Paralleling transfomers NEC

[mbrooke]

Is it permitted to have 2 breakers originating from a 480 volt panel board, each feeding a separate transformer then have them both feed into the same 208 volt panel board? Or separate panel boards each but both tied via a bus tie? Would I need kirk key interlocks for this? This idea is for redundancy in that any transformer can be taken off line while the other carries all the load.

 

[ron]

Is it permitted to have 2 breakers originating from a 480 volt panel board, each feeding a separate transformer then have them both feed into the same 208 volt panel board? Or separate panel boards each but both tied via a bus tie? Would I need kirk key interlocks for this? This idea is for redundancy in that any transformer can be taken off line while the other carries all the load.

If each is sized to handle full load and tied together, then you would need a key interlock either for the transformer or the tie. If left all tied together, the double sized incoming wouldn't protect the 208V downstream bus.

 

[mbrooke]

If each is sized to handle full load and tied together, then you would need a key interlock either for the transformer or the tie. If left all tied together, the double sized incoming wouldn't protect the 208V downstream bus.

Nothing lets you do this manually without a key if I was to upsize the buss (400amp bus for 200amps per trnsfomer)?

 

[petersonra]

Is it permitted to have 2 breakers originating from a 480 volt panel board, each feeding a separate transformer then have them both feed into the same 208 volt panel board? Or separate panel boards each but both tied via a bus tie? Would I need kirk key interlocks for this? This idea is for redundancy in that any transformer can be taken off line while the other carries all the load.

As long as the 208V panelboard has a main, I am not even sure you would need an interlock assuming there are no prohibitions on paralleling xfmrs in the code (and I don't think there are).

I think you would need a disconnecting means on the secondary sides of the xfmrs so you could isolate them.

 

[mbrooke]

As long as the 208V panelboard has a main, I am not even sure you would need an interlock assuming there are no prohibitions on paralleling xfmrs in the code (and I don't think there are).

I think you would need a disconnecting means on the secondary sides of the xfmrs so you could isolate them.

There will be a breaker when the transformer feeds into the planelboard.

 

[petersonra]

There will be a breaker when the transformer feeds into the planelboard.

I think you would need a breaker to protect the panelboard bus.

 

[mbrooke]

I think you would need a breaker to protect the panelboard bus.

Explain. Both the primary and secondary will have breaker protection.

 

[petersonra]

Explain. Both the primary and secondary will have breaker protection.

What protects the pb bus if both xfmrs are in service?

 

[GoldDigger]

What protects the pb bus if both xfmrs are in service?

Only the fact that it is rated for twice the transformer output current.
Conservative design would suggest that it be overrated that way if no interlock (hardware or electronic) is used.

 

[winnie]

I wonder if the requirements for 'secondary tie circuits' would provide guidance here. My understanding is that there is a history of separate transformers with their secondaries 'tied' together for continuity of service.

If each part of the system is protected from overload (transformer secondary, secondary conductors, and panelboard) then why would interlock be required? AFAIK transformers may be operated in parallel.

-Jon

 

[petersonra]

Only the fact that it is rated for twice the transformer output current.
Conservative design would suggest that it be overrated that way if no interlock (hardware or electronic) is used.

What difference would it make if they're only there for redundancy?

 

[iwire]

You may want to use interlock keys to keep the fault current reasonable.

Having the trans in parallel when not needed is going to raise the fault current for no reason.

 

[Smart $]

This may have some effect on the matter.

450.7 Parallel Operation. Transformers shall be permitted
to be operated in parallel and switched as a unit, provided
the overcurrent protection for each transformer meets the
requirements of 450.3(A) for primary and secondary protective
devices over 1000 volts, or 450.3(B) for primary
and secondary protective devices 1000 volts or less.

Not exactly sure how, though. Does it mean if they operate in parallel at any point in time, they must be switched as a unit? And if they must be switched as a unit can they also be switched separately (we have the technology)?

 

[ggunn]

Aren't the secondaries of the the two transformers separately derived, and as such should not be tied together? It seems to me that here could be minute differences in the secondary voltages of the transformers and a very low resistance connection between them. Wouldn't that be bad?

 

[mbrooke]

This may have some effect on the matter.

Not exactly sure how, though. Does it mean if they operate in parallel at any point in time, they must be switched as a unit? And if they must be switched as a unit can they also be switched separately (we have the technology)?

Yes, that has weight, but just as puzzled as you on the switching part.

 

[iwire]

This may have some effect on the matter.

Not exactly sure how, though. Does it mean if they operate in parallel at any point in time, they must be switched as a unit? And if they must be switched as a unit can they also be switched separately (we have the technology)?

Just thinking out loud here. A single switch with or without integral OCP feeding into separate fuse blocks to provide correct primary protection for each unit?

 

[winnie]

Aren't the secondaries of the the two transformers separately derived, and as such should not be tied together? It seems to me that here could be minute differences in the secondary voltages of the transformers and a very low resistance connection between them. Wouldn't that be bad?

You are correct; mis-matched transformers will cause circulating currents.

My understanding is that 2 transformers manufactured to the same turns ratio and impedance (with normal manufacturing tolerance) will be 'close enough'.

There are ways to calculate out how much of a problem different impedance or different turns ratio will cause. Different turns ratio will cause circulating current; different impedance will cause unequal load sharing.

-Jon

 

[mbrooke]

You are correct; mis-matched transformers will cause circulating currents.

My understanding is that 2 transformers manufactured to the same turns ratio and impedance (with normal manufacturing tolerance) will be 'close enough'.

There are ways to calculate out how much of a problem different impedance or different turns ratio will cause. Different turns ratio will cause circulating current; different impedance will cause unequal load sharing.

-Jon

Correct. See this for details:


http://www.facilitiesnet.com/whitepapers/pdfs/schneiderelectric_111711.pdf

 

[Smart $]

Just thinking out loud here. A single switch with or without integral OCP feeding into separate fuse blocks to provide correct primary protection for each unit?

I was thinking one unfused disconnect switch for both transformers to be switched as a unit followed by two fused disconnects or enclosed MCCB's, one for each transformer primary. Of course, there'd be a need for disconnecting means on the secondary in order to disconnect either transformer while maintaining energized operation of the other. OCPD on the secondary is TBD.

 

[iwire]

I was thinking one unfused disconnect switch for both transformers to be switched as a unit followed by two fused disconnects or enclosed MCCB's, one for each transformer primary. Of course, there'd be a need for disconnecting means on the secondary in order to disconnect either transformer while maintaining energized operation of the other. OCPD on the secondary is TBD.

I don't see that as 'switched as a unit'.