ECN Electrical Forum - Discussion Forums for Electricians, Inspectors and Related Professionals
ECN Shout Chat
ShoutChat
Recent Posts
Do we need grounding?
by tortuga - 03/18/24 08:39 PM
240V only in a home and NEC?
by tortuga - 03/18/24 04:29 PM
Cordless Tools: The Obvious Question
by renosteinke - 03/14/24 08:05 PM
Test Post
by sabrown - 03/06/24 05:29 PM
Solar PV Wiring Errors
by renosteinke - 03/02/24 09:12 PM
New in the Gallery:
This is a new one
This is a new one
by timmp, September 24
Few pics I found
Few pics I found
by timmp, August 15
Who's Online Now
0 members (), 94 guests, and 11 robots.
Key: Admin, Global Mod, Mod
Previous Thread
Next Thread
Print Thread
Rate Thread
Page 2 of 3 1 2 3
Joined: Jun 2014
Posts: 145
Likes: 2
D
dsk Offline OP
Member
Originally Posted by dsk
The reason are probably in another part of the building, and the combined breaker of the trade mark Chint, was renamed by one I do not know, but overheard calling it Shit. It may be right renaming, but I am not 100% sure. :-)

I do not believe so, but it may just be bad English from another Norwegian :-)

Another trust-able electrician tells me that other makes has almost the same error rates too.

Joined: Jun 2014
Posts: 145
Likes: 2
D
dsk Offline OP
Member
Originally Posted by Trumpy
dsk,
Are you really sure that that is the issue though?
Have you tested the actual earth leakage circuit breakers to rule them out?
What voltage and size pump motor is it that caused this fault to occur?
I find it very strange that a single motor could cause multiple RCD's to trip at any given time,
unless there is an issue with the neutral-earth link at the switch-board or some other poor
neutral connection somewhere in the system.
Anyone doing electrical work these days should have at the very least one of them either Fluke,
Kyoritsu or Megger multi-function testers, they give you so much in the way of things you can test.
My Kyoritsu tester allows me to upload to my laptop and it's instant test records.
I have insulation resistance to 1000V, Earth-Loop, Continuity, Earth Resistance (2P, 3P), RCD Tester.
You only have to buy it once.


They have some pretty impressing testers, but I do not know how well they know how to use it. Our supply here is an really old system called IT witch is a delta transformer with absolutely no point connected to ground/earth, only 3 wires with 230V between each. (like your 208V system, but no neutral to gnd.)
The motor to the pump is rated only 21 Amp. The main breaker for the building is on 2500A !!!

IT system: https://www.electrical-installation.org/enwiki/Characteristics_of_TT,_TN_and_IT_systems
These systems will not be used in new systems, but here we keep them up and running so far.

When testing the insulation between line and ground on the loops that is tripping it has always been more than 7 mega-ohms at 1 kV. The tester tells that the GFCI marked 30 mA release at around 17mA

I am not a certified electrician, only an engineer. The owner of the building does not allow me to do anything my-selves.

In earlier jobs I have designed, and built equipment that has been officially approved. Among that, equipment needed for building the Oakland San Fransisco east span bridge. That is why I may compare US and Norwegian systems.

Last edited by dsk; 11/05/21 07:47 AM.
Joined: Jun 2014
Posts: 145
Likes: 2
D
dsk Offline OP
Member
The GFCI's here are quite sensitive, the max current unbalance allowed is 30 ma. As you say , it may trip on far less, and they do. The wires in metal tubing has an capacitive unbalance. The transformer is pretty strange seen from the experience of modern systems. It is so called IT system, and that could be looked at as a US 240V delta system where you forgot the grounded Neutral! All voltages are floating in hence of ground!

The fault was made by a frequency converter, The converter was well grounded, but the local GFCI did not release, but it is tested and OK. A new converter solved the problem, so my guess is that it was a non mains frequency fault.

That is so far I can get in this case.

https://www.electricalclassroom.com/ungrounded-system%ef%bf%bc/

Last edited by dsk; 08/23/22 03:57 AM. Reason: Added link describing the system
Joined: Jul 2004
Posts: 9,919
Likes: 30
G
Member
I have been away from the IT biz for a while but we never brought a "neutral" to the computer room panels. The center point of a wye was still grounded at the service disconnect. I suspect there may be some people who misunderstand what that advice really means tho. "Isolated Ground" is also misunderstood. In our planning manuals, they suggested that the wire that would be the neutral be used for the isolated ground but it is still solidly connected to the grounding point in the service disconnect where the main bonding jumper lands. The only thing that makes it "isolated" is it does not connect to buildiing steel or any other grounded point after it leaves the service disconnect enclosure. It only connected to the machine frames.The equipment grounding conductor is still run normally. There end up being so many other unintentional grounding paths that this is really more theory than fact. IBM abandoned the isolated ground recommendation back in the 70s but some misunderstanding won't go away.
I have argued with installers who think "isolated Ground" means a separate grounding electrode system and that will actually cause more problems that it fixes.


Greg Fretwell
Joined: Jun 2014
Posts: 145
Likes: 2
D
dsk Offline OP
Member
It is avtually not grounded, it has a kind of spark gap between th transformer secondary and ground. I have the same system at home, and the had it one place I lived before, there they got a ground fault on on phase, and suddenly I had a 230V and a Neutral. I was a student and the meter did only meter the current trough the line that worked as a neutral so I ran the entire apartment with no metering for a month :-).. (Youngsters :-) )

Joined: Jul 2004
Posts: 9,919
Likes: 30
G
Member
Ungrounded systems are actually fairly rare in the US. You won't see them in residential at all. It is usually a critical industry where they can't handle a shut down due to a fault. They are required to have fault detectors so they know they are one more fault from a shut down.
Some lab situations may have an ungrounded supply and IBM did not ground the secondary of their "convenience transformers" that gave the techs 120v for their test equipment. It was really our little secret tho because the NEC/OSHA/Inspectors would not approve. Most of the techs didn't know it either.


Greg Fretwell
Joined: Jul 2004
Posts: 9,919
Likes: 30
G
Member
The typical Class A GFCI in the US trips at ~5ma. We consider 30ma as only protecting equipment (Class B protection)


Greg Fretwell
Joined: Jul 2002
Posts: 8,443
Likes: 3
Member
Right,
For a start, let's stop calling this a GFCI, that term is only relevant in the US.
What we have outside of the US is a Residual Current Device (RCD), it is meant to trip with a current imbalance of 30mA, with a maximum trip time of 300 mS (milliseconds).
Unless you have the relevant test equipment to test this equipment, you are wasting your time trying to fault-find this, get someone that has this equipment to test it, you will not be able to test this installation with a multi-meter.
You could possibly have a faulty RCD or one that has not been installed correctly feeding current back into a Earth busbar or a Neutral busbar.
Does this happen at any particular time of the day or week?

Joined: Jul 2002
Posts: 8,443
Likes: 3
Member
Also,
It seems that the whole installation needs to be disconnected from the supply and all circuits tested with a Megohm-meter at 500V to ensure all of the circuit wiring integrity is what it should be.
Again, if you cannot do this, by way of lack of test equipment or experience, get someone who can.
Having safety equipment tripping when it shouldn't be, means there is the risk that people will bypass it.

Joined: Jun 2014
Posts: 145
Likes: 2
D
dsk Offline OP
Member
No risk that anybody will bypass it, the weakness is the combination of an early version of the units and the old-fashioned system together with great capacitive "feedback" from the wires in steel tubing in the walls.

A less than 100 mA fault made around 15 of these elements to trip at the same time. Those are rated to trip at max 30 mA we have tested (all of) them to trip between 12 and 15 mA.

Only units of this make are tripping in these cases. I have been told that thees are lo-cost but approved.

I have been told that this supply-system was abandoned in the UK in 1948.

Attached Images
2022-09-26 10.28.46.jpg
Last edited by dsk; 09/26/22 04:39 AM. Reason: adding text
1 member likes this: Trumpy
Page 2 of 3 1 2 3

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5