Current Measured On Service Grounding Conductors

Status
Not open for further replies.

jebidieh

Member
Location
New Florence, PA
Hi Folks!

I need some education. While balancing 3-phase loads in a large high school, a testing company discovered that the main electrical service earthing conductors were carrying 32 amperes of current. The service is a 4,000 ampere, 480/277 volt, 3-phase, 4-wire service. The bare copper earthing conductors run from the main switchboard ground bus to a 3 ground rod grounding arrangement. The current was measured using a clamp-type ammeter. The current measured on the EGC bonded to the switchboard ground bus and running to the water pipe was 0. The current measured on the EGC bonded to the switchboard ground bus and running to the building steel was 10 amperes. To isolate the cause of the current, different portions of the electrical service were de-energized until they discovered that most of the current measured was coming from electrical distribution equipment serving HVAC equipment. The EGC current was reduced by 20 amperes when 2 large rooftop air handlers serving the pool area were de-energized, 10 amperes for each unit. Almost all of the remaining current stopped incrementally when each remaining piece of HVAC equipment was de-energized. The testing company was satisfied that the current was ?normal? leakage current. I have not heard of this before. First question, is it normal for equipment to ?leak? current such that it causes current on the bonding conductors? Second, could the current measured simply be the result of how a clamp-type ammeter measures current by measuring a magnetic field? Since the bonding conductor is routed in close proximity to current carrying conductors, and therefore have a magnetic field around them, could that current measured simply be a result of the implied magnetic field? Is the current even there? Third, should there be further investigation?

Thanks for helping me understand!

Jebidieh
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
First, if the jaws of the clamp-on ammeter were properly closed, there is little to no chance of a spurious reading from nearby high current wires. You can test this easily enough by clamping the meter next to the ground wire instead of around it.
I will leave the question of "normal" current to others, except to note that it is not uncommon to set the ground current/residual current trip point as higher than 10% of the total load current in some feeder situations.
The fact that the A/C units in question are ones serving the pool area makes me less willing to accept large current as normal without more investigation though.
 

kwired

Electron manager
Location
NE Nebraska
Make sure you have no neutral-ground bonds downstream from your service equipment.

Make sure any separately derived systems are properly grounded/bonded, you could be reading current from those systems.

You can be reading current from upstream separately derived systems as well, like the POCO primary system, but the changes in current from changes in load within your facility seems to suggest most of your current is originating within your facility.
 

jebidieh

Member
Location
New Florence, PA
Thanks for your input!

Thanks for your input!

Make sure you have no neutral-ground bonds downstream from your service equipment.

Make sure any separately derived systems are properly grounded/bonded, you could be reading current from those systems.

You can be reading current from upstream separately derived systems as well, like the POCO primary system, but the changes in current from changes in load within your facility seems to suggest most of your current is originating within your facility.

Thanks for your input kwired. The first course of action by the testing company was to confirm that all neutral to ground connections were installed correctly and that no additional neutral to ground connections were installed further downstream. They and we were convinced that there were no bonding straps left in or inadvertently installed where they are not permitted or intended.

Thanks

Jebidieh
 

jebidieh

Member
Location
New Florence, PA
That Answers the Ammeter Question

That Answers the Ammeter Question

First, if the jaws of the clamp-on ammeter were properly closed, there is little to no chance of a spurious reading from nearby high current wires. You can test this easily enough by clamping the meter next to the ground wire instead of around it.
I will leave the question of "normal" current to others, except to note that it is not uncommon to set the ground current/residual current trip point as higher than 10% of the total load current in some feeder situations.
The fact that the A/C units in question are ones serving the pool area makes me less willing to accept large current as normal without more investigation though.

Hi GoldDigger!

Thanks for your contribution. My experience has led me to try to understand what the test instruments are reporting before trusting the information that I read. Am I reading a potential on the conductor or simply an implied potential by a magnetic field? Your response makes it easier to trust the measurements reported by the testing company while using a clamp-type ammeter.

Thanks again.

Jebidieh
 
Hi Folks!

I need some education. While balancing 3-phase loads in a large high school, a testing company discovered that the main electrical service earthing conductors were carrying 32 amperes of current. The service is a 4,000 ampere, 480/277 volt, 3-phase, 4-wire service.
Lets run a reality check here. 32/4000x100 = 0.8% Not a number that raises a lot of concerns for me.
The bare copper earthing conductors run from the main switchboard ground bus to a 3 ground rod grounding arrangement.
It would interest me as to whether there were three rods because than quantity was needed to achieve a low ground path resistance number or whether they found it necessary for decreasing objectionable touch voltages? Either way it's only a curiosity. What is the touch voltage between various bare steel on the building, water pipes, exposed beams, etc, as measured to local ground?
The current was measured using a clamp-type ammeter. The current measured on the EGC bonded to the switchboard ground bus and running to the water pipe was 0. The current measured on the EGC bonded to the switchboard ground bus and running to the building steel was 10 amperes.
Without an enumeration of all the ground bus connections and the amperage on each it is hard to assign any significance to those readings other than to say the bonding is doing it's job of reducing touch voltages between the building steel and premise wiring.
To isolate the cause of the current, different portions of the electrical service were de-energized until they discovered that most of the current measured was coming from electrical distribution equipment serving HVAC equipment. The EGC current was reduced by 20 amperes when 2 large rooftop air handlers serving the pool area were de-energized, 10 amperes for each unit. Almost all of the remaining current stopped incrementally when each remaining piece of HVAC equipment was de-energized. The testing company was satisfied that the current was ?normal? leakage current. I have not heard of this before. First question, is it normal for equipment to ?leak? current such that it causes current on the bonding conductors?
Yes. Especially big motors. What's your A/C service schedule like? How often do they get around to blowing the dust and cr@p out of a given unit?

Second, could the current measured simply be the result of how a clamp-type ammeter measures current by measuring a magnetic field?
The reading is correct.
Since the bonding conductor is routed in close proximity to current carrying conductors, and therefore have a magnetic field around them, could that current measured simply be a result of the implied magnetic field?
To the extent there is unbalance in the current carrying inductors, yes. But it is still ground current.
Is the current even there?
Yes.
Third, should there be further investigation?
The currents are not a problem as long as the bonding and grounding systems are doing there job of mitigating touch voltage potentials throughout the facility. I would be much more interested in knowing the voltage numbers rather than the current readings.
Thanks for helping me understand!

Jebidieh
Thank you for a well written question and a good preliminary investigation report.
starbolin
 

kwired

Electron manager
Location
NE Nebraska
Thanks for your input kwired. The first course of action by the testing company was to confirm that all neutral to ground connections were installed correctly and that no additional neutral to ground connections were installed further downstream. They and we were convinced that there were no bonding straps left in or inadvertently installed where they are not permitted or intended.

Thanks

Jebidieh
It is one thing to look in panelboards for inadverntly installed bonding strap/screws, and it is something else to have a neutral bonded to ground in some piece of equipment where it wasn't expected. Was it ever verified there is no continuity between neutral and ground when one isolates them from the bonding jumper(s)?

If HVAC units seem to be contributing to the measured current on the grounding system, one should give those units some closer examination. They may have separately derived systems within them and maybe there is issues with those systems.
 

meternerd

Senior Member
Location
Athol, ID
Occupation
retired water & electric utility electrician, meter/relay tech
POCO secondary transformer neutral is grounded at the transformer via a ground rod (for underground services). Primary neutral will also likely be grounded at the pole. GEC is bonded to the secondary neutral and enclosure at the service disconnect. That creates a parallel path back to the transformer neutral through the neutral conductor and the earth, so you may have current returning to the transformer neutral through both the neutral and the earth, depending on resistance. Current takes ALL paths back to the source, not just the "easiest" path. Impedance of the neutral conductor is not always a perfect zero ohms, so some current returns through the earth. Pretty common to see some current on the ground, at least in my experience as a POCO meter tech. Long secondary runs from the transformer to the service panel are a common cause.
 
Last edited:

kwired

Electron manager
Location
NE Nebraska
POCO secondary transformer neutral is grounded at the transformer via a ground rod (for underground services). Primary neutral will also likely be grounded at the pole. GEC is bonded to the secondary neutral and enclosure at the service disconnect. That creates a parallel path back to the transformer neutral through the neutral conductor and the earth, so you may have current returning to the transformer neutral through both the neutral and the earth, depending on resistance. Current takes ALL paths back to the source, not just the "easiest" path. Impedance of the neutral conductor is not always a perfect zero ohms, so some current returns through the earth. Pretty common to see some current on the ground, at least in my experience as a POCO meter tech. Long secondary runs from the transformer to the service panel are a common cause.
And if you share some metallic path with a neighboring service you have a low resistance path compared to just an earth path. You will see higher current levels on the grounding electrode system in those kind of instances.

Also sometimes the current you are measuring contains current that originates from POCO primary distribution system, if you turn off all loads in your facility and still have current on your grounding electrode - it is probably not originating in your facility.
 

Conrad

Member
More common than you might think

More common than you might think

Your isolation of the leakage to the air conditioning units points you to where you should continue to look.
There is no established max total fault current specified for general use, as others have pointed out feeder GFCI trip levels often must be bumped to prevent nuisance trips.
Inevitably it boils down to a decision of "how much is too much". Checking the A/C's further could point out phase to grd leakage in a hermetic compressor (motor windings sit in, and are cooled by the refrigerant), or in any other motor. Since the A/C is likely to also be connected to building steel, metallic water or gas piping, there are multiple current paths and possibilities.
There is always a point at which whoever is paying considers a fault less of a problems than than costs to investigate, but a few minutes at the A/C could isolate the problem. Kirchhoff's law is your friend, clamp an amprobe around all the phase conductors feeding a compressor or motor, you should have zero in a perfect world. If the unit natural gas heat clamp your amprobe around the gas piping, you may find whatever fault exists is also causing current on it as well. In A/Cs never forget someone with less knowledge and perhaps a desire to get off the roof may have been "creative" in a repair, and any 120v in a 277/480 system was the result of control transformer.
Perfect systems exist mainly on paper, but fault currents you described could go unnoticed unless you are looking for them.
The mention that both a/c's have a very similar fault current is sound the ole alarm bell in the back of my head.
 
Status
Not open for further replies.
Top