Eathing continuiting of building structure 1

[AusLee]

Hi,

An Electrical contractor is installing wiring inside the apartments in a multi-dwelling apartments building block. The job is in Australia, based on AS 3000.

He is not using the boxes in the wall. He is just running the cable from the unit/apartment distribution board to the power outlet and will terminate directly in the outlet.

An independent inspector came to inspect and said that he is concerned if a cable is loose from the outlet behind the wall, and the active conductor in that loose cable comes in touch with the steel studs supporting the gyprock wall, that steel stud will become energised and will cause an electrocution hazard.

The Electrical contractor states that each circuit breaker in the distribution board is equipped with a residual current device/ground fault circuit interrupt.

The inspector is still not happy, he said the Electrical contractor needs to demonstrate that the steel studs are actually earthed.

Each steel stud is fixed to the ground with 2 screws.

Can you please provide some information/standard/link to a website/ which shows how can we test the earthing continuity of the structure? Do we get a normal ohm meter and connect it between each stud and the next? and what should be a good reading? Obviously concrete is not copper so I'm not expecting zero resistance but what is a good value?

 

[waross]

It may be safer and more productive in the long run to explain to the contractor that it is not always wise to try to win an argument with the Authority Having Jurisdiction.
Under our code, all connections in a ground path must be made in an approved fashion with approved devices.
I doubt that two screws in a metal stud will be an approved connection method for a ground conductor, by qualified electrical workers.
I would ask for a "Ductor" test on all studs. I doubt that all will pass.
You would not test between studs but between each stud and the grounding electrode.
Be sure to have a good fire watch when performing the Ductor tests.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[benta]

"He is not using the boxes in the wall. He is just running the cable from the unit/apartment distribution board to the power outlet and will terminate directly in the outlet."

Could you elaborate, please? I'm not certain what you mean.

 

[AusLee]

Hi waross,

Thanks, two questions please:

Q1: this inspector is not an AHJ, he was brought in as extra checking. So forgetting NEC and AS 3000 for a moment, given that there is RCD/GFCI protection, would there be any hazard on the tenants in the future based on law of physics not on codes/regulation?

Q2: I'm very interested in your recommended test method. In searhing for "Ductor" I came across this link: Megger
I noted on page 9 they recommend the 4-wire method, but otherwise not very relevant and they did not give me an indication of resistance to expect.

I also found: [link file:///C:/Users/CADi7/Downloads/2005_16_autumn_wiring_matters__complete_no_adverts.pdf]IEE[/url]
I note on page 8 the thing I am probably looking for is Earth Fault loop impedance; on page 9 they say that this is not applicable to RCD protected switches.

So i guess I'm looking for Earth Continuity Test. On page 17 of the above link, Sizing of circuit protective conductors: i think in the case postulated by the independnet inspector, the loose active conductor will be eventually in touch contact with the steel stud, thus electrifying it. I think that this is high impedance contact no?

Before they introduced the requirement for RCD/GFCI on power circuite, or for power circuits of high rating not required by code to have an RCD/GFCI, the earthing conductor is sized so that an earth fault will be of sufficiently low impedance to cause the fault current to be sufficiently large so i can be cleared by the circuit breaker working in the active/phase conductor. Even if the studs are solidly earthed with low resistance, I am not confidnent that a loose conductor in touch with the earth stud will not be sufficiently low to play ball, I think it will be high resistance and mess this concept. it is not a bolted connection, hardly a couple of strands of the multistranded conductor I'm thinking.

The eathing continuity is good for use of the steel in the building as down conductors for a lightning protection system. I am not seeing any purpose in solid earthing in the above an I am beginnign to think that the RCD provided is actually the best form of protection provided.

Can you please help me se the better picture?

 

[waross]

Testing:
A Multi-Meter tests continuity with a low voltage and low current. Usually less than 9 Volts and possible less than 3 Volts.
A Megger tests high resistances with a higher voltage. My favorite Megger had a choice; 250 Volts, 500 Volts or 1000 Volts.
It also had an option not always seen on Meggers,- Continuity testing with 6 Volts.
A Ductor test passes a fairly high current through the connection under test and then measures the voltage drop across the connection under test.
This test is mostly used to check connections on high current circuit such as bus bar connections and transformer connections as it gives the most representative indication of real world conditions.

But back to the question at hand.
The installation may last for 10 or 20 or more years without any issues.
After that time the screws making the connection between the studs may be corroded and or the GFI may be stuck.
I wouldn't bet my life on a 20 year old GFI working flawlessly the first time after 20 years and I would not bet another person's life on it.
Of all possible teaching aids, a dead body may be both the most effective and yet the most to be avoided.
What does your code say about mounting devices in outlet boxes?
In North America there may be some devices approved for installation without an outlet box, but most devices do require an outlet box. It's the law here.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[AusLee]

Thank you waross for the description of the various methods of testing. It seems that we are after an earth continuity test.

With regards to the RCD/GFCI:
1. The risk of gettinig stuck is mitigate in Australia by mandatory regular testing of the operation of the RCD/GFCI. This includes testing via pressing the button on the device, and also testing electronically to make sure the clearing time is also compliant.
2. This level of risk is acceptable by the applicable laws here. There is no perfectly safe installation. The hazards you mention of corrosion etc are applicable in other, accepted, situations.

I need assistance please describing the requirements for the contractor. The apartments being built are on Level 5 in the building. A suggesting was made to have a new, inpedendent earth electrode in the ground and run a cable up to the level 5 over the building facade and connect to the steel frame that we need to verify is earthed. Then measure the resistance between the vertical steel stud to the earth bar in the electrical distribution board inside the unit, and the reading should be Zero (0).

I'm confused by this.

1. Wouldn't it be simpler to just measure the resistance between the vertical stud in the wall and the earth bar in the electrical DB? why are we bonding the stud to a separate electrode in the ground? And if I do measure directly from the stud to the earth bar, what is the required reading for compliance? I'm thinking anything less than 10 ohms will be acceptable based on the fact that 10 ohms is used as a requirement for the earthing of the main switchbaord and other systems.

2. The separate earth at ground level and the main earth of the main switchboard are not going to be in the same location so there is a difference of potential between them equal to the test current x earth resistance.

Any assistance in how to measure the continuity to earth of the steel frames in a building is much appreciated.

 

[rtronics]

Refer to AS3000, section 5.4.6
This states that metal building materials possibly in contact with lives wires must be earthed to the earth bar directly via the protective earth. Maximum resistance between them shall not exceed 0.5 ohms.
It is usual to meet requirements by enclosing the termination within a nonconductive shroud, or by restraining the conductor by tying, lacing or clipping. IMO it would be difficult over time to ensure the whole frame of a metal framed building to be 0.5 ohms to earth without protective earthing each individual part. (as waross said)

 

[AusLee]

Hi rtronics, yes precisely 5.4.6.

the 0.5 ohms is for the bonding conductor itself not for the whole resistance of the bonding conductor + resistance to earth of the metal frame.

the maximum resistance of the main earthing coductor connecting the earh bar in the main switchboard to the main earth electrode is 0.5 ohms, but the resistance of the earth electrodde itself is higher. Same case here.

The code goes goes on to give the example that the resistance of the earthing for 100m of 2.5mm2 conductor should be around 0.8 Ohms. That's a botled earth fault resistance (over the protective earth conductor). I have no expectation that the resistance to earth of the metal frame will be even better than the earthing resistance to earth of the actual circuit itself. I'm expecting a much higher number, all i want to know is what would be a good method to test this earth continuty and what resistance to expect.

I understand that using a wall box will solve the problem, also the use of an RCD will solve the problem as well (I have cross checked this with an inspector).

I note the use of RCD/GFCI is not only to protect people, but also to reduce he risk of fire. AS 3000 did not require RCD for any and all equipment, they are specific in section 2.6.1 in what they want it for, and this particular circuit is not one that requires an RCD under the code. As there are hundreds of this circuit, I want to make sure that whatever criteria to meet AS 3000 section 5.4.6 is acrually safe. There is little purpose if I bond the metal part with a 4mm2 jumper from the GPO will achieve anything, I could have a terminal that goes loose (as stipulated by 5.4.6) but the impedance to earth is too high and a high, but not sufficiently high to trip, current will flow and cause fire.

I'm looking at AS 3000 Section 8.3 for required testing and AS 3017 for testing methods but I also am looking for your suggestions on how to measure the earth continuity and what to expect and why.

The (same) inspector has accepted that the satellite reception dish, being bloted to the structure over the lift overrun, is earthed and do not require the additional earthing (which would dissipate static build-up) so how come that is good enough earthing and the vertical studs forming the walls inside the apartments, using the same bloted fixation method, are not well earthed. Consider AS 3000 Part 1 not Part 2.

 

[ScottyUK]

AusLee,

RCD / RCBO protection became mandatory for all concealed wiring under BS 7671. I'm not sure how closely AS 3000 follows 7671, but that requirement was one of the big changes when the 17th Edition of 7671 was introduced.

 

[AusLee]

Hi ScottyUK

No we are not thre yet, RCD is required for many applications (AS 3000 section 2.6.1) but not for all applications and not for all concealed wiring. But I think the question is not only realted to concealed wiring, the question is whether or not the earthing of a metal part is a sure/safe/engineered solution that will actually and by itself eliminate the hazard.

I have pin pointed the problem, it is in AS 3000 Appendix B section B4.4 "Figure B5 shows an active-to-earth fault which, for the purposes of calculations, is deemed to be of negligible impedance."

That last bit of negligible impedance is the problem. Essetially the code author set the above mention scenario and went off on a calculation spree based on this assumption. The truth of the matter is that the impedance is rarely negligible. So in the worst case: breaking of insulation and touch of phase conductor with an extraneous metal part, I will be electrocuted if i touch that part under fault as there is no RCD and the fault return path to the source was never sufficiently high to cause automatic disconnection.

The loose conductor inside the wall is like a Class I equipment, the equipment being the wallwith the metal studs in it and whatever structure is electrically bonded to it.

What I think I'm getting at is: earthing of the metal frame of any Class I equipment is necessary but not sufficient to automatically disconnect the circuit. Either the circuit breaker has to be very close to the load (which is never the case especially in miniature DBs) or the RCD should be installed.

???

 

[FreddyNurk]

What equipment are they installing? AS3000 was updated a while ago, and as far as I recall, all socket outlets were required to be fed from an RCD protected circuit. Things like ovens and direct wired air conditioners were exempt though.

No chance they're fitting new USB wallplates are they? I'm not sure what you mean by "boxes in the wall" either.



EDMS Australia

 

[rtronics]

Take another look at the diagram in section 4.4
It clearly shows the current path from a properly PROTECTIVE EARTHED but but faulty equipment goes through the PROTECTIVE EARTH CONNECTION which is of (tested at installation time), low resistance, to the earth bar, then via the earth bar-neutral bar connection, (the MEN), to the incoming transformer neutral connection.
This current flow is not dependent on any connection to the soil via an earthing rod etc.
The CB or RCD will be tripped before a dangerous potential appears on the casing of the PROTECTIVE EARTHED equipment, (or building frame in your case.)
AS3000 does appear to use the terms 'earth' and 'protective earth' interchangeably throughout the document, which could lead to confusion.

 

[waross]

I have always considered earthing or bonding to be the first line of defence with an RCD as the second line of defence.
I don't believe that a connection between the metal studs is an approved bonding method. That is, however, a question for your local code.
I see the lack of a grounded outlet box as the removal of a layer of protection.
That said, I cannot say that it is wrong. It is a matter of the interpretation of your local codes.

Bill
--------------------
"Why not the best?"
Jimmy Carter