Parallel cables routed in tray-200 amp ground currents

[robthecob]

We are measuring 200 amps in the ground conductors from the 480 volt substation ground bus to the MCC ground buses. The 480 volt sub is a double-ended substation with two 3000 kVA 12.5kV-480 volt transformers. Each half of the substation powers three 1600 bus MCCs, each with 1000 amps of load. The MCC feeder cables are four 1/C-500MCM cables/phase routed in cable trays for 300'. The MCC feeder cables have been installed in the cable trays with the A phase cables grouped together (in one plane), then the B phase cables grouped together, then C phase cables grouped together with the single 1/C-4/0 ground conductor laying on top. I realize this is not the proper method for installing three phase cables in tray, but I inherited this problem. The motors powered from the six MCCs are connected to the MCC starters via 3/C with ground tray cable with approximately 15 amps measured in each ground conductor to the motor. The ground current to the motors is almost identical for each motor on the various MCCs.
The transformer neutrals are connected to the substation ground bus via busbar. The substation is located 150' behind the plant with elevated cable trays routing the 480 volt TC cables to the MCCs inside the plant. The six MCCs are located close together inside the plant electrical room with cable tray distribution from the MCCs to the motors. The cable trays are bonded to the MCC ground buses and each other.
I suspect the MCC parallel feeder cable routing is the problem, but I find it hard to believe it would be to this magnitude. Any ideas?

 

[weh3]

Are there bonding jumpers installed at the MCC's? With no neutral and no bonding jumper, the zero-sequence of any ground fault would be seen in the grounding conductors. Suspect a high-impedance fault.

William

 

[rbulsara]

Is the tie breaker between the two substations normally open or closed?

 

[robthecob]

Question to weh3: What is your definition of a bonding jumper?

Answer to rbulsara: Tie breaker is normally open

 

[rbulsara]

rob:

OK. Open tie eliminates suspicion of circulating currents.

Next question:

Does the 15A ground current in individual motor grounding conductor flow only when the motor is running or the ground current is also there when the motor is off? Please check a few motors.

 

[jghrist]

With the grounding conductor closer to one phase, and grounded at each end, there will be circulating current similar to the circulating current in the shield of single-core cable that is grounded at each end. 200A seems like a lot, though. Are the MCC ground buses connected to building steel or a ground electrode that is connected to building steel? The induced current would have to have a path to flow back to the substation. If the MCC ground bus is not connected to a grounding electrode or building steel, then maybe the circulating current is flowing through the motor feeder grounding conductors to building steel at the motors and then through earth to the substation grounding electrode. This current in the motor feeder grounding conductors would be independent of individual motor current.

If my hypothesis is correct, bonding the MCC ground bus to building steel may eliminate the motor circuit ground currents.

 

[dpc]

Do you have any way of looking at the ground current waveform? It would be interesting to know this was all fundamental current or if it contains significant harmonics.

Are these motors fed via adjustable frequency drives?

I agree with jghrist that this is likely some type of circulating current not directly related to the motor current.

 

[PowerfulStuff]

Unfortunately I expect that the 200A is a realistic current due to the cable installation method. I have come across similar situations with 10-20A currents but that has been with cable installed well (in trefoil).

What is probably happening is that all the grounded tray is and ground conductors are forming a very solid shorted turn where current is induced by the considerable magnetic assymetry of the cable installation. An easy way to verify this is to plot how your ground current changes with current through the main conductors as their should be a linear relationship.

In a theoretical sense fixing it is easy - remove the assymetry in the cable installation or open the ground loop. Unfortunately practically I don't think that you can avoid reinstalling the cables!

 

[rbulsara]

Also check the current in the neutral conductor between its first bond to the ground and the x0 point. If most of this is induced current, which is very likey, there should not be any current in the above section of the transformer neutrals.

 

[mc5w]

I take it that the 4/0 equipment grounding conductor is used as a backup for the equipment grounding path provided by the cable tray. Am I correct?

Is there anything that is trying to use the equipment ground as a neutral? If so, you need to put neutral conductors into the cable trays and separate the load neutrals at the MCCs from the equipment ground. Accidental or purposeful regrounding of a neutral and using the EGC as a neutral are 2 common violations. You have to remember that 70% is a passing score on licensing exams which means that the registered EE, the electrcians, and the electrical inspectors could all be only 70% correct or 75% correct.

If the phase wires are not quadruplexed around each other you can rent a high reach, cut the nylon cable ties, and rearrange the wires ABCABCABCABC. If not, you would need to shut down 1 MCC at a time and rearrange the conductors.

Are the 500 KCM phase conductors operating as single conductors in free air? If so, you can cut voltage drop and improve system efficiency by doubling the number of conductors and still be Code on your cable tray fill provided that the original conductors have 1 conductor diameter of space between them. Some of the extra conductors would need to be neutrals plus the equipment ground resulting in six 500 KCM per phase instead of 4 and three 500 KCM for the neutral instead of zero.

Explain to your boss that when you heat up the wires he is paying for electricity and not getting it.

 

[mplsscooter]

hi
not quit sure if I understand the problem but you should not have each phase grouped ie aaa, bbb,ccc the group should be abc this way inductances will cancell refer to nec section one conductors in mulyiple conduits
scooter

 

[PWR]

I have observed higher than 200Amps in similar situations. You can reduce the magnitude by "grounding" your ground conductor at intervals in the run of tray. As it is now, your "transformer" is 300 feet long. If you removed the insulation from the conductor at the center of the run and bonded it to the tray you should reduce the current by approximately one half. Similar proportional reductions would occur if the bonding is done at shorter distances.

 

[weh3]

Rob the Cob,

A bonding jumper (and I refer to the Main Bonding Jumper or MBJ) is the connection between the equipment grounding bus and the grounded bus (neutral) at the service entrance (your MCC). This jumper ensures that fault currents will reach a physically grounded point (the grounded bus).

Is your MCC mains section labeled as service equipment?

William

 

[drbou]

We have just comissioned a new installation in Germany and we are measuring currents on all are grounding conductors although not as high as those mentioned.

For example our fiberline building has six 30kV,725V transformers each feeding an MCC.
All distribution is with trefoil cables.
Currents up to 5 amps are measured in the motor feeder grounding conductors.

These motor currents seem to add up and up to 51 amps is measured on the switchgear main grounding conductor to building ground.

The grounding conductors for trays also have measurable currents in the order of around 10 amps.

I believe the currents measured in the motor grounding conductors (yellow-green) are caused by the non cancllation of the magnetic field in the cable.

The local authorities do not accept such currents although I could not get any value from them. The NEC & CEC describe do not give magnitudes for the allowable grounding currents, anybody has any idea?

 

[PowerfulStuff]

drhou,
These currents sound very high. I'm not in your part of the world so can't give any Standards to refer to. A helfpul comparison would be to look at sensitive earth fault protection units to see what tends to work i.e. the Merlin Gerin NS250 Vigi units have a 3 A maximum setting for a 250A breaker.

Other questions that come to mind are -
* What power rating and winding types are the transformers?
* Can you describe the neutral to earth connection
arrangement for the transformers and MCC's are?

 

[drbou]

PowerfulStuff,

The IEC and VDE standards that I have seen so far do not mention any values.

In the building I refered to there are nine transformers 30kV/0.725V delta-wye (Dy5)
3.125 MVA dry type transformers.

Also in same building there is also one 30kV/400V, 3.125 MVA delta-wye (Dyn5)arrangement, TN-S system, that is 3 phase 4 wire with separate neutral and ground conductor. I have not yet made any measurements on this MCC.

The last transformer is a 30kV/6.3kV, 6.3 MVA delta-wye (Dyn5) feeding an O2 plant located about 100m away.

The distribution system for the 725 V MCC's is an IT sytem, that is neutral not grounded. Ground fault detection is made with Bender relays that measure zero sequence voltage as part of their scheme to evaluate ground faults.

On the MCC's where I measured those continous ground currents all phase voltages to ground are balanced meaning no real ground faults.
I also intentionally made measurents on MCC's that have no VFD's, there are no significant harmonics in the currents measured so far.

I was thinking of possible interaction between the two systems IT and TN-S.

 

[robthecob]

I want to thank everyone for their comments on this issue. We checked the various items that were mentioned in this forum. The reply that we have a 300' long current transformer is correct. We installed a 4/0 jumper from one of the MCC ground buses to building steel with no effect on the ground current in the ground conductors between the substation and the MCC; however, the new 4/0 bonding jumper had 10 amps measured. We discussed this issue with various people (in addition to this forum) and determined the ground current must be a result of the incorrectly routed phase conductors in the tray. As first, I thought the magnitude was too high (200 amps in some); however, the total current in all of the phase conductors from the substation to the MCCs was around 5,000 amps, with the total ground conductor current being around 900 amps. There is one ground conductor for each of the 12 phase conductors and the ground conductor is typically routed directly over the B phase throughout the entire length of the cable tray between the substation and the MCCs. The inductance from the 12 parallel cables induces a current on the ground conductor. The only fix is to reposition the phase conductors in the cable tray. At least, we don't have all of the A phase (or B or C) cables routed through the same chase nipples at the MCCs, which would cause more headaches. Thanks to all for your advice.

 

[tommom]

There must be a second connection between ground & neutral. Otherwise you just see an induced voltage from the phase wire groupings and not a current flow.

 

[robthecob]

Reply to tommom:
Each of the two 12.5kV-480V substation transformers have a bus connection between the transformer neutral and the substation ground bus; in addition, there are two MCCs with neutral buses which are cable connected to the substation ground bus. From what we can tell, the "300' long current transformer" is causing current flow from the substation ground bus thru the ground conductor from the sub ground bus to the MCC ground bus, then thru the motor ground conductors (using TC cable) to building ground, then back to the sub ground via the underground ground grid.