ASTM E2174 has the special inspector witness at least 10% of each penetration firestop type during installation, or destructively test at least 2% (never fewer than one) of each type per floor or 10,000 sq ft area. ASTM E2393 uses 5% of each joint type’s linear footage witnessed, or one sample per 500 linear feet.
Fire Examiners field team · Updated October 2026
Most project teams know ASTM E2174 and E2393 govern firestop special inspection. Fewer know how many firestops the inspector must check, or what a failure does to that number. Our companion Insight, ASTM E2174 and ASTM E2393 Explained for GCs and Owners, covers what each standard is about and what a usable report looks like. This one is about the numbers: the percentages, how “type” drives the count, and what a failed sample triggers.
Where the code sends you to E2174 and E2393
Under the 2018 IBC, Section 1705.17 requires special inspection of tested and listed penetration firestops, fire-resistant joint systems and perimeter fire barrier systems in high-rise buildings and in buildings assigned to Risk Category III or IV. The 2021 IBC renumbered the section to 1705.18 and added fire areas containing Group R occupancies with an occupant load over 250. In both editions, penetration firestops are inspected by an approved agency in accordance with ASTM E2174, and joint systems in accordance with ASTM E2393. For when the requirement applies and how to plan for it, see our IBC 1705.17 firestop special inspection guide.
The code names the standards. The standards set the sample size.
ASTM E2174 percentages for penetration firestops
E2174 gives the inspector two methods. The method chosen sets the frequency:
- Witnessing installation: the inspector is on site while work is in progress and randomly witnesses at least 10% of each type of firestop system being installed.
- Destructive testing after installation: the inspector removes and examines completed firestops, and the installer repairs them. The minimum is 2% of each type, but never fewer than one, per floor. A floor larger than 10,000 sq ft is divided into areas of 10,000 sq ft or less, and the minimum applies to each area.
Witnessing means watching the system go in. Looking at a finished seal afterward doesn’t count as witnessing. Either way, the inspector also verifies that the firestops required by the approved documents were actually installed.
ASTM E2393 percentages for joints and perimeter fire barriers
Joint sampling is measured in linear feet instead of counts:
- Witnessing installation: at least 5% of the total linear feet of each type of fire-resistive joint system being installed.
- Post-installation inspection: at least one sample per type of joint system per 500 linear feet. The method must be approved by the authorizing authority and the AHJ, and can be destructive verification, disassembly and reinstallation, or visual verification where that establishes conformance.
Mechanical joint systems are the exception. E2393 calls for them to be inspected by witnessing installation, not after the fact.
Why “type” sets the real number
Both minimums apply per type, so the way types are defined matters as much as the percentage. The standards don’t lock in one definition. Their explanatory notes describe a type as a unique combination of parameters:
- Penetrations: penetrant (metal pipe, plastic pipe, cabling), firestop material or device (sealant, collar, intumescent), and substrate (gypsum wall, concrete floor, composite deck).
- Joints: joint type (head-of-wall, wall-to-wall, floor-to-wall, floor-to-floor, perimeter), material or system, and substrates.
Because each type needs at least one sample, many distinct listed systems can mean far more inspections than a flat percentage suggests.
A simple illustration (hypothetical numbers): one floor area of 10,000 sq ft or less has 150 plastic-pipe penetrations through a gypsum wall, sealed with one listed system. Witnessing means watching at least 15 go in. Destructive testing means at least 3. If that floor also has 2,000 linear feet of one head-of-wall joint system, E2393 calls for at least 100 linear feet witnessed, or at least 4 post-installation samples.
What happens when a sample fails
This is the part that moves schedules. Under both standards:
- The nonconforming firestop or joint must be repaired or replaced and re-inspected.
- One full additional round of inspection, at the original sample size, is required for that type.
- If 10% or more of the inspected quantity of a type doesn’t comply, inspection of that type stops. The installer has to inspect its own work and repair or replace the similar firestops in that area before the special inspector resumes.
The standards also set timing. They require inspection forms to go to the authorizing authority and installer within one working day after an area is inspected, and they require follow-up inspection of repaired work within two working days of the installer’s notice. The final report summarizes the quantity inspected per type, the method used, and deficiency percentages by type and overall.
How Fire Examiners builds the sample into the inspection
Fire Examiners inspectors are IFC-certified and inspect to ASTM E2174 and ASTM E2393 under IBC 1705.17. We are not the installer. Across 45,000+ inspections, we’ve refined how the sample gets planned and tracked so the required percentages are met while the work is still open:
- Plan the sample before work starts. We review the drawings, approved submittals and listings, define the firestop and joint types with the project team, and estimate counts and linear footage so the minimum sample per type, per floor and per area is known up front.
- Track every type by floor and area. Each type is logged against the floor and 10,000 sq ft area it belongs to, so it’s clear which locations still need samples before ceilings and walls close.
- Witness installation in progress. Most of our on-site work is witnessing: we watch firestops and joints go in and record measurements as work is placed, which also covers mechanical joint systems. How much of a project is witnessed versus destructively tested depends on the project type and on how early or late we’re brought onto the project. The later we arrive, the more work is already closed and must be verified destructively.
- Log destructive samples with photos. Each destructive sample is located, measured against the listing and photographed; the installer repairs it per the manufacturer’s procedure.
- Document failures and re-sample. We communicate discrepancies directly to the contractor and everyone else involved. Deficiencies are marked in the field and on the form, re-inspected after repair, and followed by the additional sample the standard requires. If a type reaches the 10% threshold, we stop that type and resume once the installer’s review is complete.
- Report to the people who sign off. We get reports out within one day. Each form records the type, location, method and quantities inspected, and the final report gives the AHJ, the authorizing authority and the registered design professional the counts, methods and deficiency percentages by type.
We inspect. We do not install, supervise installation or direct the crew. The standards require that independence. For the full service, see third-party firestop special inspection. If repeated failures are slowing closeout, see firestop deficiency remediation.
FAQ
Is the E2174 10% or 2% figure based on the whole building?
No. It applies to each type of firestop system. For destructive testing, it also applies per floor or per area of 10,000 sq ft or less, with a minimum of one per type in each.
Can a quick look at finished firestops count as witnessing?
No. Witnessing means observing the installation while it happens. Checking completed work is a post-installation inspection, which for penetrations means destructive verification under E2174.
Does one failed sample mean every firestop gets opened?
Not automatically. The failed firestop is repaired and re-inspected, and another full sample of that type is required. Inspection of that type stops only if 10% or more of the inspected quantity fails, and then the installer reviews and corrects the area before inspection resumes.
Who decides what counts as a “type”?
The standards let the authorizing authority set the types and raise the minimum. In practice, agree on types with the design team, installer and inspector before installation.
Plan your sample before the walls close
If your project needs E2174 or E2393 special inspection, talk to us before installation starts so the sample is planned around the work instead of reconstructed at closeout. Contact Fire Examiners · info@fire-examiners.com · 727-537-0047
