Moisture Issues with Exterior Screen Doors

This extensive thread examines wood choices and finishing methods for custom-made exterior screen doors. March 26, 2010

Question
I have a question that I hope someone has experience with. I have been contracted to build a set of exterior screen doors to match the existing doors which have rotted and are falling apart. I built a set last year (for a different customer) out of cypress. This was on a summer home, and over the winter the cypress expanded so much that you could not open the doors. I just checked on them again and they have shrunk back to normal. Was cypress a bad choice? What should I use? I was thinking of mahogany. These doors will be painted.

Forum Responses
(Architectural Woodworking Forum)
From contributor J:
Where are you located? In most parts of the US it is substantially more humid in the summer than the winter. Never heard of things swelling shut in the winter. When were the doors installed and what was the gap then? I would imagine cypress being a good choice assuming it didn't just come out of the swamp. I've used VG fir for exterior storms with no problems - except for a bad windstorm!



From the original questioner:
I am in Michigan. Both of these homes are located on the Lake Michigan shore. I also thought it was strange that they would swell in the winter. The rails and stiles are 3 5/8 wide, total gap of close to 3/8" (when you add the gap between doors and jamb, and the gap between the doors). The doors were installed about this time last year. I thought about using fir also...


From contributor J:
Any chance you've checked the MC on that cypress?


From contributor D:
Doors do swell in the winter if they're not sealed well. Take a warm door on its first cool night when moisture condenses and is absorbed up through unsealed bottoms and insides of the sticking on the stiles - often the doors fit one day and on the next have swelled. Then it never gets quite warm enough to evaporate out until next summer. The moisture likely evaporates to just under the paint, and begins to rot the wood and paint will peel eventually there. A screen door is more susceptible to temperature fluctuations out in the elements, unlike the main door which has one side at room temperature. The best prevention, regardless of species, is engineered construction of the stiles and sealing everything - under the sticking, in the hardware mortises, even swabbing into the screw holes.


From contributor B:
We build all kinds of exterior millwork and we always use Spanish cedar; never a problem.


From contributor W:
Cypress will swell big time unless it is sealed. We are in Louisiana where you can find cypress at just about any lumberyard. We make all the exterior stuff here from Spanish cedar.

Old growth cypress (of which there is none left) is wonderful material for exterior fixtures. It is rot resistant and very stable. New growth cypress is almost all soft, wide grain sapwood which does not hold up very well for exterior fixtures.



From the original questioner:
Thank you all for your input. I will look into Spanish cedar. I would like to select a species that grows locally. Sassafras is available. I was told by my local lumber supplier that this would be a good choice...

As for sealing the doors, what do you suggest as a paintable sealant? As with the doors last year, I am delivering these unfinished. Apparently the painter did not seal them properly last year. I would not be opposed to applying a sealant myself before I deliver them.



From contributor J:
I deliver mine primed; that way I'm sure all the surfaces are sealed up. If you ship an unfinished exterior door, you should have some disclaimer about proper finishing and its necessity for you to warranty the door.


From the original questioner:
If these were for a customer of mine I would take them to my painter and have them professionally done. These are for a contractor that I know. He wants them unfinished so that he can keep his painter busy I guess. It doesn't matter what I tell them about making sure they are properly sealed, his painter is just going to do it the way he wants to anyway. I certainly don't want the doors that I build to fall apart because they aren't sealed properly. What is the best paintable sealer that I could apply before I deliver them?


From contributor L:
Don't overestimate the effect of sealers on wood movement. Bruce Hoadley's book on the topic explains this notion very thoroughly. It's a must-read for anyone in the wood business.


From contributor U:
The way our work gets treated once it's left us is not just a perennial problem, but an international one from what I read here! On doors particularly, those bottoms will get left if the doors are hung before the painter gets to them, which is why I always treat that area with a chemical wood hardener (the type you use to treat rotten wood) and take it up the sides and edges for a good inch or so if the door's to be painted. Trouble with staining is that if you treat the visible areas it won't take, so just the underside. And if you're really nervous, the hinge rebates too.


From contributor S:
I have to agree with contributor W about the misunderstandings with regard to wood performance. This is especially true when the durability issue is discussed. Just because a dealer calls it cypress or mahogany or redwood or take your pick, they are not defining it as new growth material and explaining the differences between old and new growth.

We have used cypress that is old growth or reclaimed and it performs well, but if you look at the material marketed today, the growth rings are much less dense. I can't believe the lack of knowledge, even among many people in the business, when it comes to recognizing the stability and decay resistance differences between new growth and old growth.

I will add that we do use an epoxy sealer and do our own painting to give the product the best chance for long term performance. This includes using the epoxy sealer on copes and tenons in door and sash construction.



From contributor D:
Epoxy is the best moisture barrier out there, which is the reason it's used in cold molded boat hulls. It is expensive, and it is a pain to mix up just the right amount for a small job, and needs further treatment of a UV resistant coating because it breaks down rapidly with exposure to sunlight.

An alternative I have found very easy to use is plain old shellac. It's been used for hundreds of years and is proven. You mix the flakes, which are the natural secretions in bark from the lac beetle, in denatured alcohol to your desired consistency and brush it on. It is non-toxic. My technique is in two steps - first saturate with a weak one pound cut, which means a ratio of one pound of flakes per gallon of alcohol, for deep penetration. I then let this dry and coat with a thick nine pound cut.

There has been research in the epoxy world by the Gougeon Brothers that shows weak solutions of epoxy curing with thousands of tiny capillaries which the evaporating solvent (in this case acetone) created. I've never looked at thinned cured shellac under a microscope, but I thought it couldn't hurt to mimic their methods. The thinned shellac has a plasticizing effect on the wood as well, furthermore stabilizing it. Shellac also holds up well to sunlight, but still should be coated with spar varnish or paint for UV protection. You should always get dewaxed shellac, which is compatible with nearly any coating out there if you are coating the whole door. If you're just swabbing in the bottoms and hardware mortises and such, any old grade will do.



From contributor Q:
I built a 36x96 x 1.625 screen door and used Brazilian mahogany. Almost a year later (in SW Florida), the door operates fine. I used slip tenons assembled with epoxy and finished with some type of 2 part poly. (I subbed out the finishing.) I made sure that every face, edge, side and profile was covered with 2 coats of finish to prevent any warp.


From contributor M:
Very interesting post about shellac. My experience with shellac has only been in the last 15 years as a professional woodworker/finisher learning by trial and error, and reading Flexner's "Understanding Wood Finishing" and the Finishing Forum, so I am no expert on shellac by any means.

Just to clarify, shellac is not secretions "in the bark" (i.e. sap), but rather secretions from the lac bug itself that end up on the bark of branches and twigs of certain trees found mainly in India (paraphrased from Flexner's book).

Can you please expand on shellac being proven and used for hundreds of years? Any reference material? Is the plasticizing characteristic of shellac on wood directly correlated to the amount of wax in the shellac? And if so, would using dewaxed shellac negate that desired effect? I understand the idea of using dewaxed shellac under other types of finishes - which is strictly a blanket concern of adhesion to the minimal amount of wax in regular shellac (not dewaxed). One could argue that using regular shellac for its plasticizing effect would be preferable if one were using a coating on top of the shellac if it were tested (?) for proper adhesion first.

An experience I had about 3 years ago got me thinking about using shellac as you describe - I built an outdoor playset for my sons and it is bordered by 2 - 2 x 6's (redwood) on edge, capped with 1- 2 x 6 flat - similar to a raised garden planter bed. I ran out of exterior deck screws while attaching this last piece and used some plain zinc screws I had on hand. Before letting the boys break it in, I gave those zinc screw heads (about 4-5) a shot of shellac from a spray can with the thought of preventing a rust stain on my brand new redwood until I replaced them with the deck screws. Guess what? 3 years later and just now barely some signs of rust on the screw head! (Yeah, I never replaced them.) We are in SoCal so we don't get that much rain, but lots of sun and water from my sprinkler system. Like many tricks of the trade that I've learned over the years, most were discovered by mistake!



From contributor D:
Most antique furniture was finished with shellac, and although I don't have references, a couple examples of longevity that come to mind are furniture in ancient Egyptian tombs and the violins of Stradivarius. The plasticizing effect is apparent with a little experiment. Soak a small piece of wood in a weak solution of shellac and let it cure. Then take an equal sized piece and mic them both. Soak them in water then measure which one expanded the most. I also found after machining a treated piece that cross grain cuts had almost no tearout. There are many variables of course that will affect how deep the shellac penetrates, but you'll get the idea experimenting on a couple of scraps from your project.


From contributor M:
Interesting thoughts. Is the plasticizing characteristic of shellac on wood directly correlated to the amount of wax in the shellac? And if so, would using dewaxed shellac negate that desired effect?


From contributor D:
From my limited usage of shellac in this manner, for sealing door components, I can't tell if dewaxed shellac has any advantage other than from a finisher's perspective. The plasticizing effect is just the absorption of the shellac into the wood cells, which limits subsequent moisture absorption and therefore wood expansion and contraction. Outdoor furniture would definitely benefit from thinned shellac treatment. I have an obsessive interest in boatbuilding, and a few shipyards in Florida are experimenting with high tech materials like carbon fiber and Kevlar composites in a shellac matrix instead of epoxy or vinylester. Old world meets new as we constantly evolve and refine techniques.


From contributor N:
In my opinion shellac is a very poor choice for exterior woodwork. Shellac will actually draw moisture from the air and make it very hard to hold paint in an exterior application. Shellac has many uses but exterior woodwork is not one of them.

If you really want to protect the wood and have the time and money to put into the project, soak the door in linseed oil for a day and then wipe off the excess. (Be careful with the rags; either hang until dry or soak in water.) Let the door dry for a week or two, then paint.



From the original questioner:
Well, it seems that there are some varying opinions about the use of shellac here. The doors are built and I am just about to put a coat of shellac on them before they are delivered.


From contributor V:
Traditional work is with oil based finishes - varnish or prime/paint. Millions of screen/storm doors have been finished this way and have miraculously survived. Do finish the end grain twice as much as the rest of the door - don't fall for the old "no finish on the ends so the wood can breathe."

The (boiled) linseed oil soak is a valid tactic. While I am always interested in the next new, best thing, until it is proven over time, I will stay with what has gone before. The devil you know vs. the devil you don't know.

All due respect (as I have been very wrong many times before), but I found the whole shellac exterior thing pretty entertaining. I kept thinking one of the fans was going to mention that paying income tax is unconstitutional.



From contributor A:
Applying part of the AWI 8th Addition to this from Section 1400 related to Stile & Rail Exterior Doors, 1400-G-3 pp.465:

"While wood stile and rail entry doors have performed well for centuries, the selection of a wood door places a burden on the owner to maintain the door by keeping painted or sealed, protected from moisture, and properly adjusted in the opening."

I have built a few custom screen doors over the years. The other spot to watch for is painting them in the rabbet where the screen spline or stripping runs.



From contributor M:
Contributor N, is shellac drawing moisture from the air and making it hard to hold paint in an exterior application something you have actual firsthand experience with? Can you elaborate?

I took contributor D's thoughts on the use of shellac for exterior wood projects strictly as a sealer - specifically if it were going to be painted. I don't think he was suggesting using shellac as a finish/topcoat on its own on exterior wood projects.

The debate goes on, until some exhaustive, unbiased, scientific study is done that is definitive about the use of shellac as proposed in this thread. My very unscientific gut tells me that "plain old shellac" is very misunderstood and very underused and is suitable for exterior wood projects as a sealer - the freshly mixed from flakes variety with a good amount of wax (whatever that is?) being the crucial component to success.



From contributor N:
Shellac drawing moisture from the air is a fact. If you like to paint often and frequently, then shellac as an exterior sealer is for you. That is from personal experience. Shellac has been around for a long, long time. The research and the test of time has been done. The information on shellac is out there. Shellac is a very beautiful interior finish where it is not subject to high moisture or chemical exposure. The dewaxed variety is also a great interior sealer coat. It is a very poor choice for use outside or in high moisture areas.


From contributor D:
I need to clarify that my usage is not in a traditional manner. Thinned shellac penetrates deeper into the wood than oil of any type. I advocate its use not as an overall sealer, just on end grain, in hardware mortises, and so on. I work in a shop with about forty guys and we have built a few thousand doors over the past twelve years - not a lot by some of your standards, but enough for me. It evaporates rapidly and works well. Often we do reproduction work, and you can see where doors fail due to wood movement. The only way you'll ever believe it is to try it on scrap. I am convinced it seals out moisture better than any oil. We do not finish our doors, and this is only the extra insurance measure taken in case the finisher does not get every nook and cranny. Some of our doors have been oil finished - it doesn't matter to me. I don't see a problem using oil as long as it goes everywhere, but often finishers neglect the most important spots.


From contributor J:
Good enough for me, contributor D. More eco-friendly than the stuff I've been using as well. Thanks for passing on the fruits of your experience.


From contributor M:
Does shellac draw moisture only when it is freshly applied and during the curing process, or continually even after it is cured?

According to Zinnser, shellac is:
1. UV resistant
2. More durable and less brittle than lacquer
3. Remarkably water resistant
4. Great finish for floors
5. Wow!