I believe your thinking is correct. However you should read 277 volts to ground, not 480 volts. One thing I want to point out is that when you have a delta secondary, one phase is grounded. It may be possible that the person who installed this equipment thought it was delta and intentionally grounded phase B. You should look and see if this has happened.

You need to have a neutral installed to the main disconnect switch. Past that point you only need to install the phase conductors plus an equipment grounding conductor(EGC). The EGC can be conduit or a conductor.

Interesting post that I mostly understood, but having only worked on Delta Star transformers (380 to 220 volts), its a new area for me to hear that if the output side is Delta, that one phase is grounded (which would appear to fit the case in hand as well!). Can you explain to me why that is done?

(In a way I can probably guess that its the only way a reference can be made to earth and safety, so if any other ground faults develope, fuse and breakers will start to pop, but am I right? Thanks in advance)

It is not my field, but I believe that grounding of one phase on a delta supply is not uncommon in North America, but virtually unheard of elsewhere.

If you are sure the secondary of the transformer is a wye connected secondary, then the B phase on the secondary must be shorted to ground. In essence you now have a corner grounded three phase 4 wire system. If the neutral had of been grounded as well, you would have blown fuses on the primary side.

The voltage readings you saw concur with this. I'll bet you see 277v from the neutral point of the secondary to ground. A grounded B phase will read 480v to the other two phases and 277v to the ungrounded star point of the secondary winding.

Open all of the switches disconnecting the primary and the secondary and megger the windings to the case of the transformer and see what you have.

If the transformer is clean, its in the secondary conductors.

I also agree with Wareagle. Once you get the shorted phase fixed, ground the star point.

If this was in Canada, I would also tell you that the ground you use for this separetly derived system needs to be tied to all of the other "grounded" systems in the building.

Doing so ensures no potential difference between bond systems during a fault.

So, instead of a TT system as intended, this is an IT system.

Somewhere else on the network, one of the phases is grounded. Find it and rectify the problem.

Start by turning off the generator and disconnecting its output wires. Using a multimeter set to ohms, measure the resistance between phase B and earth, both looking at the generator and at the system.

If the fault is not in the generator, move downstream along the circuit methodically, diconnecting the B phase conductors at each dis-board until one disconnection makes the multimeter go meg-ohms. The fault lies on that sub-circuit.

generator transformer

Hi,

From your description of situation it is evident that B phase has a ground fault. Whenever you measure voltage between B phase and any other phase you are infact measuring line voltage, which in your case is 480 volts. As B phase is grounded there is no potential difference between B phase and ground hence you are getting zero reading on voltmeter.

There is no neutral conductor on secondary side and from description it seems you are measuring voltage at transformer bushing. In all probabilites the B phase winding in side the transformer has ground fault.

well as you said there was no neutral running along. Make sure the person who did the connections didn't grounded the phase B thinking it as neutral bcoz neutral ultimately has to be grounded.

I am not very experienced in this stuff, just a student, so I might be wrong.

Well i agree with msamad and arcflash,there is Ground Fault in the Installation. The Ground Fault is on Phase B, if you measured Fhase A & C to Ground you will have reading, because the Fhase B is Shorted to Ground. Check under Ground Cable or any other Installation in the Plant.

You must also check your Ground Fault Protection Relay setting Percentage % Amp on Secondary, if there is a protection Relay installed, if the leakage Current to Ground is less than Setting % from Ground Fault Relay, than i think the Ground Fault protection Relay won't work and trip the Breaker.

Don't forget to Ground the Secondary Neutral Point at TX and Frame to Ground

In this case, the leakage current to ground will be zero. There is no single phase load, and the B phase short to ground has created a "corner grounded" 480 volt system which is sometimes done on purpose, and does not present any "leakage" to ground then either.

The problem is that whatever caused the short to ground of one phase will most likely be just a little bit of vibration away from creating a second short to ground. THEN there will be a LOT of leakage to ground (actually shorting over to the B phase through the grounds). Kaboom!

Regards, CJM

This is an isolated 480 V system i.e Since the Wye side is not grounded

But at your case the B phase 100% is grounded.

This is safe ONLY for the first short circuit. If you have a second one then ....100% your system will be in big trouble.

My advice check your load on B phase or check which 3phase load is not operating correctly or is noisy due to motor/load running on 2 phase

OR turn off all your load one by one and check the voltage on B phase every time you are turning off a load

ACT ASAP BECAUSE A SECOND SHOT CIRCUIT WILL BE 100% DISSASTER FOR YOUR SYSTEM

GOOD LUCK

DONOT TRY TO PUT A GROUND ON THE WYE OF THE TRANSFORMER BEFORE YOU RESOLVE THE PROBLEM!!!!!!!!!!!!!!!!!!!!

With an ungrounded Wye system if one phase is truly grounded then one would expect the line to line voltage to be measured to ground on the remaining phases. Somewhere in the secondary system there should be a ground fault monitoring unit - which may only be three neon lamps - your voltage readings should be confirmed by one of these lights being out while the other two should be at approximately double the normal brightness.

There are three possible locations for the fault if one exists:

1. The transformer itself which can of course be tested by isolation from the load then measuring the secondary voltage to ground on each phase (a breaker or disconnect should be available to isolate the secondary from the load).

2. The cables from the transformer to the load - again should be testable using the transformer and the load disconnects and

3. The load itself again testable using the isolation local to the load.

These tests should be performed using a "megger" insulation tester of suitable voltage (at least 600v but preferably 1000v) rating as the insulation being tested may not show a fault under the minimal voltage applied by an ohm meter.

GA for a really good explanation, that even I fully understood.

Also, very interesting.....

I would give 2 x GA, but it won't let me....

Based on your description I would say that you are correct. You can use a megger to verify what you suspect. Disconnect the B phase at the transformer and megger A and C to ground, if those results are good megger between the disconnected B lead and ground. It will probably fail and confirm your suspicions. Testing A and C is a good idea because you may potentially have some other issues that will affect you in the future since the B cabling is probably damaged A and C could be damaged as well but not to the degree that B is. I saw a manufactured rigid pipe elbow that had a piece of galvanizing in it that was as sharp as a knife but with enough opening to allow wire to be pulled..... not a good combination!

Didn't the OP mention "4160V delta: 480V wye"? It is clear that the transformer primary is 4160 V (Delta) and secondary is 480 V (Wye). I see somebody turned it to secondary Delta and expressed the opinion. The question should be understood first before providing the expert opinion.

It is not uncommon for the secondary "Y" configuration to have the neutral side ungrounded and unused. It might depend on which part of the country you're from and the local code requirements but around here the "Y" can be used like a delta.

It could be true if only 3 phase loads are connected.

If single phase loads are to be also supplied, then a neutral is needed and it would commonly be earthed at the transformer output for safety reasons too.....but ONLY there....

It is possible, but other statement lead me to believe that it is Wye input and delta ouput, something that I personally have never seen and would appear to be continental USA only.....and very interesting too.....!

A megger, when used as a megger is the wrong instrument. A megger is only for checking insulation values using a high AC voltage similar to the working voltage of the equipment concerned.

If any single phase is grounded, then the megger will show a short to ground for each and every phase until the B phase ground connection (for example) is completely removed.

You would need to seperate each phase from the others and ground, then you can measure its insulation value with a megger, to ground.......if that was your intention....

Some Meggers have an Ohmeter as well, perhaps you meant one of these, but even then you woul only measure a phase to ground resistance, which might be less than one ohm for the phases NOT connected to ground as phase B probably is, via phase B!!!

S I really cannot understand what you are trying to achieve....so please take "another run at it" if you wish......

A megger is used to measure leakage resistance, usually at elevated voltages 500, 1000, 5000 VDC not AC. I'm not conversant with practices in Germany but electricians use a megger to check out leakage to ground in newly pulled installations and engineers routinely put minimum values for different types of installations. Usually 500K is the bottom end of the acceptable leakage resistance. The use of AC voltages is for testing insulation is referred to here as a hi-pot test and is generally specified in mA or uA as opposed to K-ohms or Meg-ohms in a test using a megger. I did indicate that the questionier should disconnect the B phase lead because to not do so would indicate a failure on A and C as well as B.

Since the tranformer is ungrounded then you don't need to disconnect A and C. I could have been more explicit in that the common of the 480V delta, according to NEC, should be grounded, even if it's not used as a neutral, hence becoming a delta appearing secondary.

Take care.....

With this installation, it will behave like a Delta 480V, and can be corner grounded by an accidental scrape or break in the insulation, no leakage, no Kaboom (yet).

Corner grounding is usually not intentionally done with a Wye secondary since if a single phase load is ever added (maybe a light for a dark pump room) and the transformer neutral gets grounded at the transformer in the process, again KABOOM.

The main, and most likely danger, is that the cause of one phase getting in contact with ground has probably also damaged the other 2 phases, and a second phase to ground will be a violent phase to phase (not phase to ground) event.

This situation has increased the danger to humans, since a grounded human coming into contact with one of the other ungrounded conductors, will complete a 480v short circuit to the B phase (not 277v) through the earth.

CJM

Our plant is filled with ungrounded delta. This was done for reliability and safety. You need to have daily PM and you would have earlier found the time when the phase B became grounded. The ground is probably very small, like maybe a pinched wire. You will need to pulse the phase and use a ground tracer.

Otherwise install a HRG system (Post Glover and others) This can be set up to sound an alarm. This will put you in compliance with NEC.

Be aware you are in a hazardous state now - if the second phase is accidentally grounded by a person an arc flash will occur.