Passive fire protection
Firestopping Penetrations
Every penetration is a defeat of an assembly that was tested whole — how to name what was breached and install the listed system that puts it back.
Published · Last reviewed
The Assembly Was Tested Whole
A rated wall or floor earns its number in a furnace, tested as a continuous specimen under ASTM E119 / UL 263, Standard Test Methods for Fire Tests of Building Construction and Materials. Nothing passed through it. The moment a plumber core-drills the deck or an electrician saws a slot for conduit, that tested condition no longer exists, and what remains is an assembly carrying a defect the original listing never covered. Firestopping is not sealing a hole. It substitutes a second, separately tested assembly for the fragment of barrier that was destroyed, and the two listings have to agree about what they are attached to.
Trades that treat the work as caulking produce failures that look fine and behave badly. Red sealant smeared over a gap proves nothing on its own; the tested system is the entire arrangement — substrate type and thickness, opening size, penetrant material and diameter, annular space range, backing material and its compression, sealant depth, and whether the seal lands on one face or both. Alter any one of those and the listing no longer describes the work in front of you. Inspectors who know the trade read the system number first and the sealant colour last.
Two questions therefore organise every penetration on the job: what did the opening remove, and what does the listed system put back? A 2-hour concrete floor and a 1-hour gypsum shaft wall demand different repairs for the same 4-inch pipe, because the material removed and the way heat travels through what is left differ. Depth of packing, the need for a collar, whether a temperature rating applies — all of it falls out of naming the breach precisely before anyone opens a case of product.
Read the Breach, Then the Catalogue
Establish four facts about the barrier before selecting anything: what kind of separation it is (fire wall, fire barrier, fire partition, smoke barrier, horizontal assembly), the hourly rating, the construction type and thickness, and whether the assembly must resist smoke as well as flame. Those come from the life-safety plan and the assembly schedule, never from the look of the wall. Two identical-looking gypsum partitions on the same corridor can be a 1-hour fire partition and a 2-hour fire barrier, sheeted the same way and rated differently.
Ratings reported under UL 1479, Fire Tests of Penetration Firestops, or ASTM E814, Standard Test Method for Fire Tests of Penetration Firestop Systems, are not interchangeable. The F rating measures how long the seal resists flame passage. The T rating adds a limit on temperature rise on the unexposed side, measured on the penetrant and the surrounding seal. An L rating quantifies air leakage; a W rating covers resistance to water. Codes call these out selectively — temperature ratings commonly for through-penetrations of floors outside of walls, leakage ratings for smoke barriers — and the governing edition of the International Building Code, NFPA 101 Life Safety Code, or the local amendment decides which apply here.
Outside North America the vocabulary shifts while the logic holds. EN 1366-3, Fire resistance tests for service installations — Penetration seals, feeds classifications expressed under EN 13501-2 as E for integrity and I for insulation with a duration in minutes. Canadian work references CAN/ULC-S115, Standard Method of Fire Tests of Firestop Systems. Approvals from one regime do not automatically satisfy another, and the authority having jurisdiction settles which family of evidence it will accept — a question worth asking at submittal rather than at inspection.
Through-Penetration or Membrane: Two Different Defeats
A through-penetration passes clean out the far side. Both faces of the barrier are open, and the seal has to handle heat conducted along the penetrant into the adjoining space as well as flame and smoke through the annulus. A membrane penetration stops inside — an outlet box, a recessed cabinet, a bracket anchor — defeating one skin of a multi-layer assembly while the cavity and the opposite face stay intact. The failure modes diverge enough that the two carry separate listings and separate code limits.
Membrane penetrations get abused more than any other category, because every trade fitting out a rated corridor drills them and nobody logs them. Codes in the IBC family cap the area of an individual box, the aggregate box area within a given area of wall, and the horizontal separation between boxes on opposite sides of the same partition, unless a listed system or a tested putty-pad configuration permits otherwise. Those figures move between code editions and local amendments, so confirm them against the edition in force rather than the one you learned on your first rated job.
Steel boxes mounted back-to-back with no separation and no pads remain the classic punch-walk finding. Remedying it afterwards is disruptive: relocate a box, or install listed pads suited to that box type, which means pulling devices and often opening finished board. Catching the layout at rough-in is a five-minute coordination item between the electrical foreman and the wall framer; catching it at handover is demolition in an occupied corridor.
The Annulus Carries the Specification
Listed systems state a minimum and a maximum annular space, and crews lose more rework to that pair of numbers than to any product choice. A 6-inch core around a 4-inch pipe can sit outside the tested range even though the gap looks generous and easy to fill. Point contact — the penetrant resting hard against one side, annulus zero there and doubled opposite — is permitted by some systems and prohibited by others, and only the listing says which. Guessing produces a seal that is unrepairable without re-drilling.
Depth of fill is the second number that gets estimated by eye. Systems call for backing material, typically mineral wool packed to a stated depth at a stated compression, with sealant over it to a specified thickness on one or both faces. Under-pack and the sealant slumps into the void and thins below rating. Over-pack and material is wasted while the finished face ends up too shallow. Compression is the point people miss: the wool's density once installed, not its nominal density in the bag, is what the furnace test actually saw.
On a floor with dozens of similar penetrations, the gap between listed fill depth and habitual fill depth shows up as material running out three risers short of the end of the run. Working the annular volume before the order goes in keeps the crew from stretching product across openings it was never going to cover, and keeps the last seals on a level from being the thin ones.
Annular volume decides whether a whole floor of penetrations gets sealed to listed depth or to whatever is left in the last case, so it belongs on the order sheet before the first bead goes in.
Firestop caulk needed
0.986 gal
Always use a firestop sealant system that is UL-listed for your specific penetration type (cable, pipe, conduit) and the fire-rated assembly being penetrated — this calculator estimates volume only, it does not verify UL system compliance.
For the dimensions entered, expect a firestop caulk needed of 0.99 gal. Moderate confidence — sound arithmetic, but allow for the variation any real site introduces. Set for United States. The market selector changes the units and the trade terminology; any standard behind the formula is cited under sources.
Estimated cost — your price
This site holds no price list for this material — local prices vary too much to publish honestly. Enter your supplier's price and the result is costed with it.
What the Penetrant Does When It Gets Hot
Sorting penetrants by fire behaviour, rather than by which trade installed them, reaches the right system faster. Non-combustible penetrants — steel pipe, copper, EMT, rigid conduit — hold their shape and conduct heat. The seal has to stay bonded while the pipe grows and pushes, and any temperature rating depends on the metal's conductivity and diameter. Combustible penetrants — PVC, CPVC, PEX, insulated line with a combustible jacket — simply leave. They soften, then melt, and what is left is an open hole the diameter of the pipe unless something closes it.
Intumescent devices exist for exactly that disappearance. Collars fastened to the barrier face, or wrap strips buried in the opening, expand under heat and crush the void where the plastic used to be. Selection turns on pipe material, diameter, wall schedule, orientation, and whether the pipe is open or closed to atmosphere at its ends — a detail that changes the tested condition and is written into the listing. Anchoring a collar into a substrate other than the tested one voids the system as thoroughly as fitting the wrong device.
Cables and trays bring a third behaviour into play: fill ratio. Tested systems cap the visual fill of the opening, because a dense bundle traps heat inside itself and the interstitial voids between jackets carry smoke straight through. Insulated pipe adds another decision, since the insulation may be required to run continuous through the seal, stop at the seal, or be replaced locally with mineral wool. The listing settles it, and field improvisation on that single point is among the most common causes of a rejected seal.
When the Opening Outgrew Every System
Sometimes the breach is simply too large for the trade. A demolished duct leaves a square metre of open deck where listed penetration systems top out well below that, or someone sawed a continuous slot to route six conduits and no tested geometry covers it. Work of that kind stops being firestopping and becomes reconstruction: rebuild the barrier down to an opening a listed system covers, then penetrate the rebuilt barrier properly and seal it.
In gypsum construction that means framing the opening back with studs or a header, re-sheeting with the same Type X layer count and board thickness the assembly schedule specifies, staggering joints between layers, and taping and finishing so the patch reads as continuous with the wall around it. Layer count is not negotiable. A 2-hour partition built two layers each side stays two layers each side across the patch, with fastener type and spacing matched to the assembly. A single-layer strip hidden behind paint is invisible until a fire finds it.
Concrete and masonry breaches follow identical logic with different materials — dry-pack grout, cast infill, or a listed mortar system rated for the thickness involved and installed to its tested depth. Whichever route, the reinstated barrier has to reach full assembly thickness. A flush patch thinner than the surrounding deck will fail on temperature rise long before it fails on flame, and nothing on the finished surface will tell an inspector that it is thin.
Rebuilding a rated partition around an oversized breach stands or falls on returning the Type X layer count and board thickness to what the assembly schedule demands, which is the count to settle before any sheet is cut.
Total Type X sheets needed
16 sheets
The number of layers required for a specific fire-resistance rating must be taken from a tested UL or GA fire-rated assembly design for your exact wall construction — never assume a layer count without checking the tested assembly.
- Sheets per side (one layer)
- 8
At the values currently entered, the total type x sheets needed works out to 16. Confidence is moderate: the method is sound, but real materials and site conditions vary. Figures are shown in United States units and terminology; switch the market above if you are building elsewhere.
Estimated cost — your price
This site holds no price list for this material — local prices vary too much to publish honestly. Enter your supplier's price and the result is costed with it.
Breaches That Never Stop Moving
Not every defeat of a rating is a round hole with a pipe in it. Head-of-wall joints, wall-to-wall joints, floor-to-floor joints and the perimeter gap between slab edge and curtain wall are linear breaches that open and close as the structure deflects, loads and cools. They are tested under UL 2079, Tests for Fire Resistance of Building Joint Systems, or ASTM E1966, Standard Test Method for Fire-Resistive Joint Systems, with perimeter conditions tested under ASTM E2307 using an intermediate-scale, multi-story apparatus.
Joint systems carry a movement classification — the tested percentage of joint width and the number of cycles the system survived before fire exposure. Install a static system in a deflection head that moves under live load and it will crack, and the crack shows up as smoke transfer long before anything burns. Matching movement class to the structural condition is an engineering decision recorded on the drawings; it is not a field substitution to be made because the static product was on the truck.
Where a penetration lands inside a joint, or where a joint turns a corner around a column, the tested geometry generally runs out. Manufacturer engineering judgements bridge that gap with a written opinion extrapolating from tested systems, and their acceptability rests entirely with the authority having jurisdiction. Requesting one during submittals is an email. Requesting one after an inspector has already tagged the condition is a schedule item with a hold on the ceiling above it.
Who Breaches the Barrier After You Leave
Sequencing determines whether firestopping happens once or three times. Seals installed before the final conduit is pulled get cut open by the next trade. Seals installed after ceilings close cannot be inspected without removing tile. The order that survives contact with a real job runs: barrier complete and inspected, services roughed and dressed to final position, firestop installed and photographed, then ceilings and finishes.
Coordination failures cluster at predictable interfaces. Low-voltage and security trades arrive late and drill through work already signed off. Sprinkler fitters relocate a drop after the seal has cured. Owner vendors — access control, AV, medical gas — appear after turnover with no idea the corridor wall carries a rating. Stencilled identification on rated walls above the ceiling line, backed by a barrier management plan, is the cheapest defence against later trades treating a fire barrier as ordinary partition.
Sleeving changes the economics entirely. A sleeve set before the pour, or a cast-in device with an integral intumescent element, converts a future demolition into a future cable pull through a re-enterable seal. On projects heavy with IT or medical infrastructure, over-sleeving at design and capping the spares is cheaper in labour than core-drilling a live slab later, and it keeps the structural engineer out of a conversation about drilling near reinforcement.
Proving the Repair Exists
An unlabelled seal is, to an inspector, an unknown seal. Listed systems call for identification adjacent to the penetration — system number, installer, date, and the rating achieved — in a form durable enough to last the building's service life. Across a project carrying hundreds of penetrations, photographic records keyed to a floor plan and a penetration schedule turn a week of ceiling-tile removal into an afternoon reviewing files with the inspector at the table.
Third-party inspection follows ASTM E2174, Standard Practice for On-Site Inspection of Installed Firestops, with ASTM E2393, Standard Practice for On-Site Inspection of Installed Fire Resistive Joint Systems and Perimeter Fire Barriers, covering the linear work. Sampling rates and destructive-verification percentages are set by the specification or the jurisdiction, and healthcare and high-rise occupancies attract the tighter end of both. Contractor qualification programmes operated by UL and by FM Approvals are specified as a bidding condition on some projects, which changes who may legally do the work.
Long after handover the same discipline keeps the rating alive. Every retrofit, every added cable, every relocated pipe reopens the question the original furnace test answered. A facilities team holding the penetration schedule and the system numbers repairs a new breach the same week it appears. A team without those records surveys an entire building to rediscover what its walls were meant to be, and pays for that survey twice — once in consultants and once in the work that follows.
Before the first bead goes in
What to have settled at the wall, in the order a penetration actually forces the questions.
- Assembly schedule and life-safety plan — Confirms separation type, hourly rating, construction thickness, and whether smoke resistance and a leakage rating are demanded.
- System number recorded on the drawing — The listing fixes opening size, annular range, backing depth and which faces get sealed — carry it to the wall, not to memory.
- Mineral wool at the packed density the system names — Depth and compression in place are what the furnace test saw; nominal bag density proves nothing.
- Intumescent collars or wrap strips for combustible penetrants — Sized to pipe material, diameter and orientation, and fastened only into the substrate the system was tested on.
- Type X board matching the existing layer count — For breaches larger than any listed system covers, rebuild the barrier to full thickness before sealing anything.
- Labels, photographs and a keyed penetration schedule — An unlabelled seal is unverifiable at inspection and unrepairable at the next retrofit.
Opens the calculators above on one screen with the dimensions from this article already filled in. Quantities only — this site publishes no price list, because local prices vary too much to publish honestly.
Drawn from
- ASTM E119 / UL 263, Standard Test Methods for Fire Tests of Building Construction and Materials
- UL 1479, Standard for Fire Tests of Penetration Firestops
- ASTM E814, Standard Test Method for Fire Tests of Penetration Firestop Systems
- CAN/ULC-S115, Standard Method of Fire Tests of Firestop Systems
- EN 1366-3, Fire resistance tests for service installations - Part 3: Penetration seals
- EN 13501-2, Fire classification of construction products and building elements - Part 2: Classification using data from fire resistance tests
- UL 2079, Tests for Fire Resistance of Building Joint Systems
- ASTM E1966, Standard Test Method for Fire-Resistive Joint Systems
- ASTM E2307, Standard Test Method for Determining Fire Resistance of Perimeter Fire Barrier Systems Using Intermediate-Scale, Multi-story Test Apparatus
- ASTM E2174, Standard Practice for On-Site Inspection of Installed Firestops
- ASTM E2393, Standard Practice for On-Site Inspection of Installed Fire Resistive Joint Systems and Perimeter Fire Barriers
- International Building Code, Fire and Smoke Protection Features
- NFPA 101, Life Safety Code
Guidance, not a specification. Local codes, the engineer of record and the product manufacturer’s instructions govern where they differ from anything written here.