|Home » Forums » Architectural Woodworking » Message||Login|
You are not logged in. Consider these WOODWEB Member advantages:
Engineered Stiles Veneer Failure7/18
Those of you that are concerned with such things may recall that I am a bit of a Luddite on the subject of engineered stiles for passage doors. Or "cost-accounted" stiles since that is, in my opinion, more important than the quality of the product. I have always made solid stiles and recommend same for others. Some folks disagree, and I am fine with that. It is a big world, and vive la diferance. What works is what works.
Well, I broke type and made some built up stiles for a pair of exterior Walnut doors recently. And today I get a call to come and see why the doors are 'falling apart".
I opted for 1/16" thick veneer for the faces, over ripped and glued top grade walnut for the cores. Since it is so hard to get decent looking wood, this seemed like a good time to go with the engineered/built up stiles (and rails),
We used powdered urea resin for the veneering. We have had a few bad experiences with Unibond 800 since it was reformulated, so were spooked by it. I wanted low/no water, but did not want to use epoxy since I felt the vacuum press would create too thin of a glue line. I swore off Urethane/Gorilla glue a few years ago, after learning it was not as good as other glues - even though I secretly loved it for veneering. The PUR was fresh and well mixed, temps were well over 70 degrees, etc. This all worked very well and the stiles and rails are were flawless.
Everything was tight and right. looking good and went to the finisher. Once complete, the carpenters set the doors and I went out to look at them for touch-up and look over the hardware and weatherstrip. We also used Tectus hinges and a corner silicone bead instead of the old standby- Q-lon. Nothing like keeping it interesting.
All was fine, except I noticed a fine crack along the grain - <1/32" wide and about 6" long, on a bottom rail, exterior face. I mentioned this to the finisher and he made it go away. These doors are finished as almost black, so the surface is about as hot as wood can get.
Now, 3 weeks later, the phone rings so I go out and find about 4-5 cracks in the stiles and the rails all about 12" long, less than 1/32" wide, in the grain of the veneer. The cracks are not above the joints in the core, just sort of near the center of the 5" rail and stile parts. All are on the exterior and more so on the lower third of the doors than just anywhere. The veneer is tight - not loose or peeling, but is very much open.
So.......where did we go wrong?
Wrong glue? That is what I'm thinking - it was the water in the powdered urea resin glue that readily went into the veneer and made it swell, then cure. Now, as the doors bake out in the Eastern exposure, the MC is decreasing and -c r a c k- the veneer splits. I did not take a MC reading during/after fabrication, so it may be pointless now, except as academic.
Meanwhile, I'm honing up my "Never a veneered door part again" speech and will stick to my guns. But currently I'm sick at the thought of how to fix this and what the future may hold. I also have been fantasizing about a job at McDonald's, where there is no risk, no stress.
I'm one of the engineered stile guys here.
I've always used epoxy, and though I never vac bagged, I've used cauls for pressing and never put that much pressure on that I worried about a starved surface.
The second thing is I've always used thicker shop sawn veneers for my faces. Always at least an 1/8" to 3/16" thick.
My staves were always close to being quarter sawn to minimize movement, that is I ripped them from flat sawn lumber, as I'm sure you probably did too.
So to answer your question, I think the veneers may have been too thin for an exterior door.
Sorry to hear about what happened and I hope this helps a little.
As an aside, I've never used an engineered core to save money. The added labor eats up any material savings quickly. What I've always liked about then when laid up with epoxy is they stay dead nuts flat and straight out of the cauls. The rigid glue line from the epoxy keeps them from ever bowing.
I think David has probably forgotten more about door making than I know, but my little experience agrees with what Mark said. I build my doors in a similar way and it costs more to do them that if I were just to buy flat stock and mill. So no cost savings for me. I also re-saw my own skins and sand down to about 1/8" before gluing. I however only build interior doors so use TB3 and vac press for glue up.
I also wonder if the thin veneers are part of the problem? Walnut is pretty porous so I can definitely believe it would soak up plenty of moisture. Maybe this is a case of several small things combining to create a larger problem?
Anyway I hope you find a good way to resolve the situation. I know what you mean about "day job fantasizing" though, I'm now 6 weeks into my new shop and almost ready to start working for pay again!
David has probably done many times the doors we have. I'm also an engineered core guy. Not to save $ but for the stability I hope it gives. I'm old fashioned and use resorcinol for exterior doors (but try to avoid them.) Great glue, no water but dark glue lines & slow. I also resaw faces and try to finish @ 1/8" thick. Vacuum bag with the parts inside of plastic sheeting because of the glue mess.
The main thing you did wrong was that you violated your own policy.
Then you tried something new and expected it to go off without a hitch.
I think your rails and stiles were doomed to split regardless of how thick the veneer. Nothing at all to do with how you made the doors but rather that very dark stain on a sun baked door made out of a not particularly stable wood species. I suspect that even solid rails and stiles would give you trouble in this installation scenario.
With regard to a McDonald's job I have a friend who was CEO of a pretty heavy duty company. Before he retired a few years ago he used to always say he wanted a job at the dollar store. All he would have to do is press the $1.00 key on the register all day !!
I would never do an exterior door with 1/16” veneer. Most of the door failures I have seen in the field from other companies are doors with veneer faces. I do not know what glue would be best for this application but probably nothing water based is good for thin veneer.
As many have said there is no cost savings for the small shop to build this way. The motive is to keep things straight. We always used solid stave cores butcher blocked to produce rift and quartered faces with 1/8” to 3/16” faces. See pictures. Ones on left are interior doors. For exterior we usually do a same species core.
This is nothing new. We have doors in our town built this way that are over 100 years old.
The only failure we had with this was a Sapelle door on a garage. We did the same doors on the house with no issues. The customer was parking her snow covered car with a overhead heater blowing across the car on to the door. Solid rails and panels cracked, And the 4mm thick veneer delaminated in several places. Part of the problem was the oil finish she requested offers no protection and we were not able to caulk the joints like we do with our water base finish. Anyway, we fixed the door and she stopped putting her car in wet.
My preferred method for exterior doors is 3 or 4 layer solid face glued with grain reversal. This way no exposed glue lines are to the weather and you get stability from the laminations. This only works on 2 ¼” and thicker doors and is more suitable for the Euro constructed doors we do.
The only glue failure I have encountered in 35 years of wood working was with plastic resin glue. I caught it before it went out the door and re-glued with titebond 3. I read a lot of theory on why the glue failed and in the end I thought it was due to over clamping a tight joint which resulted in a starved joint.
I think if you had followed the procedure you did with Titebond 2 or 3 you would be fine.
Hope you can sort it out but probably going to be a redo.
Things that work:
*Liquid resin glue. Use heat to catalyze the glue.
Thanks for the good responses. Not that any of it is what I want to hear. I really feel dumb all of a sudden. Humbled may be a bit more charitable, tho no more accurate. I feel like I am in a confessional....
I knew about thicker -shop made veneer - since that is what we always use on this type of application. I did not cross band since it is all narrow parts - 5" at the most. I cross band when we need to stop all movement, especially in wide rails. I assumed the movement would be consistent with veneer and core all moving together. I like Gorilla glue for quick and simple veneer projects due to its lack of or low water content, but rejected it for this job.
I'm going to wait and see how much further things deteriorate. I am thinking we will have to rework the doors with 3/16" thick Walnut, both sides.
Here is the catch. There is always a catch. The glass in this project is a layered silk art glass that is made into an insulated panel. It's cost is about $8,000 all in, for 10 pcs of glass. We put glass in so it will not leak - silicone on both the exterior stops and the loose interior stops, so it is almost impossible to remove glass without breaking it. So we will have to bring the 200# doors back into the shop, route down the entire surface both sides, apply new veneer both sides, and then machine for hardware, then finish.
Oh yeah, did I mention Tectus hinges and Rocky Mountain custom hardware with a rabbeted strike plate?
Would you like fries with that?
That is a huge amount of work to fix the doors. I would strongly suggest taking the opportunity to do some test panels while you wait for any further possible deterioration to develop on the door.
With all the labor involved to remove and replace the veneer the last thing you want is another failure. I wouldn't just assume that the thicker veneer is going to solve the problem. A few test panels with various veneer thickness set out in the sun near the door would give you a good idea of what is going to happen.
Good luck with the repair.
Bernie -Thanks for the grounding. I do have a history of going off full speed ahead in the interest of getting something unpleasant behind me, causing yet more grief in the process.
A level head is always needed.
"I do have a history of going off full speed ahead in the interest of getting something unpleasant behind me, causing yet more grief in the process."
Gee........sounds just like me. I need to learn to follow my own advice !
Looking at your website, you do nice work, way above the average.
When a person breaks his or the companies policy:
"I am a bit of a Luddite on the subject of engineered stiles."
"I have always made solid stiles and recommend same for others"
"I knew about thicker -shop made veneer"
"I assumed the movement would be consistent with veneer and core all moving together."
"I really feel dumb all of a sudden. Humbled may be a bit more charitable, tho no more accurate. I feel like I am in a confessional...."
A person will always say to himself after an episode such as this, damn I knew I shouldn't have...
At least I always do...
Dark finish & direct sun, ouch. Don't know that there is a perfect fix. Buy them some shade?
We have all been here at some point. You have our sympathies.
Obviously we are not looking at the doors. We have removed double insulated panes from doors without destroying them. They are very durable. The problem is glass actually scratches easier than one would think. That said I would take a couple of breaths before I dove into the veneer with a router.
Personally I would take the glass out and start over if at all feasible.
We have done a ton of vac bagging with epoxy. The thin line problem has never been a problem as long as both surfaces are abraded with 80 grit paper. This provides a decent width of glue line and plenty of surface area. Typically epoxy glue lines fail because of lack of abrasion and/or starvation from too much hand clamping force. The vac bag evenly distributes the clamping force and you would need to crank up the mercury to starve the joint.
Had similar deal, customer used dark red paint. On a 100 degree day that panel measured 187 with our laser thermometer, a test of the white painted side in direct sun stayed right at ambient temp. It was so hot it re-liquefied the pitch in the vg fir and actually made the paint soft. Interesting enough the stiles and rails built engineered over fingerjoint staves stayed good other than a little pitch bleeding through.
I would also check to see if the stiles with the cracks in them are cupped at all ? If cupping is substantial it will crack the veneer on the convex side. Under the extreme heat conditions it would not surprise me. I refuse to fabricate doors with a dark finish that are south facing and ,like you, I learned the hard way. That said there is always risk in the nature of our work and that is one reason we thrive on it. Go easy on your self; we are craftsman but we are not Gods. Dig deep and learn well from everything. Including the mistakes.
I can't advise you on doors, Dave (and anyway you've gotten a ton of good advise already). But in addition to 80x sanding for epoxy, just throttle back your vacuum a bit for less pressure.
David, if you end up re veneering find a local shop with a cnc nester and let them mill the old veneer down. Safe and accurate.
its gremlins, it could be combination of bad glue , ive used it I know, heat, veneer too thin/no cross banding, ive never had any faith in veneers where theres fluctuation in temp/humidity . I build my doors with solid wood and triple lamination on my stiles and rails in order to avoid warp . im not really a specialist in door manufacture so I keep it simple