MV Feeder Application 3

[mpparent]

All,

I just want your view on an installation challenge. I have an existing generator system (13.2kV), and I need to install a new UPS system on it. However, I do not have a spare output breaker, so do you think it's wise to use an existing breaker, and "double-lug" so to speak on it to feed the new load? Gen. capacity is more than adequate. I'm thinking there could be a problem with the protective relaying (GE MDP's).

Thanks in advance,
Mike

 

[rbulsara]

I do not think this will meet NEC or any other sane Code. Parallel conductors has to terminate in single set of equipment (breaker/motor etc.). You cannot feed two separate circuits off a single breaker. This is different than feeding multiple loads on single circuit.

 

[mpparent]

rbulsara,

Yeah, I think you're right as I was looking at 240.101, and reread that section. We'll have to find a plan "b".

Thanks,
Mike

 

[rcw retired EE]

Putting new CT’s and a relay on the new feeder cables so each feeder has its own protection relay, may meet code. Each circuit will be protected individually. But any problem on one will affect the other.

Think about the effect on total reliability.

The system would probably not have a UPS or a generator if there weren’t a perceived need for continuous power. Placing two circuits on one breaker may reduce the reliability and will limit the flexibility. For example, both loads may have to go down every time you do maintenance on the UPS.

 

[matwong]

It is unsafe to put two different applications in one breaker control. Rather, it will be better for sharing the breaker from the two existing out-going feeders (same type of loads - if any) where you can spare one feeder out for your UPS system. In this case, the protection is easier to set but you may face difficulties in isolating individual load (if no extra equipment to be installed).

 

[alehman]

I don't think there's anything in the NEC that would prohibit feeding two circuits from one breaker so long as both are sized such that they are properly protected by the breaker and the total load doesn't excced the rating of the breaker or upstream circuits.

You may need separate overcurrent protection for the loads.

 

[rbulsara]

It also depends if you want to be proud of your work or not... do a good design. If the installation is worth having a UPS, have the Owner pay for the modification for having its own protection so that a fault on other circuit does not take out UPS feed.

I just cannot promote bad engineering...

 

[mpparent]

I agree that taking the new feeder of an existing breaker ("double-lugging") is not a good idea. At the time, it seemed like the only possibility, excluding the fact I don't think it meets code (after reading it again). I did consider that we would have to have separate relays for each feeder.
My only option is to do a feeder tap off an existing feeder. Again, it's not my first choice, but all considered, my only one.

Thanks all for your comments.

Mike

 

[alehman]

I agree with rbulsara's point about 'good engineering', especially considering the apparent critical nature of the application. You should seriously consider separating the circuits for improved reliability and maintainability if possible.

Unfortunately real-world existing conditions and budgets sometimes require less than an ideal arrangement so long as it can be done safely, which I think it can with proper engineering.

 

[rbulsara]

One thing I forgot to ask..you mentioned 13.2kV generators.
What voltage is the UPS? I assumed the UPS is low voltage. Not that this changes my basic opinion.

But if it is a MV feeder as the heading suggests, there are different practices for protecting MV feeders and taps thereof.

 

[mpparent]

Yep,

It'll be a 13.2kV feeder from generator paralleling gear (the gear is both the "utility" feed and the standby backup feed with switching via breakers), to a padmount xfmr, which will then feed some input switchgear to a UPS.

Mike

 

[alehman]

You will need overcurrent protection for the transformer. You may have difficulty making the breaker settings work for that and whatever else it is feeding.

 

[mpparent]

alehman,

I'm going to use an S&C switch w/ electronic fuses for the OC protection of the xfmr. I think I'm covered there.
Although I still think it's a horrible idea, I still don't find a reference in the code prohibiting me from double-lugging. If anyone can point me to it,...I'd appreciate it. I just spent a better part of the evening at home looking through my good ole 1999 codebook.

Mike

 

[rbulsara]

OK..this is little different now. I should have asked for clarification earlier.

I just don’t like the idea of lugging two feeders. But for MV feeders the overcurrent device (breaker or fuse) is generally meant for short circuit protection only. Overload is generally controlled by design. That is one reason MV fuses are rate (say E rating) such that they open in 300 sec at 200% ampere current of the E rating. It takes forever to open at current anything below that.

A better solution in my opinion would be:

1. Make the new transformer loop feed type. Rearrange MV feeder to feed the transformer and loop out and then pick up the existing feed. Or vice versa, but daisy chain the two loads if you can.

2. Provide a separate fusible switch (air interrupter) near the existing gear and you can either double lug for a short cable to the new switch.

3. Double lug only if the transformer is not too far from the gear otherwise tap the feeder.

You can even make the tap smaller as long as the MV OCPD protects it for the short circuit. Check the cable damage curve.

4. Overload for the transformer can be provided at the transformer using dry-well or bay-o-net fuses. I would also plan for a primary switch for the transformer

 

[alehman]

I don't think you'll find anything to prohibit double-lugging. It should be possible to size the cable so both sets are adequately protected.

I agree with all of rbulsara's comments except I'm not sure about the loop feed. Seems like two separate radial feeds would be more maintainable, esp. with isolation switches or load-break el's at the xfmr.

How about a PME-9 with two fuse bays, one for the existing load and one for the new xfmr? That will give you one unfused switch for future extension as well.

 

[rbulsara]

I agree with alehman that code does not appear to be prohibiting double lugging even for LV, as it meets the requirment of having the OCPD at the origin of the circuit. Otherwise tap rules apply. But still it not a good design. Code does not care for that, it only deals with safety.

 

[dpc]

On the general question of "double lugging" on MV breakers: In my experience bifurcated feeders are really pretty common in the industrial sector in the US. As long as feeders are adequately sized, there are no NEC issues. I don't see this design as particularly deficient when compared with looping a single MV feeder to four or five (or more) substations in series through a bunch of loop-feed fused switches.

For this particular installation, connecting the UPS to the Generator breaker doesn't sound like a very good approach, for a lot of the reasons previously mentioned.