Diversity Factor applied to Short Circuit - I need help

[mityeltu]

I am working on the justification for switching a 480V board from normal to alternate feed. My load breakers are rated at 30kA but my short circuit current, per ETAP, is 38kA. I'll try to make my intro short. I have an existing ETAP calc that says this switch is ok because there is a 60% diversity factor on the boards that will feed this SC. My question is about the application of diversity to SC. I have never seen this done before. Is it permissible to apply a diversity factor on a board to its contribution to a short circuit on a different board?

In other words, if I have 480V board A receiving a SC contribution from 480V board B of 10kA (from ETAP whoch does not take into account any diversity durng SC), and board B has a diversity factor of 60%, can I then say that the SC contribution to board A is 6kA? Is there any white paper or anything that leads to this justification?

 

[jim dungar]

Available fault currnet is simply the addition of all of the possible sources of fault current. In your case this would be the current from the normal source + the current from the alternate source.

I have never applied a diversity factor to SC calculations. In fact I cannot even picture what you are describing.

I have taken into account the 'risk of a bolted fault occuring during switching', as well as closed transition manual vs automatic switching.

 

[topgone]

I am working on the justification for switching a 480V board from normal to alternate feed. My load breakers are rated at 30kA but my short circuit current, per ETAP, is 38kA. I'll try to make my intro short. I have an existing ETAP calc that says this switch is ok because there is a 60% diversity factor on the boards that will feed this SC. My question is about the application of diversity to SC. I have never seen this done before. Is it permissible to apply a diversity factor on a board to its contribution to a short circuit on a different board?

In other words, if I have 480V board A receiving a SC contribution from 480V board B of 10kA (from ETAP whoch does not take into account any diversity durng SC), and board B has a diversity factor of 60%, can I then say that the SC contribution to board A is 6kA? Is there any white paper or anything that leads to this justification?

Did you play around with the cabling sizes and lengths to reduce the possible short-circuit values? I've solved a similar project problem by changing the supply cables a bit. 38 kA is very near the 30kA rating, IMO.

 

[kingpb]

I am working on the justification for switching a 480V board from normal to alternate feed. My load breakers are rated at 30kA but my short circuit current, per ETAP, is 38kA. I'll try to make my intro short. I have an existing ETAP calc that says this switch is ok because there is a 60% diversity factor on the boards that will feed this SC. My question is about the application of diversity to SC. I have never seen this done before. Is it permissible to apply a diversity factor on a board to its contribution to a short circuit on a different board?

In other words, if I have 480V board A receiving a SC contribution from 480V board B of 10kA (from ETAP whoch does not take into account any diversity durng SC), and board B has a diversity factor of 60%, can I then say that the SC contribution to board A is 6kA? Is there any white paper or anything that leads to this justification?

Where is the fault location that the 38kA is available? What I am getting at is looking to see how much through fault current will the breaker actually see. Say if the fault is on the load side, then you can deduct the fault current contribution from the load. Run ETAP and look at contributions from 1-bus away.

Also, as pointed out, are your cable lengths correct?

Unless the equipment cannot physically be connected at the same time through either electrical interlocks, kirk-key set-up, or equipment physically cannot operate simultaneously, I would advise extreme caution in applying a load diversity.

 

[mityeltu]

This is an existing installation and cable sizes are correct for the installation. I cannot change these for the switch from norm to alt.

The boards primary and alternate supplies are break before make, so the SC I'm looking at is only for the already switched configuration in ETAP.

Perhaps I ma not being clear. I'll try to illuminate. The 480V board I am studying has 2 down stream boards and is fed by a 6.9kV boards with 1.5MVA 6.9kV/480V transformer with z=8%. All the boards have a diversity factor of 60%. According to the person who set up these boards in ETAP, this means that none of the boards ever sees more than 60% load at any given time - this is taken from empirical data over the period of a year.

A short circuit study using ETAP indicates that the load breakers on my 480V board are over dutied. They have a 30kA SC rating and the available SC current on the board is 38kA. The "justification" I am being told to use is that the 60% diversity factor applied to the boards feeding the fault on my 480V board limits theit contibution so that the available fault current is actually below the 30kA rating of my breakers.

This logic seems to be that since there is never more than 60% loading on the boards, there would be less than the 100% SC current that ETAP calculates since ETAP does not apply any diversity factor to its SC current calculation.

My problem with this is that I have never heard of using a diversity factor in conjunction with a short-circuit study. What I am hoping to find is either definitive justification or denial of this method.

I hope this makes it bit less muddy.

 

[jim dungar]

The load currents have nothing to do with available fault currents.

 

[T.M.Haja Sahib]

A short circuit study using ETAP indicates that the load breakers on my 480V board are over dutied. They have a 30kA SC rating and the available SC current on the board is 38kA. The "justification" I am being told to use is that the 60% diversity factor applied to the boards feeding the fault on my 480V board limits theit contibution so that the available fault current is actually below the 30kA rating of my breakers.

What is the SC Rating required if only line side of the board is considered?

 

[rbalex]

Where is the fault location that the 38kA is available? What I am getting at is looking to see how much through fault current will the breaker actually see. Say if the fault is on the load side, then you can deduct the fault current contribution from the load...

This isn't quite true (although it should be - I quit writing proposals to make it so) See Section 110.9, first sentence. Interrupting rating is always based on available line-side fault no matter where the fault location is.

 

[mityeltu]

i understand what you are saying when you make the comment about there being no correlation between lod current and fault current, but the point that was made to me is this, if there is no device connected (switched off), then there is no fault current associated with that device. That is, if there is a diversity factor on a particular board such that only 60% of the load is ever "on" then that should have an impact on the available fault current. What I can't understand is how to justify this. In fact I can find NO way to justify this without knowing exactly WHICH loads are "off" at any given moment, and then using the worst case.

hat I hear from the group is this: there is no justification that can be made for using the diversity factor during a short circuit condition. Is this accurate?

 

[rbalex]

...
I have taken into account the 'risk of a bolted fault occuring during switching', as well as closed transition manual vs automatic switching.

In light of Section 110.9, what would be the basis for this in a non-utility installation? That was specifically one of API's Proposals that CMP1 rejected. (I helped author that one too.)

Maybe it applies to the "Available fault currnet ..." in your first sentence. (Check my spelling, will you?)

 

[topgone]

i understand what you are saying when you make the comment about there being no correlation between lod current and fault current, but the point that was made to me is this, if there is no device connected (switched off), then there is no fault current associated with that device. That is, if there is a diversity factor on a particular board such that only 60% of the load is ever "on" then that should have an impact on the available fault current. What I can't understand is how to justify this. In fact I can find NO way to justify this without knowing exactly WHICH loads are "off" at any given moment, and then using the worst case.

hat I hear from the group is this: there is no justification that can be made for using the diversity factor during a short circuit condition. Is this accurate?

Yep, forget about that diversity factor thing.

Since you said yours is an existing setup, have you ever thought of using FCLs? (fault current limiters). Or in-line inductors (though you will have problems with transient recovery voltage using this).

 

[rbalex]

...
What I hear from the group is this: there is no justification that can be made for using the diversity factor during a short circuit condition. Is this accurate?

I don't have the Standard in front of me now and I don't remember its number, but there is an IEEE Standard that recognizes it for certain utility applications (it isn't the NESC - that I do have) API referenced it in the Proposal I mentioned earlier. Dr Shokooh, the original developer of ETAP, was also a member of the Standard's Technical Committee.

 

[mityeltu]

I have access to all (I think) of the IEEE standards, but I have no idea what it might be. I know it's not in the red book.

What is API? Perhaps I can find the proposal you're referring to.

 

[rbalex]

I have access to all (I think) of the IEEE standards, but I have no idea what it might be. I know it's not in the red book.

What is API? Perhaps I can find the proposal you're referring to.

American Petroleum Institute. The Proposal was in the 1996 or 1999 (more likely) cycle. Can't remember that one either - creeping senility.

 

[mityeltu]

@ topgone:

This is apparently the first timwe this issue has been raised. When we moved to ETAP (long before I got here) the calc was issued with this verbiage that the breaker SC rating is fine because of the diversity factor. I have never seen this before this past Monday, so I was completely baffled.

I may run the idea of using FCL's, but that will require a design change that will take months as opposed to the justification that is requested by tuesday.

Looks like they won't get it.

Thanks for all the input.

 

[rcwilson]

A downstream distribution board can only be a source of SC current is if it has generation, motor loads or a loop feed from another facility. Your boards probably feed motor loads that act as a source of SC current during a fault.

The amount of SC current from an induction motor is essentially the same whether it is loaded 5% or 100%. The total contribution from all the motors could be less only if some of the motors are not connected. Historical load factor could be 60% if all motors operate at 60% of nameplate or only run 60% of the time but the SC contribution would still be 100%.

Assuming 40% of the motors are not running might be OK if the process controls are designed so there is never a chance that more than 60% of the motors will be on line at once, even during switch over from Pump A to Pump B. But I would not put my PE stamp on an assumption like that.

You might check the ETAP files for motor feeder cable sizes and lengths on the downstream boards. Many times to reduce study costs, motor wiring is ignored since the motor contribution is small in relation to the total SC and ignoring the individual cables adds a safety factor. Including motor lead impedance will reduce the calculated fault contribution from the motors.

 

[topgone]

@ topgone:

This is apparently the first timwe this issue has been raised. When we moved to ETAP (long before I got here) the calc was issued with this verbiage that the breaker SC rating is fine because of the diversity factor. I have never seen this before this past Monday, so I was completely baffled.

I may run the idea of using FCL's, but that will require a design change that will take months as opposed to the justification that is requested by tuesday.

Looks like they won't get it.

Thanks for all the input.

Would lengthening your secondary cables (480V side to the DP) make things work for you? All you needed is to insert a little impedance of about 0.0019 ohms and your breakers will be just fine. See it here:

Impedance with 30kA = 480V/[(sqrt(3)(30,000)] = 0.00923 ohms;
Impedance with 38kA = 480V/[(sqrt(3)(38,000)] = 0.00729 ohms.
To bring the fault current level back to the 30kA level, you needed to insert:
0.00923-0.00729 = 0.0019 ohms!
Please tell us how many parallel cables you have for the secondary lines?

 

[kingpb]

I would take a hard look at the model and make sure it is correct.

A 1500KVA Xfmr at 8% impedance only gives you about 22.5KA, and that is assuming that you have an infinite bus on the HV side of the transformer. Which means you have to account for around 15kA contribution from your boards.

Under the short circuit study module in ETAP, go to the Study case, and select the Standard tab. There you can elect to look at Protective Device Duty Based on Max through fault. I would run it and compare it to Total. Then you can see what circumstances lead you to the higher current.

BTW: if you have to account for 15KA fault current, if each motor at 480V has on average a Xsc of 0.20, multiplying 15kA by 0.2 = 3000A of motor FLA current, on line during a fault. A 1500KVA transformer only has a rated current of 1800A, so I would start verifying the one-line set-up. Seems to be some misleading data.

 

[rbalex]

1- 120 - (110-9): Reject
SUBMITTER: Mark Goodman, BP (ARCO)
COMMENT ON PROPOSAL NO: 1-211
RECOMMENDATION: Add the following text after the first
sentence: "During the momentary paralleling of an automatically
controlled power transfer, the single source condition interrupting
rating shall be considered adequate provided there is no
intentional time delay in the transfer and the parallel condition
cannot be maintained".
SUBSTANTIATION: This comment supports the concept of the
proposal, but suggests alternate language.
The condition identified in the proposal is common in double
ended substation arrangements where maintaining power to the
loads is critical for operation or system safety. It is also a common
practice in generating stations where continuity of power flow may
be required for public safety. These systems are designed for only a
momentary parallel condition, typically only long enough for the
closure of the paralleling breaker to initialize the opening of the
designated breaker returning the system to single source
configuration. During the brief parallel time (often only a few
cycles), the short circuit rating of the switchgear feeder breakers
may be exceeded. This added provision in 110-9 would recognize
this arrangement and establish limitations under which it can be
applied.
IEEE 666, Design Guide for Electric Power Service Systems for
Generating Stations, section 4.6.1 specifically allows this
arrangement and equipment rating.
The Explanation of Negative provided by Mr. Floyd is also
supported in this comment.
PANEL ACTION: Reject.
PANEL STATEMENT: See panel action and statement on
Comment 1-116.
NUMBER OF PANEL MEMBERS ELIGIBLE TO VOTE: 13
VOTE ON PANEL ACTION:
AFFIRMATIVE: 13

I remember now - It was the Standard's number that scared every one.

 

[kingpb]

It may have been rejected, but this is a pretty standard method of operation. It is usually impractical to size the swgr for a short time event, and most owner's will accept the risk in order to save the cost.