Delta Distribution

mbrooke

Batteries Not Included
Location
United States
Occupation
*
The 120-volt delta is for all 120-volt loads and I should have mentioned that I also have a 240-volt 3-phase delta for my larger loads (two services coming into the building)

I believe installing a 3-phase panel and install 2-pole circuit breakers for my lighting and receptacle loads is correct. The grounding (could be bonding) of metal raceways and devices will be connected to a grounding bar in the electrical panel with a #6 to the water service copper line.

Also I never want to see this again -

If you can, please get some pics of the transformer bank as well as the panel boards. It will make great tool for teaching code classes:D I will worship you:cool:



If you do go ahead with the additions yes, 2 pole breakers, and ground/bond as you would with a regular wye service including the 2 ground rods and water bond. Use white wire up to the lighting and outlet neutrals so it will be ready to change over down the road. Don't freak over it though. Most Norwegian factories have a 230 volt ungrounded service for lights and outlets plus another 690 volt ungrounded service for the motors and like. Ungrounded is the norm 120 volt hospital power in the United states and Canada. All working without issue.:)
 

kwired

Electron manager
Location
NE Nebraska
I am surprised National grid hasn't asked this to be changed out yet. Even the common 120/240 high legs are being forced out of service working or not.
We are still seeing new installations of this system around here and it is fairly common. When the bulk of the load is three phase motors, it kind of makes more sense than a 208 volt system.
 

Smart $

Esteemed Member
Location
Ohio
...If you do go ahead with the additions yes, 2 pole breakers, and ground/bond as you would with a regular wye service including the 2 ground rods and water bond.
NO. While I have no objections to grounding the system, a wye service is typically grounded with an MBJ. However, you are correct that a grounding electrode system and equipment grounding system are still required.

Use white wire up to the lighting and outlet neutrals so it will be ready to change over down the road.
Also NO. An outright violation on an ungrounded system.

Most Norwegian factories have a 230 volt ungrounded service for lights and outlets plus another 690 volt ungrounded service for the motors and like.
Norwegian factories are not under the purview of the NEC... :happyno:

Ungrounded is the norm 120 volt hospital power in the United states and Canada. All working without issue.:)
And aside from this being limited to healthcare facilities under the NEC, wire colors are required to be brown, orange, and yellow with tracer other than white, green, or gray. Additionally, the system is required to have 1) an isolation monitor, and 2) the orange conductor is required to be connected to the normally-grounded connection of 125V 15 and 20A receptacles. See 517.60 for more detail.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
We are still seeing new installations of this system around here and it is fairly common. When the bulk of the load is three phase motors, it kind of makes more sense than a 208 volt system.
Are the primaries wye or delta in cases like this?
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
NO. While I have no objections to grounding the system, a wye service is typically grounded with an MBJ. However, you are correct that a grounding electrode system is still required. Without a ground detector it won't serve any purpose.

I was not referring to intentionally grounding down a phase. My point was the system needs to have panels with ground bars, conduits need bond bushings, there must be a water bond, rods, an egc to all metal frames. Ground detector aside, you still need these to keep equal potential, If phase A faulted in one appliance and phase B in another there needs to be some way to generate fault current to trip a breaker. Even if only 1 phase was faulted it still service to prevent potential differences.



Also NO. An outright violation on an ungrounded system.

If this were a code compliant install yes it would be a violation, but keep in mind the very system we are working with from the start is not code compliant nor has it been for 90 years. Some how Its better this not be further complicated. I see no reason to waste wire and further confuse someone when this system will most likely be upgraded in a few years. The customer will hate the electrician when all the wire has to be pulled out and redone. Chances are the customer will look for an electrician who will bid without repulling the wire, and there are many who are in that type of business. Then we will have a grounded system with the wrong wire colors. I see it in the real world all the time regarding other code issues.




Norwegian factories are not under the purview of the NEC... :happyno:

Correct, however the laws of physics remain the same all over the world. I was trying to prove a point that an ungrounded system when done right is neither dangerous nor inoperable. There is a misconception that they are which is not the case. In fact Norway is regarded as having one of the safest electrical systems in the world, so its not likely they would allow something dangerous to exist.



And aside from this being limited to healthcare facilities under the NEC, wire colors are required to be brown, orange, and yellow with tracer other than white, green, or gray. Additionally, the system is required to have 1) an isolation monitor, and 2) the orange conductor is required to be connected to the normally-grounded connection of 125V 15 and 20A receptacles. See 517.60 for more detail.
Correct, if grandfathering is not included there will need to be a ground monitor. However, 517 specifically states health care. Balanced audio power is also required to follow the color sequence, but in technical terms we are talking about neither in this case.
 

kwired

Electron manager
Location
NE Nebraska
Are the primaries wye or delta in cases like this?
Fairly certain they are almost always wye primary, does it matter? I do understand that with loss of a primary phase a delta is likely to continue to operate with little or no problem as long as it is not loaded too much for the remaining two coils to handle, but fail to see that as reason to not have a high leg secondary at all. It is common to even see 480 volt open delta systems with a high leg on the secondary on irrigation equipment services if it is a lower capacity service. If it has over 30 hp it will usually be a wye secondary though.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
Fairly certain they are almost always wye primary, does it matter? I do understand that with loss of a primary phase a delta is likely to continue to operate with little or no problem as long as it is not loaded too much for the remaining two coils to handle, but fail to see that as reason to not have a high leg secondary at all. It is common to even see 480 volt open delta systems with a high leg on the secondary on irrigation equipment services if it is a lower capacity service. If it has over 30 hp it will usually be a wye secondary though.

A grounded wye primary with a closed delta secondary acts as a grounding transformer during a fault. A fault on the feeder can cause a fuse(s) to blow on the bank itself from fault current being pushed through the windings from the other phases. CSPs are the worst at handling this, so they are always avoided in closed delta secondary use. The solution is to leave the wye floating, however, that in itself creates a potential problem with ferroresonance. Single phase switching or an open phase can cause it. Years ago it was not much of a concern, but since distribution voltages have greatly increased along with low loss cores the risk of ferroresonance is much higher. Both cases my a be a nuisance or liability high enough to not allow closed delta banks or delta secondaries all together. Some pocos will allow delta secondaries while others have a policy to totally eliminate it.
 

Smart $

Esteemed Member
Location
Ohio
Correct, if grandfathering is not included there will need to be a ground monitor. However, 517 specifically states health care. Balanced audio power is also required to follow the color sequence, but in technical terms we are talking about neither in this case.
As noted by Don, if the system was legit per Code at time of installation, a ground detector is not required... but I highly recommend one even if not required and retaining the current ungrounded system. If not grandfathered, the system will have to be grounded to be compliant.

If you run a white conductor, and you balance loads, you're still left with 120V circuits with two non-white conductors. It only saves pulling out two thirds of the wiring when updated to a standard configuration. But I'll still not condone a non-compliant wiring method. :happyno:
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
Retired Electrical Engineer - Power Systems
Correct, if grandfathering is not included there will need to be a ground monitor. However, 517 specifically states health care. Balanced audio power is also required to follow the color sequence, but in technical terms we are talking about neither in this case.
There is no provision in the NEC that would allow an ungrounded 120V delta for new installations.
250.20(B)(1) is pretty hard to get around, as all possible combinations provide <150V to ground on the other conductors.

The hospital provisions, 517.60 and 517.160, would not apply to 'services' unless the services feed only one single operating room per 517.160(4).
 

hurk27

Senior Member
Even if ungrounded you will still get readings to ground with a high impedance meter so the 70 volts to ground mentioned is probably "phantom voltage" from capacitive coupling, as was previously mentioned. Ive got an old 600 volt service I am working on and sometimes it meters like a corner grounded and sometimes like a 347/600 WYE, sometimes something wacky. Also it could be grounded at the pole but the grounded conductor not even brought to the building (if a wye) or not grounded again at the building (if corner grounded at the transformers) so then you would be metering through dirt so depending on the impedance of your meter, you would get different readings to ground. With this 600, I am still not sure if it is grounded at the pole and the only way to know is to go up and look.
If you did have a service that had the secondaries grounded at the pole but the grounded conductor was not brought to the building or in the case of a corner ground at the pole, the phase grounded at the pole was not bonded at the building you would have a very dangerous set up, if a fault to the grounding system happened at the building like you said the earth impedance might not cause a OCPD to open and since the earth has been referenced at the pole you would have the system voltage from anything electrically connected to this buildings grounding system and earth, if it was a cornered grounded system at the pole in your case this potential would be 600 volts, if the X0 is bonded at the pole then it would be 346.42, anyone who would come into contact to anything bonded to this system at the building and earth (even a concrete slab) could be very badly injured or killed, I have discussed this in a few post on here before.

Remember once a system has been referenced to earth a low impedance fault path must be installed to provide a solid fault clearing path or you can have a very dangerous situation, from the readings you posted I would say this 600 volt service has not been bonded at the pole, and for safety sakes a ground fault detector should be installed to alert someone if a fault were to happen.
 

hurk27

Senior Member
As I pointed out in post 19 and 21, 3-phase 120 volt ungrounded deltas are common on ships and subs and I don't know if the OP is in an area that may have been a part of servicing Navy ships or even tankers, but as to why a utility would supply this type of service for any other purpose is a big mystery, if it is a existing service then I would bet this building at one time was servicing some kind of ships or equipment for them, or even providing shore power, maybe the OP'er could add a little input as to if this was the case?

There is no reason the utility could not just reconnect the tanks into a WYE with 120 volt windings and turn it into a 208/120 volt 3-phase service they would have to add a grounded conductor to the drop into the building and the meter socket would have to be changed to a seven jaw, but the same tanks could be used for this and would provide a more useful and NEC compliant service.
 

hurk27

Senior Member
A grounded wye primary with a closed delta secondary acts as a grounding transformer during a fault. A fault on the feeder can cause a fuse(s) to blow on the bank itself from fault current being pushed through the windings from the other phases. CSPs are the worst at handling this, so they are always avoided in closed delta secondary use. The solution is to leave the wye floating, however, that in itself creates a potential problem with ferroresonance. Single phase switching or an open phase can cause it. Years ago it was not much of a concern, but since distribution voltages have greatly increased along with low loss cores the risk of ferroresonance is much higher. Both cases my a be a nuisance or liability high enough to not allow closed delta banks or delta secondaries all together. Some pocos will allow delta secondaries while others have a policy to totally eliminate it.
I'm a little confused, grounding the X0 point on a WYE primary would not give you a ground reference to the secondaries??

Also I thought the ferro resonance problem was only when the a WYE primary with a common core transformer had the X0 bonded to a WYE source grounded conductor which can cause high circulating currents in this primary neutral bond and conductor when there is any current imbalance on the secondary, I've seen a few of these mistakes cause this bonding conductor to burn up as well as cause the transformer to run very hot.

From what I understand a primary WYE made up of separate tanks which doesn't have a common core would not have this problem?

Maybe I'm missing something or have it wrong?

I do understand that using a WYE primary on a delta primaries will provide a ground reference for the deltas just like a zig-zag so it is very common for utilities to use WYE primaries on 3-bank transformers to provide a close to point of use grounding when the tanks are fed from a delta primaries, our local primaries here are 12.4/7.2 kv WYE's the only deltas are the local substation 69k feeds as well as the 138kv, and the 345kv line that feed the area sub stations 69kv feeds from the generating plants.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
There is no provision in the NEC that would allow an ungrounded 120V delta for new installations.
250.20(B)(1) is pretty hard to get around, as all possible combinations provide <150V to ground on the other conductors.

The hospital provisions, 517.60 and 517.160, would not apply to 'services' unless the services feed only one single operating room per 517.160(4).

Thats what Im saying, if this were a new service it would not be allowed. However, if this service was installed in the 1900s it would actually have been allowed back then. It might also still be grandfathered as well depending on a number of rules. Back then there were minimal restrictions on the use of ungrounded systems. It was actually the norm. The very first single phase services were run ungrounded. Some still being installed into the 20s if Ive read correctly. Grounding of the center point or 1 leg at the transformer only became a requirement latter on driven by the possibility of a primary winding faulting into a secondary winding or a primary conductor falling into a secondary. Such occurrences were common back then.


If this system is grandfathered ground detection would not be needed, but it is a very good idea to have. Other rules are also up in the air, since the NEC does not recognize this system any more.

The best solution would be to ask for either a total service upgrade or if loading permits to tap off the 240volt delta service and use a 240-120/208Y transformer instead.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
I'm a little confused, grounding the X0 point on a WYE primary would not give you a ground reference to the secondaries??

Also I thought the ferro resonance problem was only when the a WYE primary with a common core transformer had the X0 bonded to a WYE source grounded conductor which can cause high circulating currents in this primary neutral bond and conductor when there is any current imbalance on the secondary, I've seen a few of these mistakes cause this bonding conductor to burn up as well as cause the transformer to run very hot.

From what I understand a primary WYE made up of separate tanks which doesn't have a common core would not have this problem?

Maybe I'm missing something or have it wrong?

I do understand that using a WYE primary on a delta primaries will provide a ground reference for the deltas just like a zig-zag so it is very common for utilities to use WYE primaries on 3-bank transformers to provide a close to point of use grounding when the tanks are fed from a delta primaries, our local primaries here are 12.4/7.2 kv WYE's the only deltas are the local substation 69k feeds as well as the 138kv, and the 345kv line that feed the area sub stations 69kv feeds from the generating plants.

Grounding a primary wye does not ground reference the secondaries, but it does change the impedance of the primary under a fault condition involving a primary wire fault, such as a phase coming down in an over head line on the neutral. Think grounded wye broken delta setup like you mentioned to detect ground faults. A phase grounding down on the primary causes a potential to change across a broken delta, normally a near zero voltage potential changes to a much higher one. This potential will activate a relay or light to indicate a ground fault. In a closed delta, its shorted together which will cause a current to circulate within the secondary delta winding. That circulating current causes the impedance of the secondary windings to change, which in turn is transferred over to the primary windings. The 2 primary windings on the non faulted phases will draw more current while the primary winding on the faulted phase actually tends to feed back into the cable/wire. In a nut shell the primary fuses end up seeing more current then they are supposed to during a fault. If conditions are right that can cause the primary fuses to blow. Also, if the line has single phase protection, ie fused cutouts protect the line in question, if a phase becomes grounded and the fuse blows but the other 2 remain closed as they normally do with only one phase faulted, the primary wye will actually back feed power into the faulted phases. The secondaries along with the building load can back feed the transformer on the faulted phase through its secondary, where power will be stepped up and appear on the primary. The potential on the primary can then back feed the downed phase. If the fuse on the bank does not blow (it may or it may not) that can leave a downed conductor with a dangerous potential, sometimes enough to electrocute someone or start a fire. Ironically, if the phase does not remain grounded, but rather just opens up, the bank will continue to supply 3 phase power without any issue at all, as long as the load doesn't go above about half of the cans original rating. This is often desirable to do when a grounded wye primary is involved. Ferroresonance is also not an issue be it a 3 phase can or 3 single phase cans in a bank, as long as the primary has a solid ground.



Ferroresonance can occur on any ungrounded primary connection be it individual single phase cans or a 3 phase unit. I believe common cores can play a roll, but Im not sure to what extent, but with out a doubt is has been seen to occur in 3 phase banks made of single phase units. An ungrounded wye is the most susceptible, as well as delta but its seems to happen slightly less in delta primaries. Ferroresonance is not a concern when the transformer is loaded, its impossible to occur on a transformer that is fully loaded. Even at only 40% load is near impossible, mostly rare at 10%. With out any load though the risk is high, and is usually triggered by an open phase or single phase switching. Ferroresonance is a type of resonant condition when transformer inductance, cable capacitance and the like start to resonate with one another. Im not a perfect expert on the subject though, I know just the basics, some one could explain it better. All I know is never energize a transformer unloaded when conditions are ripe, always add a load bank to be or some building load that's resistive to be safe.

Some utilities will actually add a cut out to the bank that they will close when energizing the bank, then open the cutout to leave the primary neutral point floating, that way both problems can be eliminated. Others just keep it simple and disallow all deltas for new services.

Utilities only use primary grounded wye deltas for grounding purposes when its necessary to derive a neutral source when another is non existent. They are used in substations when the secondary of the distribution transformer is delta, and use the grounding transformer neutral either to detect ground faults on the feeders or if the unit is big enough to have a neutral for phase to neutral loads out on the feeder. California utilities are big users of zig zags and GrY deltas to detect ground faults on their feeders since about half the substations have delta delta transformers 66kv to 16 or 12kv. The remaining uses are like noted in the beginning of the post.



If your curious about the connections in utilities the usually go like this: Generator is wye grounded through a resistor to limit fault current. 23kv is stepped up to 345kv via a delta wye GSU. 345kv is step down to 115kv or 138kv via a grounded wye autotransformer, often having a delta tertiary for circulating currents. Some utilities may use this tertiary for loads too. 115kv is stepped down to distribution be it 7/13.8kv Y 13.2/23Y 19.9/34.5Y ect via a delta wye unit, usually rated 40MVA when serving suburban towns. The secondary is solidly grounded and a multi grounded neutral is brought along with the poles. Customers are usually served by a grounded Y grounded Y giving 120/208 or 277/480. Sometimes a delta wye is used instead if the other is too unfavorable for the application at hand.


My 2 cents.
 
Top