Unbalanced delta secondary voltages causing issues with ground fault relay

A

acolella

Member
Occupation
Engineering Technician
I have two delta-delta transformers that have very unbalanced line to ground voltages on the secondary at no load. These are 480V-600V transformers.

AB: 614V A-Gnd: 560V 0 amps
BC: 610V B-Gnd: 505V 0 amps
AC: 610V C-Gnd: 104V 0 amps

I understand that line to ground to voltages can be unusual on a delta system, especially at no load. The issue I am having is that these unbalanced voltages are causing my ground fault relays to indicate a fault. I have verified that there are no actual ground faults.

I have the ground fault relays connected to shunt trip breakers. The idea is that if one of the legs goes to ground, the ground fault relay detects that and trips the breaker feeding the transformer primary. These relays are indicating a fault due to the voltage unbalance and trip the breaker as soon as the transformer is energized. If I disable the shunt trip and load up the transformer, the phase to ground voltages become much more balanced. The relay then no longer indicates a fault.

AB: 604V A-Gnd: 334V 490.9 amps
BC: 602V B-Gnd: 362V 492.1 amps
AC: 601V C-Gnd: 347V 489.2 amps

Has anybody ever encountered this before? If so, what are some solutions? These particular ground fault relays are Littelfuse EL3100. Maybe I need a different type of relay for this application or maybe I need to build something. One of my coworkers suggested the idea of perhaps getting a small delta-wye transformer, connecting the primary of that to the 600V source and connecting the secondary to the ground fault relay. The idea being that the wye secondary should be pretty balanced and then if there is a fault on one of the primary legs, the secondary voltage should go unbalanced enough to make the relays fault.

I appreciate any replies!

Thanks!
 
Does this need to be an un-grounded system at all? Usually that's used when operations must continue in the face of a single fault to ground, and if you're tripping the breaker when the fault occurs that advantage is lost. Could be that a zigzag transformer and maybe a high-resistance grounding system would be more appropriate for the installation but it's hard to tell without a lot more into about it.
 
zbang said:
Does this need to be an un-grounded system at all? Usually that's used when operations must continue in the face of a single fault to ground, and if you're tripping the breaker when the fault occurs that advantage is lost. Could be that a zigzag transformer and maybe a high-resistance grounding system would be more appropriate for the installation but it's hard to tell without a lot more into about it.
Click to expand...
It doesn’t need to be an ungrounded system but that is all I have. This is for a temporary test stand for a large electric heater. We needed to rent transformers for this project and the only ones big enough that were available were delta-delta. I would have preferred delta-wye or even wye-wye but they weren’t available. I understand the goal of continuity of service in the event of a single fault to ground on an ungrounded delta but in our situation, we want to trip the breaker if there is any fault.
 
acolella said:
It doesn’t need to be an ungrounded system but that is all I have. This is for a temporary test stand for a large electric heater. We needed to rent transformers for this project and the only ones big enough that were available were delta-delta. I would have preferred delta-wye or even wye-wye but they weren’t available. I understand the goal of continuity of service in the event of a single fault to ground on an ungrounded delta but in our situation, we want to trip the breaker if there is any fault.
Click to expand...
Corner ground it?
 
If your delta is floating by definition it has no reference to ground, therefore GF protection using L-G voltage will be unreliable.

Your only solution is to create a reference to ground.
The historical way this was done with indicating lights, was to connect three lights (rated at the full LL voltage) into a Wye configuration with the midpoint of the Wye connected through a normally closed bush button contact and then to ground. During normal operation the lights saw the L-G voltage and they all glowed dim, but during a fault the lights saw full voltage and glowed brightly, the pushbutton could be used for testing by breaking the ground connection causing the lights to go dim.

You can use 3 individual transformers to lower the L-L voltage to something more reasonable and then use your relay voltage sensing inputs instead of lights. This may have been the direction you were given when someone suggested a transformer. Your switchgear or relay vendor probably has done this for ungrounded MV installations.

If this was my equipment I would look at a high resistance ground system (HRG) because it can be helpful in tracking down the source of the GF.
 
Last edited:
ptonsparky said:
It appears that C to Gnd already has a fault with the low voltage at no load. WAG that is what is being picked up on. IDK if it would help to rotate the phases to see if the problem follows.

I like the idea of making it a corner ground, but my smarts are lacking.
Click to expand...
Possibly a resistance ground System? I’ve heard of them, but never seen one, but I think those were used on wye’s though.
 
You must log in or register to reply here.
Top