Here is the situation:

A new installation of a generator. the generator is installed on the 34th floor (roof) of a large building. There are control conductors in a conduit run down to the basement - 34 floors- with 8 junction boxes between. The control conductors are 14 THHN with a 1-2 amp load. The conductors are run from the fuel location in the basement to the control panel inside the housing of the generator on the roof.

They ran 6 conductors for the controls, and 14 spare conductors. The spares conductors were not terminated at either end.
In the junction boxes, there is a metallic barrier the same gauge as the box. One side has the control conductors, the other side has the 480v conductors from the generator to the load locations.

Without any load connected/or the generator actually running we have an unusual situation.
NOTE: the control conductors are for alarms to notify when the fuel source is low, or out - to shut down the generator.

when the control circuit is tested, all the unused (spare) conductors develop 54 volts. This is creating havoc with the alarms.
If a spare conductor is bonded to the enclosure at one end, 2 volts are shaved off of the 54 volts. Bonding the other end of the same conductor shaves off another 2 volts. Continuing to do this with the other spare conductors reduces the "Phathom" voltage until the alarms are not affected anymore. This is a temporary solution, until we can find the reason this is occurring.

Remember the 480v wiring has not been energized yet.


What could be the reason this is occurring?

Has anybody run across this before?

Thanks for the help, I appreciate any answer, as it may lead to solving this situation - even the generator techs have not come up with a solution, as they are protecting themselves and are saying this is a wiring mistake. Maybe they are correct, I am not sure.

did you any chance did the control circuits on seperated conduct pipes ?

if not sometime the induction of stray voltage can get ahold of it and btw is the control circuits on DC or AC ??


Merci, Marc

It does sound like AC induction. Is it 54V AC or DC?
When you measure 54 V, where are you connecting the meter's leads?
What is the material of the conduit? If it is metallic conduit, is it bonded to earth or building ground?
Is there a large ground fault current on the building ground?
Place a AC current meter clamp around the entire conduit if it is small enough and see if there is a current on the conduit. If you cannot do that, can you (Carefully, there is a shock hazard!!) break the conduit's ground path and use a ammeter or low-ohm-large-watt resistor to measure the current?
Measure the frequency of the voltages on the wires and look at their waveform. Tell us what the results of the tests are.
What other systems' conduits are in the same chase with the conduit for the generator?

I failed to mention that you will use a voltmeter to measure the voltage drop across the resistor.

In answer to your question -have we seen this before?- Yes, and it is just a matter of finding the source of the current and isolating the control circuit or filtering out the voltages.

One possibility is to get rid of the wires altogether and install a fiber optic line instead. But, that is a final and expensive but interference free solution.

I know it's 34 floors so this may not be feasible but replacing the 6 control conductors that are actually being used with a manufactured cable assembly with all the conductors wrapped and spiraled together thus cancelling out any magnetic fields may be a solution if no other solution is found. Probably not feasible, just brainstorming with ya.

Thanks for the answers!!!

I was not involved in the installation, but was called with the information I passed on to you.

But your answers have sparked a thought into my head, especially Spaceman's response.

This is a NYC installation, and service grounding in some of these buildings leaves some question marks. I will pursue that avenue first and see what results I get. I will let you know, when I know.