Transformer Expert Needed

Status
Not open for further replies.
infinity

infinity

Moderator
Staff member
Location
New Jersey
Occupation
Journeyman Electrician
Someone ran this by me today. They're reverse wiring a standard Delta/Wye transformer to get 480 volts with a 208 volt supply. Here are the details as I've been told:
1) Standard everyday 225 KVA transformer w/ 480 Delta primary, 208Y/120 volt secondary
2) Temporarily reverse wired to provide 480 volts to elevator motors
3) System to operate as grounded* but not a corner grounded Delta

So disregarding the NEC for a moment, :roll:
the issue is that they want to have this grounded so if someone tests from phase to ground they will get some voltage reading indicating that the feeder is energized. They do not want to operate as ungrounded in case someone only tests to ground. They do not want to use a corner ground (I'm assuming due to fuses in the motor circuits) so they've decided to *ground one of the voltage taps on the Delta primary. Yes that's right they're grounding #5 tap on one of the H windings. This will show some voltage when testing pahse to ground based on where that tap is physically on the coil.

Questions:

Will this actually work?
What will it do to the transformer when loaded?
Is the guy who designed this crazy?

IMO in theory it will work, not much different than a center tap on one winding (only this tap isn't in the center of the winding) but was wondering, aside from the code issues, what may or may not happen.
 
Last edited:
I say not a good idea. I don't think it is code compliant, but I think you already agree on that. This would seem to be no different than corner grounding except that you are not quite at the end of the winding. I don't understand how people come up with this stuff. Why don't they just get a 208 delta X 480 Y transformer a be done with it the correct way. They are not that hard to get, even on short notice.
 
The problem I see is if there is a ground fault on same phase as the tap you grounded, do you think enough current will flow to operate overcurrent devices? You may only have something like 20 volts from that phase to ground.

I think you have no choice but ungrounded system or corner grounded system. If there is a lot of concern over people not realizing that it is normal for one phase to have zero volts to ground, then don't land the grounded phase on circuit breakers, land it on neutral bars just like you would a neutral, and mark it white or gray like it is supposed to be anyway.
 
infinity said:
... So disregarding the NEC for a moment, :roll:
so they've decided to *ground one of the voltage taps on the Delta primary. Yes that's right they're grounding #5 tap on one of the H windings. Questions:

Will this actually work?
What will it do to the transformer when loaded?
Is the guy who designed this crazy? ...
Click to expand...

Yes
Be just fine
Yes (for a whole bunch of reasons)

ice
 
iceworm said:
Yes
Be just fine
Yes (for a whole bunch of reasons)

ice
Click to expand...
You sure?

Voltage between the "H" taps is likely about 2.5% of 480 which is about 12 volts. If they ground the # 5 tap and connect a phase conductor to the #3 tap there will be about 24 volts from that phase to ground. Lets say this is all done on "B" phase. If you have a fault to ground (#5 tap of B phase) from phase A or C you get similar fault level current as you get if you just grounded B phase.

If you get a fault to ground on B phase you only have a supply voltage of about 24 volts feeding this fault, you also have very little source winding between tap #3 and tap #5 - I don't know how much fault current will flow but guessing if you had a fault on say a circuit with a 100 amp breaker - you possibly don't have enough current to even trip the breaker.

Someone know how to determine about how much current will flow in this fault if it were a 225kVA transformer? I realize there will be some unknown variables that make some difference, but just a general idea of what one might expect?
 
kwired said:
You sure? ...
Click to expand...
nope. Well, except for the third question. I'm pretty sure that answer is correct.

kwired said:
... Someone know how to determine about how much current will flow in this fault if it were a 225kVA transformer? I realize there will be some unknown variables that make (replace "some" with") a lot of difference, but just a general idea of what one might expect?
Click to expand...

Well, if one were to corner ground, then go drive the mythical 25 ohm ground rod and connect one of the ungrounded phases to the ground rod, You would get about 20A.

Now, a bolted fault on the secondary of a 225KVA, 480V, 3ph, 5%Z, inifinite primary is 225000/480/Sqrt(3)/.05 = 5400A.

ice
 
Gentle?people...
As long as only measurement(s) to ground is (are) necessary, then I believe it would make far more sense to install 3 PTs! And, be on the lookout for aberrant behaving phase-to-ground capacitance!
Regards, Phil Corso
 
infinity said:
So disregarding the NEC for a moment, :roll:
Questions:

Will this actually work?
What will it do to the transformer when loaded?
Is the guy who designed this crazy?
Click to expand...

IMO

yes
no change
no

since it is temporary, if right xfmr not available and issue is to allow someone to see voltage to ground if on. if corner grounded, then a short on that phase to ground would not show any current to pop a breaker either so I see no issue there with 20v & limit current. another option would be if any 480 wye xfmr is available, any size, just put it across the 480v to get centered reference to ground too if anyone is concerned over the 20v or so to ground reading.
 
mike_kilroy said:
IMO

yes
no change
no

since it is temporary, if right xfmr not available and issue is to allow someone to see voltage to ground if on. if corner grounded, then a short on that phase to ground would not show any current to pop a breaker either so I see no issue there with 20v & limit current. another option would be if any 480 wye xfmr is available, any size, just put it across the 480v to get centered reference to ground too if anyone is concerned over the 20v or so to ground reading.
Click to expand...

I disagree with the part in bold. A phase to ground fault with set up as described in OP will result in max available current of the portion of the winding involved (something like 5% of the usual number of turns to make 480 volts). So you basically have a smaller winding within a larger winding that has short circuit across it. The normal rated full load is 270 amps @ 480 volts. We should be able to connect 270 amps @ 24 volts load to this "phase to ground" section of winding and not overload it. Short circuit that and you will get even higher current - I guess my theory that it may not trip a breaker may be wrong, but it still should be significantly less short circuit current than what would result from a fault across the full winding. - I think, maybe it would actually be higher. Where are the transformer experts?

I can possibly see the fault current possibly being significantly higher for this one phase to ground and a need for higher AIC rating for connected equipment but only because of that one phase to ground rating.
 
Phil Corso said:
... then I believe it would make far more sense to install 3 PTs! ...
Click to expand...
Uhh .... Earth to Phil -
This has to be the cheapest goofs in creation and he (or she) is making technical decisions. Okay, maybe not the cheapest. But certainly cheap and one of the least connected with industry practice.

How would you ever expect this guy to ever pony up for $300 worth of xfm? - unless you want ZZ then its more than $300

ice
 
kwired said:
I disagree with the part in bold. A phase to ground fault with set up as described in OP will result in max available current of the portion of the winding involved ....
Click to expand...
kw -
Mike didn't say that. What he did say is correct.

mike_kilroy said:
... if corner grounded, then a short on that phase to ground would not show any current to pop a breaker ....
Click to expand...

kwired said:
... . A phase to ground fault with set up as described in OP will result in max available current of the portion of the winding involved (something like 5% of the usual number of turns to make 480 volts). So you basically have a smaller winding within a larger winding that has short circuit across it. The normal rated full load is 270 amps @ 480 volts. We should be able to connect 270 amps @ 24 volts load to this "phase to ground" section of winding and not overload it. Short circuit that and you will get even higher current - I guess my theory that it may not trip a breaker may be wrong, but it still should be significantly less short circuit current than what would result from a fault across the full winding. - I think, maybe it would actually be higher. Where are the transformer experts?

I can possibly see the fault current possibly being significantly higher for this one phase to ground and a need for higher AIC rating for connected equipment but only because of that one phase to ground rating.
Click to expand...

I think your thoery is good. Post 7 was my miserable attempt to show that it doesn't take much resistance and a GF won't trip the CB. With the 20V coil ground faulted, with 1 ohm in the loop, and the fault will only draw 20A. GFs are not always bolted.

ice
 
infinity said:
Questions:

Will this actually work?
What will it do to the transformer when loaded?
Is the guy who designed this crazy?

IMO in theory it will work, not much different than a center tap on one winding (only this tap isn't in the center of the winding) but was wondering, aside from the code issues, what may or may not happen.
Click to expand...


I guess I should answer the questions - I agree that it will work, the transformer will work as it usually works with a load on it, the guy who designed it likely is crazy.

I am concerned about what happens when there is a fault from the one mentioned phase to the ground that is only going to probably have anywhere between 12 - 72 volts depending on which "H" taps are used.
 
Thanks for all of the input (keep it coming). :thumbsup:

How does this installation stack up against the NEC? I see two options, corner grounded Delta or ungrounded Delta.
 
Off the top of my head - grounding a tap sounds like an incident waiting to happen.
The items I would want to investigate include (assuming tap #5 on the H1 winding):
Circulating current between H1 and tap, due to capacitive coupling
Fault current between transformer H1 and tap, reflected to primary
Fault current of load H1 to G
Fault current between transformer H2 and tap, reflected to primary
Fault current of load H2 to G
Elevated voltage between H2 and tap
Single pole performance of secondary breaker during a load H2 to G

​
At this point I have not made an attempt to consider these items.
 
iceworm said:
kw -
Mike didn't say that. What he did say is correct.





I think your thoery is good. Post 7 was my miserable attempt to show that it doesn't take much resistance and a GF won't trip the CB. With the 20V coil ground faulted, with 1 ohm in the loop, and the fault will only draw 20A. GFs are not always bolted.

ice
Click to expand...

sorry was composing my last post, got interrupted and finally submitted it about a half hour later.

The coil we are talking about is a source coil and not a load coil, impedance of this section will be very low, lower than the full winding (480 volt) will be. May actually allow for even higher fault current than what would ordinarily flow in the full winding, I'm not fully certain though.
 
infinity said:
Thanks for all of the input (keep it coming). :thumbsup:

How does this installation stack up against the NEC? I see two options, corner grounded Delta or ungrounded Delta.
Click to expand...

As far as NEC goes:

250.26 Conductor to Be Grounded ? Alternating-Current Systems. For ac premises wiring systems, the conductor to be grounded shall be as specified in the following:

(1)
spacer.gif
Single-phase, 2-wire ? one conductor

(2)
spacer.gif
Single-phase, 3-wire ? the neutral conductor

(3)
spacer.gif
Multiphase systems having one wire common to all phases ? the common conductor

(4)
spacer.gif
Multiphase systems where one phase is grounded ? one phase conductor

(5)
spacer.gif
Multiphase systems in which one phase is used as in (2) ? the neutral conductor
Click to expand...

Is the conductor that is grounded a "common conductor"?

I guess it is the closest thing to being one but certainly is not as common as the center point of a Wye system.
 
kwired said:
As far as NEC goes:



Is the conductor that is grounded a "common conductor"?

I guess it is the closest thing to being one but certainly is not as common as the center point of a Wye system.
Click to expand...

IMO only a wye system would have one point common to all three phases.
 
Status
Not open for further replies.
Top