Concrete
Structural fireproofing field guide: SFRM and intumescent
What fireproofing does, why the listed thickness on a tested assembly is the whole rating, the difference between SFRM and intumescent, and how the thickness, density, and bond get proved by inspection.
Direct answer
Structural fireproofing is the material that insulates structural steel in a fire so it stays below the temperature where it loses strength and the building collapses, buying the hours of its rating. The rating comes from the listed thickness applied to a tested UL assembly, so under-applying it silently voids the rating. The manufacturer, spec, and AHJ control.
Key takeaways
- Fire rating equals the listed thickness of a listed material on a tested UL assembly; under-applying silently voids the rating.
- Structural steel keeps only about half its yield strength near 1,100 degrees F; ASTM E119 ends a rating at roughly 1,100 degrees F average or 1,300 degrees F at any point.
- SFRM is thick cementitious or gypsum spray for concealed steel; intumescent is a thin coating that swells and chars, for exposed architectural steel.
- SFRM bond strength is tested to ASTM E736; the IBC minimum rises with height, commonly around 150 psf up toward 1,000 psf for the tallest high-rises.
- Special inspection verifies thickness and density to ASTM E605 and bond to ASTM E736 member by member before concealment; patch all damage back to listed thickness.
Structural fireproofing, and why the thickness is the whole rating
Structural fireproofing is the insulation applied to structural steel so the steel stays cool enough in a fire to keep holding the building up. Steel does not burn, but it loses strength as it heats, and a loaded frame can come down once it gets hot enough. The fireproofing slows the heat into the steel, which buys time. That time, measured in hours, is the fire-resistance rating.
Here is the part crews get wrong: the rating is not a property of the material in the pail. It is a property of a specific thickness of that material applied to a specific steel size in a tested, listed assembly. A fire-resistance rating comes from a furnace test, commonly to ASTM E119, of one product at one thickness on one member under one set of conditions. Underwriters Laboratories publishes the result as a design number. Apply less than the listed thickness and you have not bought the rating, no matter how good the spray looks. The shortfall does not announce itself. The wall gets closed, the inspection gets missed, and the steel is under-protected behind finish where nobody will see it until the fire.
So the work is four moves done in order. Pick the right material for the exposure. Hit the listed thickness and density for each member. Bond it to clean, compatible steel. Prove all of it by inspection. The steel itself is set and connected by the erection crew, so this guide picks up where the steel-erection work leaves off, and it is a different job from firestopping, which seals the holes that pipe and cable punch through rated walls and floors.
The thickness on the listed assembly is the rating
Everything in this guide serves one fact: the fire rating you are being paid to deliver is the listed thickness of a listed material applied to a listed steel member. Change any of those three and you are no longer inside the tested design, and the rating no longer applies.
The number is not a target you get close to. It is a minimum you hold everywhere. A column called out for a 2-hour rating in a UL design might need a defined thickness of SFRM for that member size; lay it on a little thin in the corners and at the flange tips and those spots are under their rating even if the web reads fine. The furnace tested the whole member at the listed thickness, so the whole member has to get it.
This is why the inspection later measures thickness obsessively and reports both an average and a minimum. The minimum is the one that can fail you, because the fire finds the thin spot. Do not treat the listed thickness as guidance. Treat the UL design number on the drawings, the manufacturer's published thickness table for that design, and the project specification as the authority, and confirm anything you are unsure of with the architect and the special inspector before you spray.
Why steel gets fireproofed at all
Steel is strong and stiff at room temperature and it gives most of that up in a fire. By around 1,100°F a structural steel member retains only about half of its room-temperature yield strength, and the testing that sets the ratings treats roughly that range as the failure point. ASTM E119 ends a loaded beam's rating when the average steel temperature reaches about 1,100°F or any single point reaches about 1,300°F. A heavily loaded member that is already near its capacity can reach those temperatures and fail in a standard fire in as little as 10 to 30 minutes if it has no protection at all.
An unprotected frame in a real fire can soften, sag, and pull its connections apart while the building is still occupied. Fireproofing does not stop that from ever happening. It delays it. The insulation keeps the steel below its critical temperature long enough for two things to happen: people get out, and the fire service gets in to fight it. The hourly rating is how long the assembly was proven to do that in the furnace.
That is the whole reason the thickness is sacred. The hours are a life-safety promise written into the building code for that occupancy and that member, and the only thing standing behind the promise is the material on the steel at the thickness the test used. The required ratings come from the building code, commonly the IBC, by construction type and member; the engineer and architect set them on the drawings.
SFRM and intumescent, the two materials
Two families of material do this work, and you pick between them mostly by whether the steel will be seen and what it has to take. SFRM, spray-applied fire-resistive material, is the thick, plaster-like spray that builds up a soft insulating layer. Intumescent is the thin coating that looks like paint and swells into an insulating char only when a fire heats it. Both can reach the same hourly ratings. They get there by completely different physics and cost completely different money.
SFRM is cheaper per square foot and faster to apply, but it is bulky and visually rough, so it lives on concealed steel above ceilings, inside shafts, and on the underside of floor and roof decks. Intumescent costs more and goes on in controlled coats to a measured film thickness, but it is thin and can be finished to look like a painted member, so it earns its price on exposed architectural steel that the design wants people to see.
Neither one is the default. The exposure, the look, the durability the location demands, and the budget drive the call, and the spec usually makes it for you. Whichever you use, the rule does not change: the rating is the listed thickness, for SFRM, or the listed dry film thickness, for intumescent, applied to the tested assembly for that member.
| Factor | SFRM | Intumescent |
|---|---|---|
| What it is | Cementitious or gypsum spray with fiber | Reactive coating, applied like thick paint |
| How it protects | Insulates as a thick low-density layer | Swells and chars into an insulating foam when heated |
| Typical thickness | Fractions of an inch to a few inches | A fraction of a millimeter to several millimeters DFT |
| Where it fits | Concealed steel, decks, shafts | Exposed architectural steel where looks matter |
| Relative cost | Lower, faster | Higher, more controlled application |
| Proof of rating | Thickness and density and bond, per UL design | Dry film thickness per coat, per UL design |
SFRM: the spray on concealed steel
SFRM is the workhorse for steel nobody will look at. It comes as a cementitious or gypsum-based powder mixed with fiber and water and pumped through a hose to a spray gun, where it builds up on the steel in passes. Cured, it is a soft, low-density layer that insulates by being a poor conductor of heat and by holding moisture that has to boil off before the steel underneath can climb.
It is made in density grades, and the grade is part of the listing, not a free choice. Standard-density SFRM suits dry, concealed, low-abuse interiors and is the common product on bays of floor deck and framing. Medium and high-density formulations exist for exterior, semi-exposed, or high-impact spots like parking structures and industrial steel where a soft standard product would not survive handling and weather. Going to a denser product can also change the thickness the design requires, so the density and the thickness travel together in the UL design.
SFRM is fast and cheap, and that is exactly why it gets abused on the schedule. It is also the material with the bond problem covered below, because a spray that does not stick is worse than no spray: it looks rated and is not. Use the product, density, and thickness named in the listed design for the member, and confirm the grade against the spec and the manufacturer's data.
Intumescent: the coating on exposed steel
Intumescent fireproofing protects steel the architect wants visible. At room temperature it is a thin film that can be color-matched and topcoated to read as a finished painted member. It does nothing as insulation until a fire heats it. Then it reacts, commonly somewhere in the 350°F to 500°F range, and swells many times its applied thickness into a thick carbon char that insulates the steel underneath. A coat measured in fractions of a millimeter becomes an inch or more of foam when it matters.
Because the protection is in that reaction, the dry film thickness is the whole game, the same way SFRM thickness is. The required DFT for a member depends on the rating, the member size, and whether it is a beam or column, and it climbs as the rating climbs. It goes on as a system: a compatible primer, the intumescent basecoat built up in controlled coats to the specified DFT, and a topcoat that protects the basecoat from weather and wear and carries the color. Each layer has its own film-thickness window.
The discipline here is gauge work. You are measuring wet and dry film thickness as you build the coats, not eyeballing coverage, because too thin under-rates the steel and too thick in one pass can sag, mud-crack, or fail to cure. Do not extrapolate a thickness from one tested member to another size that was not tested. Apply to the listed design and the manufacturer's loading tables, and confirm the system, including primer and topcoat, with the manufacturer and the spec.
What is the fire-resistance rating?
The fire-resistance rating is the number of hours an assembly was proven to keep doing its job in a standard fire test, and for protected steel it is commonly 1, 2, or 3 hours. It is set by the building code, usually the IBC, based on the building's construction type, its height and area, and which member you are protecting. Columns often carry a higher required rating than the floor beams they support, because losing a column is worse than losing a beam.
The rating is a code requirement, not a preference, and the drawings should call out the required hours for each member along with the listed assembly that delivers them. The hours come from a furnace test to ASTM E119 in which a loaded member is protected exactly as a design specifies and heated until the steel reaches the failure temperatures. The time it lasts is the rating.
What matters in the field is that the rating is only as real as the assembly behind it. A line on the drawing that says 2 hours means nothing without the listed design that achieves 2 hours on that member, and that design comes with a specific material at a specific thickness. Treat the required ratings as the architect and the adopted code edition state them, and treat the listed assembly as the only way you are allowed to deliver them.
| Rating | Where it commonly lands | What sets it |
|---|---|---|
| 1 hour | Floor beams, secondary framing in some types | IBC construction type and member |
| 2 hour | Columns and primary framing in many types | IBC construction type and member |
| 3 hour | Columns in higher construction types | IBC construction type and member |
| The proof | A tested UL design at the listed thickness | ASTM E119 furnace test |
The thickness for this member is the number that matters
The single most important number on the job is the thickness for the member in front of you, and it is not one number for the whole building. It is calculated, member by member, from the steel's size and the rating it has to reach, and the listed design gives you the value. A heavier, chunkier member needs less material to reach a given rating than a light, slender one, because the heavier steel has more mass to soak up heat and less surface area for the fire to attack.
That is why a thickness schedule on a real job is a list, not a single figure. The same 2-hour rating might call for one thickness on a heavy column and a noticeably greater thickness on a light infill beam, both pulled from the same UL design's tables for their section sizes. Spray the light beam to the heavy column's thickness and it is under-rated, and you will not see it without a gauge.
Get the thickness for each member from the UL design referenced on the drawings and the manufacturer's thickness table for that design and that section, not from a rule of thumb and not by carrying one number across the whole frame. When the schedule and the steel do not match, which happens when a member gets substituted in the field, stop and get the corrected thickness from the engineer and the manufacturer before you spray it.
The W/D and A/P ratio that drives the thickness
The reason a thin member needs more material is captured in a number the designers use to set the thickness: the section factor, written W/D for shapes or A/P for the general area-to-perimeter form. W/D is the weight per foot of the member divided by the heated perimeter, the part of the surface the fire can reach. A high W/D means a lot of steel mass behind a little exposed surface, so the member heats slowly and needs less protection. A low W/D, a light member with a lot of surface, heats fast and needs more.
You do not usually compute this in the field, but you should know it exists, because it explains the thickness schedule and it is the basis for the substitution rules in the listed designs. UL designs let you adjust thickness for a member whose W/D differs from the tested one using a published equation tied to a minimum tested member, within limits. That adjustment is engineering, not a field guess.
The practical takeaway is simple. The thickness varies by member because the section factor varies by member, and the schedule already did that math from the listed design. Your job is to apply the thickness the schedule gives for that exact section. If a member is not on the schedule or differs from what was tested, the corrected thickness comes from the manufacturer and the engineer, not from the nearest similar beam.
The listed assembly is the basis, with no substitutions
A listed assembly is a complete tested recipe, and it is the legal and technical basis for the rating. A UL design number, for example, names the specific material, the thickness for the member size, the steel shape, whether the assembly is restrained or unrestrained, the deck or floor above, the primer condition, and the rest of the conditions that were in the furnace together. The rating belongs to that whole package, not to any one piece of it.
That is why substitutions are not a field decision. Swap the listed SFRM for a different brand because it was on the truck, change the deck type, spray over a primer the design did not include, or use a member outside the design's range, and you are outside the tested assembly. The rating from that design no longer applies, even if the new material is good and the thickness is right.
Where the field condition genuinely differs from any available listing, the path is an engineering judgment from the manufacturer or a qualified fire-protection engineer, accepted by the AHJ, not a quiet substitution. Build to the UL design number on the drawings. When reality does not match a listing, get an accepted engineering judgment before you cover it, and keep the design number and any judgment in the record so the inspector can tie what is on the steel to what was tested.
The bond is the number-one SFRM failure
SFRM has to bond to the steel, and when it does not, it falls off. This is the most common way fireproofing fails, and it is the one that hides the longest, because spray that is going to let go can look perfect on the day it is applied and then sheet off in panels weeks later when it is bumped, when the building moves, or when condensation gets behind it. Fireproofing on the floor is fireproofing that is not on the steel.
The code treats bond strength as a measurable requirement, not a hope. For SFRM, the IBC sets a minimum cohesive and adhesive bond strength tested to ASTM E736, and the required value climbs with building height: commonly on the order of 150 psf for ordinary buildings, rising to several hundred psf and to roughly 1,000 psf for the tallest high-rises, because the consequence of a sheet of fireproofing falling off a column 60 stories up is not the same as in a two-story warehouse. Confirm the required value for the project against the adopted IBC edition and the spec.
Bond depends on two things you control: a clean steel surface and a compatible coating under the SFRM. Get either wrong and the spray will not hold its rated strength. The bond gets verified by a pull test during inspection, covered below, and if it fails the value, the fix is not negotiable: remove it, correct the surface or apply the manufacturer's approved bonding agent, and re-spray.
Primer compatibility, the bond killer hiding on the steel
The most common reason SFRM will not bond is a shop primer that the SFRM cannot grip. Structural steel often arrives with a primer or paint applied at the fabricator for handling and corrosion, and many of those coatings are too slick or chemically wrong for cementitious or gypsum SFRM to bond to. You can spray a perfect-looking layer onto a primed beam and have it let go because the primer underneath never let it grab.
The code anticipates this. Where SFRM goes onto a primed, painted, or encapsulated surface and acceptable bond performance for that combination has not already been established, a bond test is required to qualify it, and if it does not make the required strength, the manufacturer's approved bonding agent has to go on first or the coating has to be removed back to a surface that bonds. None of that is optional and none of it is a field judgment call.
So the question to settle before the first pass is what is on the steel and whether the SFRM is listed and tested to bond to it. Bare, clean steel is the safe substrate. A primer the SFRM manufacturer has qualified is fine. An unknown or incompatible primer means a bonding agent or removal. Check the primer against the SFRM manufacturer's data and the spec, and test it when the combination is not already proven.
Surface prep, because the bond starts here
Whatever the bond depends on chemically, it depends first on a clean surface. SFRM bonds to steel, not to oil, dirt, loose mill scale, or rust scale, and not to ice or frost. Grease and form-release that drifts onto steel, dust from other trades, and loose mill scale all keep the spray from grabbing the metal, and they are exactly the contaminants a busy site puts on steel between erection and fireproofing.
The steel should be clean and dry, free of oil, loose scale, and anything else that breaks the bond, and at a temperature the manufacturer allows before spray starts. The manufacturer's data and the spec set the acceptable surface condition; meet it before you pull the trigger, not after the inspector flags a low pull test.
This is also where coordination shows up as a quality issue, not just a schedule one. Steel that has sat exposed collects contamination and surface rust, and a member that was clean at erection may not be clean weeks later. Walk the steel, confirm it is ready, and clean what is not, because no amount of good spray technique recovers a bond the surface gave away.
Applying it: thickness, density, cure, and weather
Application is where the listed thickness and density actually get built, and the technique is in service of those numbers. For SFRM, you spray in passes to the listed thickness, watching that you are putting down material at the listed wet density, not just covering the steel. Thin material lays on fast and looks like coverage while reading short on the gauge, and overspray and rebound waste product and foul everything nearby, so mask and protect adjacent work before you start.
Weather and temperature gate the work. SFRM placed too cold, or allowed to freeze before it cures, is damaged the same way a winter concrete pour is damaged, and most manufacturers set a minimum air and steel temperature and require protection from freezing during cure. The cure itself takes time and ventilation; rushing finish over uncured material traps it. Follow the manufacturer's temperature, ventilation, and cure requirements, because they are part of how the tested assembly performs.
Intumescent is its own discipline. It goes on by coat to a measured dry film thickness, with the primer, basecoat, and topcoat each in their own DFT window and their own recoat and cure times. You measure wet film as you spray and confirm dry film after, and you respect the overcoat windows so the layers bond to each other. Too thin under-rates the steel; too thick in a pass sags or cracks. Apply both materials to the manufacturer's published rates and the listed design, and confirm conditions against the spec.
Density: the quality the gauge does not show
Thickness is half of SFRM quality and density is the other half, because the rating assumes the material is at the density the design was tested at. Spray that is the right thickness but too low in density is full of voids and under-insulates, and it will not hold its bond either. You cannot see density in a finished layer, which is why it gets measured.
Density is verified to ASTM E605, the same standard family that covers thickness, by taking a sample of known area and thickness, weighing it, and computing the as-applied density to compare against the listed minimum for that product and design. Standard-density and higher-density products have different required values, and the value is part of the listing, so the test confirms you applied the grade the design called for, at the density it was tested at.
On the floor, density is mostly a function of mixing and spraying the product the way the manufacturer specifies: the right water, the right pump and nozzle setup, and not over-thinning the mix to make it spray easier. Over-thinned SFRM goes on fast and tests light. Mix and apply to the manufacturer's instructions, and let the E605 density test confirm it rather than assuming the thickness alone proves the rating.
How is structural fireproofing inspected?
Structural fireproofing is proved by special inspection, and it is proved member by member, not by a walk-through. The building code requires special inspection of sprayed fire-resistive material, commonly under IBC Chapter 17, and the inspector measures the things that make the rating real: thickness, density, and bond, against the listed design for each condition. This is the quality control that separates a rated assembly from a coated one.
The thickness gets measured with a depth gauge at a defined number of readings per area, reported as an average and a minimum, against the listed thickness for that member. The density gets sampled and computed to ASTM E605. The bond gets pulled to ASTM E736 and compared to the required minimum, including any test needed to qualify a primer. The code sets sampling rates: bond test samples, for example, are commonly taken at a rate of not less than one per 2,500 square feet of sprayed area, or portion of it, in each story.
Above all, this inspection happens before the steel is concealed. Once the ceiling is up and the walls are closed, an under-thickness column or a failed bond is invisible until a fire finds it. Schedule the special inspector to check and document each area before it gets covered, and treat their thickness, density, and bond records as the proof of the rating, the same way the firestop work is proved by inspection of its listed systems. The special inspector and the AHJ control acceptance.
Measuring the thickness in the field
Thickness is measured with a depth gauge, a thin pin pushed through the SFRM to the steel that reads how much material is over the metal. The method, ASTM E605, sets how many readings to take and where, so the result is not one lucky probe but a representative picture across the member and the bay. The result is reported two ways, and both matter.
The average tells you whether the area is generally up to the listed thickness. The minimum tells you whether there is a thin spot, and the minimum is the one with teeth, because a member that averages fine can still have a flange tip or a corner below the listed thickness, and the fire does not care about the average. Listed designs and the spec set how much a single reading is allowed to fall below the design thickness and still pass, so a low minimum is not automatically a reject, but it is the number to chase.
Probe the hard-to-reach geometry on purpose: flange edges, the backs of members near the deck, intersections, and anywhere the gun could not get square. Those are where thickness goes short. Take the readings to the standard, record the average and the minimum per the listed design and the spec, and add material where the minimum falls outside what the design allows.
Testing the bond with a pull test
The bond gets proved with an adhesion and cohesion test to ASTM E736. The tester bonds a cap or plate to the face of the cured SFRM and pulls perpendicular to the surface, increasing force until something gives, and reads the force at failure over the known area to get a bond strength in pounds per square foot. The failure tells a story too: it can let go at the steel, which is an adhesion problem often pointing at the surface or the primer, or it can tear within the material, which is a cohesion problem pointing at density or mix.
The result gets compared to the required minimum for the building, which as noted climbs with height from roughly 150 psf up toward 1,000 psf on the tallest structures. If the spray does not make the required value, it is not rated, regardless of how thick it is. The fix is to find out why, correct the surface, the primer, or the mix, and re-spray the affected area, then test again.
Run the pull test at the code's sampling rate across floors and assemblies, not at one convenient spot, because bond varies with the surface it went onto and the surface varied across the building. Test to ASTM E736 at the required rate, compare to the value the adopted IBC sets for the building height, and let the special inspector's bond records stand as proof that the fireproofing will stay on the steel.
Patching damage, or the rating is void where the steel shows
The biggest in-service problem with SFRM is that other trades knock it off, and every bare spot is a hole in the rating. Once fireproofing is sprayed, the steel becomes a place to hang things, and the trades that come after drill it, bump it with lifts and ladders, and scrape it hanging duct, pipe, conduit, sprinkler, and ceiling support. Every gouge, every anchor through the coating, and every panel knocked loose exposes steel that is now unprotected at that spot.
An exposed spot is not a cosmetic problem. The rating is the listed thickness everywhere on the member, so a bare flange or a scraped web is under its rating right there, and in a fire the steel heats fastest exactly where the protection is missing. It does not matter that the rest of the member is perfect. The fire finds the gap.
So damaged fireproofing has to be patched back to the listed thickness and density with a compatible patching material, on a clean surface, the same as new work. That means walking the steel after the other trades are through, finding the damage, and repairing it before the ceiling closes, and it means coordinating so that whatever has to attach to the steel gets clipped or sleeved in a way the manufacturer allows instead of just gouging through. Patch to the listed thickness with the manufacturer's repair method, and inspect the repairs the same way as the original.
Fireproofing is not firestop
These two passive fire-protection jobs get confused constantly, and they are different work that protects different things. Structural fireproofing protects the structural member itself, the steel beam or column, so it keeps holding the building up in a fire. Firestopping seals the openings where pipe, duct, cable, and conduit pass through a fire-rated wall or floor, so the fire and smoke cannot pour through the hole around the penetrant.
They share a logic, which is why both are bound to tested, listed systems and proved by inspection, but they are not interchangeable and one does not cover for the other. A perfectly fireproofed column does nothing for a wide-open sleeve full of cable in the rated floor next to it, and a beautifully firestopped penetration does nothing for an under-protected beam. A rated building needs both done right.
Keep the trades and the records straight. Fireproofing is proved by thickness, density, and bond against a UL design for the member; firestop is proved against a listed through-penetration system rated to ASTM E814 or UL 1479 for the opening. The penetration-sealing side is its own discipline with its own listings and inspection, covered in the firestop guide. This guide is the steel-member side.
Installer safety on a fireproofing crew
Spraying fireproofing is overhead, dusty work in tight spaces, and it has hazards beyond the usual site list. Most of the spray goes up, so it rains back down on the applicator, which means eye and respiratory protection is not optional even before you get to what is in the material. Cementitious SFRM contains portland cement and can contain crystalline silica, and the dry mixing and the overspray put that dust in the air, so respiratory protection and dust control belong in the plan, sized to the actual exposure.
The work happens on the steel, which means at height and often from lifts and scaffolds, so the fall hazards of the structure are layered on top of the spray hazards. Overhead deck spraying puts the crew under their own falling material and working with their heads back for hours. Shafts, decks above ceilings, and other partly enclosed spaces can be confined spaces with poor ventilation, which matters for both the dust and the cure off-gassing.
Run it like the hazard it is. Use the respiratory protection the material's safety data sheet and your exposure assessment call for, control the dust, ventilate enclosed areas during spray and cure, and keep the fall and lift protection that the steel work already required. The material safety data sheet, the manufacturer, and the applicable OSHA requirements govern the specifics.
Sequencing fireproofing in the schedule
Fireproofing sits in a narrow window in the schedule, and getting it wrong is what creates the patching problem and the bond problem both. It goes on after the steel is erected and connected and after the deck is in place, because those are part of the assembly and because welding and bolting over fireproofing damages it. It goes on before the trades that hang off the steel cover it up, because they need the steel rated before they build under it, and because anything that runs over the steel makes that steel impossible to spray cleanly later.
The tension is that the spray needs to be early enough that the steel is still accessible and clean, and the protection needs to survive everything that comes after. Steel sprayed too early collects damage from every following trade; steel sprayed too late cannot be reached behind the work already in place. There is no perfect spot, which is why the trades that touch the steel after fireproofing have to know it is there and have to attach without destroying it.
Plan the sequence with the rating in mind: spray after steel and deck, inspect before concealment, and protect and patch through the rest of the build. Coordinate the follow-on trades so attachments to fireproofed steel are made the manufacturer's way, and walk it for damage before the ceilings close. The schedule and the GC's coordination control the timing; the rating controls why it matters.
What to document
Fireproofing that cannot be proved on paper is fireproofing that gets re-inspected, re-sprayed, or rejected at closeout. The record is what ties the material on the steel to the tested assembly that gives it a rating, and it is what the AHJ, the special inspector, and the owner rely on after the steel is covered.
Capture the listed UL design number and rating for each condition, the product and its data sheet, the required and applied thickness with the average and minimum readings, the required and applied density, the primer on the steel and any compatibility test or bonding agent used, the bond test results against the required value, the locations and sampling rates, the special inspection reports, and the patching and repairs with their own thickness and bond confirmation. A field tool like FieldOS keeps those records and photos tied to the member and the bay so the proof survives the schedule and the inspection.
| Item | Requirement | Note |
|---|---|---|
| UL design number and rating | Per drawings and adopted IBC | Ties the steel to a tested assembly |
| Product and data sheet | Per spec and listing | Brand and grade are part of the listing |
| Thickness, average and minimum | Listed thickness, ASTM E605 | Minimum is the one that fails you |
| Density | Listed minimum, ASTM E605 | Low density under-insulates and breaks bond |
| Primer and bond | ASTM E736, IBC value by height | Qualify or use approved bonding agent |
| Special inspection reports | IBC Chapter 17 sampling rates | Before concealment, member by member |
| Patching and repairs | Back to listed thickness | Re-inspect repairs as new work |
Common mistakes
- Applying less than the listed thickness, which silently voids the rating where the spray is thin.
- Spraying SFRM over an incompatible shop primer so it never bonds and sheets off later.
- Substituting a material, deck, or member outside the tested listed assembly and assuming the rating carries.
- Hitting the thickness but missing the listed density, so the layer is voided and under-insulates.
- Leaving knocked-off and gouged fireproofing unpatched, so the rating is void at every bare spot.
- Skipping or rushing the special inspection of thickness, density, and bond before the steel is concealed.
- Carrying one thickness across the whole frame instead of the per-member value the schedule gives.
- Confusing fireproofing with firestop and leaving penetrations unsealed or members under-protected.
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 required ratings come from the building code, commonly the IBC, by construction type, height and area, and member, and the architect and engineer set them on the drawings. How a given rating is actually achieved comes from a tested, listed assembly, most often a UL design number, which fixes the material, the thickness for the member, the steel, and the conditions that were tested together to ASTM E119. Build to the design number on the drawings and treat it as the only path to the rating.
The field tests that prove the work are ASTM standards. ASTM E605 covers thickness and density of SFRM. ASTM E736 covers the cohesion and adhesion bond strength, with the required minimum value set by the IBC and increasing with building height, on the order of 150 psf for ordinary buildings up toward 1,000 psf for the tallest. Special inspection of sprayed fire-resistive material, including the sampling rates, is required under IBC Chapter 17. The exact section numbers and required values shift between code cycles and by jurisdiction, so confirm them against the adopted IBC edition and local amendments before citing them.
Through-penetration firestop, the related but separate trade, is tested to ASTM E814 or UL 1479 and is covered in the firestop guide. Across all of it, hedge the rating, the thickness, the assembly, and the bond to the UL design, the manufacturer's published data, the project specification, and the AHJ and special inspector, in that order, because those are what control acceptance. The three things that do not bend: the rating is the listed thickness on the listed assembly, the bond holds only on clean compatible steel, and the thickness, density, and bond all get inspected and every bit of damage gets patched.
Terms and units
Fireproofing carries a vocabulary that crosses the drawing set, the listing, and the test reports, and the same idea shows up under more than one name. The terms below are the ones that decide whether what is on the steel matches what was tested.
Thickness is given in inches or millimeters for SFRM and in mils or millimeters of dry film thickness for intumescent. Density is in pounds per cubic foot. Bond strength is in pounds per square foot. Ratings are in hours. Confirm every value against the listed design and the spec rather than a remembered number.
- Structural fireproofing
- Insulation applied to structural steel so it stays below its failure temperature in a fire and keeps carrying load for the rated time
- SFRM
- Spray-applied fire-resistive material, a cementitious or gypsum spray with fiber, for concealed steel and decks
- Intumescent
- A thin reactive coating that swells and chars into an insulating foam when heated, for exposed architectural steel
- Fire-resistance rating
- The hours an assembly was proven to resist a standard fire test, set by the building code by type and member
- Listed / UL assembly
- A tested design, such as a UL design number, fixing the material, thickness, steel, and conditions that earn the rating
- Thickness / DFT
- The applied depth that delivers the rating; dry film thickness is the intumescent equivalent
- W/D or A/P ratio
- Section factor, weight or area over heated perimeter, that drives the thickness; heavier members need less
- Density
- Mass per unit volume of applied SFRM, verified to ASTM E605 against the listed minimum
- Bond strength
- Adhesion and cohesion of SFRM to the steel, tested to ASTM E736, with the IBC minimum rising by building height
- Patching
- Repair of damaged fireproofing back to the listed thickness and density, or the rating is void at that spot
- Firestop
- The separate trade that seals penetrations through rated walls and floors, tested to ASTM E814 or UL 1479
FAQ
What is structural fireproofing?
Structural fireproofing is insulation applied to structural steel so it stays below the temperature where it loses strength in a fire. Steel softens and a loaded frame can collapse as it heats, so the fireproofing buys hours of time for escape and firefighting. That time is the fire-resistance rating set by the building code.
What is the difference between SFRM and intumescent fireproofing?
SFRM is a thick plaster-like spray of cement or gypsum and fiber that insulates as a soft layer, used on concealed steel and decks because it is cheap and rough-looking. Intumescent is a thin paint-like coating that swells and chars into a foam when heated, used on exposed architectural steel because it can be finished to look like paint.
Why does fireproofing thickness matter so much?
The fire rating comes from a furnace test of a specific material at a specific thickness on a specific steel size, published as a listed assembly. Apply less than that listed thickness and the steel is under-rated, but the shortfall is invisible once the wall closes. Too thin silently voids the rating, so the thickness is the whole game.
What is the difference between fireproofing and firestop?
Fireproofing protects the structural member, the steel beam or column, so it keeps carrying load in a fire. Firestop seals the openings where pipe, duct, and cable pass through rated walls and floors so fire cannot spread through the hole. They are different trades with different listed systems, and a rated building needs both done right.
How is structural fireproofing inspected?
A special inspector measures thickness with a depth gauge to ASTM E605, samples density to ASTM E605, and pulls the bond to ASTM E736, member by member against the listed design, before the steel is concealed. Thickness reports both an average and a minimum, and the bond is compared to the IBC value for the building height.
Why does spray-applied fireproofing fall off the steel?
SFRM falls off when it never bonded, usually because of an incompatible shop primer or a dirty steel surface. Many primers are too slick for cement or gypsum SFRM to grip, and oil or loose scale breaks the bond. The fix is a manufacturer-approved bonding agent or clean bare steel, confirmed by a pull test to ASTM E736.
What happens if another trade knocks the fireproofing off?
Every bare spot is a hole in the rating, because the rating is the listed thickness everywhere on the member. The steel heats fastest where protection is missing, so a gouged web or scraped flange is under-rated. Damaged fireproofing must be patched back to the listed thickness on clean steel and re-inspected, or the rating is void there.
How thick does fireproofing need to be for a 2-hour rating?
There is no single number, because the thickness for a 2-hour rating depends on the member size and the listed assembly. A heavier column needs less material than a light beam for the same hours, and the value comes from the UL design's table. Use the listed thickness for the exact member, confirmed with the manufacturer and spec.
What is the W/D ratio in fireproofing?
W/D, the section factor, is the member's weight per foot divided by its heated perimeter, and it drives the thickness. A high W/D means more steel mass behind less exposed surface, so the member heats slowly and needs less protection. A light member has a low W/D, heats fast, and needs more, so thickness varies by member.
Does fireproofing have to follow a tested listed assembly?
Yes. The rating belongs to a complete tested recipe, commonly a UL design number, that fixes the material, thickness, steel, and conditions tested together. Substituting the product, deck, or member outside that design voids the rating even if the work looks right. Where the field condition has no listing, get an engineering judgment accepted by the AHJ before concealment.
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.