How to test electrical continuity of a metallic raceway used as an EGC?

Status
Not open for further replies.
H

harsha1972

Member
Location
USA
We have a lighting panel 480Y/277V 3P4W fed via the MDP without an EGC. The EMT is used as the effective ground fault path between the MDP and the lighting panel. Utility service is a grounded ?Y? 3P4W system. Is there a test that I can perform to verify there?s an effective ground fault path from the lighting panel to the MDP? (Physically cannot see the EMT as it runs inside walls and pipe chases of a 5 story building) I called Fluke tech support and they said that they don?t carry any equipment for this type of testing. Will a simple impedance reading between the neutral lug and the equipment ground lug in the lighting panel is enough to make a decision?
 
harsha1972 said:
Will a simple impedance reading between the neutral lug and the equipment ground lug in the lighting panel is enough to make a decision?
Click to expand...

Even if I ohmed out to zero I would still wonder if the available fault current would be carried. Could have a coupling just touching the raceway for example.

Unless I had a pretty solid reason to question the install of the EMT I would have to trust the EMT as the EGC.
 
I did a standard ohm meter test. Measured the resistance between the neutral terminal and the lighting panel feeder raceway locknut and did the following fault current calculation.

R(EMT) = Resistance of the EMT
R = measured resistance of the neutral + EMT conduit = 6 ohms
A(N) = cross sectional area of the neutral conductor = 1-#3AWG THHN=0.0973
A(EMT) = cross sectional area of the conduit = 1-1/2? EMT = 2.036
R(EMT) = R X A(N)/[A(N) + A(EMT)]
= 6 X 0.0973/[0.0973 + 2.036]
R(EMT) = 0.27366 ohms

Line to ground fault current through the breaker = 277V/0.27366 ohms
= 1012.2 A

This is enough fault current to trip a 20A breaker in the lighting panel instantaneously. I?m convinced that not only the metallic raceway use as the EGC for this lighting panel is electrically continuous but also is capable of producing a large ground fault current to trip the breakers instantaneously.
Had the numbers been questionable, I would have done a phase-earth loop impedance test.
 
130219-1723 EST

harsha1972:

I don't understand your 6 ohm measurement or the calculations.

#12 copper wire has a resistance of about 1.6 ohms per 1000 ft. If the conduit is good, then I would expect the resistance of the conduit to be less than a #12 copper wire. Even #3 aluminum is less than #12 copper. Thus, a measurement of 6 ohms for the sum of the neutral and conduit makes no sense.

For low resistance measurements you can not use an ordinary DC ohmmeter.

Essentially you need to do a four terminal resistance measurement. This means you pass a known current thru the resistance, then measure the voltage drop between two points. These points are directly on the resistance such that the voltage drop across the interface at the current injection points is not seen by the voltage measurement.

Run a long wire from the main neutral bus to near the far end of your conduit. This could be #22 insulated wire. It will be one test lead to your high impedance sensitive voltmeter. A Fluke 27 would work. Remove any power to the circuits in this conduit.

Measure the voltage between the far conduit end and the long test lead. Should be 0.

Use a 5 to 10 V isolation transformer with 20 A capability. Control the transformer input voltage with a Variac. Start at zero V. Disable power to any circuits that use the neutral. Connect one side of the transformer secondary to the neutral. Connect the other secondary lead to the conduit. Adjust the Variac to obtain near 20 A thru the conduit.

Measure the voltage over the length of the conduit. Should be small. The conduit resistance is the measured voltage divided by the injected current.

.
 
Status
Not open for further replies.
Top