Recently, my facility has revised their lockout tagout program which has prompted an internal discussion of the "best practice" when selecting a suitable location for the zero energy verification portion of the lockout procedure. For the sake of argument, I am considering a case where you're replacing a motor which has a single source of power (fed from a MCCB within a MCC in a separate room), associated contactor, and a local disconnect physically close to the load. Currently, there are two major schools of thought:

A) OPEN the MCCB at the MCC and apply lock. Then, perform zero energy check downstream at the local disconnect
Pros:
- Local point of test (less likely to test the wrong device due to a plethora of look-a-like equipment at the facility)
- Reduced arc flash hazard (typically) due to MCCB directly upstream of the disconnect
Cons:
- Zero energy check at the disconnect could provide false indication that the EID has done its job if the contactor is not made up at the time of the test (would need to perform a try test from control system while performing zero energy check)

B) OPEN the MCCB at the MCC. Then, perform zero energy check on the load side of the MCCB in the MCC and apply lock
Pros:
- Convenient "live-dead-live" check via the line side of the MCCB
- Ensures that the contactor argument in the "cons" above is accounted for and the EID (MCCB in this case) has done its job
Cons:
- Potential for locking out and testing the wrong power source (see "look-a-like" equipment, puts a heavy reliance on test-before-touch practices)
- Higher (typically) incident energy value when performing the check at the MCC

I apologize if this topic has already been discussed, but I could not find it explicitly addressed in my search. I greatly respect the professional opinions provided on this site and would love to hear your thoughts on this topic.

Thanks in advance,
Zak

At some of our facilities we do not have disconnects located at every motor. In those cases our EC Techs only option is to shut the MCCB off, open it, and verify that the energy source is dead. I do recommend that they use a proximity tester on the circuit wiring before they take the wiring loose at the motor, but only after they have tested the voltage source.
Where we do have the disconnects, we do need to test at the disconnect after shutting the MCCB to the off position. This is only how it is done a tour facilities, you will need to come up with your own plan at yours and make it a part of the electrical safety plan.
Thanks!

Zak,

Ideally I like MCC’s with isolators and low fault level test points installed on the front door of the cell. The test points are taken from a point on the low side of the CFS or Switched MCCB and enable testing for power, operating the switch and verifying isolation and change of state and then lockout all without opening the cell door. Most older MCC’s weren't built with this as an option. Many do have phase indicators which are often used for verification for mechanical work such as changing a pump or replacing a belt. I don't like their use for electrical work and prefer a voltage test for verification. Having the lights change state is however a good indication that the isolator is definitely open on 2 phases at least (depends on circuit used for lights) and that the likelihood of you initiating an arc fault is reduced when opening the cell door to allow for the verification test.

When you are at the motor it should be pretty obvious if you have 3 phase power available, so the main thing you need is a correctly applied unique labelling system.

Best practice is to always have the worker perform a full test before you touch with test lamps or a suitable voltmeter as a separate step to isolation verification. 1 phase can be live without the motor turning.

False negatives can occur with the volt sticks, so I would never trust them as a final check.

The lowest fault level is the best point to test, however you need to ensure that it provides verification that that the main power circuit is isolated. The best isolation procedures nominate the safest point to perform the isolation verification test.

I would argue for "none of the above". If you want the live-dead-live capability, why not check it dead at the local disconnect?
That is after all the WHOLE reason for having a local disconnect, isn't it?
Far, FAR less likely to have "de-energized" the wrong thing, if you do it there.