Delta Neutral Reference

[trchambe]

It has always been my understanding that a true delta system has no neutral.

I came across a few documents today however that seemed to indicate otherwise. It says that there actually is a neutral, due to capacitance between the phases. Is this true? If so, can it be used as a reference, or is it a characteristic that has no application?

 

[waross]

No, yes and maybe. Unreliable and the indicated neutral may not be the true geometric neutral.
Use a grounding transformer or an artificial neutral.
Consider a four wire delta with a grounded center tap on one winding and a "wild" leg.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rbulsara]

It helps to narrow down the question to your specific application. It is difficult to explain a subject for which books and documents are written in a few lines.

Just a note that capacitance coupling to ground (earth) becomes significant only on very long transmission lines or very large delta distribution system with long cables. So you could measure a voltage between earth and a delta line, they could be balanced if the loads and other characteristics are balanced.

While such capacitance may have some significance and effect for setting protective devices, that virtual "neutral" has no practical use.

As Bill said, usable neutral can be derived using grounding transformer. Although that "use" is generally limited to detecting ground faults and/or limiting voltage to ground.

"Neutral" found in a 4-wire delta low voltage system is a totally different animal.

To add to the confusion, most 3 phase system analyses are done on a single phase basis, assuming they are balanced 3 phase system, regardless of they are wye or delta. "Per phase" voltage used is always line to line voltage divided by SQRT (3). This does not mean there is a physical neutral in all cases.




Rafiq Bulsara

http://www.srengineersct.com

 

[oldfieldguy]

If you're talking about industrial distributions systems, the answer is "maybe". System capacitance can provide a measurable "neutral" when measured with high impedance instruments.

Using proper protective measures, you can see this for yourself on an ungrounded low voltage system.

However, if you use a low-impedance instrument to make the measurement, you may find that the nice "textbook" virtual neutral disappears.

It is also subject to "disappear" if the connected buswork/cables/etc. is removed and the sole energized equipment is the secondary terminals on the transformer.

Due to this "here/not here" nature of the virtual neutral in a delta system, it is not to be trusted for safety measurements.

old field guy

 

[davidbeach]

I've seen (reported here in late 2006) a case where a delta 480V system had a measured phase-neutral voltage of over 300V on each phase. Even with 3-phase instruments. All because the "neutral" wouldn't stay put long enough to get a meaningful measurement.

 

[oldfieldguy]

David-

Exactly! That's ONE problem. The other is measuring with a low-impedance instrument that "pulls" the measured leg down to near ground potential, leading the assumption that the leg (and system) is "dead". One of my former places of employment had a very serious incident over this phenomenon.

old field guy

 

[davidbeach]

Doesn't have to be very low impedance. That case I have in mind the instruments involved all were high impedance. What probably contributed significantly was the use of 277:120 VTs rather than the more suitable 480:120 VTs. But in any case, the use of an undefined quantity meant that the voltage measurements couldn't be relied upon.

 

[waross]

Hi David;
I remember your post but I didn't understand at the time that PTs were involved. I ran into the phenomena when I sent some students to do some measurements on a small transformer bank in the shop. I put a scope on the transformer outputs and saw a very distorted waveform. We were using d'Arsonval based meters and I assumed that the form factor (ratio of average value to RMS value) of the distorted waveform was not the same as that of a sine wave.
I ran into the same effect in the field with an unloaded wye:wye transformer bank feeding a pumping station. As soon as we put some load on the transformer bank, the line to neutral voltages changed to the expected values.
I suspect that the discrepancy you observed may have been the result of the unloaded PTs rather than being related to the virtual neutral of the delta.
I remember your post and I have always wondered what the cause may have been. Now that I realize that wye connected PTs were involved I suspect that we may have found the answer.
Yours
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Both wye connected VTs with too low a voltage rating and measuring directly with Fluke DMMs within their calibration period, although the VTs weren't disconnected when measuring with the VTs.

 

[waross]

The two times that I observed the effect, we were measuring on the secondary side of the transformers. The pumping station was measured with several different meters. The thing more telling than exact calibration was the noticeable departure of the line to line: line to neutral ratio from the expected 1.73:1
Once the transformers were loaded the ratio became the expected 1.73:1
If you ever get a chance to put a scope on those voltages.....

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

The noticeable distortion of the wave form was a a horizontal "zero Volts" trace separating the positive half cycles from the negative half cycles. I have no idea of the algorithms used in RMS meters but if the meter just considers the half waves and ignores the zero line then I would expect a high reading even from a so called "True RMS" meter.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

[quoteIf so, can it be used as a reference, or is it a characteristic that has no application? [/quote]
The characteristic can be used in ground detection.