My co-worker and I were replacing some old wallpacks today with new. I was assembling fixtures while he put them up. He is working on the last fixture while I am cleaning up and I hear the building back-up genny fire up. Thought is was a weird time for a genny run at 3 in the afternoon. Next thing I know a women is asking me what is wrong. She explained the building lost power. My co-worker walks over and has slight arc burn on his arm. Said he was holding the fixture against the wall when it slipped down a little and the hot leg started crackling. Next thing he knew genny behind him was running. I went down to the elec room and the 1200a main is tripped. I reset the trip mechanism than reset the main switch. Everything is fine...genny cycles down and all is well. My thinking is that the lighting circuit has no, or a bad earth ground and caused the trip. Am I correct in this assessment? Theories?
Apparently, the arc at the lighting circuit tripped the GF protection,
You neglected to describe the 'arc'; hot to ground, hot to neutral, hot-hot?
What you describe as 'earth ground' is the equipment grounding conductor (EGC) which is the fault path for OCP.
The GF can trip on phase to phase, phase to ground inbalances.
I had a 3000 amp 277/480 main trip on GF traced back to a fluorescent (old magnetic) ballast. Thru the 277 LP, thru a 277/480 MDP and to the main, without any feeder or branch trips. Hence, 'selective coordination' comes to play in critical ops.
I am very aware earth ground is the grounding conductor.
Not sure exactly what the hot wire shorted on, I am assuming the metal box. He had the neutral wire (grounded conductor) wire nutted.
[quote=SafetyWiredMy thinking is that the lighting circuit has no, or a bad earth ground and caused the trip.[/quote]
Actually it sounds like all of the bonding paths and grounded conductors are working correctly.
What you have is clearly a case of poor protective device coordination, as was previously mentioned.
Most manufacturers will ship their GF devices set to the absolute minimum. If these settings are not changed the main GF can very easily trip before a 20A branch device operates.
I would guess that fixture and ballast changes result in the predominant number of main GF nuisance trips.
Had it shorted to the neutral, the GF protection would has seen it as just another load, and it wouldn't have tripped.
If you had turned off the power beforehand, as you should have, this never would have happened.
He's lucky that he just had a slight arc burn. Things could have been much worse.
If you had turned off the power beforehand, as you should have, this never would have happened.
You are absolutely right! If I had been the one working on it I would have shut it off.
The main breaker should have had an indicator showing the nature of the trip, most likely GF. Sometimes the first breaker to have ground fault sensing is the main. You should look at the trip setting and time dial of the main Vs downstream trip units. The downstream units should always win the race. Main: "A", ".3S" Branch: "A", ".1S"--> Branch should trip first under ground fault conditions.
Main: "A", ".1S" Branch: "A", ".1S" --> It's anybody's guess.
We try to check our Digitrip units every 2 to 3 years, to verify that the pickups and times are as set. Sometimes that's not the case until those little rotary switches get exercised.
Joe
The main breaker should have had an indicator showing the nature of the trip, most likely GF. Sometimes the first breaker to have ground fault sensing is the main. You should look at the trip setting and time dial of the main Vs downstream trip units. The downstream units should always win the race. Main: "A", ".3S" Branch: "A", ".1S"--> Branch should trip first under ground fault conditions.
Main: "A", ".1S" Branch: "A", ".1S" --> It's anybody's guess.
We try to check our Digitrip units every 2 to 3 years, to verify that the pickups and times are as set. Sometimes that's not the case until those little rotary switches get exercised.
Joe
I believe there is only one trip mechanism and it is just above the Main switch. I had to remove the clear plastic cover to reset. To the right are the branch disconnects and there are no trip mechanisms.
What Joe is saying is "selective coordination" in basic terms. If you have GF on the 'Main', and on panels further downstream, the downstream panels should be at a lower setting then the 'Main'.
It sounds like the facility you were in only has GF at the 'Main', which is fairly common.
Testing and adjustments to GF trip settings and times should be based on the factory specs, or engineering studies.
Bump:
Any updates on this??
Was called there this morning. Evidently someone was doing something on the 15th floor with lighting and the main tripped.
Sure enough tripped main. I opened all downstream vacu-break switches, reset trip mechanism and reset main.
Turns out idiots replaced light fixtures with 120v ballasts on 277v circuit. Turned them on...cablamo!
This needs to be rectified.
There is a dial on the trip mechanism- .1, .2 .3 .4 etc. It is set currently at .1 I suspect this is time?
Do I need to have a testing co. come out and check this trip mech. out. See if it is wired correctly?
This has been in place for years. Why all of a sudden is this happening? I am pretty sure someone else has created an arc or fault before downstream. Could the mechanism be going bad? It is a ITE SB3 Switchboard. Three cabs. One for metering, second has Mswitch and volt and amp meter, third has vacu-break switches for branch circuits.
Mark Ode is an industry respected person who has authored many articles.
That said, if you can't locate the mfg data (doubtful if older equip) you could probably find it at the mfg website. The info (curves, etc) for the exact MCB you have.
IMHO, a qualified testing company would be the way to go.
The 'age' of the MCB could be a factor, or someone played with the GF trip setting, or previous faults tripped the branch cb.
I had a old, 2 lamp, F-96 slimline, 277 volt ballast that self destructed, and took out the GF main (2000 amp). No branch cbs tripped, ad yes, this was an 'old' facility with a new 2000 amp service.
Hope my opinion helps
I will be calling test co. tomorrow. Thanks.
Occasionally even on a correctly coordinated system the main breaker beats the branch breaker on ground fault. Most molded case low current breakers have set trip curves and cannot be adjusted. clearing times can be longer than the higher quality and perhaps much faster mains.
I think a typical 10KA rated breaker clears in 4 to 8 cycles and the main might clear in 2 to 4 cycles The main ground fault setting at 30 amps could easily trip faster than a a branch breaker that needs 6 times rated current to trip in its instantaneous range. So a 15 amp breaker might need 75 to 90 amps of fault current to trip as fast as the breaker is designed and the main is tripping at 30 to 60 amps of ground fault current.
I always rely on a testing engineer to provide selectivity and coordination.
I know this is going to sound wacky, along with being Pseudoscience, but sometimes you might just get lucky with a Ground Fault not activating the Trip on a GFPE Disconnecting Device (Main Circuit Breaker or BP Switch for 480Y/277V 3PH. 4W. Services of 1000A and greater).
Personal Experience case in point:
Back in the Days, when I actually worked hard (in the Field), there was a Large Commercial "Park", with 8 separate 120,000 sq. ft. Two-Floor Tilt-up Buildings (Occupancy B), of which We had T.I. Contracts with several Clients in several of these Buildings.
Each Building has a Four Thousand Amp 480Y/277V 3PH. 4W. Service, and each Service Disconnect included GFPE.
No Downstream GFPE was used, and the settings on all GFPE Trips were minimal delay, minimal time pickup, minimal threshold.
On One particular "Hellish" T.I., a crew of "MC Jockeys" known as "The Kings", were hired to install MC in the Steel Stud Office's Walls for each Tenant Space.
(BTW; "MC Jockeys" = something similar to "Romex Jockeys" or "Ropers"; in this case, a crew of Belligerent Beer-Fueled, normally fighting piece workers, whom are rocket fast at installing MC Cable, but have questionable Circuitry Make-Up Skills, and do not pay too much attention to details - such as Color of Wires being Terminated)
The Kings had completed Prewiring for Lighting (277V) and Receptacles (120V), and a separate (and competent) crew brought the Branch Circuits from the Suites to the Panelboards, where I Terminated the Circuitry.
After Devices & Fixtures were installed, I energized the Branch Circuits.
Beginning with the 120V Circuits, only found One Fault - a Drywall Screw penetration that occurred after we checked for Ground Faults via Ohm Meter (Cabinet Installers used long screws, missed the stud but hit the MC Cable).
When the Lighting Circuit Home runs were installed, they were also tested for Faults and Continuity. All was good at the time, several days prior to energizing Circuits.
So I throw the Breakers for the Suite Lighting Circuits.
All is OK, until I throw the Second to last Breaker - which trips-free as I throw the handle.
Lights dim out during the Fault, and I am stunned the Main did not trip!
Crew performs Troubleshooting on the Lighting Circuits, to locate this issue.
After extensive searching, the issue is discovered in a Light Fixture.
Switch Leg was connected directly to the EGC - and everyone feels this was a "King-Typical mistake".
Remaining Fixtures are checked, Circuitry checks out good, and Circuit Breaker is reset - no Trip issues. Same with final Breaker - no Trip.
Next morning, I go to turn on the Lighting, and this time the Second Breaker trips free!
Lighting dims, and once again I am stunned that the Main has not tripped!!!
Someone is sabotaging the Lighting Circuit! (turns out to be one of the Kings, disgruntled from an earlier intoxication issue).
The Troubleshooting crew must now verify all Circuitry before ANY Circuit Breaker is closed.
(read: major overhead headache, loss of productive labor, etc,)
I inform the Security Staff of the issues, which brings up the instance of "One of the 'Kings' was seen entering the Building yesterday, around 4:30PM" (we leave at 3:00PM, I leave maybe by 3:30PM at best - normally it will be 4:00PM).
So, eventually the T-Shooters finish up, and once again, the Lighting Circuits are energized.
This time, the Breakers are thrown by someone else; I was discussing installs with Two of my Crew Leads (Top Journeymen), when a familiar "Pop and Buzz" was heard, followed by darkness.
This time, the Main Tripped!!!
Strangest thing ever!!!
I had Two separate occasions, with no Trip.
The Troubleshooter was not so lucky;
first Device thrown - Bam! Down goes the entire Building!
Oh, forgot to mention the best part!
Of course, the P.M. of the complex has their Office at this Building, and - of course, One of the Staff has a Superiority Complex (AKA "I used to be an Electrical Engineer" kind of an Attitude), and is instantly on my Butt, due to this Outage issue. (read: "Nasty-Grams and Penalties A'-Coming).
I was able to relax this Person, and actually avoided all Nasty-Grams, and any Penalties from being issued!
Nevertheless, there is no legitimate reason (other than the Electro-Gods smiled happily upon me), why after Two consecutive Ground Faults, the GFPE Main did not trip for me, yet it did trip for the Journeyman!
Each instance was the same Fault condition: Switch Leg terminated directly to the EGC - via Wirenut - in a 2x4 Troffer.
Each was due to Sabotage performed by the same King, sneaking on the Job. (He was finally apprehended by a few of the more "Muscle-Enhanced" Journeymen on my crew, and dealt with as necessary; ultimately putting an end to the Ground Fault issues).
There is no Moral to this Story, and by no means am I hinting towards "sometimes it is OK to work on Hot Circuits".
Sometimes things just seem "Biased" towards a certain Person.
I have no idea what allowed for the Selective Coordination, then Non-Selective Coordination between the Two separate Devices!!! Believe me, I researched the heck out of this one!!!
--Scott (EE)