Feeder tap rules under NEC 240.21 for electrical crews

What a feeder tap is, and why there are rules for it

A feeder tap is a conductor connected to a feeder at a point where there is no overcurrent device, and where the tap conductor is smaller than the feeder it lands on. The breaker or fuse ahead of that feeder is sized to protect the feeder, not the tap. So the tap is hanging on a circuit whose overcurrent device is rated well above the tap's own ampacity. Under any other section of the code that would be a violation on its face.

Walk it back to the principle. The general rule is that every conductor gets an overcurrent device sized to the conductor's ampacity, sitting at the point where the conductor gets its supply. A #6 fed from a 60 A breaker is protected. The same #6 lands on a 400 A feeder and the 400 A breaker will let the #6 cook long before it trips on an overload. Nothing about that breaker protects that conductor.

The tap rules are the narrow set of exceptions that let you do it anyway. They exist because the alternative is often impractical: you would have to run the full feeder size all the way to a small load, or set a fused disconnect at every tap point on a busway, or oversize gear that never needed it. Instead the code trades the missing breaker for a set of hard conditions. Keep the tap short, keep its ampacity above a stated fraction of the feeder device, land it on a single overcurrent device that protects it from there on, and keep it out of harm's way. Meet the conditions and the tap is legal. Miss one and it is not, no matter how good the workmanship looks.

The general rule, and where the tap rules sit as the exception

Two sections set the frame. NEC 240.4 is the protect-at-ampacity rule: conductors are protected against overcurrent at the ampacity they carry. NEC 240.21 says the overcurrent device has to sit at the point where the conductor receives its supply, with a list of exceptions that follow. The feeder tap rules are those exceptions, grouped under 240.21(B). The transformer secondary rules sit under 240.21(C).

The mental model that keeps people out of trouble: a tap is not protected by the breaker, it is protected by the conditions. On a normal circuit the breaker is the safety device. On a tap the breaker upstream only handles a hard short or ground fault, because a dead short draws enough current to trip even a large device. What the breaker will not do is protect the tap against an overload, the slow current creep that sits above the tap's ampacity but below the feeder device. The tap rules close that gap by limiting how long the unprotected conductor runs and forcing it to terminate in a device that does protect it.

That is why every tap rule reads as a list of conditions rather than a calculation. There is no formula that makes a tap safe. There is a length limit, an ampacity floor tied to the feeder device, a single-termination requirement, and a physical-protection requirement. The conditions are the protection.

What is the 10-foot tap rule?

The 10-foot tap rule, NEC 240.21(B)(1), lets a tap conductor run up to 10 ft from the feeder with no overcurrent device at the tap point, if it meets every condition in the list. The tap is short on purpose. Ten feet of unprotected conductor is a small enough exposure that the code accepts it without a device at the supply end.

The conditions, as commonly stated: the tap is not over 10 ft long. The tap ampacity is at least the calculated load it carries, and at least the rating of the device it supplies or the overcurrent device at the end of the tap. The tap does not extend past the switchboard, panelboard, or control device it feeds. And the tap is enclosed in a raceway if it leaves the enclosure where the tap is made.

There is a one-tenth nuance people miss. When the 10-ft tap leaves the enclosure or vault where it is made, its ampacity also has to be at least one-tenth the rating of the overcurrent device protecting the feeder. So a 10-ft tap off a 1000 A feeder that leaves the enclosure needs an ampacity of at least 100 A even if the load it serves is smaller. Inside the same enclosure that floor does not apply. Read the actual subsection in the adopted edition before you size it, because the wording on the one-tenth condition is exact and the AHJ reads it exactly.

What is the 25-foot tap rule?

The 25-foot tap rule, NEC 240.21(B)(2), lets a tap run up to 25 ft with no device at the tap point, but the trade is a higher ampacity floor and a hard single-termination requirement. The tap conductor's ampacity has to be at least one-third the rating of the overcurrent device protecting the feeder it taps. That fraction is the price of the extra length.

The one-third is keyed to the feeder's overcurrent device, not to the feeder conductor size. Off a 600 A feeder breaker the tap needs an ampacity of at least 200 A, no matter what the feeder conductors happen to be. Get that backward, sizing the third off the wire instead of the device, and you can land a tap that looks fine and fails the check.

The 25-ft tap has to end at a single overcurrent device, a single breaker or a single set of fuses, rated no higher than the tap conductor's ampacity. That device is what finally protects the tap at its own rating, and it is why you cannot fan a 25-ft tap out to several breakers at the far end. One tap, one terminating device. The tap also has to be protected from physical damage, which in practice means a raceway or another approved enclosure for the whole run.

Why the one-third and the single device, not the breaker, keep the tap safe

The one-third ampacity floor on the 25-ft tap is not arbitrary. A conductor sized at least a third of the feeder device is heavy enough that a short or ground fault on the tap draws enough current to open the upstream feeder device fast, and short enough at 25 ft that the fault current does not bleed away over the distance. That is the short-circuit and ground-fault protection the tap still has. The feeder device covers the catastrophic fault. The one-third floor is what makes that coverage reliable over the allowed length.

The single terminating device covers the other failure mode, the overload. Because the feeder breaker is too big to protect the tap against a slow overload, the tap has to dead-end in one device rated at or below its ampacity. From that device onward the conductors downstream are protected normally. The single-device rule is also why you cannot treat a tap point as a free splitter: the moment the tap feeds two or three breakers, no one device is limiting the tap to its ampacity, and the protection the rule depends on is gone.

Hold both ideas together and the rest of the tap rules read as variations on the same theme. Length controls how much unprotected conductor is exposed. The ampacity fraction guarantees the upstream device still clears a fault. The single termination handles the overload. Physical protection keeps the unprotected run from being damaged. Every subsection is a different mix of those four levers.

Transformer secondary taps under 240.21(C)

Secondary conductors off a transformer are their own problem, handled in NEC 240.21(C), not the feeder-tap subsections. The reason is that the transformer changes the current. A primary overcurrent device sees primary amps. The secondary conductors carry secondary amps, which on a step-down transformer are larger. So the primary device, even sized correctly for the primary, does not protect the secondary conductors at their ampacity. The secondary is unprotected at its supply end the same way a tap is, and it gets its own set of rules.

The shape mirrors the feeder taps. There is a 10-ft secondary rule and a longer secondary rule, each with a length cap, an ampacity floor, a single-termination requirement, and physical protection. The ampacity floors carry the voltage ratio, because you are comparing a primary-side device rating to a secondary-side conductor. A common form of the 10-ft secondary condition sizes the conductor against the primary device times the primary-to-secondary voltage ratio, and the longer secondary rule works off the one-third of the primary device through the same ratio. There is also a primary-plus-secondary form that caps the combined length.

The exact subsections and their conditions move between code cycles, and the voltage-ratio arithmetic is where people slip. We keep a separate transformer-secondary write-up for that math. For this guide the point is the boundary: if there is a transformer between the feeder device and your conductor, you are in 240.21(C), not the feeder taps, and the device on the secondary is required to protect the secondary because the primary device cannot.

Can a tap conductor be unlimited length?

Yes, but only outside. NEC 240.21(B)(5) allows an outside feeder tap of unlimited length with no overcurrent device at the tap point, because the exposure that the indoor length limits are guarding against, an unprotected conductor running through occupied space, does not apply when the conductor is outside the building. The conductors run outdoors the whole way except where they land.

The conditions are specific. The tap conductors are protected from physical damage. They terminate at a single overcurrent device, a single breaker or single set of fuses, that limits the load to the conductor ampacity. That terminating device is an integral part of, or located immediately next to, a disconnecting means. And the disconnect sits at a readily accessible spot outside the building, or inside nearest the point where the conductors enter. The single device at the end is permitted to feed any number of downstream overcurrent devices from there.

This is the rule behind a long run from a pad-mounted transformer or an outdoor switch to the building's service or distribution equipment. The conductor can run hundreds of feet unprotected outside, then has to be cut off at the first device the moment it enters, at or near the point of entry. The nearest-point-of-entry requirement is what inspectors check hardest here. Bring the conductor deep into the building before the first device and the outside-tap allowance no longer covers it.

The longer industrial taps: 25 ft to a transformer and the 100 ft high-bay tap

Two more feeder-tap subsections cover cases the short rules do not. NEC 240.21(B)(3) is the tap that supplies a transformer with the primary plus secondary together not over 25 ft. The primary conductor ampacity is at least one-third the feeder device, the secondary ampacity carried through the voltage ratio also meets the one-third, the combined primary-plus-secondary length stays under 25 ft, the conductors are protected from damage, and they terminate in a single device. It is the way you tap a feeder to a small transformer without setting a primary breaker.

NEC 240.21(B)(4) is the 100-ft tap, and it is fenced into a narrow occupancy. It applies only in high-bay manufacturing buildings where the walls are over 35 ft high, and only where the installation is maintained by qualified people under the kind of supervision a plant has and an ordinary building does not. The horizontal portion of the tap is limited, commonly to 25 ft, while the total length can reach 100 ft. The tap ampacity is at least one-third the feeder device, it terminates in a single device, it is protected from damage, and there is a minimum conductor size floor. There are more conditions on this rule than on any other tap, which tells you how reluctant the code is to allow 100 ft of unprotected conductor at all.

Do not reach for the 100-ft rule on a standard commercial job. The 35-ft wall height and the qualified-maintenance condition are not suggestions, and an inspector on a normal building will reject a 100-ft tap that should have been a 25-ft tap with a closer device. When you genuinely have the high-bay case, read the full subsection in the adopted edition, because the condition list is long and every item has to be met.

The tap still gets sized to its load and its conditions

The tap rules set a floor, not the final conductor. The one-tenth or one-third fraction is a minimum tied to the feeder device. The tap still has to carry its actual load, and it still has to be sized for the conditions it runs in, the same as any other conductor. The fraction and the load are two separate checks and the conductor has to pass both.

That means the derating still applies. A tap in a hot space loses ampacity to ambient correction. A tap bundled with other current-carrying conductors loses ampacity to the more-than-three adjustment. The termination temperature still caps the usable column. So a tap that clears the one-third floor on paper can still be too small once you correct and adjust it for where it actually runs. Size it through the full ampacity procedure, then confirm it also meets the tap-rule fraction, and take the larger of the two.

The order that keeps you honest: figure the load, size the conductor for that load through ampacity and derating, then check it against the tap-rule floor and the single-termination device. If the floor drives it bigger, the floor wins. If the load and derating drive it bigger, they win. The tap is legal only when it satisfies the rule that demands the most copper.

Why not just put a breaker at the tap point

Sometimes you should, and on a clean design you often do. The tap rules are not a goal, they are a relief valve for the cases where a device at the tap point is genuinely impractical or wasteful. Knowing which case you are in is the difference between an elegant install and a code dodge that an inspector smells.

The practical cases are real. A busway running across a plant has plug-in or bolt-on taps every few feet to small loads, and setting a fused disconnect at each tap would defeat the point of the busway. A feeder passing near a small transformer or a single piece of equipment can be tapped short rather than running a full-size feeder to a 30 A load. Service-entrance and outside runs from a pad transformer land on the first device inside, with no practical place for a device out at the transformer. In all of these the tap rule lets the conductor make the short, unprotected jump that the geometry forces.

Where the tap rules get abused is the lazy case: tapping a big feeder to dodge ordering the right gear, or stringing a 25-ft tap to three panels because nobody wanted to set a proper distribution section. If you find yourself stretching to fit a tap rule, the right answer is usually a breaker at the tap point and a normally protected conductor. The rules are there for the cases where that is not possible, not for the cases where it is inconvenient.

Physical protection and routing of the unprotected run

Every tap rule requires the tap to be protected from physical damage, and on the short taps the wording is tighter than that. The 10-ft tap has to be enclosed in a raceway where it leaves the enclosure it was tapped in, and it cannot run on past the equipment it supplies. The reason is plain once you remember the conductor is unprotected: damage to a normal conductor trips its breaker, while damage to a tap may not, because the upstream device is too large to react to anything short of a dead short.

So the routing matters more on a tap than on a protected conductor. Keep it in conduit or a listed enclosure for the full run. Keep it out of reach of forklifts, doors, moving equipment, and anything that gets worked on. On the 10-ft tap, do not let it wander out of the enclosure and across the room to chase a convenient path, because both the length and the leaves-the-enclosure conditions are watching that run.

The single hardest habit to break is treating the unprotected length casually because it is short. Ten feet feels like nothing. But it is ten feet of conductor that the feeder breaker will not protect against an overload or a partial fault, so it earns the raceway and the careful route that a protected branch circuit might not.

Where taps show up: busway, transformers, meters, and gutters

Taps are everywhere in commercial and industrial work once you know the shape. The busway tap is the textbook case: a plug-in unit bolts to the busway and a short tap runs from it to a panel or a machine, sized to the tap rule rather than the busway's amperage. The transformer tap feeds a dry-type transformer off a feeder without a primary breaker, then the secondary lands on its own device.

Gutters and wireways are full of taps. A power distribution gutter or a trough takes a feeder in and splits it to several sets of conductors leaving to different panels, and each of those leaving sets is a tap that has to meet a tap rule on its own. A splitter or distribution block does the same job in a smaller box. Meter stacks and CT cabinets feed taps to the individual meters. In each case the conductors leaving the common point are unprotected at that point and live or die by the tap conditions.

The tell that you are looking at a tap and not a branch is simple: trace the conductor back to where it gets its supply and ask whether there is an overcurrent device there sized to that conductor. If the answer is no, and there is a larger device upstream, you are on a tap and one of the 240.21 rules has to be carrying it.

Taps in data centers: busway, PDUs, and remote power panels

Data center power distribution leans on taps because the layout demands it. Overhead busway runs down the rows, and each rack or PDU taps the busway through a plug-in unit and a short run of conductor. The busway might be rated 400 A or 800 A while the tap to a single PDU is far smaller, so that tap is living on the 10-ft rule and the one-tenth or load-based floor that goes with it.

The density is what makes the discipline matter. You can have dozens of taps off one busway run, each one a short unprotected conductor that has to meet its rule, terminate in a single device at the PDU or remote power panel, and stay in its raceway or plug-in housing. One tap fanned to two panels, or one run stretched past the busway tap rule length to reach a rack across an aisle, is the kind of error that multiplies across a room.

Treat each tap as its own compliance item even when they are nearly identical. The repetition tempts crews to copy a layout that was wrong once into a room full of the same mistake. Confirm the tap length, the ampacity against the busway device, and the single termination on the first one, document it, and the rest follow a verified pattern instead of a guess.

What does the inspector check on a tap?

An inspector works a tap in a fixed order, and knowing the order lets you have the answers ready. First, the length. They will identify which tap rule you are claiming and measure or confirm the run against its limit, 10 ft, 25 ft, or the high-bay and outside cases. A tap that is six inches over its rule is a failure with no judgment call to make.

Second, the ampacity ratio. They check the tap conductor's ampacity against the rating of the overcurrent device protecting the feeder, looking for the one-tenth on the 10-ft tap that leaves the enclosure or the one-third on the 25-ft and longer taps. Third, the termination: a single overcurrent device at the end of the 25-ft, outside, and longer taps, rated at or below the tap ampacity, not a fan-out to several devices. Fourth, physical protection, the raceway and the routing. And on the outside tap, the location of the first device relative to where the conductors enter the building.

The fastest way to pass is to label the tap on the drawings with the rule it is built to and the numbers that prove it. When the one-line says 25-ft tap, 240.21(B)(2), feeder OCPD 600 A, tap ampacity 230 A, single 200 A device at end, the inspector is confirming your work instead of reverse-engineering it. The reverse-engineering is where they find the problems.

The tap and the load calculation

A tap feeds a load, and that load still has to be calculated like any other. The tap-rule fraction does not replace the load calculation, it sits on top of it. You total the load the tap serves, size the conductor for that load through the ampacity and derating, and then check the result against the tap-rule floor and the terminating device.

The two numbers interact in a way worth watching. On a 10-ft tap, the ampacity floor is partly the calculated load itself, so a heavier load drives the tap bigger directly. On a 25-ft tap, the one-third of the feeder device can be larger than the load demands, which means the tap conductor is sized by the rule rather than by the load, and you carry more copper than the load alone would call for. Either way the larger requirement wins.

This is also where the estimate can drift from the install. The takeoff prices the conductor the load needs, the field finds the tap rule forces a larger conductor and a specific terminating device, and the difference shows up as a change or an eroded margin. Catch the tap at takeoff, size it to the rule there, and the number stays where it belongs.

Showing the tap on the one-line

A tap that is not drawn is a tap nobody can defend. The one-line is where the tap lives for the inspector now and for the next person who opens the gear in five years. Show the feeder, the overcurrent device protecting it, the tap point, the tap conductor size and ampacity, the tap length, and the single device at the end.

Call out the rule by name on the drawing. Writing 240.21(B)(1), 10-ft tap next to the run tells everyone which set of conditions the tap is built to satisfy, which turns the inspection into a check rather than an investigation. Note the feeder device rating right there too, because every ampacity floor is keyed to it, and a reviewer cannot confirm the one-third or one-tenth without that number on the page.

The record outlives the crew. When a tap gets re-fed, extended, or has a load added years later, the next electrician needs to see that the original was a tap on a specific rule, not a normally protected feeder, so they do not blow the length or the single-termination condition with their addition. The note is what keeps a legal tap legal through the second and third change.

The violations inspectors actually write up

The same handful of tap violations show up over and over, and all of them come from treating a tap like a normal conductor. The tap too long is the most common: a 10-ft tap that grew to 14 ft chasing a route, or a 25-ft tap stretched to reach across a space. The length is a hard line, and a tape measure settles it.

The ampacity ratio not met is next. A tap sized to its load but under the one-third of the feeder device, or under the one-tenth on a 10-ft tap that leaves its enclosure. Then the multiple-device violation: a 25-ft or outside tap fanned out to two or three breakers instead of dead-ending in a single device, which destroys the overload protection the rule depends on. And the unprotected tap, run in free air or in a spot where it can be damaged, when the rule required a raceway and a safe route.

The quieter one is the wrong rule for the case. Claiming the 100-ft high-bay tap on an ordinary building that is not a high-bay manufacturing space, or treating an indoor run as an outside tap. These pass a casual glance and fail when the inspector reads which rule actually applies. Pick the rule that fits the real installation, not the rule that makes your length legal.

What to document

Capture enough that a reviewer can confirm the tap against its rule without walking the run. The tap rule you built to, the feeder overcurrent device rating that every floor is keyed to, the tap length, the tap conductor size and ampacity, and the single device at the end with its rating. If derating drove the conductor instead of the rule, note that too, so the next person sees why the tap is the size it is.

Common mistakes

• Running the tap past its length limit, a 10-ft tap over 10 ft or a 25-ft tap over 25 ft.
• Sizing the ampacity floor off the feeder conductor instead of the feeder overcurrent device rating.
• Missing the one-third on a 25-ft tap or the one-tenth on a 10-ft tap that leaves its enclosure.
• Fanning a 25-ft or outside tap out to more than one overcurrent device instead of a single termination.
• Letting a 10-ft tap leave the enclosure and wander, or skipping the raceway on any unprotected tap.
• Forgetting that ambient and bundling derating still cut the tap's ampacity below the table value.
• Claiming the 100-ft high-bay rule on a building that is not a high-bay manufacturing space.
• Bringing an outside tap well into the building before the first device instead of the point of entry.

Field checklist

Standards and references

The framework is the NEC, NFPA 70. The general protect-at-ampacity rule is 240.4. The requirement that the overcurrent device sit at the point of supply, with the tap exceptions, is 240.21. The feeder taps are 240.21(B): the 10-ft tap at (B)(1), the 25-ft tap at (B)(2), the 25-ft tap supplying a transformer at (B)(3), the 100-ft high-bay tap at (B)(4), and the outside tap of unlimited length at (B)(5). Transformer secondary conductors are handled separately at 240.21(C).

Conductor ampacity, which the tap still has to satisfy, comes from the ampacity tables, commonly Table 310.16, with the correction and adjustment factors in 310.15. The exact subsection numbers and their detailed conditions move between code cycles, and some of these rules have been reworded or renumbered across recent editions. Confirm every section against the edition the jurisdiction has actually adopted, along with any local amendments, before you cite it on a submittal.

Equipment listings and manufacturer instructions can add their own requirements, and the AHJ has the final read on whether a given installation meets a tap rule. Cite the section that controls the point, size the conductor to the rule that demands the most copper, and let the adopted edition and the inspector settle anything the wording leaves open.

Units, terms, and the language of taps

Taps come with their own vocabulary, and the same conductor can be described a few different ways across a drawing set and a code book. Knowing the terms keeps the conversation with the inspector clean.

A tap conductor is the conductor connected to a feeder with no overcurrent device at the connection. The feeder OCPD is the overcurrent protective device, the breaker or fuse, protecting the feeder the tap lands on, and it is the number every ampacity floor is keyed to. Ampacity is the current a conductor carries continuously without exceeding its temperature rating, given in amps. Length is the routed conductor length in feet, measured along the path, not the straight-line distance. The terminating device is the single overcurrent device at the end of the tap that protects it from there downstream.

Tap conductor

A conductor connected to a feeder with no overcurrent device at the connection point, smaller than the feeder device protects at its ampacity

Feeder OCPD

The overcurrent protective device protecting the feeder; every tap ampacity floor (1/10, 1/3) is keyed to its rating

Ampacity

The current a conductor carries continuously without exceeding its temperature rating, after correction and adjustment

Single overcurrent device

One breaker or one set of fuses at the end of a tap, rated at or below the tap ampacity, that protects it downstream

Transformer secondary conductor

A conductor on the secondary side of a transformer, unprotected at its supply because the primary device cannot protect it at secondary ampacity; governed by 240.21(C)

AHJ

Authority having jurisdiction, the inspector or office that adopts the code edition and makes the final compliance call

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is an electrical load calculation?
• What is conductor ampacity?
• What is the difference between a breaker and a fuse?
• What is selective coordination?
• What is a three-way match in accounts payable?