Ze Readings bouncing all over the place

[old_sparky_RE]

At the sub feeder panel I have a reading of 0.15 on L1 & L2 and on L3 a reading of 0.16 ohms at the start of the circuit off the 50A MCCB.
System in TN-S

At the CDP the reading is bouncing between the following
L1 0.6 to .88, L2 0.68 to 0.8 and L3 0.62 to 0.78 ohms. but remains at 0.15 at the feed end. This is causing the readings on the radials to be between 1 and 4.08 ohms Zs. So pass and fail dependent when you hit the test button!!!

The distance is 97m, 35mm 4core SWA with the SWA used as the earth return. The cable is incassed in concreat along parts of its route and is under the main sewer pipe for about 30m too.

R1 + R2 is high for this length of run at just below 0.8 ohms.
IR readings are all greater than 500Mohms between all conductors and the conductors and the SWA.
All nuts / bolds / screws have all been cleaned off and re tightened.
Glands replaces paint ground off, bango's bolted through the body with 25mm pig tails used and the Incrediable Hulk hired to re tighten everything.

But the results at the CDP are still bouncing all over the place.

So my reasoning is that there is a joint somewhere along the line that is failing, but then the result would be just high not bounching all over the place.

Note: 4 different test meters have been used (Fluke 1652/3) and all show the reading going all over the place.

Other than replacing the SWA and adding an additional earth.... what is the cause and another solution please?

 

[telectrix]

i'd suspect that the outer sheath has been damaged and the armour is corroding. try an armour - soil test. ( forget the technical term ). obviously, diss both ends of the armour from the system earthing.

 

[old_sparky_RE]

I thought that as the R1 + R2 value was high (and it ia high Z reading too for a TN-S) but why would the readings be bouncing around? Surely the readings would be constantly high not 0.72 then 0.78, 0.88 ,0.69 etc.

What could be happening to cause this fluctuation has mebaffled?

 

[UKMeterman]

I thought that as the R1 + R2 value was high (and it ia high Z reading too for a TN-S) but why would the readings be bouncing around? Surely the readings would be constantly high not 0.72 then 0.78, 0.88 ,0.69 etc.

What could be happening to cause this fluctuation has mebaffled?

If you are testing live then distorions in the AC waveform could account for the changes. What happens when you dead test the cables?

 

[old_sparky_RE]

R1 + R2 was done dead so was the IR testing.

 

[Simonslimline]

i'd suspect that the outer sheath has been damaged and the armour is corroding. try an armour - soil test. ( forget the technical term ). obviously, diss both ends of the armour from the system earthing.

Sheath test?

 

[telectrix]

Sheath test?

that's the one. a sort of sophisticated IR test between the steel armour and the soil surrounding it.

 

[UKMeterman]

R1 + R2 was done dead so was the IR testing.

Ok, using a senstive mulitimeter just before doing the resistance test check that it is 0V between the two test points, you will need to use the millivolt scale on the multimeter. Check for both DC and AC voltages. What test current are you using to make these measurements, 200mA?

 

[old_sparky_RE]

Sorry for the delay in replying but I was way on another job.
I have a well worn and abused Fluke 1652 and 3 so no sensitive kit here.
From the tec plate on the tester 4VDC <Uo <24VDC Rlo< 2ohms In>200mA.

I've spoke with the powers that be and a new cable will be laid in the next few days.
I'm going to put a call out see if we have a sheath tester at one of the locations and if it turns up in time I'll see if i locate the damaged area.

My worry is that because this place was maintained by American 'electrician' though the system in 50Hz there are Earth rods all over the place and injecting 3-5kV into the area may be a bit ...fun.
So I may never get the chance to test.

 

[UKMeterman]

Sorry for the delay in replying but I was way on another job.
I have a well worn and abused Fluke 1652 and 3 so no sensitive kit here.
From the tec plate on the tester 4VDC <Uo <24VDC Rlo< 2ohms In>200mA.

I've spoke with the powers that be and a new cable will be laid in the next few days.
I'm going to put a call out see if we have a sheath tester at one of the locations and if it turns up in time I'll see if i locate the damaged area.

My worry is that because this place was maintained by American 'electrician' though the system in 50Hz there are Earth rods all over the place and injecting 3-5kV into the area may be a bit ...fun.
So I may never get the chance to test.

Was the armor disconnected at the load end? I strongly suspect that the "many earth rods" are causing a circulating current in the armour and this is causing a small voltage drop that is upsetting your measurements.

I would suggest
1. Investing in 114 - FLUKE - MULTIMETER, DIGITAL, HAND HELD, 6000 | Farnell element14 far cheaper than a new cable!
2. Also using the megger disconnect the armour at both ends and then test at 500V for 1 min and see which values you get at start and end of test.

 

[davesparks]

Sorry for the delay in replying but I was way on another job.
I have a well worn and abused Fluke 1652 and 3 so no sensitive kit here.
From the tec plate on the tester 4VDC <Uo <24VDC Rlo< 2ohms In>200mA.

I've spoke with the powers that be and a new cable will be laid in the next few days.
I'm going to put a call out see if we have a sheath tester at one of the locations and if it turns up in time I'll see if i locate the damaged area.

My worry is that because this place was maintained by American 'electrician' though the system in 50Hz there are Earth rods all over the place and injecting 3-5kV into the area may be a bit ...fun.
So I may never get the chance to test.

You don't need a 'sheath tester' just an IR tester. Isolate the armour from earth and test at 250V between the armour and the actual ground via a test spike or convenient bit of buried metal.

There shouldn't be any need to be testing LV cables at 3-5kV!

 

[conrad]

Thats not high for tn-s system.

 

[old_sparky_RE]

Was the armor disconnected at the load end? I strongly suspect that the "many earth rods" are causing a circulating current in the armour and this is causing a small voltage drop that is upsetting your measurements.

I would suggest
1. Investing in 114 - FLUKE - MULTIMETER, DIGITAL, HAND HELD, 6000 | Farnell element14 far cheaper than a new cable!
2. Also using the megger disconnect the armour at both ends and then test at 500V for 1 min and see which values you get at start and end of test.

The cable was IR tested with both glands / pig tails disconnected from the DB's and >500MOhms was recorded between phases and Earth.
Each building in the area has multiple A/C units and Signals equipment so there is current and voltage flooding the local area. so that would account for the discrepancy's.

CAble has already been laid and will be installed this afternoon or this evening ( we have cable everywhere).
Testers and good tools, we don't have, and its a very long supply chain to get what we want (4000miles and usually 3-4months).