The five-question diagnostic behind the tool, the five-indicator framework that distinguishes pre-1940 cylinder glass from modern float and restoration-grade reproductions, the configurations that route out to glaziers and specialists, and the conservation-grade protocol that protects panels in continuous service for a century-plus from the single most preventable failure in heritage residential. The pre-war glazing reading I run on the I-95 corridor, ported to software.
What the Heritage Glass Triage does, in five points:
The tool exists to put the reading of the panel in front of the operator before any chemistry touches the glass. Heritage panels have been in service for a century-plus and the scratches a steel scraper leaves on weathered cylinder glass are permanent. Identifying the panel correctly is the entire job.
There is a particular phone call I get from operators new to the Northeast heritage residential market, usually within their first year of working the I-95 corridor neighborhoods where pre-1900 stock concentrates — the brownstone blocks in Brooklyn, Federal Hill in Providence, Beacon Hill and Back Bay, Old Town Alexandria, the West End in Hartford. The call goes some version of this: I did a clean on a historic property in Society Hill last week, the homeowner just called me, she says there are scratches on the front parlor windows. Did I do that?
You probably did. Not because you were careless. Because you treated a heritage panel as production glass, and the production toolkit — the steel scraper, the ammonia cleaner, the squeegee with a corner blade — is the exact toolkit that damages pre-1940 cylinder glass in ways that no operator at production pacing notices until the homeowner sees the morning light hit the panel three days later.
The Heritage Glass Triage is the tool we built to put that reading of the panel in front of the operator before any chemistry touches the glass. It walks through the same five-indicator protocol I run on every pre-war property I work — building age, waves, tactile character, edge color, and glazing system — and produces a verdict on whether the panel is heritage, restoration glass, modern replacement in an original sash, modern replacement in a modern frame, or damaged beyond what the cleaning protocol can recover. Each verdict carries its own protocol. The heritage verdict is the protocol I have been running on the Northeast corridor for two decades; it is also the protocol Wade Marler runs on the Old Louisville Victorian heritage stock and the protocol Cal Hatcher runs on the East Nashville Craftsman belt and the Memphis Midtown Victorians.
This piece is the methodology behind the tool. What each step is testing for, why the five-indicator framework converges on the right answer when no single indicator does, what the three classes of modern panel are and why the differences matter for cleaning, and the configurations that route out to a glazier or a leaded-glass specialist rather than back into the cleaning workflow. If you want the longer narrative treatment of cylinder-glass history, manufacturing process, and the substrate properties that drive the protocol, Wade Marler's article is the deep reference. This piece is for the user who wants to understand why the tool recommended what it recommended.
Pre-1940 cylinder glass is softer than modern float glass, thinner than modern float glass, and one mistake away from a panel that cannot be replaced. The reading of the panel is the entire job. Once the chemistry or the scraper is on the wrong panel, the damage is already done.
The asymmetry of consequences is what shapes the protocol. The cost of running the heritage protocol on a panel that turns out to be modern is a slower clean and slightly conservative chemistry — a recoverable inefficiency. The cost of running the production protocol on a panel that turns out to be heritage is a permanently scratched panel that the homeowner will see for the rest of the time the house stands. The tool is designed around this asymmetry: it biases toward the conservative protocol on ambiguous indicators, and it routes uncertain cases to careful inspection rather than to a confident-but-possibly-wrong verdict.
This is the rule I want every operator entering the heritage segment to internalize. The pacing is the discipline. The pricing follows. And the reading of the panel comes before the chemistry, every time, without exception.
The first question is the largest filter. The cylinder-glass process that produced pre-1940 residential window glass was replaced by the Pilkington float-glass process over the period roughly 1959 to 1975. The float process was patented in 1959; commercial US adoption was substantially complete by the mid-1970s. The implication: properties built after about 1975 will not have original cylinder glass, regardless of how much they architecturally suggest a heritage style. Pre-1900 properties almost certainly have at least some surviving original cylinder glass; the 1900–1940 properties have it on original panels; the 1940–1975 transitional period is mixed and benefits from careful visual inspection.
The pre-1900 vs 1900–1940 split is finer-grained than the tool's verdict structure requires — both ranges produce the same diagnostic path through the rest of the questions — but the split is preserved in the question because it matters editorially. A pre-1900 building with surviving cylinder glass is a different conservation conversation than a 1920s Sears-kit house with surviving cylinder glass, and the homeowner conversation about replacement options diverges accordingly.
The "I don't know the building age" answer routes to a different terminal state if the visual indicators do not all converge. Building age can be looked up through public property records (most US counties make this freely accessible online), or through the original deed, or by asking the property owner. The tool's age-uncertain verdict directs the user to do the lookup before applying any chemistry; the visual indicators carry a softer-confidence diagnosis when all four converge, but the age anchor is the cleanest path to a confident verdict.
This is the diagnostic move. The cylinder-glass manufacturing process produced sheets that were never perfectly flat — the splitting and flattening operation left subtle waves and ripples that no float-glass process replicates. Stand to the side of the panel at a low angle, ideally with the panel reflecting something with straight lines, and the cylinder glass will visibly distort the reflected scene. Float glass will not.
The waves are the single most distinctive visual indicator of pre-1940 cylinder glass and the indicator the tool weights most heavily. The absence of waves is conclusive in the negative direction: if the reflected world is undistorted in the panel, the panel is not cylinder glass and the heritage protocol does not apply.
The presence of waves is not conclusive in the positive direction, because modern restoration glass — most notably Bendheim in the US market — is manufactured by a partially-revived cylinder process specifically to match the appearance of pre-1900 panels for preservation work. The restoration glass shows waves. This is why the tool treats the waves answer as the first of multiple indicators rather than a single decisive test, and why the tactile and edge-color tests at the next step exist.
The "subtle waves" answer is a softer-confidence positive than "strong waves." The "mixed waves — some panels have them and others don't" answer is the diagnostic signature of a mixed-glazing property and the tool routes accordingly.
Two tests at this step, both run on the panel itself rather than from across the room.
Tactile surface character. Run a clean fingertip across the surface of the panel. Cylinder glass has subtle surface unevenness that you can feel as faint texture — a slight rippling, a barely-perceptible roughness that modern float glass does not have. The texture comes from a century-plus of weathering on a surface that was already not perfectly flat from the manufacturing process. Modern float glass is flat to the touch in a way that is industrially impressive.
Edge color. Look at the edge of the panel where it meets the putty or gasket bevel. Pre-1940 soda-lime glass typically had higher iron content than modern float glass (the sand sources and refining technology of the period produced glass that ran slightly greener at the edges), and you can see the faint greenish tinge along the cut edge through the glazing. Modern low-iron glass does not show this. The greenish tinge is subtle — the edge does not glow green, it just is not perfectly clear — and seeing it reliably is a calibration that comes from doing it several times.
Neither test is conclusive on its own. Together they are diagnostic. The combination of "rough texture and greenish edge" is the strongest tactile/visual confirmation of heritage glass; the combination of "smooth and clear edge" is the strongest negation. The mixed cases — rough but clear, smooth but greenish — push toward "verify with the other indicators" rather than committing to a verdict.
This step is also where the tool's restoration-glass detection logic lives. Restoration glass shows waves (the manufacturing intent), but it does not have the century-plus of weathering that produces tactile texture, and it usually does not have the pre-1940 iron-content edge color. When the waves answer is positive but the tactile-and-edge answer is negative, the tool concludes restoration glass rather than heritage glass, and the protocol switches accordingly.
The glazing system is the structural fingerprint of the era. Pre-1940 cylinder glass was almost always bedded into wood sashes with linseed-oil glazing putty. The putty deforms slightly with seasonal humidity, which is why cylinder glass survived in service for a century-plus — the system absorbed the thermal cycling that a rigid mounting could not have tolerated. Modern aluminum-framed and vinyl-framed glazing uses rubber gaskets and silicone, which are mechanically different and were not in residential service before the 1960s.
The four indicator-relevant configurations the tool tests:
Wood sash with putty bevel is the heritage glazing system. When this is present on a pre-1900 or 1900–1940 building, the original glass is probable.
Wood sash with modern rubber gasket or silicone indicates a reglazed original sash. The sash survived, the original glass did not — almost always replaced with modern float glass during a previous maintenance pass. The tool's verdict for this configuration is "modern float in an original sash" with the protocol note that the sash and the putty (where any putty survives) still warrant conservative handling even though the glass does not.
Aluminum or vinyl frame indicates full window replacement. Both the sash and the glass were replaced together, almost always as part of an energy-efficiency upgrade. The window is modern in every respect; standard cleaning protocol applies.
Mixed frame — some panels original wood-putty, some modern frames — is the diagnostic signature of a property that has had progressive replacement work over the decades. The tool routes this to the panel-by-panel-triage verdict.
The fifth answer option — leaded glass or stained glass — routes out of the cylinder-glass diagnostic entirely. Those panels have their own conservation universe (lead came chemistry concerns, museum-conservation-derived protocols, different damage modes) and the tool sends the user to a leaded-glass conservator or stained-glass studio with conservation experience.
The sixth option — "I can't tell, the frame is painted or sealed" — routes the user to the Frame Substrate Identifier tool first, on the grounds that the frame question is solvable with the existing diagnostic toolkit and is worth resolving before applying chemistry.
The condition question routes severe-damage cases out of the cleaning workflow entirely and biases the protocol for the routine-cleaning cases.
Clean and intact runs the conservation-grade protocol on whatever the diagnosis turns out to be.
Faint surface haze is treated as a cautionary case. On a confirmed heritage panel, the haze is most often century-old weathering — part of the panel rather than contamination — and attempts to remove it succeed by polishing the surface, which destroys the heritage character. The tool's protocol copy is explicit that surface haze on heritage glass is not a cleaning problem to solve.
Mineral spots or sprinkler residue can be cleaned on heritage glass with conservation-grade descaling chemistry — citric acid solutions or dilute oxalic acid, not commercial mineral removers, and not at the concentrations that production residential uses on modern panels.
Organic residue (sap, pollen, light dirt) runs the gentlest rungs of the solvent ladder only. The Solvent Ladder Selector handles the general case; on heritage glass, the available rungs cap at Rung 2 (isopropyl alcohol applied with a cloth) regardless of what the residue chemistry wants. The trade-off — slower work, sometimes incomplete removal — is the trade-off the protocol accepts.
Visible scratches route to the damaged-heritage-glazier verdict when the other indicators are heritage-positive. Scratches on cylinder glass are typically permanent — cerium-oxide polishing can sometimes remove them, but the polish removes the wavy-glass character along with the scratch, leaving a panel that is technically clean but no longer historic. The next consultation is a glazier with heritage experience, not a cleaner.
Cracked or structurally compromised routes universally to the glazier verdict regardless of whether the panel was confirmed heritage. Cleaning a cracked panel is the wrong question; the right question is replacement.
The tool's diagnose function counts heritage-positive indicators across the first four questions: heritage-era building (1 point), visible waves (1 point), rough-with-greenish-edge tactile (1 point, with mixed cases earning 0.5), and wood-putty glazing (1 point). The maximum is 4. The verdict thresholds are 4.0 for confirmed heritage and 2.5 for likely-heritage-verify; below 2.5 the panel routes to modern-replacement or post-1975-modern depending on the building age.
The condition question and the post-1975 short-circuit operate before the counting logic. A post-1975 building can never produce a heritage verdict regardless of visual indicators. A cracked panel always routes to the glazier verdict. A scratched panel on a confirmed heritage building with wood-putty framing routes to glazier; on other configurations the scratch is treated as a substrate observation rather than a verdict-driver.
The restoration-glass detection logic is an explicit override: waves + smooth-tactile produces the restoration-glass verdict regardless of building age or frame, because the tactile signature of restoration glass is distinct enough from heritage cylinder glass to be diagnostic.
The mixed-glazing-property verdict fires on either "mixed waves" or "mixed frame" — both configurations indicate a property that has had progressive replacement, and the panel-by-panel triage approach is the right protocol response.
The triage has honest limits.
It cannot identify the specific manufacturer or year of a heritage panel. Pre-1940 cylinder glass was produced by hundreds of US and European glassworks across more than a century; the panel-level attribution is generally not recoverable. The tool produces a substrate-class verdict rather than a maker-class verdict.
It cannot distinguish 1830s panels from 1920s panels. The cylinder-glass process was substantively unchanged across that span. A 1920s panel and an 1840s panel both produce the same diagnostic indicators, and the cleaning protocol is the same for both. The age difference matters for preservation value but not for cleaning chemistry.
It cannot assess panels behind storm windows or interior shutters without removal of the obstruction. The visual inspection requires direct light access and tactile access to the panel surface. When the configuration prevents inspection, the right answer is to default to the heritage protocol on the assumption that the surviving original panel is what is being protected; the inspection can be deferred to a later visit when the storm windows are removed for season.
It cannot identify the specific restoration-glass grade in use. Bendheim produces several restoration grades corresponding to different periods and different intensities of optical distortion; specifying the grade for replacement is a glazier-and-preservation-board conversation rather than a cleaning-protocol decision. The tool's verdict for restoration glass treats all grades as functionally modern from a cleaning standpoint, which is correct.
It cannot run on leaded or stained glass. Those panels have their own conservation universe and the tool routes out of the workflow entirely when this configuration is detected.
It cannot assess the property-level preservation value of a heritage panel. The same heritage panel may be worth replacing in a low-visibility position and worth preserving even with damage in a visually-prominent position; this is a preservation-board and homeowner conversation that goes beyond what a five-question diagnostic can resolve. The tool's job is to produce the substrate verdict and the cleaning protocol; the broader preservation question is a separate consultation.
The tool's job is the substrate diagnosis that a careful operator can run with their own eyes and a few minutes of inspection. The remaining decisions — glazier consultation, preservation-board engagement, replacement-glass selection — are downstream consultations that the tool routes to honestly when the cases match.
The five-indicator framework is not a textbook protocol. It is the version of substrate identification the working trade has converged on after decades of getting it wrong in instructive ways. Each indicator was selected because it survives without specialized equipment, can be run in five minutes by an operator on an unfamiliar property, and produces consistent results across operators when run carefully. The framework converges on the right answer not because any single indicator is decisive but because the combination of four moderately-reliable indicators is dramatically more reliable than any single test.
I learned the framework over my first decade in the Northeast corridor, mostly by watching operators damage panels on properties I subsequently worked on. The pattern was always the same: an operator at production pacing, working without the inspection step, applying chemistry and technique that would have been fine on float glass and were not fine on the panel they were actually working on. The framework exists because the inspection step is the difference between a heritage panel surviving the cleaning and a heritage panel not surviving it.
The Heritage Glass Triage is the version of the framework that scales — it puts the same five-indicator inspection in front of every operator on every property, with the routing logic for mixed-glazing properties, restoration-glass detection, and damaged-panel cases that took two decades of fieldwork to assemble. The tool is, in a meaningful sense, the article's voice in interactive form. Wade Marler's article tells you the framework at the end of a 2,900-word piece. The tool tells you the verdict in five minutes of clicking, and it works whether or not you read the article first.
Derek Giordano is part of the Giordano Inc. editorial team and covers the Northeast corridor editorial beat for Window Washing Guide, with a particular focus on pre-war and pre-1945 glazing diagnostics and handling protocols. Editorial content is researched and reviewed in collaboration with the Giordano Inc. editorial team and informed by interviews with practicing window-washing operators in the region, plus published trade, IWCA, and historic-glazing references.
Derek Giordano is part of the Giordano Inc. editorial team and covers the Northeast corridor editorial beat for Window Washing Guide, with a particular focus on pre-war and pre-1945 glazing diagnostics and handling protocols. Editorial content is researched and reviewed in collaboration with the Giordano Inc. editorial team and informed by interviews with practicing window-washing operators in the region, plus published trade, IWCA, and historic-glazing references.