How to find it?

[kme]

This might get interesting - or I`ve had a "brain fart", and am starting to lose the plot.............

Imagine an intake panel for a TPNE TNC-S; 800A/phase property; metallic in structure.

On one end is a serviceway; with ten SWAs of various sizes, terminated to the top surface, running vertically down to their respective fuses. (from 50mm up to 150mm 4 core)

There are both Neutral and earth busbars running down the side, and along the bottom, of the chamber.

The earth busbar is, at regular intervals, bolted to the chassis. The neutral is bolted to the earth, through insulating spacers.

If I clamp the earth busbar where I can get the jaws around it, I can read a maximum of 28A returning current - I can clamp around two of the supports for the busbar - each one has in excess of 3A flowing between the busbar and chassis.

I cannot clamp around all of the individual earth cables which attach to the busbar (due to proximity to other cables).

I cannot isolate ANY of the outgoing circuits.

I did manage to clamp the supply cables (including neutral), and found 2 with approx 5A each "missing".

This still leaves me with around 18A (at least) unaccounted for.

Thoughts / suggestions would be appreciated ...........:coffeeheadbang

 

[NozSpark]

Can you clamp the L & N cables at all?

 

[kme]

Can you clamp the L & N cables at all?

Mate:

I did manage to clamp the supply cables (including neutral), and found 2 with approx 5A each "missing".

That was what I was getting at there bud - I`ve "found" 2 of the cables which have an imbalance; but, of the rest, they either don`t show an imbalance, or I can`t get to all the conductors. Sorry -I didn`t explain that too well - blame messrs. Artois

 

[Jono Pashley]

N-E reversal somewhere? may be talking rubbish cant explain how you can have more coming out than going in tho.

Another thought one or more of the incoming cables all live conductors not brought in through the same hole in the metal enclosure and causing a eddy current again may be total rubbish but..

Some chimp has used one of the SWA armours as a neutral conductor? seen that...

 

[kme]

1. We don`t have more coming out - I can`t measure all cables; but there is at least 28A of returning earth current to the main panel- - I`m trying to find out where from!

Most of the supplies that leave are SWA.

2. The SWA`s in question are (correctly) terminated through the roof of the panel - so not an eddy per se ; but there may be a circulatory effect causing some compound issue - good thought.

3. I`m 99% sure the earths are just that. Being TN-C-S, and unable to isolate makes testing the hypothesis rather difficult though

keep it up - this is all helpful. When you`ve discounted everything else, you`ll have something left over which must be the root cause (blatant paraphrase of A.C.Doyle)

 

[sb_89]

Is there a high number of electronic devices ? I'm thinking appliance earth leakage. Pc's for example. 26A is very high but with an 800A 3p supply I'm guessing it's a rather large building, whats the usage of the building ?

 

[apprentice87]

Hi kme,

Just a newbie idea here...... You say the supply is TNCS to a metal framed building.... Sooooooo, the frame of the building is connected to neutral........

What else is built in the area?? I presume the point where you are measuring the earth currents is between what ever passes for an earth terminal in a humongously ginormous installation, and where it connects to whatever passes for a MET in a humongously ginormous installation. In other words, halfway along what would pass for an earthing conductor in a "normal" sized installation.

Anyway, this returning earth current. Is there any possibility that it is actually NOT "your" current at all????

Could it be currents from someone elses PME installation, one with possibly a not very good neutral, or ANY installation, with perhaps, an earth fault, taking a "short cut" up into your metal frame, through your earth conductor, and down your neutral.....

I know "diverted" neutral currents do flow though the ground and through "other peoples" bonding conductors under certain conditions, as, in the case of where you have, say, two separate structures in contact with the ground and bonded together, certain precautions have to be taken in respect of the cabling to provide for this eventuality.....

Then again, i might be talking nonsense.....

[Just an idea for you...]

john..

 

[Sidewinder]

App,

I don't think there is anyone else's N current flowing as the building has its own 11kV transformer outside IIRC.

However, that "may" feed other premises, but they will all be similar in nature.

There will be good earth bonding due to the structure.

I would think about harmonics.

KME,

Which meter did you use?

What was the N current? (Measured or Calculated?)

I'm guessing you measured the earth current in the NE link at the read of the DNO section of the panel board?

Same as we did in the other big job we looked at?

---------- Post Auto-Merged at 06:53 ---------- Previous post was made at 06:50 ----------

When you do your part sectional shut down can you nip in and measure any currents?

This would at least eliminate the area where you are changing the dis board, which could be a good shout for leakage?

BTW,

Roughly what was the phase current when you did your measurements?

 

[sb_89]

I might be totally wrong here, and shoot me down in flames if I am but, could you put your clamp around each armoured cable after the gland inside the unit. So you would have the inner cores of the cable - line,neutral and earth inside the clamp at the same time. Live and neutral would cancel each other out and you would be left with what is on each earth. Providing you have the space and that each cable is'nt relying solely on the armouring for an earth.

 

[kme]

Is there a high number of electronic devices ? I'm thinking appliance earth leakage. Pc's for example. 26A is very high but with an 800A 3p supply I'm guessing it's a rather large building, whats the usage of the building ?

There are a few computers & servers around - its a 140 bed hotel, with associated laundry, kitchen, services etc. 26A is too high a leakage for the number of items, IMO

Hi kme,Just a newbie idea here...... You say the supply is TNCS to a metal framed building.... Sooooooo, the frame of the building is connected to neutral........

Whoa John. The building isn`t metal frame. The main distribution and intake panel is a large metal cabinet. The building is brick mate.

What else is built in the area?? I presume the point where you are measuring the earth currents is between what ever passes for an earth terminal in a humongously ginormous installation, and where it connects to whatever passes for a MET in a humongously ginormous installation. In other words, halfway along what would pass for an earthing conductor in a "normal" sized installation..

Good thoughts within your post mate - we have a dedicated 11KV sub, so it has to be coming from us ( I think)

I would think about harmonics.KME,Which meter did you use?What was the N current? (Measured or Calculated?)I'm guessing you measured the earth current in the NE link at the read of the DNO section of the panel board?Same as we did in the other big job we looked at?

Yup, that was it.The original fault was picked up on your power analyser. The readings now are coming from the Fluke clamp or the Alphatek clamp, depending on which jaws will go round which cables / busbars. FYI the Alpha is TRMS.The bigger cables - I have to use the Skytronic clamp, cos the jaws are massive, but low current readings are unreliable. Where possible, I clamped all three on the same thing simultaneously, to verify readings.

When you do your part sectional shut down can you nip in and measure any currents?This would at least eliminate the area where you are changing the dis board, which could be a good shout for leakage?

Agreed - hopefully can be done

BTW,Roughly what was the phase current when you did your measurements?

I`ve got the original data from the analyser - stand by. Iirc, there was quite a bit of harmonic; and some pure DC coming back.

I might be totally wrong here, and shoot me down in flames if I am but, could you put your clamp around each armoured cable after the gland inside the unit. So you would have the inner cores of the cable - line,neutral and earth inside the clamp at the same time. Live and neutral would cancel each other out and you would be left with what is on each earth. Providing you have the space and that each cable is'nt relying solely on the armouring for an earth.

Well, you`d really want to get around each SWA, without it having an earth conductor mate - any imbalance would have to be "lost" current - none of these cables have an internal earth core - they all use either an external 6491X, or rely solely on the SWA armour. So lives and neutral are effectively cancelling - although when I was doing this, I found a 160A SP dist board, with a 17A phase supply, and 36A neutral current. That`s an oops somewhere, then.........

It could be normal leakage for a supply of that size, like sidewinder says....Harmonics would be one thought and any Drives (or anything else with EMC filters) on the supply that have leakage currents to earth ?Is your clamp meter big enough for supplies that size ? :slap

Just about mate - the biggest problem is where three of four SWAs enter close together - I can`t "move one out of the way" to clamp the next one; and the neutrals separate off to the N busbar.Right - added info - from the power analyserower usage:55.4 KW / 69.4 KVA / 41.8 KVAR0.78pf53A imbalance current returning on N conductor - including 10A 3rd harmonic, and 10A DC21.2A current measured on earthing conductor, at 28% t.h.d. (total harmonic distortion) ; )Individuated load current (kitchens & laundry closed

)L1 : 99A 11% thd

L2 : 29.5A 22% thd

L3 : 27.6A 30% thd (predominantly 5th)

KME

ps thanks to all for your thoughts & ideas

 

[Sidewinder]

I'm guessing the power readings were with my 1ph power analyser then with the clamp on 1ph and the V probes on the other 2?

I would not rely on this on an unbalanced 3ph supply as being accurate TBH.

I'll have to get hold of the 3ph analyser for you I think.

Canoe, does your Fluke have 3 or 4 clamps mate?

The one I can get hold of only has 3, so it calcs the N current.

Can't afford to buy one yet!

IIRC you have loads of computing equipment, drives on the pool, drives on the kitchen extract, loads of discharge & fluorescent lighting, and tons of other electronic kit around, so, probably loads of harmonics.

I suspect that the results you have are not showing the true picture due to limitations of the kit.

 

[Evans Electric]

Martyn , with an 800A incomer and , I guess, a lot of single phase outgoing sub mains I'd certainly expect to find that sort of current on the neutral. Not sure about the earth though. Lots of IT stuff ?

Are the earth cables on the bar just the jumpers up to the banjos on the armour glands ? Shame you can't get to them individually.

 

[kme]

Snakehips - I`ve spoken to our friend (and special friend) about this already - I can get the KT unit you had without a problem - if required.

At the moment, they`ve only allowed a few hours time to investigate it, so I was undecided as to its usefulness in this scenario .......

Deke - all bar 2 of the sub-mains are TPN, though some unbalanced loading is on the ends of em. I was just reporting the neutral current as an aide, not flagging it as a problem

Snakey- VLT on the kitchen extract, but not on the pool, or anywhere else for that matter. There aren`t THAT many computers, TBH - prob about 14 or 15 workstations; plus the servers (I believe there are four of `em - in the same place).

Discharge lighting - yes, but not operating during the day - except the 11 no. SON-T in the pool area. Florries - yup - quite a few.

 

[Jono Pashley]

has been known for fridges/freezers and other stuff around kitchens and bars to have some earth leakage as well.

 

[sb_89]

This is a really good explanation of what causes earth leakage for anyone who wants to read it. If your reading along and wondering what could cause a large current on the earth ....

http://www.marcspages.co.uk/pq/3220.htm