Paralleling Switchgear Short Circuit Rating

Status
Not open for further replies.

Sahib

Senior Member
Location
India
The rest of your "review" is irrelevant to the OP.
May be irrelevant from your own point of view. But there are other PE's who consider it relevant because according to them, ''Based on a review of applicable codes and standards, there is no need to consider both sources of fault current when sizing equipment for closed-transition transfer schemes.''

I guess you still aren't reading too carefully
Yes. I do not think it very relevant.
 
ANALYSING RELIABILITY, A SIMPLE YET RIGOROUS APPROACH (PCIC-2003-25) wasn't my best IEEE paper IMO, but it was the only one that has been elevated to IEEE Transactions status.

One of the things revealed in the underlying FMEA study was that Figure "g" was both the safest and most reliable configuration if proper load auto-starts/restarts are included in the control system. Since auto-starts/restarts are almost always in the control system in some form or another, eliminating the tie altogether is often the best solution since it eliminates several potential failure modes and it doesn?t really provide the additional perceived ?flexibility?.

In fact, while I was on the NFPA 70E TC, I said I could design a distribution system that virtually never needed to be worked ?live? if dual feeds were provided without a tie. It has all the redundancy necessary for a continuous process.


View attachment 8863

While your claim for figure 'g' is true, in most cases the objective is that if A1 was running, then A1 should be restarted when figure 'e' is utilized. In order for A2 to start under figure 'g', not only automatic restart of A2 is required but all neecssary valve lineups would need to be automated, which they normally not. Quite the cost difference. Not to mention the valve sequence operation time addition and the complexity of equipment sequencing.
 

rbalex

Moderator
Staff member
Location
Mission Viejo, CA
Occupation
Professional Electrical Engineer
May be irrelevant from your own point of view. But there are other PE's who consider it relevant because according to them, ''Based on a review of applicable codes and standards, there is no need to consider both sources of fault current when sizing equipment for closed-transition transfer schemes.''
Unless and until the OP author explains why your ramblings are relevant to him, I don?t intend to waste anymore time responding to them.

Yes. I do not think it very relevant.
I think we can both agree you don?t believe reading carefully is relevant.
 

rbalex

Moderator
Staff member
Location
Mission Viejo, CA
Occupation
Professional Electrical Engineer
While your claim for figure 'g' is true, in most cases the objective is that if A1 was running, then A1 should be restarted when figure 'e' is utilized. In order for A2 to start under figure 'g', not only automatic restart of A2 is required but all neecssary valve lineups would need to be automated, which they normally not. Quite the cost difference. Not to mention the valve sequence operation time addition and the complexity of equipment sequencing.
The dominant failure mode concern is simply that A1 may fail for any number of reasons. If A2 is not ready to automatically assume the load the process usually fails OR you have enough time to set it up without the tie anyway.

Double service / fed systems are always expensive. However the cost of the tie is eliminated and often the short-circuit ratings are significantly reduced throughout the system. In my experience, the additional cost of electronic controls has been negligible in modern systems. The cost of piping controls are may be significant but are often offset by the reduced electrical costs and increased safety / reliability. Statistically, the tie severely degrades reliability as a major single-point failure in its own right.
 

Sahib

Senior Member
Location
India
Unless and until the OP author explains why your ramblings are relevant to him, I don?t intend to waste anymore time responding to them.

I want to stress again it is not irrelevant to the OP, because it is not simply my opinion but based on sound engineering practice.

I think we can both agree you don?t believe reading carefully is relevant.

I disagree.
 
The dominant failure mode concern is simply that A1 may fail for any number of reasons. If A2 is not ready to automatically assume the load the process usually fails OR you have enough time to set it up without the tie anyway..

There is the difference of what is the primary concern. Our primary concern is the upstream transformer/feeder/bus failure.

Perhaps this is the fundamental difference between the truly continuous refinery business as opposed to purely Chemical Processes?


Double service / fed systems are always expensive. However the cost of the tie is eliminated and often the short-circuit ratings are significantly reduced throughout the system. In my experience, the additional cost of electronic controls has been negligible in modern systems. The cost of piping controls are may be significant but are often offset by the reduced electrical costs and increased safety / reliability. Statistically, the tie severely degrades reliability as a major single-point failure in its own right.

Just looking at a two pump + recirculation leg feed system it would involve the addition of and automation of 6 on/off 4" valves, the addition of limit switches on block valves and other proess safety concerns that I am sure would come. Additional real estate for the access of the equipment, etc. woul be rather significant cost adder IMO.
 

JoeStillman

Senior Member
Location
West Chester, PA
I want to stress again it is not irrelevant to the OP, because it is not simply my opinion but based on sound engineering practice.

Sound Engineering Practice and what the code requires are sometimes two different things, hence my orignal quandry and the point of all of Mr. Alexander's RFCs.

What I gather from the responses of the CMP (in the attachments) was that they didn't want to touch the issue unless somebody shows them the math. One could probably make a case for the out-of-phase scenario, but the only case that can be made for the fault-during-transition problem is statistical improbability. Until someone can quantify the improbability, I gues the jury is still out.
 

rbalex

Moderator
Staff member
Location
Mission Viejo, CA
Occupation
Professional Electrical Engineer
SoundEngineering Practice and what the code requires are sometimes two differentthings, hence my orignal quandry and the point of all of Mr. Alexander's RFCs.

What I gather from the responses of the CMP (in the attachments) was that they didn't want to touch the issue unless somebody shows them the math. One could probably make a case for the out-of-phase scenario, but the only case that canbe made for the fault-during-transition problem is statistical improbability. Until someone can quantify the improbability, I gues the jury is still out.
Actually, the math has been done. I wish Mr. Goodman had included it in his Comment 1-120, May 2001 ROC, Page 38, included in Pages from 110.9 History-6.pdf‎. Having been somewhat involved with the development of IEEE-666, he was the one familiar with it. Statistically, there hasn?t been enough time in the accepted age of the universe for it to have happened once.

To make it simple:

IF the installation is within the Scope of the NEC [Section 90.2(A)] which CMP1 and the TCC controls too.

THEN you have two options:

ONE: Include the available fault from all potential sources at the line terminals whether they will be interrupted or not.

TWO: Sell it to the local AHJ.

Where I have attempted TWO (more than 6, but probably less than 10 times) I have never yet ?sold it?. See my Comment 1-116, May 2001 ROC, Page 37,Pages from 110.9 History-5.pdf‎
 

rbalex

Moderator
Staff member
Location
Mission Viejo, CA
Occupation
Professional Electrical Engineer
There is the difference of what is the primary concern. Our primary concern is the upstream transformer/feeder/bus failure.

Perhaps this is the fundamental difference between the truly continuous refinery business as opposed to purely Chemical Processes?

Just looking at a two pump + recirculation leg feed system it would involve the addition of and automation of 6 on/off 4" valves, the addition of limits witches on block valves and other proess safety concerns that I am sure would come. Additional real estate for the access of the equipment, etc. woul be rather significant cost adder IMO.
Sorry for the tardy response; I hit the ?go to last post button? and only saw your support post (#30) Thanks, BTW.

I had to chuckle a bit. Yeah, even refineries draw the line somewhere; e.g., a 19,000hp, 4160V hydrogen recycle compressor with a 3000A LCI ASD and 20MVA close-coupled bypass transformer all on a little bit more than a ? acre lot. Today, we probably would use a higher voltage and more compact switchgear, but there?s still no way they?d make a redundant setup for that ? even though it?s the heart of the process unit (the whole refinery actually) and they had the real estate for it. The original plot did have enough real estate, but there wasn?t even a second thought to redundancy. The process folks did consider steam as an alternate.

But for most smaller running/standby loads, redundancy is the key to process load reliability since the local mechanical units are statistically more likely to fail on their own than the electrical upstream. Of course, an upstream failure ALSO takes out the process loads without mechanical and electrical redundancy.
 

Sahib

Senior Member
Location
India
Sound Engineering Practice and what the code requires are sometimes two different things, hence my orignal quandry .

There is, by no means, universal agreement on what ratings to use for a closed transition. It is ultimately the responsibility of the engineer of record to make the decision.
 

Sahib

Senior Member
Location
India
Actully it is up to the AHJ to make the decision.

If the switch gear with marginal interrupting capacity installed with the permission of AHJ blows up injuring people nearby, the Engineer of record would be held responsible and not the AHJ. That is the meaning intended.
 

iwire

Moderator
Staff member
Location
Massachusetts
If the switch gear with marginal interrupting capacity installed with the permission of AHJ blows up injuring people nearby, the Engineer of record would be held responsible and not the AHJ. That is the meaning intended.

Nice spin. :thumbsup:
 

Sahib

Senior Member
Location
India
Nice spin. :thumbsup:

No. I have not changed sides.:happyno:

To use or not to use closed transition with switch gear of marginal breaking capacity depends on meeting carefully defined set of conditions: the Engineer's responsibility is to meet all those conditions not only during the inspection by the AHJ but also throughout the service life of the electrical installation.
 
Status
Not open for further replies.
Top