Industries / Construction and property / Commercial and residential construction / Drywall
Fire and sound walls
This is the work of building walls that hold back fire for a rated time, or keep noise from passing between rooms, by following a tested design layer for layer and screw for screw.
7
Milestones achieved, from 1918 to 2018.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1918
A standard fire test
An international fire prevention congress in 1910 failed to agree on one temperature-time history for fire tests. In 1918 ASTM published standard C19, now E119, which heats a wall or floor in a furnace along a fixed temperature curve; Britain's first standard with such a curve followed in 1932. Hourly ratings for gypsum walls are still set by this test, which UL runs as UL 263. A 2021 review in Fire Technology found that no surviving historical literature supports the temperature points that define the curve.
LessonThe yardstick behind every rated wall is an agreed convention, not a measured fire.
Credit
ASTM; the review by John Gales, Bronwyn Chorlton and Chloe Jeanneret
1939
Loose layers stop sound
In March 1939 the US National Bureau of Standards gathered about 16 years of its sound measurements on walls and floors into one report. It found that a solid, uniform wall must be "excessively heavy" to insulate well, that a wall built in loosely connected layers insulates far better, and that small openings "may almost completely destroy" a wall's sound insulation.
LessonHow a wall is put together matters more than what it is made of.
Credit
V. L. Chrisler, National Bureau of Standards
1946
Fire-resistant board
By the account of a 1974 National Bureau of Standards report, experiments with additives in the gypsum core in the early 1940s led to the first patent on fire-retardant gypsum wallboard, filed in 1943 by Michael Croce and Clarence Shuttleworth of Certain-teed Products and issued on 11 October 1950, and the first Underwriters' Laboratories listing of such board appeared in 1946. Glass fibers spread through the core hold it together after the fire has driven off its water, and ASTM C36 came to define the product as "Type X". By 1974 about a fifth of the US industry's output of more than 14 billion sq ft (1.3 billion m²) of gypsum board a year went into fire-rated walls, shafts and ceilings, by that report's count.
Credit
Michael Croce and Clarence Shuttleworth (Certain-teed Products); Underwriters' Laboratories
2003
England tests finished homes
England's sound rules, Part E of the Building Regulations, have required sound tests in the finished building since 1 July 2003 for rooms for residential purposes, such as hotel rooms, and for flats and houses made by converting other buildings, and since 1 July 2004 for new houses and flats, with at least one set of tests for every ten homes in a group. A new wall between homes must reach 45 dB DnT,w + Ctr, a field measure of how much quieter the next home is, adjusted for low-pitched noise. Since 1 July 2004 builders of new homes may instead use "robust details", high-performance wall and floor designs from Robust Details Ltd that are expected to be reliable enough not to need the test.
LessonA test in the finished building checks the wall as it was built, not only as it was drawn.
Credit
UK government (Approved Document E); Robust Details Ltd
2008
Shaft walls must resist blows
Shaft wall, a system of special studs and thick gypsum liner panels that encloses stairs and elevator shafts and can be built from one side, was patented by USG and first used in the World Trade Center towers, by the Cold-Formed Steel Engineers Institute's account. NIST's investigation of the towers' collapse recommended that stairwells and elevator shafts be designed to survive impacts and blasts as well as fire. In September 2008 the International Code Council approved changes for its 2009 codes that harden exit stair and elevator shaft enclosures in buildings over 420 ft (128 m), in buildings of 75 to 420 ft where their failure would put lives at risk, and in essential buildings such as hospitals. In 2009 NIST pressed three common two-hour gypsum shaft walls with air until their studs buckled: the wall with 2.5 in (64 mm) studs held 2 psi (14 kPa) and the two with 4 in (102 mm) studs held 3 psi (21 kPa).
LessonA wall rated for fire can still fail under a blow, so the walls around escape routes now need a strength rule as well.
Credit
NIST, with tests by H. S. Lew; International Code Council
2010
Wrong screws undo the channel
Resilient channel is a thin, springy metal strip screwed across the studs, to which the board is screwed so that it does not touch the studs directly. In laboratory tests reported in 2010, John LoVerde and Wayland Dong replaced the 1 in (25 mm) board screws in a resilient channel wall with 1-5/8 in (40 mm) screws long enough to reach the studs, a "short circuit", 5 to 35 at a time. The wall's laboratory rating (Rw, the weighted sound reduction index, an international cousin of the US sound transmission class, STC) fell almost in step with the share of shorted screws, from 60 to 51 when all 35 were shorted. They wrote that before their study there was very limited systematic data on the effect.
LessonA few wrong screws hidden behind the board can cost a wall much of its sound rating.
Credit
John LoVerde (Veneklasen Associates) and Wayland Dong (Paul S. Veneklasen Research Foundation)
2018
Missing walls between flats
On 27 April 2018 Leopalace21, a Japanese apartment builder and landlord, disclosed that walls between units, which Japan's rules require to block sound and to reach up into the attic or ceiling space to stop fire spreading, were missing or incomplete in buildings it had built. Disclosures in February and May 2019 added walls filled with the wrong material for sound insulation, and walls, exterior walls and ceilings that did not match their approved fire-resistant designs. In Saitama Prefecture alone, 3,237 buildings were affected by the end of June 2019, by the prefecture's count. In July 2019 Japan's Fire and Disaster Management Agency listed 267 buildings that might breach fire law, with 3,964 more under investigation.
LessonA wall hidden above a ceiling can be seen only before the ceiling is closed, so that is when it has to be checked.
Credit
Saitama Prefecture; Fire and Disaster Management Agency, Japan
Watch for
Fire tests modeled first
UL Solutions launched a service that uses thermal models of its UL 263 furnace, including how gypsum gives up its water as it heats, to let board makers screen new formulations before full-scale fire tests. It calls the models "validated" but gives no accuracy figures, and it says modeling does not replace the tests required for certification.
Achieved whenUL Solutions or a board maker publishes modeled and measured fire-resistance times for the same boards, or an outside laboratory reproduces the predictions.
Claimed by
UL Solutions, October 2026
Open
01
Find shorted channels after boarding
A handheld method that finds, after the board is up, the screws that have gone through a resilient channel into a stud, tested blind on mock walls with short circuits planted by someone else.
The briefCredit · unclaimed02
Screw spacing check by photo
A phone tool that photographs a fire-rated wall after the board is hung and before it is taped, finds every screw head, measures the spacing along each board edge and across the field, and flags any gap wider than the tested design allows. Test it on mock walls against a tape measure.
The briefCredit · unclaimed
Open 01 · the brief
Find shorted channels after boarding
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A handheld method that finds, after the board is up, the screws that have gone through a resilient channel into a stud, tested blind on mock walls with short circuits planted by someone else.
Why it is still open
A December 2025 quality bulletin from AECOM Hunt tells crews to watch for board screws reaching the studs and notes that a 1-5/8 in screw creates a short where a 1 in screw is correct, so the check is made by eye while the board goes up. LoVerde and Dong measured how much each short costs in the laboratory. This research found no field tool that finds shorts behind finished board.
Built with what exists
Mock wall panels about 1.2 by 2.4 m (4 by 8 ft) on wood and steel studs with resilient channel; 1 in and 1-5/8 in drywall screws; a handheld wall scanner that senses metal and studs, such as a Bosch D-tect 120 or Walabot DIY 2, to map screws against stud lines; a small rubber-tipped tap hammer and a USB accelerometer or phone to record tap responses at each screw; Python with SciPy to compare the responses.
First step
Build one panel on wood studs with resilient channel, drive five long screws at recorded spots, and see whether a scanner and a tap test pick them out.
Done looks like
10 panels with 0 to 35 shorted screws each, placed and recorded by another person; the method scores every screw, and the detection rate, false alarms per panel and time per panel are published with the raw readings.
Open 02 · the brief
Screw spacing check by photo
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A phone tool that photographs a fire-rated wall after the board is hung and before it is taped, finds every screw head, measures the spacing along each board edge and across the field, and flags any gap wider than the tested design allows. Test it on mock walls against a tape measure.
Why it is still open
A Walls & Ceilings explainer on the Gypsum Association's GA-600-2024 fire manual says a wall's fire performance rests mainly on the number and type of board layers, how they are attached and how far apart the fasteners are. National Gypsum prints GridMarX guide marks on some boards so that inspectors can check its wider fastener patterns by eye, and Roboflow, which sells computer vision tools, describes checking fastener spacing on factory-built wall panels but publishes no test. Image work on drywall so far finds panels, frames and insulation for progress tracking (Trzeciakiewicz and colleagues, 2025), and open drywall projects on GitHub segment cracks and taped joints, not screws. This research found no tested tool that checks screw spacing on a rated wall on site.
wconline.comnationalgypsum.comroboflow.comarxiv.orggithub.com
Built with what exists
A phone camera; printed ArUco markers for scale, read with OpenCV; an open object detector of the YOLO family, or a promptable segmenter such as SAM 2, fine-tuned on hand-labeled screw heads; the screw pattern from a published fire-rated design entered as a simple rule file.
First step
Hang 10 boards on a bench frame, tape ArUco markers to the corners, and photograph and hand-label every screw head.
Done looks like
30 mock boards hung to known patterns with 60 planted gaps over the limit, scored by the tool; a published detection rate, false alarms per board and spacing error in millimeters against tape measurements, with the photos and labels released under an open license.