Electrical
Electrical service entrance and metering installation field guide
Bring the utility power in, set the meter, land the disconnect nearest the entry, bond the neutral once at the service, and get the green tag before the utility energizes.
Direct answer
The service entrance is where utility power enters a building: the service drop or lateral, the service entrance conductors, the meter, and the service disconnect, governed by NEC Article 230. The load calculation sizes it, the utility supplies it, and the AHJ inspects it before the meter is set.
Key takeaways
- NEC Article 230 governs the service entrance: service drop or lateral, service entrance conductors, the meter, and the service disconnect.
- NEC 230.70 requires the service disconnect at a readily accessible location outside or inside nearest the point of entrance of the service conductors.
- The neutral bonds to ground only once, via the main bonding jumper at the service disconnect per NEC 250.28 and 250.24; never rebond downstream.
- Gear AIC and SCCR must equal or exceed the utility's available fault current per NEC 110.9 and 110.10; shorter conductor runs raise fault current.
- Self-contained metering runs up to about 200 A; CT metering is required on larger services, commonly above 400 A single-phase or 200 A three-phase.
The service entrance, and the four parts that make it
The service entrance is the point where the utility's power crosses onto the building and becomes the customer's wiring. NEC Article 230 governs everything from the connection at the utility to the load side of the service disconnect, and it is one of the few places where the line between what the utility owns and what you own is drawn in the code, not just on a drawing.
Four parts make up the service, and naming them straight keeps the rest of the job in order. First, the service drop or the service lateral, the conductors from the utility that land at the building, overhead or underground. Second, the service entrance conductors, your conductors from that connection to the service equipment. Third, the meter, the utility's measuring point, sitting in a socket or a CT cabinet you provide. Fourth, the service disconnecting means, the switch or breaker that kills the whole building, located near where the conductors enter.
The reason the parts matter is that the rules change at each boundary. The conductors ahead of the service disconnect have almost no overcurrent protection in front of them, because the only protection is back at the utility transformer. That is what makes the service entrance different from everything downstream of it, and it is why the disconnect location, the conductor length inside the building, and the available fault current all carry more weight here than anywhere else in the system.
Overhead or underground service?
Overhead and underground are the two ways the utility power reaches the building, and the choice is usually made for you by the utility's distribution in the area and the local requirements, not by preference. Overhead uses a service drop from a pole to a weatherhead and mast on the building. Underground uses a service lateral, conductors run in conduit below grade to the meter or service equipment.
Overhead is cheaper to install and easier to repair, and it is still common on rural and older services. The cost is exposure: the drop is in the weather, the mast has to take the mechanical pull of the conductors, and the clearances over the roof and the grade have to be held. Underground costs more up front because of the trench, the conduit, and the pull, but it is out of the weather and out of sight, and most new commercial and subdivision work runs underground because the local code or the developer requires it.
Whichever path the service takes, the utility publishes a service handbook or service requirements document that controls the details: who furnishes and who installs the drop or lateral, the conduit size and the point of attachment, the meter location, and the clearances. Read that handbook before you set anything, because the utility will reject a service that does not match it, and a rejected service holds up the meter and the occupancy. The handbook and the AHJ both have to be satisfied, and they do not always say the same thing.
Sizing the service from the load calculation
The service size starts with the load calculation, not with a round number off the panel schedule. NEC Article 220 totals the connected load, applies the demand factors, and returns the minimum service size in amps, and that number sets the service entrance conductors, the meter equipment rating, the disconnect, and the gear you order. Anvilfield's load-calculation guide walks that procedure; this guide takes the amps it produces and turns them into an installed service.
From the amps you get the voltage and phase, the conductor size and material, and the raceway. A 200 A, 120/240 V single-phase service is the common dwelling case. Commercial work runs 120/208 V or 277/480 V three-phase, and the larger the building the more likely the service climbs into the hundreds or thousands of amps, which changes the metering, the gear, and sometimes the voltage delivered.
Size the service entrance conductors for the calculated load and confirm the ampacity against the conductor tables and the terminal temperature rating, the same way you would a feeder. The one trap specific to the service is that the conductors ahead of the disconnect are protected only by the utility, so they are sized to the load and the disconnect rating, and the available fault current, not the load, often drives the gear selection. Size the conductors for the amps, then check the fault current for the gear.
What is the difference between self-contained and CT metering?
Self-contained metering runs the full service current through the meter itself. The meter plugs into a socket, the line and load conductors land on the socket jaws, and the meter measures the actual current passing through it. This is the standard up to a service size the utility sets, commonly around 200 A, with 320 A continuous rated sockets used on what the trade calls a 400 A class service. Above that the current is too large to run through a plug-in meter safely.
CT metering, or instrument-transformer metering, is what larger services use. Current transformers wrap around the service conductors and step the current down to a small, proportional signal, and that signal, not the full current, runs to the meter. The CTs and often potential transformers live in a CT cabinet or metering compartment you furnish, and the utility installs and seals the instruments. The exact crossover from self-contained to CT varies by utility, commonly landing near 400 A single-phase or above 200 A three-phase, so confirm the threshold in the utility's service handbook before you order the gear.
Getting the metering type wrong is an expensive mistake because it changes the equipment, the cabinet, and the conductor terminations. Order a self-contained socket for a service the utility wants metered with CTs and the whole front end is wrong. Decide the metering type off the service size and the utility's table first, before the socket or the CT cabinet is on the order.
| Metering type | How it measures | Typical service size | Who sets the threshold |
|---|---|---|---|
| Self-contained | Full current through the meter socket | Up to about 200 A, 320 A continuous on a 400 A class socket | Utility service handbook |
| CT-rated | Current transformers step current down to the meter | Larger services, commonly above 400 A single-phase or 200 A three-phase | Utility service handbook |
| Potential transformers added | Steps voltage down too, on higher-voltage services | Higher-voltage and large services | Utility service handbook |
Meter location, height, and the utility seal
The meter location is the utility's call, written in their service handbook, and it is one of the first things the utility checks. Most utilities want the meter outdoors, readily accessible from the property without going through a locked gate or a yard with a dog, and at a height that lands the meter center in a common window, often around 4 ft to 6 ft above finished grade. Confirm the exact dimension and the access rules with the serving utility, because they differ from utility to utility and a meter set too high or too low gets rejected.
The meter socket or CT cabinet is equipment you furnish and install, and it has to be on the utility's approved equipment list. Mount it plumb and at the right height, land the line and load conductors on the right lugs, and leave the meter itself for the utility. The socket has a lever bypass on many ringless designs so the utility can pull and replace the meter without dropping the building, and the cover takes a utility seal once the meter is set. Once it is sealed, that seal is the utility's, and breaking it without authorization is a problem you do not want.
On a multi-occupant building the meters group into a meter stack or a metering center with a meter for each tenant. Each meter has to be marked with the unit it serves, durably, so the utility and the next electrician can tell which meter feeds which space. Unmarked meter stacks are a callback waiting to happen, and the inspector looks for the labeling.
Service entrance conductors, SE cable, and parallel sets
The service entrance conductors run from the utility connection to the service disconnect, and they are sized for the calculated load and the disconnect rating, with the ampacity confirmed against the conductor tables and the terminal temperature column. They can be individual conductors in a raceway, or service entrance cable, the assembly the trade calls SE or SER, where the construction allows it. In a wet location, which includes most of the run on the outside of a building and conduit below grade, the conductors have to be listed for wet locations and the raceway has to drain or be sealed so water does not collect in the gear.
Service entrance cable is common on dwelling services. SE cable has the grounded conductor and the ungrounded conductors in one jacket; SER adds a separate equipment grounding conductor for feeder use. Used as service entrance conductors the cable is fine, but the moment it becomes an interior feeder or the bonding rules change, the type and the conductor count have to match the use. Watch the transition from service conductors to feeder, because the neutral-ground bonding rule changes at the disconnect and the cable type has to follow.
On large services the conductors run in parallel: multiple conductors per phase, run in separate raceways or bundled, sharing the current. The NEC allows paralleling at 1/0 and larger, and the catch is that every parallel set has to be the same length, material, size, and termination so the current divides evenly. A short conductor in a parallel set carries more than its share and runs hot. Where the conductors land in a wireway or gutter and tap off to multiple disconnects, the tap and the gutter fill have their own rules. Keep parallel sets matched, and keep the gutter from becoming a heat trap.
Where does the service disconnect go?
The service disconnect goes nearest the point where the service conductors enter the building, and that location rule is enforceable, not advisory. NEC 230.70 requires the service disconnecting means to be installed at a readily accessible location either outside the building or inside nearest the point of entrance of the service conductors. The reason is the conductors ahead of the disconnect have no overcurrent protection in the building, so the code limits how far those unprotected conductors travel inside before something can shut them off.
The NEC does not give a single fixed distance for that run inside the building; it says nearest the point of entrance, and the AHJ interprets how far is too far. Some jurisdictions hold it to a few feet, some allow a run if it is in conduit or encased. Do not run unprotected service conductors deep into a building to reach a disconnect in a back electrical room. If the gear has to live inside and away from the entry, the common fix is a disconnect outside where the conductors enter, then a protected feeder to the interior gear.
Readily accessible means a person can get to it to operate it without climbing over obstacles or using a ladder. A disconnect behind stored material, above a dropped ceiling, or behind a locked door the occupant cannot open is not readily accessible, and the inspector will fail it. The disconnect is the thing a firefighter or the occupant reaches for in an emergency, so it has to be reachable, and that is the lens the inspector uses.
One disconnect, the six-disconnect history, and the emergency disconnect
The service disconnecting means has been allowed as a group for decades, but the rule tightened. The older six-disconnect rule let up to six switches or breakers in a single service enclosure serve as the disconnect, so killing the building meant throwing up to six handles. The 2020 NEC changed 230.71 so each service has one disconnecting means, and where two to six are still allowed they have to be in separate enclosures or separate compartments, not crammed in one box with shared live bus. The reason was worker safety: even with all six off, the old single enclosure still had energized bus inside, and the only way to make it dead was to have the utility cut power at the transformer.
Confirm which edition the jurisdiction has adopted, because this is exactly the kind of rule that differs by where you are working. A jurisdiction on an older code still allows the six in one enclosure; a jurisdiction on the 2020 or later does not. The adopted edition and local amendments control.
Separately, the 2020 NEC added 230.85, an outdoor emergency disconnect for one- and two-family dwellings. It requires a disconnecting means in a readily accessible outdoor location on or within sight of the dwelling, so a first responder has an exterior way to kill the house in a fire or a flood. It can be the service disconnect itself, a meter disconnect built into the metering equipment, or a listed disconnect marked suitable for use as service equipment ahead of the service disconnect, and where there is more than one it has to be marked and grouped. Later editions have refined the marking and the rating language, so verify the requirement against the adopted edition.
Where does the neutral bond to ground?
The neutral bonds to ground at the service, and only at the service. The connection is the main bonding jumper, and it ties the grounded conductor, the neutral, to the equipment grounding conductor and the metal of the service disconnect enclosure. NEC 250.28 covers the main bonding jumper and 250.24 requires it at the service. This is the single point in the whole system where the neutral and the ground are deliberately connected, and Anvilfield's grounding and bonding guide covers the electrode side of that connection in depth.
Downstream of the service disconnect, the neutral and the equipment grounding conductor stay separate. NEC 250.24(A)(5) prohibits reconnecting the grounded conductor to ground or to equipment grounding conductors on the load side of the service disconnect, except where the code specifically allows it. That separation is what keeps normal neutral current off the equipment grounding system. Bond them again at a downstream panel, even with a bonding screw left in by accident, and you put neutral current on the conduit and the equipment grounds, where it makes stray voltage and defeats the protection.
This is the rookie mistake that hides for years. A subpanel installed with the factory bonding strap still in place looks fine and works, until someone measures current on a ground conductor or a piece of equipment bites somebody. Pull the bonding screw or strap in every panel except the service. The main bonding jumper goes in at the service disconnect, once, and the inspector checks for it there and checks that it is missing everywhere else.
The grounding electrode system at the service
The grounding electrode system connects the service to earth, and the grounding electrode conductor, the GEC, is the wire that does it. The GEC runs from the service, the neutral or the bonded enclosure, out to the grounding electrodes: the concrete-encased electrode the trade calls a Ufer, ground rods, metal underground water pipe, building steel, and any ground ring. NEC Article 250 requires every qualifying electrode present at the building to be bonded into one system, and Anvilfield's grounding and bonding guide covers electrode selection, the rod resistance rule, and the GEC sizing in full.
At the service the two things to get right are where the GEC lands and how big it is. It lands at the service, on the grounded conductor or the bonded service enclosure, ahead of or at the service disconnect, so the earth connection and the main bonding jumper share the same node. The size comes from the GEC sizing table against the service conductor size, with a maximum that applies when the only electrode is a rod, where a smaller conductor is allowed because a rod cannot carry more than that anyway.
Do not treat the rod in the dirt as the fault-clearing path. The breaker trips on the bonding back to the source, not on the earth connection. The grounding electrode system is there for lightning, surges, and a stable voltage reference, and it has to be present and bonded, but the conductor that clears a fault is the bonding jumper and the equipment grounding conductor, not the GEC to the rod.
What is available fault current, and why must the gear match it?
Available fault current is the maximum current the utility can push into a bolted fault at the service, and the service gear has to be rated to take it. Every overcurrent device has an interrupting rating, the AIC, the maximum fault current it can safely clear, and every assembly like a panelboard or switchgear has a short-circuit current rating, the SCCR, the maximum it can withstand without coming apart. NEC 110.9 and 110.10 require those ratings to equal or exceed the available fault current at the point of installation. Undersize them and a fault does not trip the breaker cleanly, it blows the gear open.
The available fault current comes from the utility. You request it, and the utility provides a value or a transformer impedance you calculate from, in what the trade calls the utility letter. NEC 110.24 then requires service equipment in other than dwellings to be field-marked with the maximum available fault current and the date the calculation was made. That marking ties the installed gear to a documented number so the next person can check it.
The catch that bites people is that the number moves. A shorter service conductor run has less impedance, so it raises the available fault current at the gear, not lowers it. Shave the conductor length from the approved drawing and you can push the fault current above the gear's rating without realizing it. Get the utility number in writing, size the AIC and SCCR above it with margin, and recheck if the conductor run changes. The fault current is the thing that gets value-engineered down by accident and kills people later.
Working clearance at the service gear
The service gear needs working clearance in front of it, and the inspector checks it before the wiring. NEC 110.26 sets the working space: a clear depth in front of the equipment, a width at least the width of the equipment or a set minimum, and headroom, all so a person can work on energized or recently energized gear and get out. The common depth is 36 in for the usual service voltages against a grounded or ungrounded wall, but the exact figure depends on the voltage and what is across from the gear, so confirm the dimension for the condition against the adopted code.
The space has to stay clear, dedicated, and it cannot be storage. The width is commonly 30 in or the equipment width, whichever is greater, and the door has to be able to open and a person able to stand in front of the gear. Headroom is commonly 6.5 ft or the height of the equipment. Above the gear there is a dedicated space the code keeps clear of piping and ducts that are foreign to the electrical equipment.
This is the cheapest thing to get wrong and the most annoying to fix. The gear gets set in a closet that meets the size on paper, then the plumber runs a pipe through the dedicated space above it, or the owner stacks boxes in front, and the inspector fails the clearance. Protect the working space on the drawing and on the job, because moving set gear to gain a few inches is real money.
Weatherhead, mast, and overhead clearances
On an overhead service the drop lands at a weatherhead, the fitting at the top of the mast that lets the conductors in while keeping rain out. Below it you form a drip loop, a downward bow in the service conductors so water runs off the loop and drips before it reaches the weatherhead, instead of running down the conductors into the gear. The drip loop is a small detail that keeps water out of the meter, and leaving it off is a way to put water where it does not belong.
The mast carries the mechanical pull of the service drop, and on a tall mast or a long span that pull is real. The mast has to be strong enough and braced or guyed so the drop does not bend it over, and the utility handbook usually specifies the minimum mast size and where guying is required. A bent mast pulls the clearances down and stresses the roof penetration, which is where leaks start.
The clearances over the roof and the grade are enforceable and the inspector measures them. NEC 230.24 requires service drop conductors to clear a roof by at least 8 ft, with reduced clearances allowed in specific cases, such as a steep roof or a mast that only penetrates the overhang. Over the ground the common figures are 10 ft above finished grade at residential service voltages where the conductors are accessible to people, 12 ft over residential property and driveways, and 18 ft over public roads and areas subject to truck traffic. Confirm the exact clearances for the voltage and the location against the adopted edition, because the table has conditions the round numbers do not capture.
| Location | Common minimum clearance | Note |
|---|---|---|
| Over a roof | 8 ft | Reduced in specific cases per 230.24, such as a steep roof |
| Over residential grade, accessible | 10 ft | At 120/240 V class service voltages |
| Over residential property and driveways | 12 ft | Areas not subject to truck traffic |
| Over public roads and truck areas | 18 ft | Confirm against the adopted edition |
The underground service lateral
An underground service runs the lateral in conduit from the utility, a transformer pad, or a handhole to the meter or service equipment. The conduit, the size, the depth, and who pulls the conductors are all in the utility handbook, and they vary by utility, so the handbook is the first read on an underground job. The conduit that comes up out of the ground to the meter, the riser, takes physical abuse, so it is commonly run in Schedule 80 PVC or rigid where it is exposed, not the thinner Schedule 40 used below grade.
Burial depth follows the cover requirements in NEC Table 300.5, measured as the shortest distance from the top of the conduit or cable to finished grade. The common figures are 24 in of cover for direct-buried conductors, 18 in for conductors in PVC conduit, and 6 in for rigid metal conduit, with 24 in required under driveways and areas subject to vehicles regardless of the method. Confirm the depth for the wiring method and the location against the table, and remember the utility may require deeper than the NEC minimum.
The pull is where underground services go sideways. Long laterals, several bends, and large parallel conductors make for a hard pull, so plan the conduit with sweeps instead of tight bends, pull boxes or handholes where the run is long, and pulling lube and a tugger sized for the job. Pulling large service conductors through an undersized or bend-heavy conduit damages the insulation, and you do not find that until it faults. Size the conduit for the pull, not just for the fill.
Intersystem bonding and surge protection at the service
The service is where the other building systems bond to the electrical grounding, and the NEC requires an intersystem bonding termination at the service so the phone, cable, and other utilities have a place to connect to the grounding electrode system. It is a small bar or set of terminals, accessible, near the service, with capacity for the systems that need it. Without it, the low-voltage trades land their grounds wherever they can, which is how you get separate, unbonded ground references that put a difference of potential across equipment during a surge.
Metal water piping, metal gas piping, and structural steel that is likely to become energized get bonded too. The metal water pipe is both a possible electrode and something that has to be bonded; the gas pipe is bonded, commonly through the equipment grounding conductor of the circuit likely to energize it, not used as an electrode. Building steel gets bonded into the system. Anvilfield's grounding and bonding guide covers which metal gets bonded and how it is sized.
Surge protection at the service is increasingly expected and in some occupancies required. A service-level surge protective device, a Type 1 or Type 2 SPD, clamps the transient that rides in on the utility before it reaches the panel and the equipment downstream. It needs the earth connection to have somewhere to send the energy, which is one more reason the grounding electrode system has to be real and bonded, not just present on paper. Recent editions have expanded where an SPD is required, so check the adopted edition for the occupancy.
Generators, transfer switches, and large or medium-voltage services
When a generator backs up the building, the tie point is a transfer switch, and on a whole-building setup it is often a service-rated automatic transfer switch, an ATS that includes the service disconnecting means. A service-rated ATS satisfies the disconnect requirement and switches the building between utility and generator without a separate service disconnect ahead of it. The neutral handling at the transfer switch matters: whether the neutral is switched or solidly connected changes where the neutral-ground bond lives and whether the generator is a separately derived system. Get that wrong and you either lose the bond or create a second one. Cross-check the bonding against the transfer-switch listing and the grounding rules.
On large commercial and industrial work the service can come in at medium voltage. The utility brings primary voltage to a pad-mounted or vault transformer, often owned or maintained under a different arrangement, and the building's service starts at the secondary. Medium-voltage work brings its own clearances, its own qualified-person requirements, and arc-flash exposure that is a different category from 600 V work. A data center or a large campus may have multiple services, paralleling gear, and tiers of redundancy, and the metering, the available fault current, and the gear ratings all scale up with it.
The common thread across these is that the basics do not change: the service is sized from the load, the disconnect goes near the entry, the neutral bonds once, the gear meets the fault current, and the AHJ and utility both sign off. The scale and the equipment change. The rules underneath them are the same Article 230 and Article 250 rules, applied to bigger iron.
Utility coordination, inspection, and the green tag
A service is one of the few installations where two authorities have to be satisfied before power flows: the AHJ that inspects to the NEC, and the utility that owns the connection. The order matters. On most jobs the AHJ inspects the service and issues the approval, commonly called the green tag, and the utility will not energize or set the meter until that approval is in hand. The green tag is the gate. No green tag, no meter, no power.
Coordinate early, because the utility's lead times and requirements drive the schedule more than the wiring does. The utility handbook tells you who furnishes the drop or lateral, the meter equipment they approve, the location and height, and the clearances. Apply for the service, get the available fault current in writing, and confirm the equipment is on their approved list before you order it. A service built to the NEC but not to the utility's handbook still gets rejected, and that rejection is measured in weeks, not days.
Temporary power usually comes first so the trades can work, and the temp service is its own permitted, inspected, grounded installation, not an afterthought run off an extension cord. Anvilfield covers temporary power separately. When the permanent service is ready, the inspection confirms the disconnect location, the bonding, the grounding electrode system, the working clearance, and the labeling, then the green tag releases the utility to set the meter and energize.
What to document
A service that nobody documented is a service the next electrician has to reverse-engineer with the cover off and the meter live. The record is what lets an inspector close the job, a utility set the meter, and the next person work safely. Capture it as the service is built, not from memory afterward.
Record the service size and the load calculation it came from, the voltage and phase, the conductor material and size and whether it is parallel, the metering type and the meter equipment, the disconnect type and location, the available fault current from the utility and the gear AIC and SCCR ratings, the main bonding jumper, and the grounding electrode system with the GEC size and the electrodes connected. If the service is in other than a dwelling, the available-fault-current field marking with its date is part of the installation, not optional.
| Field to record | Why it matters |
|---|---|
| Service size and voltage/phase | Sets conductors, gear, and metering type |
| Conductor material, size, parallel sets | Ampacity, terminations, and even current division |
| Metering type and equipment | Self-contained vs CT changes the whole front end |
| Disconnect type and location | Proves the nearest-entry and access rules |
| Available fault current and date | Required field marking on non-dwelling service equipment |
| Gear AIC and SCCR ratings | Must equal or exceed the available fault current |
| Main bonding jumper and GEC | Proves the single bond and the electrode system |
Common mistakes
- Bonding the neutral to ground downstream of the service, with a bonding strap left in a subpanel, putting neutral current on the equipment grounds.
- Gear AIC or SCCR rated below the utility's available fault current, often after the conductor run was shortened and the fault current rose.
- A service disconnect that is not nearest the point of entrance, or not grouped and readily accessible, with unprotected conductors run deep into the building.
- Ordering self-contained metering for a service the utility requires to be CT-metered, or the reverse.
- GEC landed in the wrong place or undersized, or electrodes present at the building left unbonded.
- Service drop clearances over the roof or grade short, or no drip loop at the weatherhead.
- No utility letter for the available fault current, and no field marking on non-dwelling service equipment.
- Building the service to the NEC but not to the utility handbook, then waiting weeks for a rejected service to be fixed.
- No green tag before expecting the meter, or the working clearance blocked by storage or foreign piping above the gear.
Field checklist
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Standards and references
The NEC, NFPA 70, is the framework. Article 230 covers services from the utility connection through the service disconnecting means: the overhead and underground supply, the service entrance conductors, the disconnect location and number, and the drop clearances. Article 250 covers the grounding and bonding, including the main bonding jumper and the grounding electrode system. Article 110 covers the install conditions that matter at the service: 110.9 and 110.10 on interrupting and short-circuit ratings, 110.24 on field-marking the available fault current, and 110.26 on working clearance. Underground cover comes from Table 300.5.
Specific points to verify against the adopted edition, because they have changed and they vary by jurisdiction: the single versus six-disconnect rule at 230.71, the outdoor emergency disconnect for one- and two-family dwellings at 230.85, the service drop clearances at 230.24, and where a surge protective device is required. The exact section numbers and the thresholds shift between code cycles, so confirm them against the edition the jurisdiction has adopted and any local amendments before citing them on a submittal.
Two authorities sit alongside the NEC on a service. The serving utility's service handbook or service requirements document controls the connection, the metering equipment, the location, and the clearances, and it can be stricter than the NEC. The AHJ inspects and issues the approval that releases the utility to energize. Where the equipment is listed, the UL listing and the manufacturer's instructions govern the terminations and the ratings. Cite the authority that controls the point, and let the utility handbook and the project specification override habit.
Units, terms, and the names on the drawing
The service entrance carries a stack of terms that show up differently across a utility handbook, a drawing set, and the code, so the same part can read three ways. Keeping the names straight keeps the coordination clean.
The service drop is the overhead utility connection; the service lateral is the underground one. The meter base and the meter socket are the same thing. CT metering is also called instrument-transformer metering. The service disconnecting means is the service disconnect or the main. Available fault current is also called available short-circuit current or fault duty. AIC is the device interrupting rating; SCCR is the assembly short-circuit current rating. The GEC is the grounding electrode conductor; the main bonding jumper is the single neutral-to-ground bond at the service. Service size is in amps, voltage in volts, conductor size in AWG and kcmil, and burial cover in inches.
- Service drop / lateral
- The utility's overhead (drop) or underground (lateral) conductors landing at the building
- Service entrance conductors
- The customer's conductors from the utility connection to the service disconnect
- Self-contained vs CT metering
- Full current through the meter vs current transformers stepping it down for larger services
- Service disconnecting means
- The switch or breaker nearest the entrance that disconnects the whole service
- Main bonding jumper
- The single connection of neutral to equipment ground, at the service only
- Available fault current
- The maximum fault current the utility can deliver, which the gear AIC and SCCR must exceed
- GEC
- Grounding electrode conductor, from the service to the grounding electrode system
FAQ
What is a service entrance?
A service entrance is where utility power enters a building and becomes the customer's wiring, covered by NEC Article 230. It includes the service drop or lateral from the utility, the service entrance conductors, the meter, and the service disconnect. The load calculation sizes it and the utility connects it.
Where does the neutral bond to ground in a service?
The neutral bonds to ground at the service only, through the main bonding jumper at the service disconnect, per NEC 250.28 and 250.24. Downstream of that disconnect the neutral and the equipment ground stay separate. Bonding them again at a subpanel puts neutral current on the equipment grounds and is a common, dangerous error.
What is CT metering and when is it required?
CT metering uses current transformers to step the service current down to a small signal the meter reads, instead of running full current through the meter. It is required on larger services, commonly above 400 A single-phase or 200 A three-phase, where self-contained metering cannot carry the load. The serving utility's handbook sets the exact threshold.
What is a service disconnect and where does it go?
A service disconnect is the switch or breaker that shuts off the whole service. NEC 230.70 requires it at a readily accessible location outside the building or inside nearest the point of entrance of the service conductors, because those conductors have no overcurrent protection inside ahead of it. The AHJ interprets how far the interior run can be.
How deep does an underground service lateral have to be buried?
Burial cover follows NEC Table 300.5: commonly 24 inches for direct-buried conductors, 18 inches for conductors in PVC conduit, and 6 inches for rigid metal conduit, with 24 inches under driveways and vehicle areas. Cover is measured to finished grade, and the serving utility may require deeper than the NEC minimum.
What is available fault current and why does it matter for service gear?
Available fault current is the maximum current the utility can push into a fault at the service. The gear's interrupting rating (AIC) and short-circuit current rating (SCCR) must equal or exceed it, per NEC 110.9 and 110.10. Get the value from the utility in writing, because a shorter conductor run raises the fault current.
How high above a roof does the service drop have to be?
NEC 230.24 requires overhead service drop conductors to clear a roof by at least 8 feet, with reduced clearances allowed in specific cases such as a steep roof or a mast penetrating only the overhang. Over residential grade the common minimums are 10 to 12 feet. Confirm the figures against the adopted code edition.
Do I need an outdoor emergency disconnect on a house?
The 2020 NEC added 230.85, requiring an outdoor emergency disconnect for one- and two-family dwellings in a readily accessible location on or within sight of the dwelling, so first responders can kill the house. It can be the service disconnect, a meter disconnect, or a listed disconnect ahead of the service. Verify the requirement against the adopted edition.
Who inspects a new electrical service, the utility or the AHJ?
Both. The AHJ inspects the service to the NEC and issues the approval, commonly called the green tag. The serving utility controls the connection, the meter, and the clearances through its service handbook and will not set the meter until the green tag is in hand. Coordinate the utility early, since lead times drive the schedule.
What size is the service disconnect, and how many are allowed?
The disconnect is rated for the calculated service load and the available fault current. The 2020 NEC 230.71 allows one disconnecting means per service, and where two to six are used they must be in separate enclosures or compartments, not one shared box. The adopted edition and local amendments control how many are permitted.
People also ask
Codes cited in this guide
This guide is written and reviewed against the published standards below. Always confirm the current adopted edition with the authority having jurisdiction.