SDS or not?

[rojay]

I’m reviewing a set of plans for a new generator install. Gennie is 3 phase 277/480 with a 1000 amp feeder breaker. The feeder supplies a 480 volt generator distribution switchboard where emergency & optional standby loads are split off to step down transformers (emergency & optional standby loads are 120/208 volt) before hitting their separate transfer switches. The optional standby transfer switch is solid neutral & contains the service disconnect. Questions- how can the generator be considered a non separately derived system and rely on the building service grounding when there’s a transformer between it and the transfer switch? Shouldn’t the neutral be bonded at the genset & connected to a separate grounding electrode system? If the generator is solidly grounded isn’t a ground fault signaling device required too?

 

[jaggedben]

First, all systems in a building that need a connection to a grounding electrode system should be connected to the same GES. Do not start creating separate electrode systems. Each system having its own grounding electrode conductor to the GES is okay.

I agree that the generator is separately derived, as you've described it, and needs it's neutral connected to the GES with a GEC.

I believe it's okay for the transformers' neutrals on the 208V side to be grounded through the service MBJ, although I wouldn't be surprised if that became a question. Alternatively, if the transfer switches switched the neutrals, the each transformer would be an SDS requiring its own GEC to the GES.

(Why is it being done this way? Is the service 208/120 but they could only find a 480/277 generator?)

 

[rojay]

First, all systems in a building that need a connection to a grounding electrode system should be connected to the same GES. Do not start creating separate electrode systems. Each system having its own grounding electrode conductor to the GES is okay.

I agree that the generator is separately derived, as you've described it, and needs it's neutral connected to the GES with a GEC.

I believe it's okay for the transformers' neutrals on the 208V side to be grounded through the service MBJ, although I wouldn't be surprised if that became a question. Alternatively, if the transfer switches switched the neutrals, the each transformer would be an SDS requiring its own GEC to the GES.

(Why is it being done this way? Is the service 208/120 but they could only find a 480/277 generator?)

Thanks for the feedback! With the generator being located outside, I was not aware that NEC 250.30(C) required the connection to the building GES, I thought a separate GES was required at the outdoor source? This is a really strange set up here, and it seems to be motivated by economics. The generator is customer owned and being shipped over from a different location.

Since the transformer secondary for the optional standby system terminates in the transfer switch where the service disconnect is located, I figured the main bonding jumper for the service & the system bonding jumper for the xfmr would be the same animal- no need for the SBJ at the xfmr end.

 

[jaggedben]

The generator being 'outside' doesn't mean much. Like, if it sits next to the wall of the same building I'd still say any required GEC should still go to the building GES, in my interpretation. But if you're saying it's remote from the building, meaning it's really at its own separate building or structure, then yes, you would install a new GES there, and run an EGC with the feeder to the building being served.

Take this next bit with a grain of salt because it's a very different application: I install battery back systems that utilize autotransformers to form a neutral when operating off grid. The neutral is connected to the the feeder neutral from the grid side and grounded by the GEC (and bonded by the MBJ) at the service 'upstream'. So I think it can be okay to have a transformer in a backup system where the neutral is grounded at the service. But, I don't know for sure if the systems I install have a solid or switched neutral connection to the transformer. And it's an autotransformer, not an isolation transformer.

What I think is clear in your described setup:
- The generator is an SDS.
- With a solid neutral transfer switch, you must not install system bonding jumpers at the isolation transformers because that would create objectionable current and a parallel neutral paths. (You said that above. I agree.)

What I think is not clear:
- Whether it is a 'good idea' and completely code compliant to have solid-neutral transfer switch downstream of an isolation transformer.

 

[wwhitney]

Questions- how can the generator be considered a non separately derived system and rely on the building service grounding when there’s a transformer between it and the transfer switch?

As I understand, there are two potential neutral paths between the generator and the service: generator - emergency system transformer - emergency system transfer switch, and generator - optional system transformer - optional system transfer switch.

So if either path has a solidly connected neutral, the generator is not an SDS. If neither path has a solidly connected neutral, then the generator is an SDS.

Usually transformers are configured as SDSs, but it is not required. Since you mentioned the optional system transfer switch has a solidly bonded neutral, if you want the generator not to be an SDS, you could set up the optional system transformer as a non-SDS.

It would presumably be a code violation to configure both transformers and both transfer switches with solidly bonded neutrals, as then you'd have a neutral loop.

Cheers, Wayne

 

[rojay]

As I understand, there are two potential neutral paths between the generator and the service: generator - emergency system transformer - emergency system transfer switch, and generator - optional system transformer - optional system transfer switch.

So if either path has a solidly connected neutral, the generator is not an SDS. If neither path has a solidly connected neutral, then the generator is an SDS.

Usually transformers are configured as SDSs, but it is not required. Since you mentioned the optional system transfer switch has a solidly bonded neutral, if you want the generator not to be an SDS, you could set up the optional system transformer as a non-SDS.

It would presumably be a code violation to configure both transformers and both transfer switches with solidly bonded neutrals, as then you'd have a neutral loop.

Cheers, Wayne

I think the generator definitely fits the definition of a SDS. Both the optional standby & EM system transformers are delta primary, wye secondary. Without the neutral to chassis bond at the generator, there doesn’t appear to be a ground fault return path.

I don’t think NEC 250.20 allows any choice in grounding the transformer secondaries either- it appears to be a must. I think the 3 pole optional standby/service disconnect transfer switch will work out fine. The problem will end up being the EM transfer switch. If a three pole solid neutral is used- there will end up being a neutral loop like you mentioned.

 

[jaggedben]

I think the generator definitely fits the definition of a SDS. Both the optional standby & EM system transformers are delta primary, wye secondary. Without the neutral to chassis bond at the generator, there doesn’t appear to be a ground fault return path.

I don’t think NEC 250.20 allows any choice in grounding the transformer secondaries either- it appears to be a must. I think the 3 pole optional standby/service disconnect transfer switch will work out fine. The problem will end up being the EM transfer switch. If a three pole solid neutral is used- there will end up being a neutral loop like you mentioned.

If you don't install a system bonding jumper, I don't see the neutral loop. Can one of you describe it in detail? Am I missing something?

Are the two sets of loads on separate service disconnects? Or does the transfer switch you mentioned with the service disco serve both when connected to the service?

(The service is 208/120, right?)

The thing I'm hung up on is this:

If you're EM transfer switch doesn’t contain a service disconnect and main bodning jumper, then 250.30 seems to require a system bonding jumper at that location or at the transformer. But that's what you can't do. If the neutral is bonded and grounded upstream on the utility side of the transfer switch, I don't know what the actual problem is but it just kinda seems wrong and possibly not compliant with 250.30.

 

[hillbilly1]

I think all of the transferswitchs would need to be four pole, if I’m following you correctly. It appears you have two service disconnects on the utility side? One for the building, and one for the emergency loads? The generator would be a SDS, transformers would be bonded as normal, they don’t make any difference, it’s the neutrals being paralleled on all of the different transferswitches if three pole switches are used.

 

[wwhitney]

I think the generator definitely fits the definition of a SDS.

If the voltage system the generator is creating has a grounded conductor, then that entirely depends on how you wire it, just like it does with transformers. See below.

Both the optional standby & EM system transformers are delta primary, wye secondary.

I take it here you mean primary = 480V, secondary = 208Y/120V. The generator you described as 480Y/277V.

Say you had only one transfer switch (for simplicity, to start), and you didn't want any SDSs. You could do that as follows: The generator has no N-G bond. You bring 4 wires plus EGC from the generator to the transformer. The EGC passes through, just bonding the case, and the primary neutral gets connected to the secondary neutral. The primary/secondary ungrounded conductors are connected as usual. Then at the transfer switch, only the ungrounded conductors are switched. The neutral-ground bond occurs at the service as usual. When on either utility power or generator power, all voltage systems depend on the service neutral-ground bond for ground faults.

I'm not say this is a good idea, or a bad idea; I'm just saying it's an option. Well, maybe I'm overlooking an NEC rule that says that when operating on generator power, the N-G bond can't be so far away from the generator. If there is a rule that would require a neutral-ground bond at the transformer secondary, then you'd need to use transfer switches that switch the neutral. Otherwise, if you want a neutral-ground bond at the generator, and don't want to switch the neutral in the transfer switches, then you better set up the transformers as SDSs, or you'll have two neutral-ground bonds in the system.

Cheers, Wayne

 

[jaggedben]

I think all of the transferswitchs would need to be four pole, if I’m following you correctly. It appears you have two service disconnects on the utility side? One for the building, and one for the emergency loads? The generator would be a SDS, transformers would be bonded as normal, they don’t make any difference, it’s the neutrals being paralleled on all of the different transferswitches if three pole switches are used.

How exactly are the neutrals paralleled? I'm still not seeing it.

The transformers totally make a difference. Without them it would be similar to these threads. But with isolation transfomers it seems to me that this is no longer the case at all. Neutral current would only go back to its respective transformer. Right?

(I agree that bonding the transformers 'as normal' and using four-pole switches is an option. Perhaps it's the better option. I just don't clearly see the danger of doing it the other way, i.e. not bonding and using three pole switches. But maybe an engineer should chime in on this thread and set the record straight.)

 

[hillbilly1]

How exactly are the neutrals paralleled? I'm still not seeing it.

The transformers totally make a difference. Without them it would be similar to these threads. But with isolation transfomers it seems to me that this is no longer the case at all. Neutral current would only go back to its respective transformer. Right?

(I agree that bonding the transformers 'as normal' and using four-pole switches is an option. Perhaps it's the better option. I just don't clearly see the danger of doing it the other way, i.e. not bonding and using three pole switches. But maybe an engineer should chime in on this thread and set the record straight.)

The transformers make no difference, it’s the utility side. The transformers are bonded to the utility neutral because the utility neutral is bonded to the same GEC the transformers are. If the generator was 120/208, it would still be the same issue. Sure when on generator, the current flow would be on their respective transformers, but when on utility, every neutral is paralleled because of the neutral bond on the generator/transformers side. I don’t think you could have the transformer bond using the neutral instead of a GEC is the problem. It would probably be a hard sell to the inspector.

 

[jim dungar]

Do you have a line-line or three a line diagram of the system which shows the neutral and grounds not just the hot lines?

If the generator is on the primary side of the transformers, their secondary Wye grounding/bonding connections are a separate discussion.

It seems the generator is 480Y/277 and the utility is 480Y/277 and the transfer switches are 480Y/277 with a solid neutral. If this is correct your generator is not an SDS.

 

[hillbilly1]

Do you have a line-line or three a line diagram of the system which shows the neutral and grounds not just the hot lines?

If the generator is on the primary side of the transformers, their secondary Wye grounding/bonding connections are a separate discussion.

It seems the generator is 480Y/277 and the utility is 480Y/277 and the transfer switches are 480Y/277 with a solid neutral. If this is correct your generator is not an SDS.

From what I gather though, the service is 208, not 480, just the generator is 480. Electrically what @jaggedben is suggesting would work, treating each transformer as a non-sds source, but rules for transformer bonding seem to be different than for generators.

 

[jim dungar]

From what I gather though, the service is 208, not 480, just the generator is 480. Electrically what @jaggedben is suggesting would work, treating each transformer as a non-sds source, but rules for transformer bonding seem to be different than for generators.

If the generator is 480V there must be a transformer between it and the utility service. Again the grounding/bonding on the secondary of this transformer is independent of the primary side. In this case the generator would be an SDS.

 

[rojay]

From what I gather though, the service is 208, not 480, just the generator is 480. Electrically what @jaggedben is suggesting would work, treating each transformer as a non-sds source, but rules for transformer bonding seem to be different than for generators.

That’s correct, the service is 208. For me, that’s where all the confusion started. Adding a SDS in between the generator & utility seems to change the game.

As proposed, a four wire feeder from a 277/480 volt wye connected generator hits a 1000 amp generator distribution panel. Why? From the distribution panel separate 3 phase 3 wire feeders supply the two delta/wye transformers- one for EM loads, one for optional standby loads.

I should have provided more detail in my original post- utility supply for EM loads comes from a different service than the optional loads. Essentially the optional standby is backing up the entire load of service B. Since the OSS transformer secondary terminates at the service disconnect in the transfer switch, I think the unswitched neutral there is appropriate as long as the SBJ is not installed at the transformer.

The EM system transfer switch is supplied by a feeder from another service where presumably it’s neutral has already been grounded. Since the EM system transformer secondary is required to be connected to the building GES, I think the transfer switch has to switch the neutral.

 

[jaggedben]

The transformers make no difference, it’s the utility side. The transformers are bonded to the utility neutral because the utility neutral is bonded to the same GEC the transformers are. If the generator was 120/208, it would still be the same issue. Sure when on generator, the current flow would be on their respective transformers, but when on utility, every neutral is paralleled because of the neutral bond on the generator/transformers side.

What bond? I'm saying you don't bond or ground the transformer neutrals again, just rely on the service bonding and grounding.

If there's factory bonds in the transformers that can't be removed then yes that's a problem.

I don’t think you could have the transformer bond using the neutral instead of a GEC is the problem. It would probably be a hard sell to the inspector.

A GEC is for grounding, not bonding. But see above.

 

[jaggedben]

... Since the OSS transformer secondary terminates at the service disconnect in the transfer switch, I think the unswitched neutral there is appropriate as long as the SBJ is not installed at the transformer.

I agree. I see no reason the MBJ and SBJ can't be one and the same if the location is compliant for both.

The EM system transfer switch is supplied by a feeder from another service where presumably it’s neutral has already been grounded. Since the EM system transformer secondary is required to be connected to the building GES, I think the transfer switch has to switch the neutral.

Again, not sure what precisely the problem is with having the neutral unswitched and grounded and bonded remotely at the service. But agree that switching the neutral and having it bonded and grounded normally as an SDS is clearly okay and more straightforwardly compliant.

 

[rojay]

I agree. I see no reason the MBJ and SBJ can't be one and the same if the location is compliant for both.


Again, not sure what precisely the problem is with having the neutral unswitched and grounded and bonded remotely at the service. But agree that switching the neutral and having it bonded and grounded normally as an SDS is clearly okay and more straightforwardly compliant.

I was thinking that not switching the neutral may be a violation of NEC 250.142(B)?

 

[jaggedben]

I was thinking that not switching the neutral may be a violation of NEC 250.142(B)?

I don't see how, as long as the neutral isn't bonded or grounded at the transformer.

 

[rojay]

I don't see how, as long as the neutral isn't bonded or grounded at the transformer.

The EM transformer secondary conductors terminate at the transfer switch. Without an overcurrent device at the EM transfer switch (unlike the OSS transfer switch) aren’t I stuck having to add the SBJ at the EM transformer and do my grounding there?