ground fault current

[jssilvia]

I Have a 1200 amp 277/480 volt 3phase service with a gfi main breaker,the breaker has a digital ampmeter. The 3 phases all show about 90 amps but the ground fault amperage goes up to about 280. It seems to go up more as in duct heaters start the heaters are 277 volt 1 phase 40 amp. Could anyone give me a reason why the grond fault amperage is so high

 

[don_resqcapt19]

The most likely cause of that much current on the grounding system is an illegal neutral to ground bond at a panel. On second thought, I'm not sure, you said 90 amps on the phase conductors and 280 on the grounding system.
Don

 

[ramsy]

When you say ground, if you mean grounding or green, then you have a serious objectionable current path problem.

If you mean grounded or neutral, maybe three phases each @ 90 Amps combine in the main neutral feed, the excess Amps could be zero-sequence harmonics.

In this 3-Phase Imbalance thread:

If your asking for the current in the neutral for an unbalanced system it is the square root of (Ia^2+Ib^2+Ic^2-IaIb - IaIc - IbIc).

Correct me if I'm wrong, but if A=B=C=90A, this neutral works out to 156 Amps. 24 Amps is left over if said meter indicates 180 Amps on the neutral.

If that meter uses a true-RMS current transducer, that is potentially 15% zero-sequence current harmonics present in the neutral. But, harmonics are not usually associated with heater loads.

 

[don_resqcapt19]

Roger,

Correct me if I'm wrong, but if A=B=C=90A, this neutral works out to 156 Amps.

If all three phases are 90 amps, the current on the neutral is zero amps.
Don

 

[catchtwentytwo]

I believe some duct heater units use SCRs rather than contactors. If they do, they can cause harmonics (especially when not running at 100%).

If jssilvia has a clamp-on ammeter with a Hz function, put it on a phase conductor and note the reading. If it isn't around 60 hz, there are harmonics. A reading of 180 hz shows the presence of triplens (3rd harmonics).

 

[ramsy]

I believe some duct heater units use SCRs rather than contactors. If they do, they can cause harmonics (especially when not running at 100%).

What are those, SCR's?

If jssilvia has a clamp-on ammeter with a Hz function, put it on a phase conductor and note the reading. If it isn't around 60 hz, there are harmonics. A reading of 180 hz shows the presence of triplens (3rd harmonics).

What an excellent poor man's harmonics meter. Does the cl/amp Hz function really measure the highest Hz rather than largest magnitude (fundamental)?

 

[catchtwentytwo]

Quote:

"What are those, SCR's?"

SCRs are Silicon Controlled Rectifiers - http://www.americanmicrosemi.com/tutorials/scr.htm . Try a Google or Yahoo Search for Duct Heaters SCRs .

Quote:

" What an excellent poor man's harmonics meter. Does the cl/amp Hz function really measure the highest Hz rather than largest magnitude (fundamental)?"

See Fluke Application Note: Troubleshooting Power Harmonics- Basic troubleshooting using multimeters and current clamps page 4 Section 3. The whole article shows ways to detect the presence of harmonics without elaborate PQ analyzers.
http://www.informationstore.net/fluke/e-fullfill-bot.asp?link=10603-eng&Email=

I'm sure there are more articles on this method, try variations on "180 hertz current triplens" with Google and/or Yahoo searches.

 

[ramsy]

SCRs are Silicon Controlled Rectifiers (applications include) Duct Heaters.

(For cl/amp Hz function) See Fluke Application Note: Troubleshooting Power Harmonics The whole article shows ways to detect the presence of harmonics without elaborate PQ analyzers.

Thank you for brining this to the table. That Fluke PQ guide was an excellent read for me, which answered some long overdue questions.

The 800# in that article got Fluke tech.support, which explained Fluke's 337 cl/amp meter Hz function. This detects the largest magnitude Hz only. That means it only shows 50/60Hz fundamentals on phases, not harmonics, so clamp Hz diagnostics is limited to the neutral.

Efforts to find a poor mans harmonic detector on shared neutrals may confirm harmonics, but won't trace the source to specific loads. Neutral indicators also suffer the limits of common mode voltage (last part).

Measuring a shared neutral @ 50/60Hz would confirm an unbalance, and Fluke support said the 337 clamp measures neutral harmonics if zero-sequence sums are the predominate magnitude (ie) 180Hz. At this writing, the 337 retails for U$D 340.00. Fluke's 175, 177, 179 meters (similar Hz function) with the seperate I400 clamp accessory, retail for about the same.

Fluke support confirms tracing phase harmonics directly require PQ analyzers, currently their cl/amp LH1060 is U$D 749.00 or 2060 U$D 1095.00. Nevertheless that Fluke PQ guide suggested another technique with a much simpler True-RMS cl/amp.

While Flukes PQ guide describes using special peak measurements as well, if the poor mans PQ meter exists, its simplest form may be comparing the phase current against the load or kVA nameplate (E*I). Excluding other variables, if the RMS current of phase(s) exceed its load nameplate or xfmr nameplate, harmonics may be assumed. Fluke's lowest end True-RMS clamp is that I400 attachment, (400A Max) @ U$D 100.00

 

[ramsy]

If my read of GFPE requirements is correct, an unballance of 1000A Max trip setting is possible for that 1200 Amp service. If this is correct, it appears a ground fault amperage well above 280 could persist for quite some time without tripping a properly installed GFPE.