When the Teak You Are Matching Is Really a Finish

What the Match Target Actually Is

The color a restoration is asked to match is almost always a coating, not the teak underneath it. A caprail reads deep reddish brown. A stair rail reads honey. A run of interior doors reads close to chocolate. None of those colors came out of a log. They came out of a stain, a tinted varnish, a sealer that has ambered, or decades of oil wiped on every spring, and the wood under all of it is a different color.

The order goes wrong at the first step. Someone photographs the existing surface, carries a chip of it to the lumber desk, and asks for teak that matches. The sample search then chases a target the species cannot reach in its bare state. No board of bare teak is that dark, that red or that even, so boards get turned down for failing to match something that was never wood color in the first place.

Separating the substrate from the coating before the order goes in is the whole job. Get that done and the rest of the restoration is ordinary work. Skip it and the replacement arrives looking wrong, gets finished the way the original was finished, and then looks wrong in a new direction.

Reading a Coating Before You Write the Order

Three quick tests on the existing material tell you what the color is made of. Start with raking light across the surface. A film finish sits on top and shows a continuous skin with its own reflection, while a penetrating oil leaves the grain texture open and the sheen low and uneven. Worn edges give it away faster than flat fields do, since a film wears through at a corner and leaves a visible shoreline you can feel with a fingernail.

Next, a solvent test in a spot nobody sees. Denatured alcohol softens shellac. Lacquer thinner softens lacquer. A cured oil based varnish or a polyurethane shrugs both off, which narrows the answer by elimination. Work on the back of a stile or the underside of a rail, use a cotton swab, and stop the moment something softens.

Last, look at a cross section. A chipped corner, an old screw hole or a hinge mortise shows the film in profile under a loupe. Pigment sitting in a discrete layer above the wood is a stain or a toner, and it tells you the substrate was never that color. Color that has gone into the fiber itself is a dye or a deep penetrating stain, and that one will not come off with a light scuff. The National Park Service preservation briefs work through the same reading exercise on historic surfaces at greater length, and the Secretary of the Interior's standards set out how much of an original surface a project is supposed to keep before replacement is on the table at all.

Where to Cut a Patch Back

A cleaned patch the size of a business card, cut in a sheltered area, settles the argument in about ten minutes. Pick a spot the eye never lands on. The back face of a shutter stile, the underside of a caprail, the rebate of a cabinet door, the stile under a hinge leaf, the bottom rail of a door where the film ran thin anyway. An area 2 inches by 3 inches is plenty.

Cut back mechanically rather than chemically. A sharp cabinet scraper takes a film off cleanly and tells you by feel the moment it reaches wood. Chemical strippers can drive residue into the open grain of teak and leave a patch that reads lighter or blotchier than true bare wood, which defeats the point of cutting it.

Then photograph the patch next to the finished surface with a gray card in frame, once in daylight and once under the lamps the room actually uses. Wipe half the patch with mineral spirits and leave the other half dry. The wet half is a fair preview of what a clear coat will do to bare wood. The dry half is the honest substrate color, and that is the half the lumber order gets written against.

One caution on a sheltered patch. A back face or a rebate has seen less light than the exposed face beside it, so the bare wood there may sit a shade lighter than the bare wood under the show surface. Cut a second small patch on the exposed face where a trim piece or an escutcheon will cover it, and read the two together.

Ambering in the Film and Patina in the Wood

Two separate changes are stacked in that old color, and only one of them is in the wood. Oil based varnishes and oils yellow as they cure and keep yellowing under ultraviolet light. Teak oxidizes on its own account, moving from the color it had off the saw toward a deeper, warmer brown. A light cut back reaches the patina. A deeper cut reaches the color the board had the day it was milled.

Both changes matter, and they matter differently to the purchase order. The film amber is not reproducible in new stock and should not be chased in the lumber at all, since it belongs to a coating that is about to be reapplied anyway. The substrate patina is partly reproducible, because new teak oxidizes too, just not on the restoration's schedule.

So the useful target is the patina, read as a family rather than as an exact value. Buy into that family and sit slightly light rather than slightly dark. Correcting a board that landed too dark means stripping and bleaching, which on an oily species is slow and unreliable. Correcting one that landed a shade light means another pass of toner. The Wood Database entry for teak is a reasonable starting reference for the species properties a restoration has to work with.

A Finish by Finish Reading of the Target

Every common finish on old teak hides something specific, and each one changes what the replacement order should describe. The table below runs the finishes a restoration actually meets against what sits under them. Read the right column as a specification input rather than as a color, since in every row the lumber order is written against bare wood and the color is built later.

The bottom two rows catch people out most often. A silvered exterior rail sets an expectation of gray, and no supplier ships gray teak. A painted element sets no expectation at all, and more than one restoration has ordered teak to replace woodwork that turned out to be a different species entirely once the coating came off.

Writing the Order Against Bare Wood

A teak order that can be filled reliably describes the board, not the room. Name the species and the origin first. Teak is Tectona grandis, and the plantation teak J. Gibson McIlvain carries is Indian plantation teak. J. Gibson McIlvain imports it as an FSC certified importer and miller, and the sourcing documentation travels with the material, which matters on institutional and publicly funded restoration work where the specification carries sourcing requirements of its own.

Then the geometry. Thickness is quoted in quarters of rough thickness, so 4/4 is a nominal 1 inch rough and 8/4 is 2 inches, and every surfaced dimension comes off that rough number. Say which you need and say it in words, since teak comes in against export grades and the written description on the order rather than against the National Hardwood Lumber Association rules that run domestic hardwood. Quartersawn and rift faces travel less across the width than plainsawn ones, which is why a wide flat face that has to stay flat gets specified that way, and it is also why a quartersawn original and a plainsawn replacement will never read alike no matter how good the finisher is.

Moisture content belongs on the same line as thickness. Interior work lands near 6 to 8 percent. Exterior work runs near 12 to 16 percent. Air dried stock settles near the local equilibrium and is not a substitute where an interior tolerance matters, and the Forest Products Laboratory research on equilibrium moisture content is the place to check what a given service environment will pull a board toward. J. Gibson McIlvain mills teak at the millwork shop on site at the lumber yard to the in service moisture content the job calls for.

Describe color the way a yard can act on it. Light straw to medium golden brown. No heavy dark streaking on exposed faces. Sapwood excluded. Tone matched within the run so adjacent pieces read as one. Those are instructions somebody can pull stock against. A photograph of a varnished rail is not.

"When a customer sends a photo of the existing woodwork, the first question we ask is whether anyone has looked under the finish. Most of the time the color in that photo is a coating, and once we are selecting against bare wood the order fills quickly."

Norm Moton, Director of Sales, J. Gibson McIlvain

Where the Bare Color Starts in the Boards You Buy

New teak does not start where the old teak started, and the order has to account for that gap instead of pretending it away. A rotation length log carries a larger share of the younger outer wood than a log that grew slowly for a very long time, and that younger wood runs lighter and more even in bare color. Nothing is wrong with it. It simply sets a different baseline for the finisher to work from.

Older installations frequently hold Burmese teak, which has its own bare baseline, usually a touch deeper and more variable board to board. That is a second gap sitting on top of the finish question, and it is closed in the finish schedule rather than in the purchase order. Chasing it with stock selection alone sends a buyer through lift after lift looking for something that is not in any of them.

Selection depth is what makes the stock half of the match tractable. J. Gibson McIlvain pulls restoration teak from a Maryland yard holding over seven million board feet of exotic and domestic lumber, which is what allows an order to be filled on bare color and grain spacing rather than on whatever happens to sit at the top of the nearest unit. Ask for the selection criteria in writing on the acknowledgement, so the second phase can be pulled against the same description.

What Goes Wrong When the Order Chases the Coating

Matching to the coating produces a replacement that looks close on delivery day and wrong the moment it is finished. Take the common case. The original reads dark reddish brown under forty years of tinted varnish, so the buyer hunts down the darkest, most heavily streaked teak available. It arrives and the comparison looks promising. Then the finisher applies the same schedule the original carries, the new boards amber on top of an already dark substrate, and the replacement goes noticeably darker and redder than the thing it was supposed to disappear into.

The opposite failure shows up outdoors. A silvered caprail reads pale and cool, so the order goes in for the palest stock in the house. One coat of oil on new teak and it comes up warm and orange against the gray around it, and now the match is further off than it would have been with ordinary stock.

Light causes a third kind of miss. Two surfaces that match under the showroom lamps can separate under north daylight, because the colorants involved are different and respond differently across the spectrum. Approve samples under the lighting the finished room uses, and look again at a different hour.

There is a quieter failure too. A pigmented stain levels out grain contrast, so the original shows less figure than the wood under it actually has. Match the stock to that calmed down appearance and the replacement, once stripped of any expectation of stain, can read busier than anything else in the run.

Who Owns Which Half of the Match

The lumber order owns the substrate and the finish schedule owns the color, and most failed matches are a result of one side being asked to do the other's work. The substrate half is species, origin, sawing, thickness, moisture content, width and length, sapwood, figure and bare color range. All of that is specifiable, verifiable on arrival, and sits inside the purchase order.

The color half is dye, stain, toner, glaze, sheen, film build and the number of coats. None of it is a lumber question. The finisher builds the target on a panel made from the delivered material rather than on a generic teak sample, because a toner that lands correctly on one piece of teak can land warm on another from a different lot. Architectural Woodwork Institute finishing system guidance covers how transparent finish systems are specified and judged.

Say this out loud at the start of the project so nobody discovers it during punch list. The finish schedule, the sample panels and the final approval belong to the restoration's own finishing contractor. The supplier's responsibility ends at a board that matches the written description, and that is the only half of the match anyone can guarantee in advance. For a wider view of how coatings behave on exterior wood over time, Building Science Corporation has published extensively on moisture and coating performance.

Sample Stock, Lot Continuity and Phased Work

Order sample stock with the job and out of the same lot, or the finisher develops the match on a different piece of wood than the one that gets installed. A few extra board feet buys a step panel with the full schedule laid out in stages across it, cut from the same boards the parts come from. That panel is the approval document. It travels better than a verbal description and it survives a change of personnel on either side.

Restoration approvals rarely move fast. A historic district review, an owner's committee and a finishing contractor's schedule can stretch a sample cycle out for weeks, and lumber sitting in a job trailer through that period is lumber picking up moisture and surface dirt. J. Gibson McIlvain holds finished material in storage for inventory control, so the balance of a lot stays put while the color question gets settled.

Phased work raises the same problem at a larger scale. J. Gibson McIlvain ships nationwide and sends restoration material to California regularly, which means a project running in stages can draw phase two against the written description that produced phase one rather than starting the search over. Where the restoration needs more than a board, ICD, the company's architectural millwork division, builds the custom part itself from the matched stock.

How J. Gibson McIlvain Would Specify This

The specification separates the two halves on paper before anyone prices anything. Open with the substrate. Teak, Tectona grandis, Indian plantation origin. Thickness in quarters with the surfaced dimension stated. Sawing pattern named where flatness matters. Moisture content near 6 to 8 percent for interior work or near 12 to 16 percent for exterior. Bare color range written in words, sapwood excluded on exposed faces, tone matched within the run.

Then the evidence. A photograph of the cut back patch, dry half and solvent wiped half, with a gray card, keyed to its location on the building. A note identifying the existing finish by the tests described above. A statement that the lumber is being selected against bare wood and that the color is the finish schedule's responsibility.

Then the finish schedule as a separate document, owned by the finishing contractor, with a step panel required from the delivered lot before any production finishing begins. Keep the two documents apart. The moment a color chip migrates into the lumber specification, the order becomes unfillable.

J. Gibson McIlvain selects teak against that kind of written description every week and mills it at the lumber yard to the thickness and moisture content the job requires. To talk through a restoration match, see the millwork services at J. Gibson McIlvain or call 800-638-9100 with the patch photograph in hand.