Ground Fault Interupter on 3 wire 480 system?

[1Millwright]

I just started as an electrical supervisor for a hot strip mill. On my first week we were coming up from an outage and a ground in a screw compressor motor tripped a 4000 amp breaker with GFI fault causing a loss of power in two entire departments.

This compressor is fed off a downstream load center but the older panel did have a chance to trip before the 4000 amp kicked. I want to prevent future nuisance trips and isolate the downstream issues. I have contacted the manufacture and they are looking up the build specs and going to see if they can recommend any add on equipment for our old load centers feeding off this main to catch the ground as close to the fault as we can.

The 4000 amp breaker is the main on the secondary of one of our transformers. It is a 3 wire 480 volt system and it looks like they created an earth ground there for the GFI. I will look into it further, but I believe that it is either a wye - delta or delta - delta step down transformer.

In the code book it looks like you only need the GFI on a wye (4 wire) system above 1000 amps. They make no mention of using it on a 3 wire system. Is there something that I am missing or is it possible we ordered the GFI version when it was not necessary/required?

If the GFI wouldn't have tripped the equipment breaker should have when there was a spike phase to ground. If not it then the main on that subpanel should have caught it.

Thanks for any help.

 

[jraef]

We cannot second guess your issues without a CLEAR understanding of EXACTLY what your power system is, 3 wire Delta or 4 wire Wye, and what your grounding arrangement is, i.e. Corner Grounded Delta, Ungrounded Delta, Solidly Grounded Wye, Resistance Grounded Wye, etc.. It makes a HUGE difference with regard to ground fault tripping issues. What I will say however is that you undoubtedly need what is called a "System Coordination Study" of all of your protective devices, which will be the tool to do exactly what you say needs to be done; have the fault cleared at the lowest possible level. This is typically a site specific service preformed by a registered PE who can investigate everything involved in detail and make recommendations for changes.

Also, semantically, "Ground Fault Interrupter" is not really the correct terminology and may lead to confusion and misunderstanding. You can have Ground Fault tripping on large services, called "Equipment Ground Fault" protection (abbreviated as GFE or GFP for "Protection"), but a Ground Fault (Circuit) Interrupter, abbreviated as GFCI, is a specifically defined device for PERSONNEL protection in North America that implies a level of tripping current that is totally inappropriate for large equipment. Depending on what type of distribution system you have, GFP trip levels can be in the range of 30-300 mA, amps, or even a percentage of the full load current, i.e. 5% of 1000A, which is a lot of current compared to the 5mA allowable for GFCI personnel protection.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[Parchie]

@1millwright,
Please consider first things first. Any ground fault protection system needs to "sense" the ground fault before it "trips". If your system is not grounded at some point, any line that will be grounded will just clip that line to ground potential while the other lines experience elevated voltages (x 1.732). Remember that a single line-to-ground fault on an ungrounded system does not change the line currents much. Only after the occurence of a second line-to-ground fault on either of the remaining healthy lines will your phase overcurrent sees the problem and trips. But, if you have a "GFP" relay or your breaker is equipped with ground fault protection element, it's just a matter of setting the GFP to expected level of ground fault in your system.

We can't see enough information from your post and what jraef said, we can't help but guess what's happening out there.

 

[1Millwright]

Thanks for your posts! I too want to have a clear understanding of our system. Thanks for correcting my nomenclature as well, I prefer to use proper names to remove any ambiguity.

jraef: On the disconnect is says three wire 480. From my experience that is either a ungrounded delta or a floating wye on the secondary. On the line drawing it is T9. They are showing a Delta-Wye transformer and I believe that it is accurate. I will look to see if we have any wiring diagrams on the install itself. For now, I am under the assumption that there are three wires to the transformer and nothing connected to the center tap. We will have to wait for a downturn to physically inspect the breaker cabinet.

Parchie: I am going to have to think about that more. You raised some good points. From mill rumors I believe they created a ground point at the breaker for the GFP, but how would it sense it. I will have to read up on some GFP setups. I will update when I learn some more.

Thanks again for posting.

 

[waross]

I have seen grounded wye systems where there are no line to neutral loads, and so there is no point in spending the extra money running the neutral to the switch or MCC or supplying a neutral bus bar in the MCC or switchgear. This is probably a grounded wye system feeding a three wire main switch. Note that this switch may be easily converted to a 4 wire switch by the addition of a neutral terminal. (Installs with two screws in pre tapped holes.)
If this is the case then the X0 terminal will be grounded at the transformer and the ground fault protection will work normally.
Note; A four wire designation refers to the presence of the neutral conductor/connection. There should still be a grounding or bonding conductor.

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

 

[melspuds]

1Millwright,
Is this your first hot strip mill? I spent 35+ years working for 2 steel mill companies as an EE, with a total of 11 hot rolling mills, plus EMS melt shop, and finishing mills.
You my have a 3 wire delta system with a small zig-zag transformer to develop a neutral point. The neutral can be used in several ways to detect a ground on one of the 3 phases. The most common way to protect against grd faults, and arc flash (for OSHA) is use 5 Amp grounding resistor. When you have ground on a phase, the current flows from the fault to the neutral, then the zig-zag to the grounded phase wire. The current flow is picked up by a CT on the neutral to trip your breaker.
I believe this is how your GFI works.
Hope this helps and good luck with the mill foreman job.
Dave

ps: what size is the mill?

 

[1Millwright]

waross: As far as I have been told we have no neutral loads in the plant. The center tap may be grounded at the transformer, but I believe that is as far as it goes.

melspuds: This is my first hot strip mill. I am excited to get on this side of the steel making process. For 5 years I worked in a mill that processed secondary steel. I started out on the mechanical shop, but had a great mentor that got me into electrical. We would pull out coils anywhere from .75-1.5" and level them. Then using multiple head gas thermal machining centers we would burn our parts for mobile equipment or elevators. Some of them were laminated stacks that were tacked back together before getting shipped out. At this facility they had a true 3 wire system. We got 480 from three wires coming from the pole mounted can transformers. They didn't bring a neutral or ground in with our service entrance. I was responsible for cutting down 3 DC cranes and replacing them with AC ones. The lack of a ground was reeking havoc on the wireless remote/drive systems and we were have really weird problems. We ended up solving it by driving in new ground rods and providing a solid ground for the cranes.

From there I went to a forged roll shop for about 5 years as lead electrician/millwright. I worked with DC drives and CNC systems for the large lathes, grinders, and boring machines. I was responsible for all of the electric repairs and worked with contractors on some of the larger mechanical rebuilds. It was a good experience because I went from working on machines where our part tolerance was +-1/8 to a roll shop making some of the most perfect cylinders under the influence of gravity.

I then moved on to a closure stamping facility as an electrical engineer for a little over 2. I helped direct the union guys and recommended replacements for obsolete components. I was also responsible for program alterations and safety improvements. This was a high speed manufacturing facility making anywhere between 4-6 million closures a day with around 20-32 production lines running daily. Another good experience for learning programming tricks and tracking parts moving though turret machines around 300-500 parts per minute.

When I interviewed at this mill they had me at the smell of mill scale and seeing that orange glow moving through the mill. Raw power is the best way to describe it. I guess once you have steel in your veins it is hard to shake. We have a 85MW EAF that melts our 100% scrap shot. It feeds our metallurgical furnace which pours into our continuous slab caster. Once cooled they are fed into our reheat furnace and then move on to the hot strip mill. Roughing mill, coil box, 5 stand finishing mills and then cooling and the coiler. We band them up and that's all for our process here. We just have one hot strip mill at this facility. We are classified as a mini mill. I figured it would be the best place to learn the new machines and processes. Our other plant has 3 or 4 hot strips and some cold strips as well. I figure ones I learn the ropes I may bounce between both places.

I am a differential equations class away from my associates in engineering science and will be continuing to go for my electrical engineering degree. It is rough to work and go to school, especially in the steel world. As they say, "you are married to the mill and get to visit your family." I am trying to find a program that will give me credit for my practical experience, but not many do.

 

[waross]

The problem may be a coordination issue.
We had a ground fault on a 30 Amp circuit. The cable had been disconnected at the load end. The three phase wires and the ground wire were twisted together and taped. A lock dog was installed on the 30 Amp GFI breaker feeding the cable.
An electrician (who no longer is employed at that plant was explaining to another electrician that that type of lock dog was not very effective and the breaker could be turned on despite the lock dog. He demonstrated by turning the breaker on. A 1200 Amp breaker went out on a ground fault trip. The plant was down for four hours while crews tracked down the cause and determined that it was safe to re-energize.
The 1200 Amp breaker had been installed "Out of the box". The default GFI setting was instantaneous. The settings were reviewed and a short time delay added to the GFI trip to allow co-ordination with downstream devices.
Your issue may be similar.
The three wire designation refers to the lack of a neutral. It does not imply an ungrounded system
The one line drawing shows a wye secondary winding. The wye point is almost certainly grounded.
With no line to neutral loads the ground method may be either a solid ground or an impedance ground.
Ungrounded systems haven't been used for years.
Search this site for "arcing ground fault" and "discontinuous ground fault".
With ungrounded systems, an arcing ground fault may act as a combination of a spark gap radio transmitter and an auto-transformer and leave a high frequency, high voltage signal superimposed on one of the phases. A symptom of this issue was widespread motor burn out throughout a plant.

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