Electrician Talk banner
1 - 18 of 18 Posts

·
Registered
Joined
·
6 Posts
Discussion Starter · #1 ·
Just wondering. Utilities can size their service laterals much smaller than the NEC would require for the premises wiring. They don't have to have the ampacity of the service disconnect. And they can deal with voltage drop by just upping a tap setting on the xfmr a little to make up for small conductors.

So, is it ever a possibility that a long service lateral run can result in enough resistance to cause available fault current to be too low to instantaneously trip the main or other breakers if there's a ground fault event? Anyone heard of any type of scenario such as this?
 

·
Moderator
Estwing magic
Joined
·
26,312 Posts
From what I remember from Bussmann school is that conductor size and length is included in available fault current calculations. In normal situations I don’t think it makes much of a difference, though.

I could never really understand the term “instantaneous trip” since it’s a time/current relationship and nothing is instantaneous. Probably a term invented by the breaker guys.
 

·
Bilge Rat
motors and controls.........
Joined
·
8,407 Posts
During a bolted fault, s small long service lateral with a small transformer could easily have high enough impedance to not cause the magnetic trip to engage but most likely the thermal would trip before anything burnt up.

Even if it does burn something up, it's the PUCOs responsibility.

I do know for a fact that a #6 triplex being fed from a 75 KVA pole-mount transformer and feeding a 200 amp split-bus panel with 5 - 50s and 1 - 60 all fairly heavily loaded will indeed burn up..........not that anyone intentionally loaded this panel to the gills to see what would happen.........nope, not us.........
 

·
Registered
Ready Mix concrete plant electrician
Joined
·
2,494 Posts
During a bolted fault, s small long service lateral with a small transformer could easily have high enough impedance to not cause the magnetic trip to engage but most likely the thermal would trip before anything burnt up.

Even if it does burn something up, it's the PUCOs responsibility.

I do know for a fact that a #6 triplex being fed from a 75 KVA pole-mount transformer and feeding a 200 amp split-bus panel with 5 - 50s and 1 - 60 all fairly heavily loaded will indeed burn up..........not that anyone intentionally loaded this panel to the gills to see what would happen.........nope, not us.........
Were the breakers the optional no-trip ever GE’s, or the old standard FPE by chance? These are a favorite of carpenters and welders everywhere.
 

·
Registered
Joined
·
557 Posts
Probably part of the reason the smallest overhead I see run these days is #2 alum.

My PoCo shares the #2 feed with my neighbour, i'm watching and waiting to see how long the common portion lasts :sneaky:
 

·
Registered
Joined
·
632 Posts
on a like note: I have had 12/2 uf 800' long with a dead short cause a small fire and not trip a 20 amp breaker. 480 volts.
 

·
Registered
Scada Supervisor
Joined
·
3,535 Posts
Probably part of the reason the smallest overhead I see run these days is #2 alum.

My PoCo shares the #2 feed with my neighbour, i'm watching and waiting to see how long the common portion lasts :sneaky:
Growing up in Philadelphia, they would run a city block ( 18 houses ) off of shared #2 copper.
 

·
Chief Flunky
Field Service Engineer
Joined
·
2,386 Posts
Just wondering. Utilities can size their service laterals much smaller than the NEC would require for the premises wiring. They don't have to have the ampacity of the service disconnect. And they can deal with voltage drop by just upping a tap setting on the xfmr a little to make up for small conductors.

So, is it ever a possibility that a long service lateral run can result in enough resistance to cause available fault current to be too low to instantaneously trip the main or other breakers if there's a ground fault event? Anyone heard of any type of scenario such as this?
Yes.

In fact regardless of what they do detecting faults is a huge problem, Rather than using overcurrent on distribution systems utilities use what are called distance relays. In simple terms a distance relay measures V/I or impedance. If the impedance drops too low it trips. Distance relays have limits as to the “reach” or distance that they can detect faults. Beyond that distance another relay is required.

Distance relays dominate distribution systems just as time overcurrent relays are the work horses of industrial and power plants.
 

·
Chief Flunky
Field Service Engineer
Joined
·
2,386 Posts
From what I remember from Bussmann school is that conductor size and length is included in available fault current calculations. In normal situations I don’t think it makes much of a difference, though.

I could never really understand the term “instantaneous trip” since it’s a time/current relationship and nothing is instantaneous. Probably a term invented by the breaker guys.
Lower cost gear such as MDPs, MCCs, and panel boards only has a 3 cycle AIC unlike switchgear which has a 30 cycle rating. Breakers take typically up to 3 cycles to clear a fault at the distribution level (detecting fault, physically moving and quenching the arc). Thus for all intents and purposes this is instantaneous...no intentional delay in tripping. This is for electrical faults, shorts and arcing faults.

An overload condition is a little different. The change in load may happen over time or instantly but equipment has a thermal limitation...time is on our hands and we have to allow temporary overload conditions from temporary load increases as well as starting surges to pass while stopping actual overloads from causing thermal damage, the fault usually isn’t related to time but the effects are.

When it comes to coordination we’d like to simplify things by assuming that except at the lowest level (branch circuit protection) where the highest trip speeds should occur that time-current can be used for coordination everywhere. In the case of main breakers protecting low end distribution equipment, Houston we have a problem! An MCC for instance needs 3 cycle protection from shorts for the bus. But using breakers would cause miscoordination with the MCC breakers and fuses over breakers nearly always causes this. The solution is series ratings...tested combinations of fuses and breakers where we know that the combination will work even if the devices all have instantaneous tripping. Overload protection still uses time-current but the instantaneous side has to be handled differently.

Once we are into true switchgear the much longer short circuit withstand ratings allow us the luxury of turning off instantaneous protection and using time-current for all types of faults.

This works until impedance becomes a problem as OP pointed out. At that point distance relays are used which are based on detecting impedance rather than just current.
 

·
Chief Flunky
Field Service Engineer
Joined
·
2,386 Posts
Growing up in Philadelphia, they would run a city block ( 18 houses ) off of shared #2 copper.
It’s called diversity. As long as it’s not a hot summer afternoon when all the HVAC kicks on at one time, the chances of everyone doing laundry at the same tine is low.

Why don’t you run one breaker per receptacle? Diversity.
 

·
Bilge Rat
motors and controls.........
Joined
·
8,407 Posts
Were the breakers the optional no-trip ever GE’s, or the old standard FPE by chance? These are a favorite of carpenters and welders everywhere.
It was a split-bus panel so the top 12 spaces were main lug. The 60 fed the bottom half of the panel. The 50s fed various temp loads, one of them was pretty much loaded with 300 watt incandescent lamps.

After a while, the lights started flickering so we all went outside to watch the show........

Quite a bit of smoke and an occasional flash-bang then a larger flash and the triplex was blown in two. When the transformer side of it hit the pole, there were a fair number of flash-bangs then both hots were blown off the transformer bushings.

Then it was quiet........lol.
 

·
Registered
Joined
·
9,118 Posts
It was a split-bus panel so the top 12 spaces were main lug. The 60 fed the bottom half of the panel. The 50s fed various temp loads, one of them was pretty much loaded with 300 watt incandescent lamps.

After a while, the lights started flickering so we all went outside to watch the show........

Quite a bit of smoke and an occasional flash-bang then a larger flash and the triplex was blown in two. When the transformer side of it hit the pole, there were a fair number of flash-bangs then both hots were blown off the transformer bushings.

Then it was quiet........lol.
I'd watch that instead of the 4th of July fireworks in the background...
 

·
Band Member
DIYer Extrodinaire
Joined
·
7,721 Posts
Probably part of the reason the smallest overhead I see run these days is #2 alum.

My PoCo shares the #2 feed with my neighbour, i'm watching and waiting to see how long the common portion lasts :sneaky:
I think you'll wait an awful long time !

#2 ACSR is rated at 185A @30˚C ambient. And it will do a lot more than that for a long time without melting.
 

·
Band Member
DIYer Extrodinaire
Joined
·
7,721 Posts
So, is it ever a possibility that a long service lateral run can result in enough resistance to cause available fault current to be too low to instantaneously trip the main or other breakers if there's a ground fault event? Anyone heard of any type of scenario such as this?
Hydro sizes the transformer protection to protect the transformer, not the service lateral.
And even then, their not all that worried about the transformer .... think 600-1000% overloading allowed.

So, even on a short run, the wires can melt before the cutout opens.
 

·
Registered
Electrical Contractor
Joined
·
2,809 Posts
Just how often do you guys run into a main from an overhead service trip from a fault?
 

·
Registered
Residential, lite comm., Industrial
Joined
·
193 Posts
Just wondering. Utilities can size their service laterals much smaller than the NEC would require for the premises wiring. They don't have to have the ampacity of the service disconnect. And they can deal with voltage drop by just upping a tap setting on the xfmr a little to make up for small conductors.

So, is it ever a possibility that a long service lateral run can result in enough resistance to cause available fault current to be too low to instantaneously trip the main or other breakers if there's a ground fault event? Anyone heard of any type of scenario such as this?
keep in mind ... they are in free air, but i agree, they do undersize drastically in some cases.... i personally think it is do to rookie linemen who dont have enough training/experience
 

·
Chief Flunky
Field Service Engineer
Joined
·
2,386 Posts
Just wondering. Utilities can size their service laterals much smaller than the NEC would require for the premises wiring. They don't have to have the ampacity of the service disconnect. And they can deal with voltage drop by just upping a tap setting on the xfmr a little to make up for small conductors.

So, is it ever a possibility that a long service lateral run can result in enough resistance to cause available fault current to be too low to instantaneously trip the main or other breakers if there's a ground fault event? Anyone heard of any type of scenario such as this?
Yes before it even gets to the customer end of things. Forget about ground faults. That’s small potatoes.

Utilities use overcurrent protection at generating stations and locally in substations but outside substations it simply doesn’t work even after a couple miles. If a tree fell on a line or lightning hit and caused a dead short and they relied on overcurrent protection alone as we do at the loads, it would never trip.

What they actually use is called a distance relay. These work by measuring the voltage and current and calculating the ratio (Z=V/I, Ohms Law) and tripping if this ratio gets too low. Even distance relays have a distance limitation so they may program a recloser every few miles as the distance limit sets in.



And grounding is pretty much a joke compared to what you expect. First off in a utility system they practice “peg” grounding. The distribution system is 3 wire. The poles are grounded and there is often a static line but it is in no way a 4 wire system, despite the fact that utilities most often run wye wye transformers. The reason is that the impedance on a cable is proportional to its length. But the impedance between two ground rods is proportional to the inverse of distance. This sounds nonsensical but you have to remember that the Earth is effectively a 2 dimensional sheet. It is slightly greater than 2D but over “short” distances we can ignore the curve. The resistance along any possible path through the Earth grows proportional to distance just as it does on a wire. The number of possible paths grows with the square of the distance. So we get d/d^2 for resistance. Two of the d’s cancel, so the resistance is proportional to 1/d, the inverse of distance. Engineers can read IEEE Green book to verify. Nonengineers will see this math equation in any 3 wire ground test which uses two grounds spaced a long distance apart. So grounding itself is very good but phase conductor impedance drives everything.

As far as the impact looking through a transformer though it’s not as bad as it sounds. We can model the transformer impedance as %Z. Similarly if we assume the transformer is the ideal model then we are looking at only the turns ratio N1/N2 which is called a. The secondary side voltage is V/a and the secondary side current is Ia. We can also transform primary side impedance. Z=V/I (Ohms Law) on the primary side. On the secondary side by substitution we get Z=(V/a)/(Ia) = (V/I) / a^2. So for example with 12,470 to 480 a=26 and a^2 = 626. To put this in perspective transformer %Z varies from a low of 1-3% on small transformers to 5-10% on large ones (1 MVA+). So utility losses are barely noticeable.

On the other hand on say a 1000 kVA transformer with 5% Z at 480 V is 0.0115 ohms. Or looking at short circuit current at the transformer and ignoring utility losses we are looking at 1000 / 480 / 1.732 / 0.05 = 24 kA. This is enough to trip a UL class B breaker curve (10x rating).

500 MCM has an AC impedance if about 0.0278 ohms per 1000 feet so with 4/C 500 MCM to give us 1200 A ampacity the impedance is about 0.007 ohms per 1000 feet. So to double the impedance we need about 16,000 feet. That’s not a long feeder…that’s operating our own utility! Generally it becomes a problem in longer distribution lines (hundreds of feet) with fairly high currents. Using this example going down to a 15 A branch circuit breaker feeding #14 the impedance is much higher, about 3 ohms per 1000 feet, but the instantaneous trip is now at only 150 A.
 
1 - 18 of 18 Posts
Top