CABLE DAMAGE CURVE AND NEC 240.4 (B)

[PE (always learning)]

Hey everyone,

I have a question concerning the ampacity of a conductor not corresponding with the standard ampere rating of the circuit breaker protecting it. The project in question has an 80A transfer switch being fed by (4)#6 THWN and the breakers supplying the transfer switch on both emergency and normal are rated for 70 amps. These breakers are 70A FXD6 Siemens breakers and when they are plotted on SKM with the #6 cable it shows the cable damage curve crossing into the breaker curve. I am considering telling them to beef up the cable size to a #4, but technically the #6 would be acceptable according to NEC 240.4 (B) which allows you to round up when the ampacity of the conductor does not correspond with the standard ampere rating of a circuit breaker. A #6 THWN is good for 65A which is technically allowed to round up to 70A. Should I still tell them to beef up the cable because the breaker curve and cable damage curve are crossing? This would normally not be an issue, but the cable has already been put in place and I've been asked to evaluate the project. Pictures attached for reference.

 

[suemarkp]

Is it really THWN, or THWN-2/THHN? If so, the conductor is 90C and rated at 70A. But the terminations are not rated at 90C. Do you have a load calc? How close to 65A are they? Since there is an emergency circuit here I would think you have electricians who know what they are doing and would evaluate the circuit when adding any new load. Our panel schedules have conductor sizes written on them because you can't always read the wire to see what it is. With decent documentation, and perhaps even a label on it indicating "65A Conductor Amapcity Limit" should be sufficient if you have a load calc and fixed loads.

 

[mbrooke]

Hey everyone,

I have a question concerning the ampacity of a conductor not corresponding with the standard ampere rating of the circuit breaker protecting it. The project in question has an 80A transfer switch being fed by (4)#6 THWN and the breakers supplying the transfer switch on both emergency and normal are rated for 70 amps. These breakers are 70A FXD6 Siemens breakers and when they are plotted on SKM with the #6 cable it shows the cable damage curve crossing into the breaker curve. I am considering telling them to beef up the cable size to a #4, but technically the #6 would be acceptable according to NEC 240.4 (B) which allows you to round up when the ampacity of the conductor does not correspond with the standard ampere rating of a circuit breaker. A #6 THWN is good for 65A which is technically allowed to round up to 70A. Should I still tell them to beef up the cable because the breaker curve and cable damage curve are crossing? This would normally not be an issue, but the cable has already been put in place and I've been asked to evaluate the project. Pictures attached for reference.

What is the max fault current involved?

 

[jim dungar]

If it is a breaker listed per UL498, like the FXD is, then it has been tested as protection for conductors properly sized and applied per the NEC. For all intents the cable damage curve is irrelevant.

 

[mbrooke]

If it is a breaker listed per UL498, like the FXD is, then it has been tested as protection for conductors properly sized and applied per the NEC. For all intents the cable damage curve is irrelevant.

Tested but not for all scenarios likely to be found in the real world.

 

[kwired]

Is it really THWN, or THWN-2/THHN? If so, the conductor is 90C and rated at 70A. But the terminations are not rated at 90C. Do you have a load calc? How close to 65A are they? Since there is an emergency circuit here I would think you have electricians who know what they are doing and would evaluate the circuit when adding any new load. Our panel schedules have conductor sizes written on them because you can't always read the wire to see what it is. With decent documentation, and perhaps even a label on it indicating "65A Conductor Amapcity Limit" should be sufficient if you have a load calc and fixed loads.

What he said. what is the actual load?

80 amp transfer switch is sufficient to handle a circuit with 70 amps overcurrent protection.

70 amps overcurrent is sufficient on a conductor rated 65 amps (unless this is a situation where 60 C ampacity would apply), but the load in such case must still be 65 or less as that is the maximum ampacity of the conductor.

Current during a short circuit or ground fault is going to be much higher, but for very little duration compared to overload protection trip curve.

Also consider that you would only need an 8 AWG for equipment grounding conductor

 

[jim dungar]

Tested but not for all scenarios likely to be found in the real world.

What proof do you have?
How many instances are you aware of where conductor applied per the NEC where not protected by a UL listed breaker?
What real world scenarios are likely to be an issue?

Proving that breakers protect actual conductors is one of the reason the UL testing includes 4' of conductor.

 

[ramsy]

..These breakers are 70A FXD6 Siemens breakers and when they are plotted on SKM with the #6 cable it shows the cable damage curve crossing into the breaker curve..

As long as NFPA-70 idiot tables remain the adopted standard to determine overloads or overheated wire, engineers must convert superior formulas & models into the relevant code violations; in section 220 load calcs, section 310 table ampacities/limits/assumptions, with cross references to adjustments for ambient temperatures, wiring method, CCC's, harmonic heating, reactive ferrous-wiring methods, and any inductive-load adjustments for motors, etc..

No inspectors, electricians, or planners can take proper corrective action without reference to the idiot-table or code violation.

Voltage drop that increases current & heat with inductive loads, is among the rare exceptions where NFPA-70 must punt to superior engineering models like SKM, to prove significant error in field methods.

 

[mbrooke]

What proof do you have?
How many instances are you aware of where conductor applied per the NEC where not protected by a UL listed breaker?
What real world scenarios are likely to be an issue?

Proving that breakers protect actual conductors is one of the reason the UL testing includes 4' of conductor.

The equation in Table 240.92 (B).

Above a certain short circuit value the conductor will be damaged by the time the breaker has made the decision to trip, unlatch and break the current flow.

Here is an example, at just 15,000 amps:

There is also concern whether or not the ATS is 3 cycle or 30 cycle rated.

 

[Fred B]

The equation in Table 240.92 (B).

Above a certain short circuit value the conductor will be damaged by the time the breaker has made the decision to trip, unlatch and break the current flow.

Here is an example, at just 15,000 amps:

There is also concern whether or not the ATS is 3 cycle or 30 cycle rated.

One word: WOW

 

[paulengr]

The equation in Table 240.92 (B).

Above a certain short circuit value the conductor will be damaged by the time the breaker has made the decision to trip, unlatch and break the current flow.

Here is an example, at just 15,000 amps:

There is also concern whether or not the ATS is 3 cycle or 30 cycle rated.

I think you are confusing things a bit.

The normal ampacity rules in article 310 apply under normal operating conditions and even overloads where there is sufficient time for a cable to dissipate the thermal load. Standard Neher-McGrath approach. The formula is given under engineering supervision but about the only time it applies is buried cable where the given conditions just aren’t enough for every situation. Neher-McGrath makes good money supplementing it.

Under short circuit conditions this can’t happen. Hence the formula in 240.92(B). This is all per ICEA P-32-382-2007 which does a much better job of explaining it. Unfortunately 240.92(B) isn’t quite as forward about this as it should be but it easily does the job.

The big issue that it sounds like you are alluding to is the “back feed” issue that occurs with lots of inductive loads. Most electricians are just using the infinite bus assumption and calculating short circuit with that alone, or maybe with something similar to the Bussmann point to point method. This misses inductive loads but honestly they only raise short circuit by about 10% in some cases. The IEEE Color books series goes far more into detail and demonstrates that they are ALL guesses and assumptions. That is why for instance SKM uses the impedance method instead of the simplified X/R method for its arc flash module, but even that is a guess.

If you truly want to look at short circuit conditions EMTP models are the way to go. sKM is sparse matrix nonlinear solver nonsense that gets there with a lot of assumptions too, not actual time domain analysis. It is more refined but you can do pretty good with just SKM.

 

[mbrooke]

I think you are confusing things a bit.

The normal ampacity rules in article 310 apply under normal operating conditions and even overloads where there is sufficient time for a cable to dissipate the thermal load. Standard Neher-McGrath approach. The formula is given under engineering supervision but about the only time it applies is buried cable where the given conditions just aren’t enough for every situation. Neher-McGrath makes good money supplementing it.

Under short circuit conditions this can’t happen. Hence the formula in 240.92(B). This is all per ICEA P-32-382-2007 which does a much better job of explaining it. Unfortunately 240.92(B) isn’t quite as forward about this as it should be but it easily does the job.

The big issue that it sounds like you are alluding to is the “back feed” issue that occurs with lots of inductive loads. Most electricians are just using the infinite bus assumption and calculating short circuit with that alone, or maybe with something similar to the Bussmann point to point method. This misses inductive loads but honestly they only raise short circuit by about 10% in some cases. The IEEE Color books series goes far more into detail and demonstrates that they are ALL guesses and assumptions. That is why for instance SKM uses the impedance method instead of the simplified X/R method for its arc flash module, but even that is a guess.

If you truly want to look at short circuit conditions EMTP models are the way to go. sKM is sparse matrix nonlinear solver nonsense that gets there with a lot of assumptions too, not actual time domain analysis. It is more refined but you can do pretty good with just SKM.

Back feed or no back feed 100,000 amps of fault current can damage smaller cables before the breaker opens.

There is a curve, high fault currents lower final temps of the conductor, but beyond a certain point you see a sharp rise once you hit the maximum clearing time.

Can you give more info as to how time domain plays into this?

 

[mbrooke]

I think you are confusing things a bit.

The normal ampacity rules in article 310 apply under normal operating conditions and even overloads where there is sufficient time for a cable to dissipate the thermal load. Standard Neher-McGrath approach. The formula is given under engineering supervision but about the only time it applies is buried cable where the given conditions just aren’t enough for every situation. Neher-McGrath makes good money supplementing it.

Under short circuit conditions this can’t happen. Hence the formula in 240.92(B). This is all per ICEA P-32-382-2007 which does a much better job of explaining it. Unfortunately 240.92(B) isn’t quite as forward about this as it should be but it easily does the job.

The big issue that it sounds like you are alluding to is the “back feed” issue that occurs with lots of inductive loads. Most electricians are just using the infinite bus assumption and calculating short circuit with that alone, or maybe with something similar to the Bussmann point to point method. This misses inductive loads but honestly they only raise short circuit by about 10% in some cases. The IEEE Color books series goes far more into detail and demonstrates that they are ALL guesses and assumptions. That is why for instance SKM uses the impedance method instead of the simplified X/R method for its arc flash module, but even that is a guess.

If you truly want to look at short circuit conditions EMTP models are the way to go. sKM is sparse matrix nonlinear solver nonsense that gets there with a lot of assumptions too, not actual time domain analysis. It is more refined but you can do pretty good with just SKM.

Here is an example curve:




The higher the fault current, the cooler the conductor will be once the breaker clears. Assuming a 20 amp breaker, the wire will be the coolest once you just enter the magnetic trip region around 100 amps for this graph. From that point on, the final temp will actually start to increase. Beyond 5-10,000 amps you could damage the insulation. Going further (ie 25,000 amps) the copper will simply melt.

This is due to the fact that once you hit the magnetic pickup threshold (typically 10x the handle rating), the breaker will not open any faster.

 

[paulengr]

Back feed or no back feed 100,000 amps of fault current can damage smaller cables before the breaker opens.

There is a curve, high fault currents lower final temps of the conductor, but beyond a certain point you see a sharp rise once you hit the maximum clearing time.

Can you give more info as to how time domain plays into this?

Over very short time intervals, SKM, ETAP, etc., model asymmetrical fault current using a simplified approach that usually works, but it relies on assumptions that don’t always apply. SKM does better than ETAP in that at least it gives you a transient coefficient but that’s it. One way to truly mess things up as an example is introduce capacitors and transients which are common in medium and higher voltages during switching. Combine that with your cable issue and SKM is completely outmatched. This is especially important in HV and EHV where transients are a major switching problem and where opening times are often measured in seconds, not cycles.

 

[jim dungar]

The equation in Table 240.92 (B).

Breakers have been protecting conductors for more than 80 years, even if the fuse manufacturers want to claim something differently, the UL489 test procedure confirms this.

Please provide examples of a properly applied NEC conductor that was not protected by a properly applied breaker.

The formula in 240.92 is for specific applications in supervised industrial locations and applies to feeder tap conductors where the conductor is sized smaller than what would have been used in the UL test.

 

[mbrooke]

Breakers have been protecting conductors for more than 80 years, even if the fuse manufacturers want to claim something differently, the UL489 test procedure confirms this.

Please provide examples of a properly applied NEC conductor that was not protected by a properly applied breaker.

The formula in 240.92 is for specific applications in supervised industrial locations and applies to feeder tap conductors where the conductor is sized smaller than what would have been used in the UL test.

"It has been done that way X years" doesn't mean anything as history has shown tradition wasn't always right about everything.

UL actually tests the smallest conductor allowed by code under the full AIC rating of a breaker with voltage likely to be sustained by such a source in the real world?

The formula is granted for specific applications under the NEC, however the physics hold true for all conductors.

 

[mbrooke]

Over very short time intervals, SKM, ETAP, etc., model asymmetrical fault current using a simplified approach that usually works, but it relies on assumptions that don’t always apply. SKM does better than ETAP in that at least it gives you a transient coefficient but that’s it. One way to truly mess things up as an example is introduce capacitors and transients which are common in medium and higher voltages during switching. Combine that with your cable issue and SKM is completely outmatched. This is especially important in HV and EHV where transients are a major switching problem and where opening times are often measured in seconds, not cycles.

Thanks

 

[jim dungar]

"It has been done that way X years" doesn't mean anything as history has shown tradition wasn't always right about everything.

But it doesn't make it wrong either.

You have yet to show a history of actual failures with properly sized and applied breakers and conductors.

The formula should only come into play when trying to protect conductors outside of the general article 240 applications.

As far as I am concerned, there is no advantage in continuing this discussion with you.

 

[mbrooke]

But it doesn't make it wrong either.

You have yet to show a history of actual failures with properly sized and applied breakers and conductors.

The formula should only come into play when trying to protect conductors outside of the general article 240 applications.

As far as I am concerned, there is no advantage in continuing this discussion with you.

Right, because you deny the existence of ohms law and I2R. I can't change that.

 

[jim dungar]

Right, because you deny the existence of ohms law and I2R. I can't change that.

I don't respect your DIY engineering enough to consider changing my mind, which is why I have asked for real examples.

I have over 40 years of experience as a licensed engineer, so I am comfortable with my ability to apply engineering fundamentals.