Calculate Short-Circuit Currents

[dahualin]

I am doing a tenant fit-out project. Since I don't have the utility transformer information, I cannot do the short-circuit current calculation. One senior electrical engineer told me that I can use 600A RK-1 fuses at service disconnect means then I don't need to do any short-circuit current calculation, just put AIC 10,000A for each panel. I have checked the information of current limiting RK1 fuses that says its interrupting rating is 200,000A RMS Sym. I am confused about the AIC rating of each electrical panel and the interrupting rating of fuses.
Is the senior electrical engineer's idea correct? How do you approach the AIC rating or short-circuit current calcuation without transformer's information?
Any help is really appreciated.

David

 

[hmspe]

That wouldn't fly where I live.

The plans reviewers want to see the calcs. I'm not sure why it's necessary with current Code. At one time the 110-22 label had to state the available fault current at the panel. That's changed to showing the required rating, so all the calculation does is tell you whether a 10KAIC panel is fully rated based on fault current reduction due to feeder impedence.

In any case, I'm not sure using 600A RK-1 fuses works for series rating in this case. The 600A fuses would be in the main, so any series rating would be from the main OCPD to the feeder mains, or there would have to be a laboratory verified three tier series rating available. This sounds like the old "up, over, and down" method. It is no longer considered acceptable in most cases. Any circuit with "dymanic impedance", essentially a condition where when a breaker's contacts start to open an arc is created which limits the fault current to a value too low to open the upstream OCPD, can no longer use the fuse's let-through current to limit fault current. In my opinion this is more a marketing thing than a real problem, but it's the way it is.

In this area the utility provides a "worst case" fault current at the service, and we're generally not allowed to calculate fault reduction across a service drop or a service lateral. When I'm faced with a problem like you have I generally use 65KAIC for 120/208 or 42KAIC for 277/480 at the service. These are a standard AIC ratings for breakers at those voltages, and they work most places for services up to around 200A services.

Martin

 

[coulter]

... This sounds like the old "up, over, and down" method. It is no longer considered acceptable in most cases. Any circuit with "dymanic impedance", essentially a condition where when a breaker's contacts start to open an arc is created which limits the fault current to a value too low to open the upstream OCPD, can no longer use the fuse's let-through current to limit fault current. ...

Martin -

I've heard this, but I can't find any references - just inuendo and anecdotal information. What I am looking for would be IEEE papers, regulations, other codes or standards. Can you help me out?

 

[jim dungar]

The up-over-down method has never ever been acceptable in ANY case that involves molded case circuit breakers. Bussmann, in their SPD literature, refers to these as devices with dynamic impedance.

Look at 240.86.

240.86(A) requires your engineer to apply his stamp to his calculations and design.

240.86(B) requires you to only use listed combinations of devices. But it does not tell you whose to follow, the fuse manufacturer's listing or the circuit breaker manufacturer's. Be careful of just simply using the UL Yellow book information, it is intended for manufacturer's (OEMs) not users.

And while you are at it don't forget to follow 110.22.

 

[ron]

In summary, you cannot simply install current limiting fuses and ASSUME that all down stream equipment is 10kAIC.
You need utility information, or at least the size of the transformer to go anywhere with AIC ratings of downstream equipment.

Or .... you could just specify everything to 200kAIC and you're covered in most situations.

 

[dahualin]

So I have to find the AIC or transformer information from local utility company to do short circuit current calculation. I read from another thread that says there are two AIC rating panels, fully-rated panel and series-rated panel. That thread didn't explan it very clearly. Could anybody explan it again for us what the fully-rated panel and series-rated panel are? Thanks.
For the most cases, which system you normally use, fully-rated system or series-rated system? Why?

Thanks.

David

 

[ron]

David,
I only specify fully rated systems.
Fully rated means that the device/equipment is rated on its own to interrupt the calculated fault current at its position in the distribution system.
Series rated means that the device/equipment is rated for a particular value, only when another device is further upstream of it in the distribution system. See 110.22

 

[jim dungar]

I usually go with series rated equipment for non-industrial and non-service entrance applications.

My experience is that after the intial project, devices (in particular breakers in branch circuit panels) are added with no consideration given to the available fault level. So the more standard (i.e. cost effective) equipment I can design in the less likely that a misapplication will occur in the future.

 

[coulter]

The up-over-down method has never ever been acceptable in ANY case that involves molded case circuit breakers. ...

Maybe I don't know what the "UOD" method is. I had thought that it refered to using the mfg curves for current limiting fuses.

...240.86(A) requires your engineer to apply his stamp to his calculations and design. .../QUOTE]

Yes. However, there is no specification about this is to be done. If the registered engineer wants to use the mfg curves on current limiting fuses,, why wouldn't that be acceptable?

...240.86(B) requires you to only use listed combinations of devices. ...

My interpretation of 240.86 says "A" (engineered, stamped) OR "B" (un-engineered, un-stamped, selected from MFG catalog for end use items.)

...240.86(B) requires you to only use listed combinations of devices. ... Be careful of just simply using the UL Yellow book information, it is intended for manufacturer's (OEMs) not users. ...

240.86(B) says the tested combination is marked on the end use device such as a panelboard or a switchboard. That suggests the intent is to select from a group of tested combinations offered by a mfg in a panel. It ceertainly does not suggest it is okay to pick a fused disconnect from one mfg and a cb panel from another.

... And while you are at it don't forget to follow 110.22. ...

Which, of course has to do with 240.86(B) not (A).

So, asking again - Assuming "UOD" refers to using the mfg curves for current limiting fuses: What prohibits a registered engineer from using the mfg curves, with mfg notes, limitations, and directions, for current limiting fuses, to provide a series rating for a panel board?

 

[jim dungar]

Coulter

You are wrong in your interpretation of UOD. The UOD method is used to determine the required short circuit WITHSTAND rating not the INTERUPPTING (AIC) rating of downstream equipment.

The is no fuse manufacturer that recommends the use of the UOD method for protecting molded case circuit breakers.

240.86(B) requires only that the combination of devices to be listed. It does not require them to be from the same manufacturer not that they be in the same enclosure. In fact UL listing requirements for series rated fuse-breaker combination must not be dependent upon any specific fuse manufacturer.

110.22 is applicable to both 240.86(A) and (B).

For more infomation on the use of fuses to protect breakers go to:
www.bussmann.com for the view point of a fuse manufacturer and www.squared.com for that breaker of a breaker manufacturer.

In 2005, IAEI even had an article on the viability of 240.86(A), http://www.iaei.org/subscriber/magazine/05_e/nema.htm where they say:
"One analytical method used in the past for estimating some series ratings was the Up-Over-Down method for protection of some equipment by current-limiting fuses. However, as related to series ratings, it was found that if the downstream circuit breaker contacts were to open before the upstream fuse clears, the protection is unknown. The reason protection cannot be predicted in this way is that the two devices are sharing the energy of interruption. Contact separation is the beginning of circuit interruption by the downstream circuit breaker. Remember this breaker is not rated for the circuit with the increased fault current that it is attempting to clear. This action by the circuit breaker introduces dynamic impedance in the circuit and takes some of the energy of interruption from the upstream device, which slows clearing by the upstream device. The result is that the downstream circuit breaker often attempts to interrupt more energy than it is designed to take and it is therefore not protected. Publications in endnote references 2, 3, and 4 describe this interaction."

 

[coulter]

...You are wrong in your interpretation of UOD. The UOD method is used to determine the required short circuit WITHSTAND rating not the INTERUPPTING (AIC) rating of downstream equipment.

Just so I am clear - are you absolutely sure I am crossed up on my thinking?

...The is no fuse manufacturer that recommends the use of the UOD method for protecting molded case circuit breakers.

I certainly haven't checked all fuse mfgs like you have, so I'll take your word for that. And, you have certainly convinced me I don't know what the UOD method is. And after reading your post and checking your references, I have come to the conclusion I still don't have a definition of the UOD method. I promise, this is really how I feel, I absolutely, uncategoricaly, without hesitation, in a loud clear voice state, "I don't have a clear definition of the UOD method".

What I am trying to understand is if you saying that using the mfg curves for current limiting fuses to calculate the available downstream fault current, for the purpose of keeping a CB panelboard within its SCC ratings is a bad engineering choice. As I recall, I asked for IEEE papers, regulations, other codes or standards. An IAEI article on bad engineering causes bad operation, is not what I had in mind.

Interestingly, Bussmann, in their SPD-02, http://www.bussmann.com/apen/pubs/spd/index.asp
pages 71 - 73, discusses currently limiting fuses, clearing times, and operational parameters. And they discuss how these currently limiting fuses can limit the short circuit to protect downstrean components and how to predict the clearing time to coordinate. They stopped short of saying one could use this method to limit short circuit current to a downstream CB panelboard.

... 240.86(B) requires only that the combination of devices to be listed. ... 110.22 is applicable to both 240.86(A) and (B)....

Uh, no, 240.86(B) says "tested", doesn't say anything about listing being required.

And 110.22, para 2, discusses series combination ratings marked on the equipment by the mfg. That is 240.86(B) - has nothing to do with 240.86(A).

 

[hmspe]

In the 2002 SPD there's a section on "Component Protection: Current-Limitation starting on page 72. The method they show in the graph on page 72 is "UOD". You can also see this on-line in section 10 of the document at http://www.bussmann.com/apen/pubs/spd/index.asp

For the "up", you enter the table at the available fault current at the line side of the fuse (about 86,000 on the page 72 chart). Go up until you hit the load line for the ampacity of fuse you're using (800A in the chart). Go "over" to the left until you hit the A-B line, then go "down" to the bottom of the chart to get the "let-through" current (about 21,000A on the chart).

The 1996 SDP had a very different presentation, and had prospective short current charts (like on page 72 of the 2002 SPD) and let-through tables for all the common fuses.

10+ years ago UOD was the accepted method of limiting fault current to panels, at least in the area where I live. My recollection of the history here is that many years ago circuit breakers were relatively slow to start opening, so it was highly unlikely that the contacts in a breaker would move in the first 1/2 cycle of a fault, so the fuse would current-limit, protecting the downstream breakers. About 10 years ago either circuit breakers suddenly got faster, or someone realized that circuit breakers had gotten faster, or the breaker manufacturers came up with a creative way to encourage people to use main breakers rather than main fuses. I've heard all three quoted as fact. In any case, the use of UOD with respect to down-stream breaker panels came to an abrupt stop. For a while we could use the data points listed in the let-through tables because that "were the result of lab testing". Then that went away. I've heard various things about why UOD was eliminated for fuse/breaker systems, most frequently that the breaker manufacturers pulled their listings/approvals.

As has been previously posted, the theory for eliminating UOD for fuse/breaker combinations is that circuit breakers can start to open before the upstream fuse can clear. The resulting arc at the breaker contacts may limit the current to the point that the current limiting fuse may not see enough current to clear. The term I've heard use for this is "dynamic impedance". UOD can still be used if there are no devices with dynamic impedance downstream of a fuse.

In reality there's not much difference between the old UOD method and the current series rating method except for the work on the part of the designer. With UOD we could just cite the let-through chart for the fuse in most cases. With series rating we have to do fault calcs and have to call out 110.22 labels.

You might want to go to page 81 in the SPD and read the section on circuit breaker protection.

2005 240.86(B) may not use the term "listed", but here's the commentary from the handbook:

"Section 240.86(B) requires that, when a series rating is used, the switchboards, panelboards, and load centers be marked for use with the series rated combinations that may be used. Therefore, the enclosures must have a label affixed by the equipment manufacturer that provides the series rating of the combination(s). Because there is often not enough room in the equipment to show all the legitimate series rated combinations, UL 67, Standard for Panelboards, allows a bulletin to be referenced and supplied with the panelboard. These bulletins typically provide all the acceptable combinations. Note that the installer of a series rated system also has to provide the additional labeling on equipment enclosures required by 110.22, indicating that the equipment has been applied in a series-rated system."

 

[jim dungar]

Coulter,
Did you look at the sources referenced in the IAEI article?

Have you called Bussmann (or any other fuse manufacturer) and asked if they would "bless" using the UOD method to protect molded case circuit breakers.

And I agree the NEC does say tested not listed. But the installer of series rated equipment must still ensure that labels are applied per 110.22.

HMSPE,
When I started in the business in '78 the UOD method was already being discouraged and promoted as unsafe for devices with "dynamic impedance".

 

[coulter]

Martin -

Thank you for your explanation, that is the part I was unclear about - the connection between the idomatic phrase, "UOD", and a technical description of the process.

... My recollection of the history here is that many years ago circuit breakers were relatively slow to start opening... About 10 years ago either circuit breakers suddenly got faster, or someone realized that circuit breakers had gotten faster, ...

My recollection, from 35 years ago (might have only been 30 years), in using the mfg curves for current limiting fuses (I didn't hear the phrase "UOD"), is there are several factors involved:
1. The available short circuit current.

2. The impedance between the high SSC segment and the applicable panelboards.

3. Coordination between the CLF and the downstream CB.

For example, no one I knew would have put a CB with an instantaneous trip directly downstream from a CLF, where the system impedance was too low to allow the CB to open within its limits. No, you might not be able to use an inexpensive thermal-magnetic. One may have to spec a CB with a decent trip unit. Even thirty years ago we knew that one had to pay attention to the fuse clearing time and trip curve settings.

Some of the articles, like the IAEI quoted in a previous post, make it sound like no one ever looked at the required coordination to make an engineered series combination a reliable, safe system. Well, 35 years ago, I was still in the junior engineers' club and the guys and girls I was learning from sure knew about coordination and clearing time. I don't think the CBs are any faster, and I don't think the laws of physics changed. It mostly appears that the group knowledge base changed.

... In any case, the use of UOD with respect to down-stream breaker panels came to an abrupt stop.

This is one of the things I am trying to figure out. What changed? Did we as a group forget how to do coordination and equipment specification? Was there a regulatory change? I haven't worked on one of these since 1991, and it is likely I will never do another. If I did, I likely would select from the mfgs tested equipment bulletins. Why wouldn't one? The data is all there, no liability - if one can read. Just pick the combination from the cookbook.

...As has been previously posted, the theory for eliminating UOD for fuse/breaker combinations is that circuit breakers can start to open before the upstream fuse can clear. The resulting arc at the breaker contacts may limit the current to the point that the current limiting fuse may not see enough current to clear....

Yes, as I mentioned earlier, if one is sloppy enough to allow an instantaneous trip CB downstream of a CLF, where the system impedance is too low to limit the SSC to the cb interupt rating. No question, sloppy engineering is bad.

...You might want to go to page 81 in the SPD and read the section on circuit breaker protection ...

I did. I had. That was one of Jim's references. And I do tend to read the posts and the references.

In the 2002 SPD there's a section on "Component Protection: Current-Limitation starting on page 72. The method they show in the graph on page 72 is "UOD". You can also see this on-line in section 10 of the document at http://www.bussmann.com/apen/pubs/spd/index.asp
...

Yes, I knew about that. In my previous post, this is exactly the same section and link I listed. I guess you missed that.

 

[coulter]

Did you look at the sources referenced in the IAEI article? ...

One of them. The abstract on another appears to be MOS. The third I have to order. My subscription to IEEE doesn't allow instant access to papers. Company said they would pay for the change in subsctiption - so my fault for not changing it yet.

... Have you called Bussmann (or any other fuse manufacturer) and asked if they would "bless" using the UOD method to protect molded case circuit breakers.

Nope. Don't plan on it either. As I said, it is apparent you did, and I believe you completely.

...When I started in the business in '78 the UOD method was already being discouraged and promoted as unsafe for devices with "dynamic impedance".

Yes, and now that I have a definition for the idomatic phrase, "UOD", I see a mis-communication. Thank you Martin for taking time to read my post, and see that I needed a definition.

You always use the term "Molded Case CB". Well, they come in lots of flavors.

Small, molded case CBs have an inherent instantaneous trip. As I understand, the small thermal mass makes this hard to get around. By small I mean 250A frames or smaller.

So, putting CLFs ahead of a QO panel probably won't work. That is not news. I think I was hearing that when I was a member of the junior engineer's club - Dirt was old by then, and we had stopped using wooden electrons.

However, once the CBs get big enough to get decent adjustable trip units, then one can select a trip curve that will not interact with the clearing time of the fuses. But they are still molded case CBs.

The last one of these I did, had a 500kVA, 2.5%Z, 208V xfmr going in on a remodel. The available SSC was 56kA. We were ok until we got to the first NQOD panel. As I recall, the system impedance had the SSC down to 22.300kA. The panel was rated at 22kA. I'm sitting there looking for a calculation change that I can make that will lower the SSC a little. The Chief Engineer looks over my shoulder and says, "If the SSC is 22,001A, you don't have a defensible position." Ah-so. Lesson heard and learned.

Now, do you think that appropriately chosen CLFs, installed right next to the xfmr, was a poor engineering choice?

Would I do that today? Maybe, but then again, why not select one from the mfgs bulletin? Easy, convenient, cookbook. But I'm likely dumb enough that I would check the available SSC and system impedance anyway

 

[jim dungar]

Coulter,

All circuit breakers (even 2000A ones) with thermal-magnetic trip units have an instantaneous trip point and so should not be used with fuses in a series rating application unless they have been tested.

Molded Case Circuit Breakers is a term used by UL http://www.ul.com/regulators/circuit600.pdf

 

[coulter]

Mr Jim -

...All circuit breakers (even 2000A ones) with thermal-magnetic trip units have an instantaneous trip point ...

Yes, that would be absolutely true. The way I can tell that is you used the word "magnetic", which implies the trip unit has an instantaneous function that can not be turned off

...Molded Case Circuit Breakers is a term used by UL http://www.ul.com/regulators/circuit600.pdf

Interesting reference - but nothing that said a molded case CB had to have an instantaneous trip.

I only have a few minutes, so this is the only one I found. This is a molded case, available in 600A frame, and the literature says the instantanous function can be shut off.

http://ecatalog.squared.com/pubs/Ci...crologic Trip Units/Series B/48049-027-02.pdf

I'm not sure I see your point. If you are saying that a mis-applied molded case CB should not be used downstream of CLFs - I'd agreee with that.

 

[ron]

I think the point is, that you can't install a fully rated fuse or circuit breaker that is listed below the calculated fault current. The installation of a random current limiting fuse upstream does not reduce fault current for the underrated over current device your trying to save because of the dynamic impedance that is seen when the lower rated device begins opening before the higher rated device. This can happen instantaneously, even before you can begin seeing it on a time current coordination curve characteristic curve.
It must be a series rated tested combination if you wish to use one to protect the other in a fault condition.

 

[jim dungar]

Coulter,

In the reference you posted, there is a statement that even with the inst setting turned off this breaker will try to open in order to protect itself during a high fault condition.

Page 25
"In circuit breakers with both short-time and instantaneous trip, the adjustable instantaneous trip can be disabled by setting INSTANT. PICKUP to OFF. A high-level instantaneous override remains in effect."

In conclusion, in order to even attempt a successful engineered implementation of fuse curves/charts in a "series rating", the downstream ciruit breakers must be guaranteed not to have any dynamic impedance.

 

[ron]

Jim,
I'm not sure that the downstream device can be guaranteed to not have ANY dynamic impedance, but at least not enough to effect the operation of the tested combination with the upstream OCPD.
You point is correct regarding the Square D breaker as it still has an inst override that cannot be turned off.