SCCR rating - engineering supervision 1

[eeprom]

Hello,
I'm trying to get a solid understanding of NEC article 409.22 stating that an industrial control panel must be labeled with the SCCR rating. I've been reading through UL508A SB4 which (sort of) explains how to determine the SCCR rating of a panel.

In using their table for calculations, you could never make a panel with an SCCR of greater than 10kA, and 5kA is more likely. There are very few industrial applications with fault currents this low, and so this calculation becomes impractical.

So what I'd like to know is where in the code does is say I can use engineering supervision to make this calculation? I know I can get around this SCCR deal with a good fuse, but I'd like to know that I can refer to the code when an inspector says the SCCR is too high.

EE

 

[jraef]

...you could never make a panel with an SCCR of greater than 10kA, and 5kA is more likely.

Therein lies the flaw in your assertion. How did you come up with that conclusion? I've rarely had a problem with it.

If I can speculate, you most likely have not discussed this with quality suppliers who have been trained on what to offer. Most of the major mfrs or components have tested their power devices in series with each other so that they have a tested and listed combination SCCR that usually far exceeds the lowest level of one device in the series. This is really what the rule was set out to accomplish; avoiding the willy-nilly assembly of parts that were never tested to go together.

So for example let's say you want to make a motor starter circuit. The breaker is rated 65kA, the contactor has a 25kA SCCR, but the overload relay is only 5kA SCCR. At first glance that would appear that the highest SCCR you can apply on the circuit then is 5kA. But if you use components that are all from the same mfr, and that mfr has a table of tested series components that list them together, the series may in fact ALREADY have an SCCR of 65kA. What you can no longer do is buy a junk little Elmwood DP contactor from FleaBay and pair it with an overload relay from AutomationDestruct and a circuit breaker from Breakers-R-Us unless you can live with that 5kA SCCR.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[DRWeig]

Good on jraef.

eeprom, since you mentioned 508A SB4, read it more carefully. There's a table you are allowed to use under the UL rules that defines the peak instantaneous let-through of current-limiting circuit breakers. Items downstream of these breakers must only meet the let-through in the table.

So, choose your main breaker in the panel to coordinate with the ratings of all your downstream devices. The main breaker can be located outside the panel as a feeder breaker or branch circuit breaker, but must be supplied with the panel to comply with UL 508A). Our shop always includes the breaker within the panel.

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[djs]

Also do not forget current limiting fuses! UL loves fuses. In fact I have not found one independently rated current limiting CB. They are always listed in combinations with outher breakers. Not so for current limiting fuses!
For example type CC fuses will have a SCCR of 100KA if all downstream devices have a SCCR of 5 KA or higher. No restriction on types or manufacturers.

 

[eeprom]

djs,
Yes, it seems that fuses are the way to go. And it does not add much cost either. It's really an easy workaround.

 

[MAGTiger]

Take caution using current limiting fuses. Ul does not recognize the current limiting capability if the fuse is located in a branch.

Neil

 

[eeprom]

MAGTiger
I had the impression that (for example)... branch consisting of a breaker, overload and contactor; the whole branch has an SCCR of 14kA.

Now I protect the branch with a class CC fuse with a peak let through current of 4kA at a known fault current of 55kA.

Because the let though current is less than the SCCR of the branch, the branch rating can now be promoted to the IR of the fuse.

Do you believe this is incorrect? If so, please let me know where in the documentation you found this.

thanks,
EE

 

[DRWeig]

eeprom, if in your example you are eliminating the breaker and changing to the fuse, the fuse becomes the branch circuit protection and I believe you're correct.

If, however, you are connecting several branch circuits to the downstream side of your fuse (e.g., 100A fuse feeding three 50A breakers), you can't up-rate the SCCR of the entire device above the 14kA limit of your breakers.

Read SB4.3.3 very carefully, especially the last two lines of part SB4.3.3.a (it's confusing) and break it into individual questions.

Neil, can you elaborate a little more on what you're citing?

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[DRWeig]

Correction: I meant to point out the last FIVE lines of SB4.3.3.a, not just the last two.


Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[LionelHutz]

Last time I looked, you don not eliminate the breaker or fuse when protecting it fuses to increase the short circuit rating. You end up with a series fuse-fuse or fuse-breaker combination.

 

[DRWeig]

I think you may be right, Lionel. The breaker can then be considered supplementary protection as long as it's not required for branch circuit protection of the stuff downstream of it.

In the case of one feeder fuse protecting several smaller breakers, though, the breakers have to be considered branch circuit protection and their SCCR reflects back to the load side of the fuse. It's a confusing statement in SB4, but it eliminates the series rating (unless the breaker is listed for series application with a particular fuse in front of it).

We've actually had the same inspector call it both ways for us. Whenever it comes up (which isn't often), we try to get the inspector on the e-mail before laying the panel out and get him to agree to a quick one-line sketch.

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[eeprom]

Using a certain class of current limiting fuse (CC, J, G, T, etc) allows you to "modify" the rating of the panel. In order to do this, you need the available fault current at the panel and you need the peak let through current for the fuse you are using at that particular fault current. If the let through value is lower than the SCCR ratings of the devices downstream of the fuse, then you can use the IR rating of the fuse. This is a bargain.

The breaker won't matter unless it is upstream of the fuse. It is also true that you end up with a breaker-fuse series combination, but for the sake of increasing your panel's SCCR you will get more improvement using the fuse IR rating.

 

[DRWeig]

As this is a good professional discussion, I don't think UL would have a problem with me quoting from the standard.

eeprom, you answer is basically correct, unless you're actually going for a UL 508A listing. Check what I've bolded below. If a breaker downstream of the current-limiting fuse qualifies as branch-circuit protection, the current-limiting feature of the fuse doesn't count. The example would be if it's a 100A fuse ahead of a 50A breaker. If the circuit downstream of the breaker needs 50A protection, the breaker is a branch circuit protector and the SCCR upstream of the fuse is equal to the SCCR of the breaker. However, if the circuit downstream of the breaker is also protected by the fuse (a 100A circuit in my example), then the breaker is supplemental protection and you get to claim the current limiting feature. If the breaker is not branch circuit protection, the SCCR upstream of the fuse is equal to the fuse's SCCR.

The work-around, when forced to employ it by UL, was to simply add a fuse set for each breaker.

Note: I'm very willing to be convinced otherwise! It's a very inconvenient rule that doesn't make sense to me.

From UL 508A rev Feb 2010

SB4.3.3 For branch circuits supplied by a Class CC, G, J, L, RK1, RK5, or T fuse in the feeder circuit, the short circuit current rating on the line side of the fuse shall be one of the following:

a) The short circuit current rating of the feeder fuse when all of the components in the branch circuit have a short circuit current rating not less than the peak let-through current
corresponding to the specific fuse class employed from Table SB4.2 based on the current rating of the fuse and the available short circuit current on the line side of the fuse, and the short circuit current rating of all branch circuit protective devices on the load side are not less than the short circuit current rating of the feeder fuse;

b) The smallest short circuit current rating of any branch circuit protective device on the load side of the feeder fuse, when the conditions of SB4.3.3(a) exist except the short circuit current rating of the branch circuit protective devices on the load side are less than the short circuit current rating of the feeder fuse.

c) The smallest short circuit current rating of any branch circuit on the load side of the feeder fuse, when the conditions of SB4.3.3(a) or SB4.3.3(b) are not met.

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[LionelHutz]

I don't agree with that assessment. The standard simply says the SCCR of the panel is never higher than the interrupt rating of the lowest rated interrupting device in the panel.

Here is an example of what the standard is saying.

Fuse IR = 100kA & Ip = 4kA -> breaker IR = 22kA -> contactor WR = 5kA

The panel is 5kA before installing the fuse due to the 5kA withstand rating of the contactor. The panel becomes 22kA after installing the fuse. The fuses modify the withstand rating of contactor to be 100kA. However, the breaker interrupt rating is 22kA so the panel must be 22kA.

If you want 100kA, then the breaker used must be rated for 100kA.

 

[DRWeig]

That's what I was trying to describe, Lionel -- but you said it much more plainly.

Thanks!

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[gvasiliou]

Hi guys, interesting topic.
Although i would like to ask something about LionelHutz example.

The panel becomes 22kA after installing the fuse

I suppose you mean "after installing the breaker".

Has the breaker to be a "current-limiting" breaker, being able to effectively limit SC Current...?

If breaker is not current-limiting and 22 kA reach the 5kA contactor (till the breaker to interupt the circuit) it seems we have a problem for the low rated contactor.

 

[fab1961]

eeprom,

this is a very interesting question I have been asking myself more than a time - thank you. I have a question though (I ' m not sure I'm going off thread): why do you say 10 kA is a small value? considering feeding cables length and transformer size in small shops this could be reasonable. What do you think?

 

[LionelHutz]

However, if the circuit downstream of the breaker is also protected by the fuse (a 100A circuit in my example), then the breaker is supplemental protection and you get to claim the current limiting feature. If the breaker is not branch circuit protection, the SCCR upstream of the fuse is equal to the fuse's SCCR.

Just to be clear, this is the part I was not agreeing with. You can never claim the SSCR of the fuse if it is higher than the breaker. You always have to use the SCCR or IR of lowest rated protection device in the panel.

gvasiliou - no, the example is correct. The fuses are installed upstream of the breaker to increase the short circuit rating of the breaker-contactor branch circuit. The breaker does not need to be current limiting. 22kA will not reach the contactor due to the current limiting of the fuses.

Besides, the contactor withstand rating of 5kA does not mean it will withstand 5kA. You must understand how the testing was done to know what it means. The testing for the contactor is done with a cheap run of the mill thermal magnetic breaker. The incoming test leads are shorted and the test current of 5kA is set. Then, the test leads are connected to the breaker-contactor circuit and the test is performed. The fault current that actually flows during the test will not be 5kA due to the extra impedance added by the breaker and contactor. This means the 5kA rating refers to the bolted fault current at the incoming of the panel, not to the individual devices used in the panel. Lots of people get this wrong and believe a 5kA rated contactor means it will withstand 5kA of current.

The above is the reason you can not just use the let-through current of a breaker or fuse and install it on other devices that meet or exceed that let-through current. From my previous example, you can not argue that the fuse lets through 4kA and the contactor is 5kA so they are co-ordinated without the breaker. The contactor is only 5kA rated when it is installed with the breaker. So, you have to co-ordinate the fuses with the complete breaker-contactor circuit.

 

[DRWeig]

Lionel, thanks again -- I happen to agree with you. One of our UL inspectors disagreed, though.

Fuse IR = 100kA & Ip = 4kA -> breaker IR = 22kA -> contactor WR = 5kA

His opinion was that if the fuse and breaker are both 100A rated, the breaker is in fact unnecessary. Thus, the breaker does not have to be classified as a branch circuit device (call it an expensive switch instead) and its 22kA SCCR does not reflect back to lower the SCCR of the panel upstream of the fuse.

The way the standard is worded is poor. We never escalated it to the UL engineering folks, and in fact had another inspector call it the other way. Ah, well. We made our internal standard to fit the way you described.

Good on ya,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[MAGTiger]

EEPROM,
I can't speak to breaker use as we don;t use them in this capacity. We build 508A panels all the time and we have a letter written by a UL Engineer that clarifys our requirement. We use 2 current limiting fuses in series, per phase. Both fuses are class T. The source side fuse is considered feeder protection and thus has current limiting capability. The load side fuse, although identical can not be considered as a current limiting fuse. An expensive way to build panels but there it is.
Shoulda bought stock in Cooper before this NEC change took effect.

Neil