ground loop

Status
Not open for further replies.
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
Trying to get rid of ground loops at a dairy owned transformer 480 to 208 Y.
Existing installation had three taps coming from the secondary of the transformer with a system bonding jumper in place.. Bonding occurred again at 4 locations. Grounding electrode was not present at transformer but is at two remote locations. To be changed.

We removed the bonding at two of the panels that are in/on the same building that houses the transformer. We installed two rods and attached the GEC to both the 480v GC and the XO of the xfmr. We now have 5 wire feeders to each local panel or disconnect location. EGs to each are located in a series of stacked lugs attached to the left inside of the enclosure. I did not want to rely on the 10-32 screw threads so a #4 cu ties the EG lugs to the same location as the system bonding jumper. Bottom right.

We were really surprised to find 6 amps of current flow on that short jumper. This is with the xfmr energized and no loads connected. No current flows on the GEC to/from the rods. Lesser amounts flow on each of the EGs but don?t total 6 amps. Is this normal to have that kind of circulation at a Xfmr? If it means any thing the current waveform looks like the wave on a slop bucket.
 

lpelectric

Senior Member
To me, that seems like an awful lot of current flow. I've been reading "Equipotential Planes, A Figment of the Imagination", by Donald Zipse, and find it fascinating and very, very informative. It sure is changing the way I look at stray current. :smile:
 

brian john

Senior Member
Location
Leesburg, VA
Disconnect the service to each panel, disconnect the neutrals (feeders and branch circuits), megger the neutrals (at a low DC voltage 100-250 VDC).
 

kingpb

Senior Member
Location
SE USA as far as you can go
Occupation
Engineer, Registered
ptonsparky said:
Trying to get rid of ground loops at a dairy owned transformer 480 to 208 Y.
Existing installation had three taps coming from the secondary of the transformer with a system bonding jumper in place.. Bonding occurred again at 4 locations. Grounding electrode was not present at transformer but is at two remote locations. To be changed.

We removed the bonding at two of the panels that are in/on the same building that houses the transformer. We installed two rods and attached the GEC to both the 480v GC and the XO of the xfmr. We now have 5 wire feeders to each local panel or disconnect location. EGs to each are located in a series of stacked lugs attached to the left inside of the enclosure. I did not want to rely on the 10-32 screw threads so a #4 cu ties the EG lugs to the same location as the system bonding jumper. Bottom right.

We were really surprised to find 6 amps of current flow on that short jumper. This is with the xfmr energized and no loads connected. No current flows on the GEC to/from the rods. Lesser amounts flow on each of the EGs but don?t total 6 amps. Is this normal to have that kind of circulation at a Xfmr? If it means any thing the current waveform looks like the wave on a slop bucket.

You say there were three taps... what were they connected too?

Bonding occured at 4 locations yet you only have three taps?

Two remote locations?

#4 copper bond? Is that big enough, what size are the conductors?

Possibly a sketch would be more helpful, as I am trying to understand the set-up, but can't quite visualize what you have going on.
 
Location
NE (9.06 miles @5.9 Degrees from Winged Horses)
Occupation
EC - retired
dereckbc said:
There should be no current. Do as Brian suggest and find the problem(s)

In the process of doing just that. Just suprised the heck out of me to see 6 amps on what should have nothing. Especially when no load is connected other than feeder and branch circiut neutrals. Never had the reason or occasion to check that conductor before.

The #4 EG are good for the 200A feeders, but the system bonding jumper should be a #2, I believe. 75kva. Wire sizes depended on the phase of the moon at installation.

1 tap to 200 amp panel with main brkr
1 tap to a 42 circ MLO panel with 1 3p cb in use
1 tap to 200 A fused disc.
The MLO cb and fused disc feed 4 wire MLO panels some distance away and that is where the existing rods are.

I am sure no one read anything about transformers or taps prior to installation let alone overcurrent protection for either.

Stray current is my first goal, after that comes those minor :grin: code corrections if their money holds out.

Zipse was a good read.
 

dbuckley

Senior Member
ptonsparky said:
Zipse was a good read.

Indeed.

However, the widespread use of the multigrounded neutral abberation renders the normal design of equipotential plane ineffective, because "normal" current is designed to flow through the ground, leading to all the problems that Zipse describes. You need to keep stray current out of the equipotential zone(*) using a surrounding ring of rods, then the equipotential plane can archive it's goals.

I also note on page 6 a recommendation to not bond the primary neutral to the secondary on transformers, which I support fully, and have even suggested herein, despite the fact that such arrangements are not permitted within code; code says SDS secondaries need to be bonded to the primary. Whereas this is often a good idea, sometimes it's a really (really) bad idea. Any of the threads about dock lifting winches and shocks therein are an immediate testament to this folly.

*) Edited to add that this statement is of course rubbish; Kirchoff teaches us that current uses all available paths, and given that the voltage across the grid is proportional to both current and resistance, and resistance is pretty much fixed, the best we can do is to minimize current across the grid by providing a more attractive path for the majority of the current to flow through, ie a lower impedance path.
 
Last edited:
Status
Not open for further replies.
Top