
N - E
E - E

Tried this tripped the RCD on no trip, & high current settings.

Perhaps these were two of the options I tried and could not remember one of them.

I've slept since then!

 

[Andy™]

Other option is 3 wire and only connect up 2 wires from tester

I have done it - can't remember now which way i did it - Was no RCD on circuit i tested though

cant say for fluke, but megger will not test if the N isnt connected and its on non-trip

 

[Sidewinder]

You can do it on high current test, it only looks for two wires then, did it today.

However, I can't find a way of doing a Line (Phase) to cpc Zs test without having a neutral present with the Fluke.

Oh & found out for definite today that it is RUBBISH @ low values of Ze!!!

LTW425 is MUCH better at that,

cant say for fluke, but megger will not test if the N isnt connected and its on non-trip

LTW425 is a 2 wire non-tripping Megger loop impedance tester, can't speak for the Megger MFT's 15xx & 17xx series?

 

[steptoe]

The way I understand the fluke is it sort of charges a cap to bleed enough current for the Zs test while still keeping the RCD balanced via the N, 

So, it NEEDS the N probe on the N to keep the RCD balanced.

I don't think it's possible to test without it. 

 

[NozSpark]

The only way that I can think of that has half a chance of working would be to load up the earth impedance with an 8k resistor and then deduct that value from the result,,, but I think you'll end up losing resolution