Voltage Measurments

[ray cyr]

The plant I work in has three large, three phase transformers. Each one feeds its? own switchgear. The serving utility tells us that all three are Delta secondary configurations. I am told that each is an ungrounded system.
On system ?A? at disconnects fed from its? switchgear, when I measure voltage I read 480 volts from line A-B, line B-C and line A-C. When I measure voltage between line A and the disconnect cabinet, I read 480 volts. Same thing for line B. Between line C and the can I read 4 volts (nope, not a typo).
I get the same results on system ?C?.
On system ?B? I read 480 volts line to line and 277 volts from each phase to the can.
I know what my thoughts are on this and they don?t agree with what I am being told?
I will appreciate your thoughts on this.
Thanks,
Ray

 

[SEO]

A&C sound like corner grounded delta and b sounds like a Y system.

 

[Volta]

Sounds like systems A and C are corner grounded, and B is a wye. A and C could have been ungrounded previously . . .

 

[dbuckley]

Being as these systems are ungrounded, and in the absence of further information regarding ground monitoring and/or resitances, there is no correct voltage for anything other than line to line. The 480V line to line is correct. The can and switchgear are isolated from the secondary of the transformer, they are floating.

But.... on system A, it smells like line C is unintentionally (or maybe intentionally) grounded. Normally, there is monitoring in place to detect this, so you can investigate and fix the problem before one of the other phases gets grounded and you have a bit of a bang on your hands.

 

[dbuckley]

You guys were quick!

Sounds like ... B is a wye.

OP states transformers have delta secondaries...

 

[ray cyr]

You guys were quick!



OP states transformers have delta secondaries...

Actually I was told that they are delta....my readings make me question that

 

[dbuckley]

You need to do a visual inspection of the transformer secondary connections (or the nameplate!) to determine what is what. If you do have wyeness going on then system A may be distinctly dodgy!

 

[ray cyr]

I am told we have a powerdown scheduled in May to PM the switchgear, hopefully we can do thorough visual inspections then.

 

[don_resqcapt19]

Ungrounded systems can show voltages just like a wye when the loads and capacitance on each of the phases are very close to equal. Do these systems have ground detector lights? It appears to me that you have an accidental ground fault.

 

[RHJohnson]

You say those 3 transformers belong to the company you work for. In that case you should be able to gain access and read the nameplate data - then you will know if they have a delta secondary.

If they do have an ungrounded delta secondary: A and C transformers each have a ground on "c" phase.
Transformer "B" is reading correct......480 v ph to ph, and 277v to ground.

Sounds like heavy industrial plant......is this the mine just outside Yakima?

 

[jim dungar]

I go along with the group that says you have a ungrounded systems that has a fault on your C phase. It does depend on where you are making your measurements, but I would not expect to see 4V phase to ground on a intentionally corner-grounded system.

And as others have mentioned, the more balanced your loads the more likely you are to read 277V to ground on an ungrounded system.

 

[masterinbama]

During you next shutdown. Measure the voltage while dropping loads 1 by 1. If your voltage changes with the loss of a particular load you have isolated your ground fault. Then you can keep isolating downstream until you find the actual fault. I have found ground faults in open delta setups in odd places, had one phase to ground on a high bay fixture once, talk about a needle in a haystack.

 

[ray cyr]

Thank you very much for the replies. There is so much that I still don't know that I didn't want to close my mind to other possibilities simply because I don't know of their existance. Also, I want to trust the people that are telling me what these systems are even though my limited experience can't explain the difference between what they are telling me and what I think my meter is telling me.
RHJohnson, nope this isn't that mine... I live in Yakima but had to go to Ellensburg to find employment, this is a food processing plant.
You are all saying that 277 volts to ground would be expected on an ungrounded balanced system, great that means that B system is probably working correctly, but I sure would like to understand the theory behind this . Do any of you have any suggestions for resources that I can read up on this?
Thanks guys,
Ray

 

[LarryFine]

Ray, how about a little more detail?

For example, presuming the non-grounded Delta story is accurate, a high-impedance meter should indicate what your B showed, and act like a grounded-Y. I'd repeat the test again, but with a low-impedance tester, like a solenoid-type tester, you should read more as if every phase was loosely grounded; little voltage.

A non-grounded system acts like a center-grounded system because of the relatively-balanced capacitance between each phase and the rest of the grounded world. The capacitance acts as if there was a high-value resistor between each phase and ground. A voltmeter reads the voltage across that phantom resistor.

When you can, for the two "corner-grounded" transformers, I agree with shutting down one circuit at a time, testing all three phases each time. An accidental ground anywhere in the premises will give you what you're seeing. Or, you could start with main breakers, again testing each phase to ground with each change.

 

[ray cyr]

Thank you for the replies guys :smile:. My journeyman and I were able to spend a little bit of time with this over the past couple of days and this is what we found.
There is monitoring equipment, for 2 of the systems, in our shop. One of the monitors is for B system, it has three lights that stay on steadily, dimly. The other monitor is for A system. This is where the fun begins.
It had one light on bright. After some investigation we determined that 2 of the small transformer light sockets were bad. After replacing these, we had 2 bright lights and one not on. Ok....ground fault on that phase. A few minutes later, all 3 lights were dim, then a few minutes later back to 2 bright, one off. Ah, so this ground fault is associated with a machine, when the contacts are closed the fault appears. Well, this all happened yesterday.
When we went to work this morning we saw that the one light that had been off was now bright and the other two were dim, so I took voltage measurements.
A-G 236v,
B-G 238v,
C-G 416v.
After drawing out some circuits, I think what is going on here is a ground fault in a wye configured motor.
Any comments are welcomed :smile:
Ray

 

[dbuckley]

Good work.

After drawing out some circuits, I think what is going on here is a ground fault in a wye configured motor.

Not convinced its a wye motor in fact not convinced its a motor at all, but its something that is either intermittent or is switched on and off. See if you can figure out if there is a pattern in the times of brights to all dim, and if so, start asking around; someone will know what the times mean.

If theres no pattern you're either going to have to start dissecting the plant, by taking stuff offline, or have one of you at the panel watching the lights and the other wandering around, radio in hand, hoping to strike lucky...

 

[ray cyr]

Good work.



Not convinced its a wye motor in fact not convinced its a motor at all, but its something that is either intermittent or is switched on and off. See if you can figure out if there is a pattern in the times of brights to all dim, and if so, start asking around; someone will know what the times mean.

If theres no pattern you're either going to have to start dissecting the plant, by taking stuff offline, or have one of you at the panel watching the lights and the other wandering around, radio in hand, hoping to strike lucky...

It's looking like this is what we are going to have to do. Another idea that was mentioned is building a portable indicator to carry around with us because we have LOTS of 480 volt welding outlets throughout the plant, and they are usually near a panel. So we could plug in close to a panel and watch the indicator as we open and close circuits.
Whatever it is I have a feeling that it is going to take quite a few manhours to nail it down.
I'll let you guys know what we finally find, thanx for your suggestions :smile: I'm certainly open to more if anyone has 'em.
Ray

 

[Smart $]

It's looking like this is what we are going to have to do. Another idea that was mentioned is building a portable indicator to carry around with us because we have LOTS of 480 volt welding outlets throughout the plant, and they are usually near a panel. So we could plug in close to a panel and watch the indicator as we open and close circuits.
Whatever it is I have a feeling that it is going to take quite a few manhours to nail it down.
I'll let you guys know what we finally find, thanx for your suggestions :smile: I'm certainly open to more if anyone has 'em.
Ray

If the culprit is a switched load, measure the current at the switch gear just before, during, and after any change in indicator status (using a recorder would be best). This should help in the elimination process by following this change in current downstream, that is if occurence is somewhat frequent. Otherwise, the amount the current changes on each change of status will give an indication of possible loads involved. You may also have more than one intermittent fault, since it appears no one has been monitoring for a while