A tensioned high gloss membrane that puts a mirror image of the room, and of the light already in it, into the ceiling plane. We plan the cove, the light lines and the backlit panels around the reflection instead of fighting it.
A high gloss stretch ceiling is a tensioned PVC membrane with a high gloss surface, stretched into an aluminum perimeter track. The surface is specular. It returns an image of the room instead of scattering light across it. The result is a ceiling that reads about twice as tall and turns a cove or a light line into two of them. Gloss is also the finish that tested lowest for flame spread of the three we run. Under ASTM E84 on white specimens, gloss came back FSI 5 and SDI 300, matte 10 and 300, satin 15 and 250. The laboratory assigned no classification, and we say that plainly rather than round it into a letter grade. The specimens were white, they were not mounted per ASTM E2573, and the test covers the membrane alone, not a finished ceiling assembly. Nothing in this data permits the membrane in a return-air plenum.
The reason to build a gloss ceiling instead of adding another fixture is that a fixture lights a spot and a reflective plane changes what the entire room does with light. Uplight that a matte ceiling would scatter comes back down as an image of the source. A single perimeter cove reads as a floating edge on all four sides. A backlit panel next to a gloss field appears twice. Against a glass mirror overhead, the membrane is roughly 8 mil of PVC held in a 6063 aluminum frame, so there is no glass above people’s heads, no mirror joint grid in the field, and the whole plane can be de-tensioned and re-tensioned if somebody needs to get into the ceiling cavity. Against a tile, panel or module system, there is no module at all. The mirror runs wall to wall, wraps a curve, or takes any outline we can bend on our own CNC in Dania Beach. One drawing set, one shop, one crew, and a physical mockup before anyone commits.
A specular plane puts a mirror image of the space above the real ceiling line. The reflected room sits as far above the ceiling line as the real room sits below it, so the space reads with roughly twice the apparent depth at the center of the field. That is a visual effect, not a dimension. Small rooms benefit most, which is why powder rooms, entry vestibules and elevator lobbies are the easiest wins. Darker gloss colors read as a truer mirror because a mirror needs a dark backdrop for contrast. White gloss reads brighter and softer, closer to wet paint, and bounces more of the room’s uplight back down.
This is the fact most lighting layouts get wrong. Matte ceilings scatter a cove wash into an even glow. Gloss does not scatter anything. It shows you a sharp picture of whatever you aimed at it, including the LED strip itself. So on a gloss field you never light the ceiling. You light the wall, or a matte border, or a translucent panel next to the gloss, and let the mirror carry it. Done that way you get the glow and the depth. Done the other way you get a bright line of diodes staring back at everyone in the room.
We bend the profiles ourselves, so a gloss field is not limited to a rectangle. Curves, islands, floating shapes and freeform outlines come off the same CNC that cuts the straight track. And unlike drywall or glass, the finished ceiling comes down. A trained installer de-tensions the membrane, works in the ceiling cavity above it, and re-tensions it. That matters in Florida, where the ceiling cavity holds the AC and somebody eventually needs to reach it.
The spaces where we install high gloss and mirror ceilings most. The rest of the range is on the lighting systems page.
Lobbies, elevator lobbies and corridors, where doubling the apparent height is worth more than adding fixtures.
Black gloss over the bar, with the back bar and the pendants reflected in the ceiling plane.
A mirrored plane makes a small floor read large, and the merchandise reads twice.
Entry foyers, powder rooms, primary baths and home theaters, the spaces where low ceilings actually favor gloss.
A gloss field broken by AP-306 light lines set on the unit rhythm, so the required field break is the design.
Dark gloss reflecting the water surface, with no glass and no powered fixture over the water.
Gloss over RGBW light lines, so a color change reads as a double image across the whole plane.
Layer by layer, in the order it happens on site.
These are shop standards from our own field practice and mockups, not laboratory data. They are the numbers we will hold ourselves to on a real job.
Gloss is best above 9 ft, where the reflected room is far enough away to read as depth. Between 8 and 9 ft it works if the plan is disciplined and nothing bright hangs low. Below 8 ft, in a large open room, gloss starts reflecting people, clutter and desk surfaces, and it puts light sources in the glare zone. The exception is a small enclosed space. A 7 ft 6 in powder room, vestibule or bar alcove is where gloss is at its best precisely because doubling a small volume is the whole point. Enclosed spaces that carry their own interior finish requirements, elevator cabs and exit enclosures among them, need the AHJ to accept our ASTM E84 data first, and that data was run on white specimens, membrane alone, and the laboratory assigned no classification.
Angle in equals angle out. Anything bright hanging more than about 12 in below a gloss plane will be doubled, and the copy sits the same distance above the ceiling as the original hangs below it. Plan pendants and chandeliers as pairs or do not use them over gloss. Windows count. A west facing storefront will put a bar of afternoon sun into the ceiling of a restaurant. We check every bright object from two eye heights, 5 ft 4 in standing and 4 ft seated, from the far corner of the room, because the far corner is the worst case for a mirror.
Never aim a cove up at the gloss expecting a wash. It will not diffuse. It will show the strip. Shop rule: the cove lip return must be at least equal to the strip’s setback from the lip face, with a 2 in minimum on AP-302-W, and then we verify the sightline from the far corner at standing height before we close it up. The move that actually works is to run a matte or satin border 12 to 24 in wide at the perimeter, wash that border with the cove, and let the gloss field reflect a lit band instead of a lit diode. You get the soft edge and the depth without the glare.
AP-305, AP-306 and AP-307 sit in the ceiling plane, so they do not reflect themselves. What they do is expose alignment. A line that is 1/4 in off parallel to the wall reads as 1/2 in off in the mirror. We hold light lines to the surveyed wall line, not the drawing dimension, and we prefer an odd number of lines centered on the room axis over an even number offset to a duct.
We specify gloss as an opaque finish. Backlighting does not go through it. It goes in a translucent panel adjacent to the gloss, split by a light line or a separation profile, and the mirror then shows the glowing panel twice, which is usually better than the original. Design the translucent panel against 42.0 percent diffused transmission, measured at Poznan University of Technology by the method in CIE 130-1998. That is a test report, not a certificate. Do not use the 70.5 percent transparent figure for a backlit panel. Transparent transmits more but does not hide the diodes.
Our shop rule for an even backlit panel is cavity depth equal to or greater than strip pitch.
Strip density of 120 LED/m or higher, so the diodes do not resolve at the panel edges where the cavity is shallowest. We specify strips their manufacturer rates CRI 90+. We hold no CRI testing of our own, and we do not present the manufacturer’s rating as ours.
About 2 in for the profile and wiring. That is the floor. If ductwork or a beam forces less, we drop the field or use a shallower profile, we do not squeeze the membrane.
Cavity too shallow for the strip pitch. Strip mounted too close to the membrane. A weld seam sitting directly over a strip run. A cavity that was never painted flat white, or was lined with foil, which streaks. Transparent membrane used where translucent was needed. And mixed strip batches, where two reels from different bins put a visible CCT step in the middle of one panel. On gloss that last one costs double, because you see the step and its reflection.
24V constant voltage. We cap a single feed at 16 ft, and we feed from both ends on any run over about 10 ft. Home runs are 12 to 14 AWG and drivers sit within roughly 25 ft of the first LED. This matters more on gloss than anywhere else, because a mirror often shows you both ends of a long run at once and a dim, warm shifted tail that nobody would notice on matte becomes obvious.
Drivers, power supplies and control gear go somewhere a technician can reach without touching the ceiling: a closet, above a cabinet, an access hatch. Budget about 1 sq ft of accessible surface per 200 W of driver, with air space around it. We size drivers at roughly 80 percent of rated load. The membrane can be de-tensioned for service, but that is a service call with a crew, not a maintenance plan, and it should never be the intended route to a driver.
This is where gloss jobs are won or lost. The membrane itself is flat. What is not flat is whatever the perimeter track is fastened to. Shop standard: continuous blocking or a rigid furred plane behind the full length of the track, track plane shot to a rotating laser and shimmed to within plus or minus 1/8 in over any 20 ft run, tightened to plus or minus 1/16 in on mirror critical fields and on anything under a 10 ft ceiling. Walls must be straight in plan as well as level, because a bowed wall becomes a bowed line in the reflection and there is no way to hide it. We scribe and shim rather than accept the substrate. If the GC cannot deliver the plane, we build our own furring and price it up front instead of discovering it on install day.
A tensioned membrane over a wide open field carries a very slight center relaxation. On matte it is invisible. On gloss it bows the reflected line. We cap an unbroken gloss field at roughly 22 ft in the short direction and break anything larger with a separation profile or, better, a linear light line that turns a structural necessity into part of the design.
Our gloss stock runs up to about 10 ft wide, color dependent, so fields narrower than that are seamless. Wider fields carry HF welded seams. On matte a seam disappears. On gloss you can find it under raking light. We run seams parallel to the long axis of the room and the dominant sightline, and we keep them out of the reflection zone of the main light lines.
Dark gloss reads as a real mirror, because contrast is what makes a mirror. Black gloss over a bar or a water feature at night is the deepest version of this. White gloss reads brighter and wetter, bounces more uplight back into the room, and forgives more. Mid tones tend to read as neither. Pick one end. The full range is in the color chart.
Every sprinkler head, diffuser, speaker and sensor in a gloss field is doubled. Push them to a matte perimeter band where the design allows. Where they have to be in the field, they go on a strict grid aligned to the room, because misalignment that would be tolerable on a matte ceiling looks like an error in a mirror.
PVC tightens when cold and relaxes when warm. We install at shop temperature and let the room settle. Do not aim a supply diffuser at the membrane. Air blowing directly on a gloss field will show as a slow moving ripple in the reflection, and no amount of tensioning fixes it. Redirect the diffuser.
No IES files, no LM-79, LM-80 or TM-21 data, and no lumen, lux or footcandle output for the assembly. No acoustic absorption data for this system either, which means no NRC figure and no ASTM C423 report. On fire and code we hold the ASTM E84 report only, run per finish on white specimens, and the laboratory assigned no classification. The specimens were not mounted per ASTM E2573 and the test covers the membrane alone. We hold no NFPA 701, no NFPA 286 and no NFPA 265 test, no ASTM E2573 mounted-assembly test, no UL 2043 plenum rating, and no Florida Product Approval or Miami-Dade NOA for this assembly. If your model needs photometry, the FAQ below explains how we recommend building it.
No. We do not hold LM-79, LM-80, TM-21 or IES files for the assembly, and we will not hand you numbers we did not measure. Here is how we recommend you model it instead. Build the LED layer in your calculation using the strip manufacturer’s own published IES for the product you are specifying. Then apply the membrane as a diffuse transmitting layer. For a translucent backlit panel, use 42.0% diffused transmission from the Poznan University of Technology report, method per CIE 130-1998, which we will send. That is a test report, not a certificate. For the gloss field itself, model it as a specular reflector, not a diffuser, because that is how it behaves. If your project needs a stamped photometric report on the finished assembly, tell us early and we will quote independent testing as a line item rather than pretend the data already exists.
This is the question that decides whether a gloss ceiling looks expensive or looks wrong. We need continuous rigid blocking or a furred plane behind the entire perimeter track run, not intermittent anchors into board. We shim the track to a rotating laser and hold plus or minus 1/8 in over any 20 ft run, tightening to plus or minus 1/16 in on mirror critical fields and on anything under a 10 ft ceiling. That is our shop standard from field practice. Walls also need to be straight in plan, because a bowed wall becomes a bowed line in the reflection and there is no finish trick that hides it. If the substrate cannot deliver that, we would rather build and price our own furring at bid time than find out on install day.
Not on the data we hold. Our ASTM E84 results are per finish on white specimens, gloss FSI 5 and SDI 300, and the laboratory assigned no classification. The specimens were not mounted per ASTM E2573, and the test covers the membrane only, not a plenum assembly. We do not hold UL 2043. So the honest answer is that this system is not permitted in a return air plenum on our current data, and any engineer who tells you otherwise is reading something we did not send them.
Not if the field is narrower than about 10 ft, which covers most corridors, powder rooms, bar soffits and island shapes. Our gloss stock runs to roughly that width depending on color, and inside it the panel is one piece. Wider than that and the panel carries an HF welded seam. On a matte ceiling a seam disappears. On gloss you can find it under raking light, so we do not hide it, we plan it. Seams run parallel to the long axis of the room and stay out of the reflection zone of the main light lines. The seam plan goes to you for approval before we cut material.
It depends entirely on the size of the room. In a small enclosed space at 8 ft, gloss is excellent, and that is exactly why powder rooms, vestibules and elevator cabs are the classic application. Doubling a small volume is the whole point. In a large open room at 8 ft, gloss starts reflecting people, tabletops and clutter, and it drops light sources into the glare zone at seated eye height. If that is your situation we will usually recommend a gloss island or a gloss border with a satin field, so you get the depth without the mirror following everyone around the room. We will tell you which one your room is before you buy it.
Yes. A trained installer de-tensions the membrane, works in the ceiling cavity above it, and re-tensions it. That is a real advantage over drywall and over glass mirror panels. It is still a service call with a crew, so it should not be the planned route to anything you touch regularly. Drivers, power supplies, valves and dampers all belong in an accessible location by design, and we will flag anything on your drawings that ends up sealed where it should not be.
Yes, and this is the part a panel or tile manufacturer cannot match. We bend our own profiles on our CNC in Dania Beach, so a gloss field can follow a curve, wrap a radius wall, take an irregular outline, or float as a shaped island using AP-212 or AP-213 formed into U, C or O shapes. The radius on your drawing is the radius that gets built. There is no module to work around and no cut tile at the edge. The same shop builds our 3D ceilings and structures.
Not through the gloss. We specify high gloss as an opaque finish, and pushing light through a specular surface gives you visible diodes, not a glow. What we do instead usually looks better. The backlit area is built as a translucent satin or matte panel next to the gloss field, split by a recessed light line or a separation profile. The gloss then reflects the glowing panel, so a single luminous element reads as two. Design the translucent panel against 42.0% diffused transmission, measured at Poznan University of Technology by the method in CIE 130-1998 and issued as a test report, not a certificate, and follow the cavity to pitch rules in the design section. Or see light boxes where the panel is the whole feature.
What we hold today and issue on request. The published files sit in the lighting technical library.
Reserved and not yet published. Listed by name so nobody assumes it exists:
Our profile cross-sections, dimensions and the full color chart live in the technical library. CAD files are sent on request, usually within one business day.
A panel manufacturer can hand you photometry for a tile, but every tile is a fixed module on a ceiling. A mirrored plane that runs wall to wall, wraps a curve we bent on our own CNC that morning, comes down for access to the ceiling cavity, and is laid out from a reflection plan surveyed in your actual room is something a module system structurally cannot deliver. The same shop that draws it welds it and installs it, and we have been installing stretch ceilings in South Florida since 2014.
View system
Tell us the room, the ceiling height and what hangs below it. We will tell you whether gloss belongs there, and we build a physical mockup before anyone signs off.
Share a few details and our team will follow up within one business day.
