mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Voltage Drop to Determine Wire Size on a 480/277 Line-to-Neutral Lighting Run?
- Thread starter nmonaco
- Start date
- Status
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
I don't believe time is much of a factor for magnetic trip function, basically the time to mechanically operate once trip operation commences is about the only time factor involved. Otherwise the condition to start magnetic trip is based on current level, you either reached magnetic trip level or else it will be tripping after delay on thermal trip operation.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
I don't think there is a simple calculation for this application.
It is a MWBC and yes balanced load would calculate out about like a three phase load would, but there is hardly any situaion of near balance in any individual segment of the layout.
Would be much simpler if you had three identical lights across all three phases at each light pole, then you could ignore the neutral and treat each pole as a 480 volt three phase load for the most part.
Instead you dropping current on one phase but also adding current to neutral at each pole, next pole you are doing the same but on another phase.
There may be shortcuts to get a reasonable in the field result, but if you want to know actual voltages at each point you about have to calculate each segment individually
It is a MWBC and yes balanced load would calculate out about like a three phase load would, but there is hardly any situaion of near balance in any individual segment of the layout.
Would be much simpler if you had three identical lights across all three phases at each light pole, then you could ignore the neutral and treat each pole as a 480 volt three phase load for the most part.
Instead you dropping current on one phase but also adding current to neutral at each pole, next pole you are doing the same but on another phase.
There may be shortcuts to get a reasonable in the field result, but if you want to know actual voltages at each point you about have to calculate each segment individually
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
I don't think voltage drop would be over 3% if the run meets the requirement of 250.4 (A) 5.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
How do you get no more than 3% from that?
I see no specifications/limitations here, seems it would still be designer decision as to what is acceptable.
winnie
Senior Member
- Location
- Springfield, MA, USA
- Occupation
- Electric motor research
Quick background for the OP: mbrooke is raising the issue of an 'effective ground fault current path'. This is an unsettled topic.
He makes the not unreasonable claim that code requires that a ground fault trip a breaker at a certain speed, on the basis of NEC 250.4(a)5 in combination with other standards. In the installation being discussed, a short circuit at the far end would take many 10s of seconds to trip the breaker, and during this time all grounded metal at the far end of the circuit would be at elevated voltage and a shock hazard.
Others make the not unreasonable claim that this is an excessive interpretation. While a ground fault would cause a shock hazard, it would be transient in nature because if a circuit is designed with reasonable voltage drop the breaker will _eventually_ trip, and the vast increase in expense mitigating this risk would provide better safety benefit elsewhere. In general OCPD clearance time is not considered in these situations, so mbrooke has a bit of an uphill battle convincing the rest of us that it should be considered.
In the case of the installation being discussed, it would mean using roughly 6x the copper, solely to change the time it takes OCPD to clear a fault.
I'd suggest that the OP do a search for some of the other threads on this topic, rather than rehashing the discussion here.
-Jon
He makes the not unreasonable claim that code requires that a ground fault trip a breaker at a certain speed, on the basis of NEC 250.4(a)5 in combination with other standards. In the installation being discussed, a short circuit at the far end would take many 10s of seconds to trip the breaker, and during this time all grounded metal at the far end of the circuit would be at elevated voltage and a shock hazard.
Others make the not unreasonable claim that this is an excessive interpretation. While a ground fault would cause a shock hazard, it would be transient in nature because if a circuit is designed with reasonable voltage drop the breaker will _eventually_ trip, and the vast increase in expense mitigating this risk would provide better safety benefit elsewhere. In general OCPD clearance time is not considered in these situations, so mbrooke has a bit of an uphill battle convincing the rest of us that it should be considered.
In the case of the installation being discussed, it would mean using roughly 6x the copper, solely to change the time it takes OCPD to clear a fault.
I'd suggest that the OP do a search for some of the other threads on this topic, rather than rehashing the discussion here.
-Jon
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
I think you may find voltage drop of 3% or more easily at service equipment or first disconnect of separately derived systems during ground fault conditions, particularly on smaller capacity supplies that have pretty limited potential output to begin with.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Right- I would argue in those cases you can confidently go up to 10 seconds without physiological danger since the voltage at the transformer terminals would dip substantially during a ground fault.
On the flip side 100-200 amps of fault current at a typical 2000 amp 480 volt service would not cause any significant dip in voltage at the transformer's terminals resulting in about 138 volts to remote earth (assuming EGC and phase are the same size). In reality however a voltage gradient (shell) may form around the metal base reducing the touch voltage the person is exposed, however it may still be around 100 volts.
Examples of reduced from gradient but still dangerous touch voltage:
Anything like a fence, guard rial, bus shelter, sport equipment, ect may all become energized if sitting within the gradient- it does not have to contact the light post.
Thanks for explaining this to me. I'll definitely look into the code and other discussions on this to make sure it does or doesn't apply to my situation.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
My comment was assuming ground fault to something solidly bonded. You will have this problem you posted here with most anything that is only grounded via electrode(s) or just plain contact with soil by the object in question.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Even when solidly bonded, the EGC is nothing more than a potential divider. An EGC does not remove voltage (it never will), merely 277 volts is reduced to 138 volts relative to remote earth. If the object has rebar or metal going into the soil, a voltage gradient will forum around the footing- this typically provides some equal potential by small amount whereby 138 volts is reduced to 105 volts hand to foot.
Using ohms law and a 1000 ohm hand to foot resistance we obtain 105 milli-amps of current. According to IEC TS 60479-1 100 milliamps of current beyond 500 milliseconds has the ability to cause irreversible effects such as sustained ventricular fibrillation. At 1.5 seconds the probability of sustained ventricular fibrillation (death if no CPR and/or defibrillation is administered with seconds to minutes) exceeds 50%.
IEC makes extensive gestures toward physiological danger, while the IEEE makes extensive references toward fire, property damage, and fusing of the EGC. (see post below)
wwhitney
Senior Member
- Location
- Berkeley, CA
- Occupation
- Retired
This sounds like an ongoing debate from another thread. Surely for the purposes of this thread a simple reference to the other thread would suffice, rather than rehashing everything here?
Cheers, Wayne
Cheers, Wayne
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
The IEEE Green Book, EPRI, and Eaton software makes the case that a breaker tripping in 10s of seconds may fuse the EGC since EGCs are typically smaller than the phase conductors (less than 1/10th the mm2 area) on the basis that it will carry fault current for very short periods of time.
Table 250.122 Inadequate?
Is 250.122 Inadequate in high fault current applications or where devices have been selectively coordinated? Some eye opening statments from the IEEE- 1) 2) 3)
forums.mikeholt.com
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
We are getting off topic from the maximum of 3% voltage drop design that started this. I think an effective fault clearing path can have more then 3% voltage drop. Yes the less VD the more effective it can be, but I was merely stating even just VD in the source can be pretty significant in some cases.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Honestly: Protection of life and property takes precedence over performance. Splitting hairs over VD is pointless when having the breaker open in a reasonable amount will by itself force larger wire whereby 3% voltage drop is guaranteed.
romex jockey
Senior Member
- Location
- Vermont
- Occupation
- electrician
selective coordination would seem to fit , in the sense that a little Z ,and it's all gone china ~RJ~
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Selective coordination is just one possible factor that can increase a breaker's opening time. Long runs are another factor, even for a non adjustable breaker or one set as low as permitting/possible.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
Just to put things into perspective for those reading:
10 gauge copper wire fuses at 8900 amps of current in 32 milli seconds (this would trip a 30 amp breaker instantly (and larger handle ratings on IM) before fusing can take place) where as that same 10 gauge wire fuses at 333 amps in 10 seconds.
https://en.wikipedia.org/wiki/American_wire_gauge#Tables_of_AWG_wire_sizes
Increasing time results in a steep decrease in a conductor's ability to carry short circuit current.
Ideally the final temprature of the EGC after the breaker opens should not exceed 150*C for THHN/THWN-2 installations and 250*C for XHHW (XLPE) conductor.
mbrooke
Batteries Included
- Location
- United States
- Occupation
- Technician
An example of where Table 250.122 would become inadequate-
333 amps / 60 = 5.55
8 gauge wire 472 amps divided by 100 = 4.72
6 gauge wire 668 amps divided by 200 = 3.34
A 60 amp circuits will result in the EGC melting under acceptable worse case breaker clearing times, a 100 amp circuit has a 50% chance of the EGC melting before the breaker opens and a 200 amp circuit has a 100% chance of the EGC melting (breaking the the path back to the source) before the breaker ever opens.
Based upon these time current curves of a common brand name UL listed breaker one could make the inference that Table 250.122 assumes a clearing time faster than 10 seconds (and as such the intent {goal} of the code) is that a ground fault clear in less than 10 seconds.
333 amps / 60 = 5.55
8 gauge wire 472 amps divided by 100 = 4.72
6 gauge wire 668 amps divided by 200 = 3.34
A 60 amp circuits will result in the EGC melting under acceptable worse case breaker clearing times, a 100 amp circuit has a 50% chance of the EGC melting before the breaker opens and a 200 amp circuit has a 100% chance of the EGC melting (breaking the the path back to the source) before the breaker ever opens.
Based upon these time current curves of a common brand name UL listed breaker one could make the inference that Table 250.122 assumes a clearing time faster than 10 seconds (and as such the intent {goal} of the code) is that a ground fault clear in less than 10 seconds.
- Status