How Not to Finish an Exterior Door

Sometimes, "finished" is another way to say "all done." February 17, 2014

Against my better judgment I have been ordered to follow this finishing schedule. I am hoping to have my employer browse through the comments you leave. Here's the story.

Large, solid 6 panel mahogany exterior door. Lots of casing as well as a fancy upper window frame. Two large exterior frame and panel end panels with leaded glass inserts that are to be installed on each side of the door. This door with all its parts and pieces is very intricate, with applied moldings and fancy details.

This is the finish schedule I am forced to follow...
(1) Completely fill all pores with cherry wood putty… Yes! I said wood putty, not paste wood filler or a level sealer; but putty.
(2) When putty is dry the raw wood is sanded with 150-220 and final sand with 320. Remove dust etc.
(3) Apply a dark mahogany gelled stain.
(4) Apply 9 coats of MacLac sanding sealer sanding each coat with 220 and 320 paper in between coats.
(5) When grain is completely filled, apply an automotive satin polyurethane catalyzed 4 parts to 1 part and thinned out with Home Depot lacquer thinner.
(6) When poly has dried, sand with 220 and 320 and coat again.
(7) Repeat step 6 until finished.

I am fearful this finish will degrade shortly. Right now if you even gently tap on the surface or accidently nick an edge while blowing it off, the finish chips all the way through to the wood. Also, with all the coatings that have been applied, the finish is starting to have a blue cast.

For what it's worth, the counterperson at the paint store told the shop owner that this was the way to go. He has already completed a few parts of the job (they are all wrapped up and stored away in the shop), so he has boldly told me to complete the rest of the job in the same manner as the completed parts were finished.

Forum Responses
(Finishing Forum)
From contributor M:
Why 9 coats of sealer?

From contributor B:
Burying putty could be a problem. Building 9 coats of sealer is a problem. Topcoating with urethane over those... problem. Your boss and counterperson have a problem unless that door is encased in an airtight glass display case.

From contributor A:
9 coats of sanding sealer? It would be like looking through a foggy window. The dumb guy isn't the 1st one to jump off the bridge. It's the 2nd one who followed his mate. Unfortunately you are being pushed off the bridge even though you know it's not a good idea. I would let it be known that you will not warranty your work on this project, because it is prone to failure. Their choice - their problem.

From contributor R:
Do a cross hatch adhesion test on the sample and show there is an adhesion problem. Suggest putting the sample outside for a while to see how it holds up. Do a new sample to match this sample using proper exterior products and do the same cross hatch test and exterior exposure test. If they have any brains at all they should make the right decision. (If not I would make them sign off on the sample assuming all responsibility for failure.)

From the original questioner:
Thank you all so much for responding. I have told the shop owner that this finish schedule is bound to fail. He has already finished another set of doors using this same technique and insists that I follow the same process. It is taking so long to complete this project that the customer calls every day wanting to know when it will be installed. The shop owner then approaches me asking when it will be finished - the customer is angry it's taking so long, etc.

From day one I have expressed my concern over the products being used on this project but to no avail. All I hear from the shop owner is that it's taking too long to complete and he wants to know when it will be done so he can install it. I am aware of the various adhesion tests but right now it's impossible to take the time to fabricate a new sample. Every working hour is spent on the completion of this project.

I will direct the shop owner to this site to read your input.

From contributor K:
State your case to the shop owner, then do what you are told in a timely manner. You are an employee in this unfortunate situation.

Many in finishing, or woodworking in general, believe that there are secrets and mysteries counter to logic and science that need to be obeyed for some ancient reason. This is superstition, and will always cause problems. Outside of the shop, it causes war, starvation and numerable problems.

While we cannot effect large scale change in the world, we can do so in the woodshop. Many woodworkers have found themselves in your position and this has led to them starting their own shop. If this is your cue, be sure to separate superstition and belief from science and accurate observation.

From Paul Snyder, forum technical advisor:
My first impression is it must be a joke... Finisher's humor about how not to do a finish.

For exterior wood, all the finish ingredients must be rated for exterior use and be properly applied. Some exterior finishes are much better than others. When clear finishes are used on exterior wood, the typical good quality exterior finish requires annual maintenance. Lower quality products will fail in a few months. You will be lucky if your finish lasts a few months and it will be a nightmare to replace. Notice I said replace, not repair.

Wood putty is more difficult to use as a pore filler than the products made for that purpose. Paste wood pore filler is a better choice.

Sanding the wood surface to 320 grit is too fine and the clear coat finish will not adhere as well as it would if you sand the wood to 120 or 150 before staining.

Good quality gel stains come in handy for entry level finishers on woods that are prone to blotching (uneven coloring) when stain is applied. These woods include pine, alder, aspen, etc. Mahogany stains very well and a gel stain is not required. It would be much easier to use a spray and wipe stain.

This sealer is not rated for exterior use and will fail. Also, never use more than 2 coats of sealer (when you use it). Only use sealers that are compatible with the topcoats you're using. MacLac sanding sealer should never be used with 2K PU (2 component polyurethane). Do not use this sealer in an exterior finish.

The ratio of catalyst to finish is product dependent so follow the directions for the brand you use. Lacquer thinner is not the proper reducer for 2K PU… Use the same brand thinner as the finish and you won’t run into trouble. Note, an acrylic 2K PU can be a good product for an exterior clear coat wood finish if you choose the proper product and apply it correctly. I’ll recommend Matthews satin MAP based on firsthand experience.

Follow the recoat directions for the product you choose.

How will you know you’re finished?

For a published reference, have your employer get a copy of Bob Flexner's book "Understanding Wood Finishing." It goes into much greater detail about why most of the steps in your finish schedule are a bad idea and destined for failure. Please provide the finish schedule you propose to use including the brands and products you recommend.

From the original questioner:
If I had been the owner of the company instead of an employee I would not have taken on this particular job in the first place. As a past owner of a wood finishing business and a wood finisher with 28 years under my belt, I stick to what I am 100% sure I can put my name on, and exterior doors are not an item I choose to deal with.

As an employee one of my greatest responsibilities is to look after the welfare of my employer. I have gone well above and beyond this responsibility by expressing my objections about this finish schedule and the products being used. My objections are being ignored and I feel sad for the owner who will have to re-do this disaster sooner than later.

I am aware that exterior finishes require a maintenance schedule. This system has, however, been sold as the ultimate of finishes and I know in my heart it is not anywhere near that.

From contributor L:
That is a disaster waiting to happen. Using putty as grain fill is fine, I've done it. Sanding to 320 before sealer is asking for trouble. 180 at the most. Applying 9 coats until flat is fine - there are no build limits on nitrocellulose. However, normal practice is no more than 2 coats of sealer as it is designed for easy sanding and not durability. Putting a catalyzed product over an uncatalyzed product is where the disaster will occur. You need to use compatible products and nitrocellulose and automotive catalyzed products will not be compatible.

From the original questioner:
I have been urging the shop owner to use a polyurethane coating throughout the finishing process but all I hear is it's too expensive. When faced with a failure in the mist like this job I find myself asking aloud, "Why is there never enough time to do the job right the first time, but an endless amount of time to do it over?" Go figure.

From contributor J:
In addition to all of the horrendous mistakes and misconceptions in this finish schedule that others have already pointed out, I would never use an automotive finish for wood, inside or exterior. Yes, I know the logic behind it - if it stands up on a car it should stand up on wood. But automotive finishes do not contain the flex agents needed to accommodate the movement of the wood with changes in temperature and humidity. This finish is doomed to failure and whoever sold it to the customer or your employer as some type of ultimate finish should not be allowed anywhere near a spray gun or an architect's spec book.

From contributor I:
Seems like you are doomed to proceed with this finish so one more piece of advice. Put your concerns in writing in an email to your employer so 6 months from now when it all goes bad, you don't end up getting blamed. (Voice of experience.)

From contributor A:
Spraying two coats of liquid gold would be cheaper than 9 coats of sanding sealer. I would love to hear the original thoughts behind that decision. If 2 coats is better than 1, 9 must be better than 2. Automotive clears work very well on wood. I've seen several dining room tables that have held up well for 20 years. We used to spray 3 coats of clear Imron (acrylic 2k auto) on teak toe rails on the bow of our 32 foot racing sailboat. They take the most abuse of any wood on a boat. The finish would get scuffed down to bare wood. It would not peel like regular varnish. The stuff really sticks to the wood. Keep in mind that is teak in a full UV marine environment for 5 months a year. The lack of flex is BS. Total myth.

From contributor O:
On parts of the boat that did not get scuffed how long did the Imeron last in that fully exposed exterior environment? And when you needed to restore the quality of the finish did you take it to bare wood or just scuff and respray?

From contributor J:
Imron is pretty tough stuff but not all automotive finishes are up to that task. And even if it does not fail due to lack of flex, it will eventually need to be refinished and spraying Imron on a customer's front porch is not my idea of a fun day.

I used Imron on some structural steel for a boat many years ago. I think it was marketed as a marine grade finish and not the same formulation as their normal 2k automotive stuff? It's been a long time since I used it, so I may be wrong.

Personally I have seen 2k type automotive finishes fail when used on exterior wood work. So call it a myth all you want - but there is nothing that will outperform a properly applied and well maintained exterior marine grade spar varnish over the long haul.

From contributor A:
The toe rails were the only things that we would coat with Imron. It's not something you want to brush on a frequent basis. We would spray them every year and wood them after about 4 or 5 seasons. The bare spots would not show through like they do with varnish. I think it's because varnish ages in UV. The Imron doesn't age; however, the wood beneath it probably does to some extent (minimizing the bare spots?). I realize this is kind of backwards logic.

When a typical varnish job gets dinged, the exposed wood quickly changes color, the water gets under the film and starts to peel it. Which results in another distinct area of some protection. This results in these aura spots when you scuff sand them. The Imron never seems to peel.

The parts I'm talking about are 1.5" x 1.5" x 10' trapezoid (triangle with the top cut off). They prevent people from slipping off the bow of a sailboat that has smooth shear. People are constantly stepping on them, lines chaff them, and spinnaker poles are dropped on them. Hell of a lot of abuse.

Many people lay down a couple of coats of epoxy then varnish over it in situations like cabin soles (teak/holly ply). The epoxy prevents damage to the wood when the varnish fails.

From the original questioner:
Rather than having to re-do this door and all the other parts and pieces and having to deal with the employer's future moaning and groaning, I severed my tenure with this particular employer. I'm not an "I told you so" type of person and on top of that I really felt crappy each day I worked on the door knowing the finish would fail. The whole shebang is outa my hands now and my conscience is clear.