Neutral conductor on a 3 wire line

[Mbrooke]

IF you have a 3 phase 3 wire line with all loads connected in delta, is it better to run the neutral along the poles and ground/bond the system exactly like an MGN or simply skip the 4th conductor all together? What are the pros and cons of both?

 

[jraef]

If the loads do not need or use the neutral, no need to run it at all. Done all the time with MCCs.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[cranky108]

With no neutral, you need to make sure each of your pole grounds are good, as that is what the distribution transformers are tied to, as well as your arresters. And your PPG's.

Also a shield wire has been shown to reduce lightning strikes, so a 4th somewhat conductor is good for that.

 

[waross]

With a "-- 3 phase 3 wire line with all loads connected in delta" you do not have a neutral.

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

 

[Mbrooke]

With no neutral, you need to make sure each of your pole grounds are good, as that is what the distribution transformers are tied to, as well as your arresters. And your PPG's.

Also a shield wire has been shown to reduce lightning strikes, so a 4th somewhat conductor is good for that.

According to the NESC, don't you ground the can and arrestors to one ground, and then connect the LV neutral to another ground 8 feet away when there is no MGN? Or just tie all of them together? What does the IEC say?

With a "-- 3 phase 3 wire line with all loads connected in delta" you do not have a neutral.

You do at the substation, you can either draw it out or leave it there.

 

[bacon4life]

Are we assuming a solidly grounded wye at the substation serving medium voltage distribution? MV distribution is a much different case than a motor control center with a ground but no neutral.

Unfortunately the problems with skipping the neutral are hard to quantify compared to the obvious costs of wire and taller poles.
For a line to ground faults, fault current may flow in and damage adjacent objects such as communication wiring or fencing. If any of the circuit has tape shielded underground cable, a bare neutral is needed in parallel with the UG sections. I am not quite clear on what happens eventually there are sections alternating between 3 wire overhead and 4 wire underground. Also, line regulators have to be connected in delta, which can cause unbalanced phase angles.

 

[Mbrooke]

Are we assuming a solidly grounded wye at the substation serving medium voltage distribution? MV distribution is a much different case than a motor control center with a ground but no neutral.

Yes, a 115kv to 12kv substation serving an over head distribution. The 12kv over head is what is in question. The 115kv to 12kv transformer is low impedance earthed, though solidly in cases where the power transformer is small.

Unfortunately the problems with skipping the neutral are hard to quantify compared to the obvious costs of wire and taller poles.
For a line to ground faults, fault current may flow in and damage adjacent objects such as communication wiring or fencing. If any of the circuit has tape shielded underground cable, a bare neutral is needed in parallel with the UG sections.

This a legit concern, and one where I am at a loss. Relying on the earth alone will put inadvertent bonding (grounding) conductors like TELCO shields at risk for high current through faults. That is of course, unless, the transformer is high impedance earthed which in theory is also doable.

I am not quite clear on what happens eventually there are sections alternating between 3 wire overhead and 4 wire underground.

Would these not be treated like any other ungrounded legacy system found in most US POCOs? My understanding (though limited) is the shields are bonded together at each end and then connected to ground electrodes. Perhaps a California or IEC member could chime in (where 3 wire is common)

Also, line regulators have to be connected in delta, which can cause unbalanced phase angles.

How would these phase angle shifts compare to a wye grounded system where neutral shift is another concern? My understanding is that with delta regulators one pf the phases can be higher in voltage relative to gorund than the other two, however 15kv class equipment being run at 12kv mitigates this concern.

 

[davidbeach]

Be careful. If the protection setting process assumes a neutral then you'll wind up with settings appropriate for a system with a grounded neutral. Oddball situations generally result in inappropriate settings.

 

[Mbrooke]

Be careful. If the protection setting process assumes a neutral then you'll wind up with settings appropriate for a system with a grounded neutral. Oddball situations generally result in inappropriate settings.

None to odd ball for POCOs which use them exclusively (California for example) to which relay manufacturers are aware. However, in my mind I am imagining a greatly reduced zero sequence setting to take this into account. Would I be correct?

 

[Mbrooke]

Also, a concern for me is that when a neutral is run, faults at the substation cause the voltage rise relative to remote earth to leave the station.

 

[davidbeach]

So, if no feeders include a neutral, then a feeder without a neutral isn't an oddball situation. But for a company like mine, for example, where every feeder has a neutral, if we were ever to do a feeder without a neutral we'd undoubtedly find several unstated assumptions in our setting process that only work (100% of the time to date) on feeders with neutrals.

What would really be fun is the transition between assumptions. I grew up in an area with a 12kV delta distribution system. All three wire distribution, no neutrals; no phase-neutral/ground connected loads. Faced with capacity issues, that company converted from 12kV delta to 21kV wye. Some stations had one, other, perhaps adjacent, had the other. Worked in that area for a while, years ago, and traced out feeders to see what a particular site might have. I'd follow a circuit that was clearly 21kV with three phases on fully rated insulators and a neutral on pole clamps into an area where one of the phases dropped out and the neutral moved up into one of the phase positions. Didn't think too much of it at the time, all of my work then was on the customer side of the meter, but now that I'm wholly on the other side of the meter it gives me the willys. So the 12kV needed to follow one set of assumptions and the 21kV a different set of assumptions, but to me, at the time, the boundary was entirely random. I hope they got it right, and I hope, 25 years later, that they're long done with the transition.

 

[bacon4life]

At least when the circuits sit on top of an old salt water marsh, the lack of neutral doesn't seem to drop the fault current enough to prevent reasonable operation with standard phase overcurrent relays. The 3 wire zero sequence impedance only goes up by about 50% in 100 ohm-meter soil as compared to a multi grounded neutral.
Z0 = (0.301 +j 1.890) Z1 = (0.131 +j 0.594) Multi grounded with 100 ohm-m soil
Z0 = (0.417 +j 2.940) Z1 = (0.131 +j 0.594) substation only grounded with 100 ohm-m soil

I assume under different soil conditions, the individual pole or transformer grounds might be high enough resistance to prevent adequate sensitivity with overcurrent relays.

I can't seem to find the reference about phase angles on delta regulators. I might have been thinking of open delta regulators. We had looked into adding some to our 3 wire system, but decided to not to install regulators until we add enough of the neutral to allow wye regulators.

 

[Mbrooke]

At least when the circuits sit on top of an old salt water marshsurprise, the lack of neutral doesn't seem to drop the fault current enough to prevent reasonable operation with standard phase overcurrent relays. The 3 wire zero sequence impedance only goes up by about 50% in 100 ohm-meter soil as compared to a multi grounded neutral.
Z0 = (0.301 +j 1.890) Z1 = (0.131 +j 0.594) Multi grounded with 100 ohm-m soil
Z0 = (0.417 +j 2.940) Z1 = (0.131 +j 0.594) substation only grounded with 100 ohm-m soil

Thanks. Any idea where you re getting 100ohm meter soil from? My understanding (yet I could be very wrong) is that the bulk of the impedance comes from the electrodes and not the soil itself as the earth has a near zero ohms resistive impedance. The substation will usually be a good value (say under 25 ohms) and the primary concern is the grounding at the 12kv/480 volt transformer. However, in most case, a zero sequence pickup value of 5 amps is sufficient for a 3 wire line in regards to tripping out the feeder itself in nearly all cases?

I assume under different soil conditions, the individual pole or transformer grounds might be high enough resistance to prevent adequate sensitivity with overcurrent relays.

I can see a 50K or 25T fuse being hit or miss for a line to ground fault, however would not a 5 amp ground setting on the feeder relay (or transformer relay if protected by such) be adequate in most cases? Ignore selective coordination for a moment.

I can't seem to find the reference about phase angles on delta regulators. I might have been thinking of open delta regulators. We had looked into adding some to our 3 wire system, but decided to not to install regulators until we add enough of the neutral to allow wye regulators.

You might be, but still worth pointing out. Better to check list everything before hand then have a surprise down the road "Oh man, how did we miss that" lol.

 

[bacon4life]

The earth resistivity is just an assumed value depending on soil type used when calculating the line impedance based on Carson's equations. For a modeling a fault, the fault resistance would include both the actual arc resistance and the resistance of the ground rod.

 

[Mbrooke]

Makes sense. The ground rod itself is what we are worried about rather than soil resistivity over x amount of distance?

 

[cranky108]

Ground impedance varies from place to place. Sand and rock have high impedance, so goes my comment about carrying a neutral/ground to tie pole grounds.