Generator Tests Looses Neutral

[ed downey]

I just got a call from a building owner that has indicated that when he tests his generator once a month he transfers the building over to emergency power and is having a lot of problems with burning up ballasts & electric resistance heating coils.

I have a meeting scheduled at the site at 1:00 PM today to meet with the building owner and the electrician to figure out why this is happening.

My theory is that they are loosing their neutral when they switch to emergency power. (I am not sure if this is a 3 pole or 4 pole transfer switch. I assume it is a 4 pole transfer switch)

I was curious if anyone else had run into this problem. And if there is any other explanation for this.
-Ed Downey

 

[wanderer20001us]

Re: Generator Tests Looses Neutral

This can happen if the transfer switch is 4 pole and they have not bonded the generator neutral. I also had a situation with a 277/480Y genertor feending a 3 pole transfer switch in a 3ph, 3w set up. The generator neutral was not bonded and was floating. This caused problems with many pieces of equipment because the generator voltage had no ground refernece.

Sam A. - PE

 

[catchtwentytwo]

Re: Generator Tests Looses Neutral

Sound like an open neutral kind of issue or the generator output voltage/frequency might be out of wack. I'm assuming this is an open-transition transfer switch. Does the problem occur "going to generator" or "back to normal"? Or both ways? Have you put any recording devices on the system? You could, at minimum, use a meter with Ma Min-Max feature although a Dranetz or other PQ recorder would be better.

 

[don_resqcapt19]

Re: Generator Tests Looses Neutral

Sam,

This can happen if the transfer switch is 4 pole and they have not bonded the generator neutral.

I don't understand how the lack of the required bond when using a 4 wire transfer switch would cause ballasts to fail. They would only see the output voltage of the generator. My comment assumes that there is a 4 wire connection between the generator and the transfer switch.
Don

 

[wanderer20001us]

Re: Generator Tests Looses Neutral

If the generator is not bonded, when the 4 pole transfer switch operates, the generator system has no solid reference to ground. The whole system is free to "float" based on the capacitive coupling of the connect equipment. Electronic ballasts, especial cheap ones, cannot always handle the increase electrical stress that occurs if one of the ungrounded conducts reference closer to ground than the other due to the capacitive characteristics of the load to ground, not unlike an ungrounded delta system. The first ground doesn't cause an OCD operation but subsject the other put to higher that normal voltage to ground.

Sam

 

[dereckbc]

Re: Generator Tests Looses Neutral

Sam, wouldn't that depend on the type of neutral contact at the transfer switch, either "make before break" or "break before make"?

 

[wanderer20001us]

Re: Generator Tests Looses Neutral

It doesn't matter. The end result is the utility ground system is disconnected from the load and, if the generator is not grounded, the load neutral is ungrounded. 'Make before break' or 'break before make' still has a 'break'.

 

[dereckbc]

Re: Generator Tests Looses Neutral

OK, you are assuming the generator is not grounded, I missed that the first time.

 

[ed downey]

Re: Generator Tests Looses Neutral

Well here is what I got out of my meeting with the building owner, electrical contractor, transfer switch rep., and generator rep.

The generator does have a ground rod driven at the unit.

The transfer switches are 4 pole with overlapping neutral and closed transition (100mS type). The contacts on the neutral appeared to be a little more than 1/8" apart so the transfer switch rep. closed the gap to about 1/16".
We then tested the system after this and everything appeared correct looking at the (3) transfer switches. We checked voltages at the switches both before and after transfer with a Fluke DMM everything appeared O.K. The last test was to have someone inside the building while we made the transfer and see if they noticed anything. At this point when the life safety transfer switch was transfered the flourescent lighting basically pulsed on & off for about 3 to 5 seconds.
We have now scheduled a technician with a recording meter out to the site on friday to check everything out as we transfer loads.

I am out of possibilities so any help will be greatly appreciated.
-Ed Downey

 

[ron]

Re: Generator Tests Looses Neutral

Be sure that the "ground rod driven at the unit" is connected to the building's grounding electrode system.

 

[roger]

Re: Generator Tests Looses Neutral

Just a few questions that you have probably covered.

Are there inphase monitors on this system?

Are there motors transfering during the test.

What happens during a real event with a delay in the transfer from normal to emergency?

Is this a hospital? (questioned because of the mention of three switches)

Roger

 

[ed downey]

Re: Generator Tests Looses Neutral

Roger,
Yes, there are in phase monitors the transfer switches will not allow a transfer until they are within 8 degrees
On one of the (3) xfer switches there are motors connected but if we transfered only the life safety load (which does not have motors) we get the problem with the lighting.
This is a medical building that does surgery (mostly outpatient but they do have a few overnight beds).
I forgot to answer one of your questions. we did not simulate a utility outage.
-Ed Downey

[ February 17, 2004, 10:40 PM: Message edited by: ed downey ]

 

[ed downey]

Re: Generator Tests Looses Neutral

another item that I noticed while looking at the One-Line Drawing Is As Follows:
There is a driven ground rod at the generator. the generator then feeds to a distribution panel and that panel has another driven ground rod (Something that I don't usually see). Then the main service entrance switchboard has its own driven ground rods. I haven't traced this down in the field yet to make sure they are all connected as Ron has suggested.
-Ed Downey

 

[catchtwentytwo]

Re: Generator Tests Looses Neutral

Ed,

1. Does the problem occur "transferring to generator" or "transferring back to normal"? Or on both ?

2. Could you get two digital clamp-on ammeters with min/max features? Put one on the ground rod conductors at the main switchboard and the other at the generator's ground rod conductors.

I'd be curious to see what happens when you transfer and have the problem (possible circulating currents).

Regards,
John

 

[ed downey]

Re: Generator Tests Looses Neutral

John,

It Seems To Only Happen When We Are Transferring To Generator.

I Will See If I Can Run Your Test On The Ground Rod Conductors And See If There Is A Difference.

We Cannot Do Our Next Test Until Friday Afternoon When The Transfer Switch Technician Is Available.
-Ed

 

[highkvoltage]

Re: Generator Tests Looses Neutral

Ed, Have you checked the Hz when switching. When a load is place on the generator it will slow down for a few seconds. This is very common with generators. It is the same as when mowing your lawn and you go from short grass to very long grass if you don't stop to let the mower engine catch up the first few feet isn't cut very well. The difference here is the ballast wouldn't have the right Hz to work properly. Has anyone checked to see if the load is to great for the generator? I personally size generators at 150% for this reason.

 

[ed downey]

Re: Generator Tests Looses Neutral

Well To Wrap This Item Up We Had A Transfer Switch Technician At The Site On Friday Night To See If We Could Determine The Problem. After Testing The First Transfer Switch He Noticed That We Seemed To Loose The Neutral On Transfer To Generator Only. He Started To Trace Down The Wires And Found That The Load Neutral Was Landed On The Normal Neutral Termination And The Normal Neutral Was Landed On The Load Termination. THis Casued There To Be No Neutral When We Where On Generator. This Situation Occurred In Switch #1 & #3.