Conductor Withstand Rating

[baumancl]

Recently I was reading through the Cooper-Bussman “SPD – Selecting Protective Devices Based on the 2005 NEC”. I came upon a couple sections which through me for a loop regarding the withstand rating of conductors and the need to upsize both phase and ground conductors in order for them to carry the full rating of a possible worse case fault scenario.

It does not point out where in the NEC code that it is a requirement to do this for phase conductors but it does refer to 250.4(A)(5) and 250.4(B) (4) for grounding conductors which states that the grounding conductor be sized adequately for the short-circuit current that could be let-through. If it is stated there then why wouldn’t it be the same for the phase conductors? This whole publication does push the use of Bussman Low-Peak fuses to reduce available fault current so it does have some biases but has anyone ever applied this method of upsizing conductors based on the withstand rating. I asked one of the principals at the Electrical Engineering Consulting Firm I work at and he told me he has never applied this to any designs.

Can anyone shed some light on this or at least make it clear to me if upsizing phase and ground conductors is really spelt out anywhere in the NEC code.

Please refer to the article pages 4 & 5 of 27 (74 & 75 Page numbers) at the following link.

http://www.bussmann.com/library/docs/spd02/SPDSection10.pdf

 

[bphgravity]

Review sections 110.9 and 110.10 of the NEC.

 

[baumancl]

I am aware of these two articles which in all honesty are pretty vague as are many other articles in the code. I wanted to see if any design/consulting engineers actually include this in their design?

 

[jim dungar]

Fuses cannot reduce the amount of available fault current. Some fuses can reduce the amount of let-through fault current, but only when they operate in the current limiting region of their time-current-curve.

UL tests both circuit breakers and fused switches with an amount of appropriately sized conductor as an assembly (conductor, lugs, OCPD), therefore it is not possible to "under protect" when the circuit is within NEC parameters.

 

[don_resqcapt19]

Fuses cannot reduce the amount of available fault current. Some fuses can reduce the amount of let-through fault current, but only when they operate in the current limiting region of their time-current-curve.

UL tests both circuit breakers and fused switches with an amount of appropriately sized conductor as an assembly (conductor, lugs, OCPD), therefore it is not possible to "under protect" when the circuit is within NEC parameters.

Jim,
How does that work? The 1/2 cycle damage current for all conductors 1/0 and smaller is less than 65,000 amps. If the OCPD is not current limiting and you have a supply that can support a 65,000 amps short circuit current, would you not have to increase the size of the conductor to comply with 110.10?

 

[jim dungar]

Jim,
How does that work? The 1/2 cycle damage current for all conductors 1/0 and smaller is less than 65,000 amps. If the OCPD is not current limiting and you have a supply that can support a 65,000 amps short circuit current, would you not have to increase the size of the conductor to comply with 110.10?

The fact is every overcurrent device limits current while it is operating (arcing). The inherent dynamic impedance of protective devices is why we need to concern ourselves with series ratings being tested combinations.

A device rated for 65KAIC may actually begin to open when the current reaches 30kA and impart enough dynamic impedance that the current never actually reaches the full 65KA that is available. But, it can only be called "current limiting" if it meets specific clearing time criteria.

110.10 does not require protective devices to be "'current limiting". 110.10 says you should consider all of the circuit impedances when determing that equipment is able to withstand the available SCA. The UL test of enclosed protective devices includes cable, therefore listed devices have proved by test that they do protect conductors applied per the 'NEC'.

 

[don_resqcapt19]

The fact is every overcurrent device limits current while it is operating (arcing). The inherent dynamic impedance of protective devices is why we need to concern ourselves with series ratings being tested combinations.

I understand that.

A device rated for 65KAIC may actually begin to open when the current reaches 30kA and impart enough dynamic impedance that the current never actually reaches the full 65KA that is available. But, it can only be called "current limiting" if it meets specific clearing time criteria.

My understanding of a current limiting device is that it must act within 1/2 cycle. Currents of 30,000 amps or more for 1/2 cycle will damage conductors #4 and smaller. Currents of 30,000 amps or more for one cycle will damage conductors #2 and smaller.

110.10 does not require protective devices to be "'current limiting". 110.10 says you should consider all of the circuit impedances when determing that equipment is able to withstand the available SCA. The UL test of enclosed protective devices includes cable, therefore listed devices have proved by test that they do protect conductors applied per the 'NEC'.

I still don't understand. It still appears to me that you can have a combination of available fault current, OCPD, and conductor size that will result in conductor damage in an installation that is compliant with the NEC with the exception of the rule in 110.10 that would require an increase in conductor size to protect the conductor.

It is because a conductor that is sized to the ampacity of the OCPD has enough impedance to limit the fault current to a level that is below the damage curve for that conductor?

 

[jim dungar]

My understanding of a current limiting device is that it must act within 1/2 cycle. Currents of 30,000 amps or more for 1/2 cycle will damage conductors #4 and smaller. Currents of 30,000 amps or more for one cycle will damage conductors #2 and smaller.

The ICEA cable damage curve does not say the Ip must always remain below 30kA nor that the current be at zero in .XXXsec.

A device that limits current does not need to be a "current limiting device". A 65kAIC rated device that fails in the amount of peak current it lets through or the time it takes to clear the fault may still easily meet the cable protection requirements of limiting the current to 30kA for .008s or .016s, and in fact UL489 testing says it does.

It is because a conductor that is sized to the ampacity of the OCPD has enough impedance to limit the fault current to a level that is below the damage curve for that conductor?

It is possible that the current limitation is coming from the cable, but it is most likely due to the internal and dynamic impedance of the protective device.

 

[mayanees]

good question!

good question!

Can anyone shed some light on this or at least make it clear to me if upsizing phase and ground conductors (to handle fault current) is really spelt out anywhere in the NEC code.


http://www.bussmann.com/library/docs/spd02/SPDSection10.pdf

If you're using a Power System Analysis software package, you can plot out a cable damage curve for any given feeder. In SKM, the curve is developed from the ICEA standard.
SKM automatically plots out the damage curve for the phase conductors, which is shown to clear well within NEC overcurrent protection limits.
If you then plot out the damage curve for the associated equipment grounding conductor you'll see that the 400A phase device trips within time to avoid the damage curve for the ground conductor.
I just looked at a quick example of 400 amp phase protection against a #3 copper conductor damage curve, and the 400A CB trips well before the #3 damage curve.
John M

 

[baumancl]

I think I am getting the big picture but please clarify if I am incorrect:

Particular conductor sizes are U.L. tested and listed to be adequately protected when used with overcurrent protection devices sized per the NEC becuase they in fact do limit the fault let through current to within its withstand rating? It appears that SKM Analysis Software confirms this as mayanees has mentioned.

I know Bussmann is pushing for using fuses instead of breakers but why would they publish to oversize the conductors as in the following article link?

pages 4 & 5 of 27 (74 & 75 Page numbers)
http://www.bussmann.com/library/docs/spd02/SPDSection10.pdf

 

[rcwilson]

On 4.16kV feeders we have to upsize the cables when the client specifies breakers instead of fuses. A 250 HP motor (32 FLA) that could be fed with a #2 or #1/0 cable has to be fed with #4/0 awg minimum size to prevent cable damage during a fault. If the motor is fed by a fused contactor, the fuse clears the fault before the #2 cable damage curve is reached. With a breaker, the protective relay reaction time and the breaker clearing time are enough to damage the smaller cables during a phase-phase fault at the motor.

I have never found a need to upsize cables in low voltage systems (<600V) for short circuit withstand. The smaller cables are usually self limiting. Their impedance is enough to limit the short circuit at the end of the run to a value less than the damage curve. If the fault is not at the end of the run, the cable is obviously damaged already and needs to be replaced anyway. Short cables might be an exception.

For example, a #12 wire can handle about 4,000 amps when protected by a typical circuit breaker. If the circuit is fed from a 480V panel with 50 kA fault current available at the panel, it only takes 40 feet of #12 wire to knock the fault current down to 4,000 amps. If the circuit is only 10 feet long, uses fuses or a larger wire ( #1/0) or run around the room twice.

 

[mayanees]

I think I am getting the big picture but please clarify if I am incorrect:

Particular conductor sizes are U.L. tested and listed to be adequately protected when used with overcurrent protection devices sized per the NEC becuase they in fact do limit the fault let through current to within its withstand rating? It appears that SKM Analysis Software confirms this as mayanees has mentioned.

I know Bussmann is pushing for using fuses instead of breakers but why would they publish to oversize the conductors as in the following article link?

pages 4 & 5 of 27 (74 & 75 Page numbers)
http://www.bussmann.com/library/docs/spd02/SPDSection10.pdf

... to sell fuses.
It's certainly worth verifying that a phase device protects to an EGC's damage curve.
In the even that it didn't, the alternatives are to upsize the conductor so it has a higher withstand and fits under the breaker's protection curve, or improve the OCPD characteristic such that it protects to the EGC's damage curve... use their fuse.

John M

 

[baumancl]

Thanks for all the input it is much appreciated. If anyone else does designs for low voltage systems (<600V) and has more input on this topic please feel free to jump in.