IEEE 142 System Grounding - Question

[mityeltu]

I am new to grounding and am reading this to help get me started. I am a fairly green engineer, so please, be gentle.

in section 1.4, last paragraph in the 1991 version, it stipulates a grounding transformer should be connected directly to the bus without using any intervening fuses or breakers (to make sure it is not taken out of the system inadvertantly), then in the very next sentence, it suggests that the grounding xfmr should be connected to a feeder breaker.

Did I miss something? I understand that one would be on the MAIN bus and the other on a sub-bus, but if the xfmr is for grounding the system, what would be the effective difference? They should both be sized the same, right? I know the impedance to the breaker connected xfmr would be slightly increased due to the buswork/cbling, but would that make such a difference that it voids the point of NOT having the xfmr on a breaker?

What am I missing?

 

[skeshesh]

From IEEE 142 (green book) circ. 2007:

"It is generally desirable to connect a grounding transformer directly to the main bus of a power system, without intervening circuit breakers or fuses, to prevent the transformer from being inadvertently taken out of service by the operation of the intervening devices. (In this case the transformer is considered part of the bus and is protected by the relaying applied for bus protection.) Alternatively, the grounding transformer should be served by a dedicated feeder circuit breaker, as shown in part a) of Figure 1-17, or connected between the main transformer and the main switchgear, as illustrated in part b) of Figure 1-17. If the grounding transformer is connected as shown in part b) of Figure 1-17, there should be one grounding transformer for each delta-connected bank supplying power to the system, or enough grounding transformers to assure at least one grounding transformer on the system at all times. When the grounding transformer is so connected, it is included in the protective system of the main transformer."

So we have three suggested conditions all of which have some sort of protection scheme; breakdown is as follows:

a)Connected to main bus: "In this case the transformer is considered part of the bust and is protected by the relaying applied for bus protection".
b)Connected to a dedicated feeder circuit breaker: Self-explanatory.
c)Between main transformer and main switchgear: "Included in the protective system of the main transformer".

 

[mityeltu]

How are they really any different from the standpoint of their function? The idea from section 1.4 is "obtaining system neutral". If the idea behind putting this xfmr here or there is simply placing a system neutral in a previously ungrounded system, what difference does it make if it is connected to the main bus directly vs. another bus via breaker? Isn't the end the same, a xfmr grounded system?

That being said, it seemd to me that the placement refers to the protection of the xfmr and not the system. If that's the case, then why impose the notion that one method is preferable to another?

 

[skeshesh]

How are they really any different from the standpoint of their function? The idea from section 1.4 is "obtaining system neutral". If the idea behind putting this xfmr here or there is simply placing a system neutral in a previously ungrounded system, what difference does it make if it is connected to the main bus directly vs. another bus via breaker? Isn't the end the same, a xfmr grounded system?

That being said, it seemd to me that the placement refers to the protection of the xfmr and not the system. If that's the case, then why impose the notion that one method is preferable to another?

There are several issues being addressed here. Yes, the function of the transformer is to obtain a neutral. But then the question becomes how to protect the grounding transformer. If the transformer is protected by the relaying of the main bus (and assuming the protective scheme has been properly coordinated), since the bus protection is upstream of the rest of the system, this setup will "prevent the transformer from being inadvertently taken out of service by the operation of the intervening devices." So while you're right that the end-game is to have a xfmr grounded system, one setup will give you superior reliability. The protection of the xfmr is a part of the bigger picture that is the protection of the system, so by viewing it in that context some methods have advantages over others that you may not see if you're just isolating the case of the xfmr.

 

[mityeltu]

I guess what I'm really having trouble with is the wording. I see that the location of the transformer determines its protection, but the statements still seem in opposition to me. If you don't want the transformer removed from the system to avoid operation ungrounded, then why not ensure that the xfmr is always unprotected, from a component level (that doesn't sound right, but I hope you understand what I mean - protect the system and RTF the xfmr).

Is the real point "access"? What I mean is, if I can't gain access to the system via main bus, but I CAN get access via a sub, then place the grounding xfmr where I CAN gain access? And, if placed on a secondary bus, since I can't have a direct link to the main bus for grounding, protect the xfmr with a feeder breaker?

Here's the next question then, if the idea is grounding, and you don't want to operate the system ungrounded, then why apply a feeder breaker? Why not connect directly to the downstream bus if the main is unavailable? If the breaker is set for selective coordination, then it may trip out during a ground fault.... wouldn't that be bad?

If the idea is to trip out on an internal fault on the grounding xfmr, then why wouldn't a breaker be applied in the same manner when the xfmr is on the main bus?

Sorry for the length.

 

[skeshesh]

I guess what I'm really having trouble with is the wording. I see that the location of the transformer determines its protection, but the statements still seem in opposition to me. If you don't want the transformer removed from the system to avoid operation ungrounded, then why not ensure that the xfmr is always unprotected, from a component level (that doesn't sound right, but I hope you understand what I mean - protect the system and RTF the xfmr).

Is the real point "access"? What I mean is, if I can't gain access to the system via main bus, but I CAN get access via a sub, then place the grounding xfmr where I CAN gain access? And, if placed on a secondary bus, since I can't have a direct link to the main bus for grounding, protect the xfmr with a feeder breaker?

Here's the next question then, if the idea is grounding, and you don't want to operate the system ungrounded, then why apply a feeder breaker? Why not connect directly to the downstream bus if the main is unavailable? If the breaker is set for selective coordination, then it may trip out during a ground fault.... wouldn't that be bad?

If the idea is to trip out on an internal fault on the grounding xfmr, then why wouldn't a breaker be applied in the same manner when the xfmr is on the main bus?

Sorry for the length.

Your concern is relevant and there may be very special applications where unprotected grounding xfrm may be used (not that I know of), but the point of reliability in a grounded system is usually not to sacrifice protection to continue operation, specially considering the fact that if the conductors feeding your grounding transformer or the transformer experience equipment damage, then your system neutral is lost AND you have to experienced extensive downtime.

"Access" may be an issue. I think a more relevant concern is where you need to serve line-to-neutral loads. When you have a large system and facility, at times it makes sense to only put the grounding xfmr where you need it. And yes you can and should connect it to the main bus if possible; I think you're misreading the language to say if it's anywhere other than the very main switchgear/substation you have to connect it to a CBed feeder, whereas it's saying that's the less desired alternative.

When the grounding xfmr is connected to the main bus directly it is protected by the OCPD of the main bus. When you connect it downstream with a feeder, the feeder will require it's own OCPD.

No worries about the length. I wouldn't consider myself an expert anyway, but I have some experience with the subject and the discussion allows me to understand its application more clearly.

 

[mityeltu]

Good, well, I'm glad I'm helping us both then.

So, I guess the point is the balance between system protection and grounding xfmr protection. The xfmr's zone being decided by its relative location. That makes sense. Tradeoffs seem to permeate engineering.

I guess my focus is too narrow. When looking at the entire scheme of a grounding xfmr, it's not appropriate to look at it independantly of the protection applied to the system AND the xfmr. So, with the xfmr placed here or there, there should still be ground fault sensing relays etc to take system components off line for safety as well as to prevent damage. Running the xfmr to failure would not be acceptable (as you pointed out) because the system would be continually experiencing a ground fault that would then be essentially voiding the protection scheme if it wasn't cleared.

Thanks for the input. It has helped me understand this better.

 

[skeshesh]

Running the xfmr to failure would not be acceptable (as you pointed out) because the system would be continually experiencing a ground fault that would then be essentially voiding the protection scheme if it wasn't cleared.

Yes that's a good point also. If you allow for a ground fault to propogate it may experience a line-to-ground or doubleline-to-ground fault which usually results in a 3-phase fault. Having said that it's very possible that, specially if using an advanced relay, the ground fault would be caught, but the result is the same I guess.