Inherent danger for M-T-M ATO scheme?

[elec_eng]

This one makes me scratch my head and makes me nervous a little bit. This solution is very common. 3-pole CB transfers power between two sources without switching the neural. In 3-phase, 4-wire distribution, if the transformers (SDS, not services) are located outside, you have two bonding points at each transformer. This creates an issue for the ground fault coordination before, but the solution was the modified differential ground fault protection scheme.

However, another issue is that will create unwanted circulating current between two sources during normal operation because there are two bonding connections between two SDSs. You can draw a simple diagram to see this. This is big no-no from the safety standpoint. When we use the ATSs for the source transfer, we can always use the 4-pole ATSs and switch the neutral so we don't have this issue. But with 3-pole CB, you can't switch the neutral. Am I missing something here?

 

[ATSman]

Ground Strap GF Scheme

Ground Strap GF Scheme

If my memory serves me correctly I recall some M - T - M systems I have tested the GF schemes on
have the 2 transformers share a common neutral bus. The neutral bus is grounded at the mid-point (behind the tie), at a single point. A GFR CT is installed on the tie point and monitors GF current from either transformer. I think it was called a Ground Strap GF scheme.
I think for this to work correctly the GF protection on the feeder breakers would have to be zone interlocked with each main breaker for the logic to trip the main that is on the side of the switchboard where the fault originates.
Just a thought......not sure this would fit your application since it requires GF protection on all feeder breakers and that they are zone interlock connected with their mains.

 

[elec_eng]

If my memory serves me correctly I recall some M - T - M systems I have tested the GF schemes on
have the 2 transformers share a common neutral bus. The neutral bus is grounded at the mid-point (behind the tie), at a single point. A GFR CT is installed on the tie point and monitors GF current from either transformer. I think it was called a Ground Strap GF scheme.
I think for this to work correctly the GF protection on the feeder breakers would have to be zone interlocked with each main breaker for the logic to trip the main that is on the side of the switchboard where the fault originates.
Just a thought......not sure this would fit your application since it requires GF protection on all feeder breakers and that they are zone interlock connected with their mains.

Yes...that is another way to do it..single point grounding protection scheme. Looks like that scheme will not create any circulating current on the neutral.

But my issue is with the multi-point grounding with Modified Differential GF protection scheme. With it, it might work on the GF protection but the neutral current will flow through the EGC (parallel path). So, can't we use MDGF scheme for the MTM ATO scheme?

If this were services, not SDS, then per 250.24(A)(3), they have to ground the both at transformer and at the tie. In that case, MDGF scheme seems to be only option but it still creates parallel path for the neutral current. Not sure why the code allows this.

If this was the 3-ph, 3-wire distribution I think it will work but with 3-ph, 4-wire distribution I don't think there is a way to avoid this parallel path for the neutral current. Any thoughts?

 

[elec_eng]

Actually, there won't be any parallel path for the service so, my previous post about 250.24(A)(3) doesn't create any hazards. I stand corrected. But with SDSs, this is still an issue.