SSDriver
Senior Member
- Location
- California
- Occupation
- Electrician
Maximum Circuit Length Help?
- Thread starter mbrooke
- Start date
- Status
SSDriver
Senior Member
- Location
- California
- Occupation
- Electrician
I don't know where they get the 400% from because I know I have seen circuits much longer (well over 1000') than on those tables that clear faults.
romex jockey
Senior Member
- Location
- Vermont
- Occupation
- electrician
one would think....., more mass, .....more Z.........~RJ~
Julius Right
Senior Member
- Occupation
- Electrical Engineer Power Station Physical Design Retired
The conduit impedance it is a bit more complicate. See for instance:
moam.info
Modeling and Testing of Steel EMT, IMC and RIGID (GRC) Conduit - M.MOAM.INFO
Electrical Metallic Tubing. IMC. Intermediate Metal Conduit. GRC. Galvanized Rigid Conduit. Arc Voltage, Varc. The volta...
moam.info
The conduit impedance it is a bit more complicate. See for instance:
Modeling and Testing of Steel EMT, IMC and RIGID (GRC) Conduit - M.MOAM.INFO
Electrical Metallic Tubing. IMC. Intermediate Metal Conduit. GRC. Galvanized Rigid Conduit. Arc Voltage, Varc. The volta...
Yes but simplified...higher iron content (ferrous) and physically thicker walls makes for higher reactance. At fault levels resistance is usually a small nuisance. Almost everything is driven by reactance. That’s why the ANSI short circuit method all but ignores resistance. Computer models just get it down to the third digit in accuracy.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
I think that specific table doesn't include the conduit impedance. Just phase and EGC. At least thats the current I get calculating it with just the phase and ground Zs.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Yes. Look at OPs chart. The maximum lengths are all very similar for #6 and smaller. Then it takes a nose dive as wire gauge grows. Confirmed by this chart.
Should be no surprise why it shows roughly a little over 200 feet for #6 or smaller, then quickly drops as wire size grows.
The ANSI method normally practiced works mostly on X, not R. Remember that these methods were developed when we used slide rules not spread sheets and cell phones. Doing a full complex math problem requires between 2 and 6 calculations not just one. ANSI uses some quick and dirty ways to estimate X/R and states when you can use the simplified method and when not. Most of the current limitation for short circuit happens closest to the transformer...VD becomes your “friend”, and the ANSI method is about short circuits and so is the conduit length limit. The various short cuts that ANSI takes though are easily “broken”. Within the “ANSI method” you can do the full calculation but what is commonly referred to only uses reactance in the calculations.
Doing calculations on small wiring is one way to “break” ANSI but the rules prevent you from using it because the X/R ratio is outside the range the ANSI simplified method allows. The table from the GEMI effort though does full impedance (reactance and resistance) including conduit effects and different current paths.
Since the lengths quoted for #6 and smaller are all similar I doubt anyone is concerned. But when it tells you 50 feet as a limit, that can make a long run from the MDP to a garage subpanel too long, never mind mast to pole.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Right, but at what does Z become to much Z? Or rather why?
ramsy
NoFixNoPay Electric
- Location
- LA basin, CA
- Occupation
- Service Electrician 2020 NEC
I am having visions of an electromagnet.
The more current flows, during a short circuit, the stronger the magnetic XR against the current flow.
Since Z includes XR reactance, the more ferris-metallic mass in the conduit, the more Z increases with current.
The more current flows, during a short circuit, the stronger the magnetic XR against the current flow.
Since Z includes XR reactance, the more ferris-metallic mass in the conduit, the more Z increases with current.
Last edited:
I=V/Z. Ohms law for AC here. If Z is in polar notation (with a phase angle) in terms of magnitudes it’s just I=V/Z. But we are discussing resistance and reactance separately. The magnitude of the complex impedance is:
Abs(Z)=sqrt(R^2+Z^2)...Pythagorean here.
So which is worse, R=1 and Z=100 or R=100 and Z=1? It doesn’t matter. Same current. The only thing that changes is if it’s mostly vars or mostly watts. The apparent power does not change.
But also we are discussing fault conditions. That’s when we use the impedance (not just resistance) of one phase going out but returning on the ground conductor. So now the ground path impedance matters. Resistance is quite low but reactance isn’t. So I have a higher specific resistance in my steel conductor which is good so R is lower than in a copper ground but now Z is much higher so the net effect is a higher impedance, even though my ohm (milliohm) meter says conduit is far superior.
So why this matters is to get the breaker to trip on a fault you need to be able to recognize the fault. If we have a 1 ohm fault impedance on a 480 V circuit the fault current is 277/1=277 A. If we had say a 10 A breaker with a 10x magnetic trip, it trips out instantaneously so it’s all good. Now if we have a 30 A load with the same 10x instantaneous trip, it won’t trip and we have to rely on thermal tripping, much slower. And 1 ohm is impedance, not just resistance. If reactance is high it has the same effect...well, their geometric mean does anyway.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Breakers do not care about watts or VAR current, only magnitude.
Alright. Can you explain why the 30 amp load will take longer to trip the circuit during a fault? Maybe I misread something.
ramsy
NoFixNoPay Electric
- Location
- LA basin, CA
- Occupation
- Service Electrician 2020 NEC
In that example, a 30A load with the same 10x instantaneous trip requires 300 Amps.
With 1 ohm fault impedance on a 480 V circuit, the fault current is too low, 277/1=277A wont engage the Instantaneous trip function.
With 1 ohm fault impedance on a 480 V circuit, the fault current is too low, 277/1=277A wont engage the Instantaneous trip function.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
But it will still engage the thermal portion of the of the breaker.
When Paulengr says 1 ohm, is he talking about peak or RMS?
The fact that breakers or more specifically thermal magnetic trips care about amps is my point...the geometric mean of resistance and reactance is what matters. Just because resistance is negligible doesn’t mean we can ignore reactance.
Ok so back to the example, common breakers typically have a 10x magnetic trip with UL class C. So with a 10 A breaker on a 480/277 V system with a 1 ohm ground fault magnitude the magnetic trip is 100 A. Our fault current is 277 A (277 V / 1 ohm). With the same conditions with a 30 A breaker the trip is 300 A. Now the magnetic trip shouldn’t go off (we are on the ragged edge) and the thermal trip element applies. A better example is probably a 50 A breaker though since we don’t have to talk about breaker accuracy.
This is the reason NEC mandates ground fault on larger circuits, and one of the major arguments for high resistance grounds. If all your loads are small and cable runs are relatively short, you can expect simple thermal magnetic breakers to do double...actually triple duty; catching phase to phase overloads, shorts, arcing faults, AND ground faults both arcing and dead shorts. But as the size (ampacity) increases and/or length increases the sensitivity of the instantaneous trip is lost. Trip times increase. That’s a problem especially in UL 489 based equipment for example that is not designed to withstand 30+ cycle trip times. UL 1077/ANSI designs with tolerance to longer fault times among other things becomes a necessity. Hence conduit length and type matter. Steel works better than aluminum which works better than PVC.
The problem when it comes to the reactance and resistance calculations is that steel conduit in particular is nonlinear in terms of performance. Hence GEMI was born, although other power system analysis software does have some kind of conduit calculation. But for those on a budget GEMI and the tables is free.
This idea has been around for decades but the guy who originally came up with it was a bit off in his calculations. And it’s not required by NEC directly so it is not well known.
romex jockey
Senior Member
- Location
- Vermont
- Occupation
- electrician
an apt reply Paul 
~RJ~
~RJ~
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
I know, both matter.
My apologies if I implied that one or the other didn't matter for larger sizes. Typically in smaller sizes reactance is ignored, however.
Is there any part of UL that specifically mandates a magnetic trip? I know UL489 doesn't require it IIRC.
Agree.
Agree, however breakers can still open on thermal trip. If circuits 30 amps and under trip within 0.4 seconds for 277 volt circuits and 0.8 seconds for 120 volts; 5 seconds for circuits over 30amps, building occupants are technically protected from electrocution.
Of course the incident energy goes up when you don't trip magnetically.
But either way- why not just increase the size of the circuit? Why rely on GFP?
Right. Though thinking about... isn't this what puts mag trip in breakers? In that instant tripping prevents faults above 5-10x the handle rating from exceeding 30 cycles?
Can you expand more into this? I didn't know UL1077 effected run length and conduit material.
I know, both matter.
Typically in smaller sizes reactance is ignored, however.
Is there any part of UL that specifically mandates a magnetic trip? I know UL489 doesn't require it IIRC.
Agree.
Agree, however breakers can still open on thermal trip. If circuits 30 amps and under trip within 0.4 seconds for 277 volt circuits and 0.8 seconds for 120 volts; 5 seconds for circuits over 30amps, building occupants are technically protected from electrocution.
Of course the incident energy goes up when you don't trip magnetically.
But either way- why not just increase the size of the circuit? Why rely on GFP?
Right. Though thinking about... isn't this what puts mag trip in breakers? In that instant tripping prevents faults above 5-10x the handle rating from exceeding 30 cycles?
Can you expand more into this? I didn't know UL1077 effected run length and conduit material.
I know, both matter.
Typically in smaller sizes reactance is ignored, however.
Is there any part of UL that specifically mandates a magnetic trip? I know UL489 doesn't require it IIRC.
Agree.
Agree, however breakers can still open on thermal trip. If circuits 30 amps and under trip within 0.4 seconds for 277 volt circuits and 0.8 seconds for 120 volts; 5 seconds for circuits over 30amps, building occupants are technically protected from electrocution.
Of course the incident energy goes up when you don't trip magnetically.
But either way- why not just increase the size of the circuit? Why rely on GFP?
Right. Though thinking about... isn't this what puts mag trip in breakers? In that instant tripping prevents faults above 5-10x the handle rating from exceeding 30 cycles?
Can you expand more into this? I didn't know UL1077 effected run length and conduit material.
Short circuit ratings for panel boards and switch boards are different. Panel boards don’t have 30 cycle ratings. So even if you can thermally trip in 30 cycles (0.5 seconds) it’s way too long for a panelboard and series ratings become critical. At the distribution level from a pure coordination point of view we need longer trip times anyway and arc flash concerns are the driver for faster trip times. Tripping in 20-30 cycles is a lot of incident energy.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
If tripping in 30 cycles you're way below the magnetic pickup and in turn your fault current is low (few hundred amps).
Series ratings are to increase the AIC rating.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
In my calcs I ignore the conduit.
- Status