Face Spline problem.

Fsimard

True Grumbler
Joined
Aug 7, 2013
Posts
50
Location
Québec city
Dear Grumblers,

I have a problem with a batch of frames I did a couple of months ago.
The client ordered 15 large maple frames (42" x 64") which are all assembled using a technique called face spline and then painted white using lacquer.
I use this very often for closed corners white frames and never had a problem.
Last year, I did a 200 frames job using this method, and they still are perfectly joined. No cracks, seamless miter.
Now, the client send back 4 of those frames for a repair, saying that the triangle insert starts to appear on the surface (see the picture attached).
It's maybe due to a temperature variation during storage or rough handling..., but I think it may also be due to a difference of moisture in the pieces of maple I used for the project..
I apologized to my client and will do the paint job and delivery for free, as I always honour my warranty against any defect.

The question is: do any of you that use this technique have seen this problem before? Do you use maple for the inserts while checking with a hygrometer for a perfectly dried piece?
I join my frames with thumbnails, then make the cut for the insert and use wood glue and clamps. Then I use a Bondo kind of putty to seal the cracks before applying 2 coats of primer, sanding with 600 grit and applying two coats of white water-based lacquer.
What are your tips for face splines?

many thanks!

François
 

Attachments

  • Cadre 2.jpg
    Cadre 2.jpg
    68.5 KB · Views: 156
I've never seen splines used in precisely this way. But I agree with your self-diagnosis.

I assume the grain of the insert is running across the miter. Wood tends to swell sideways, across the grain.

I think if I were doing it I would insert the spline away from the face. That way it's glued all around, thus a stronger joint and less likely to swell.

Fortunately it's not a hard thing to put right. :D
 
I agree with Peter, is there a specific reason you have done them this way?
 
It strike me that because of the size (42 x 64) there would be a lot of stress on that corner, even with a strainer.

Doug
 
For this size, I definitively have a strainer with cross-braces to put less stress in the corners.
The main reason I use face spline is to have a seamless finish, wich is common in white lacquered frames. The second reason is a structural one, as Prospero said, the grain is running across the miter.



I did some research on the forum and found this infos about face splines, why to use them and some answers to my questions.


I quote Mr. Waldmann from Vermont Hardwood who is way better than me to explain:

"I suppose the simple answer is, yes. But it's really not that much more. Especially if the face spline is instead of the top-most spline (i.e. you move it from near the top to the face). I calculate that on a 1/2" face moulding there is 3.4" versus 3.2" of joint with a face spline compared to a normal spline and a miter joint.

But there are other (IMO) important factors that trump any possible extra labor.

Because of the physics involved there is much less likelihood of the joint being disturbed after the frame is complete. Similar to the principle of drilling a hole at the end of a crack in a piece of plastic or metal, the face spline spreads the force normally at one point (the center of the joint) across a much wider surface.

The face spline makes a much stronger joint too, at least near the face which is the most visible part of the frame. This is greatly exaggerated on narrow mouldings, because of the limitation of thickness at the stem part of the moulding. On a 1/2" face moulding with a 3/16" wide rabbet there is 2.5 times as much surface area available for a face spline as a "regular" one that goes just into the stem. We have done (unscientific) testing which corroborates the implication.

If you are seeing a line with this type of spline it's most likely that the glue or any filler at the joint wasn't completely done shrinking when the finish was applied. The other possibility is if the spline material and moulding are at different moisture contents (as they will even out with time) shrinking/expanding to equilibrium and showing the difference at the joint."


The result is that I've bought an hygrometer to measure moisture level in the timber I use for these projects.
 
I don't use any oil based paints or primers and have noticed sometimes the water based stuff has a way of seeping in and expanding certain types of glue. Sometimes wood. I have has the best of lick and no problems after using Titebond 3. The waterproof properties make it better for water based paints. Your problem may be unrelated, but worth a try to see if there is a difference.
 
You are to be commended for using face splines that aren't going to be seen.
If it were me, I would probably install that last top spline conventionally, 3/16" back from the face so it's glued on 2 sides so it stays put.
As far as the seamless look is concerned, 2 coats of primer, sanding, and 2 coats of w.b. lacquer AFTER sealing and sanding the crack with a latex filler (Bondo-type filler like you stated)
would get rid of the miter line just as well, wouldn't it?
Also, I would like to ask as an aside and not to jump your post, Francois: if basswood splines are used to join a hard or soft maple frame, would this lead to any
future difficulties like the one pictured?
 
Due to an issue one of my customers was having with face splines I have since theorized one more reason for a fault to show up in face splines, and that is large changes in Relative Humidity.

This issue could arise on a tall moulding made from flat sawn material, with splines made by resawing flat sawn material. The end result is grains generally perpendicular to each other. Since almost (if not) all woods shrink and swell at different rates radially and tangentially, if there is a large enough RH change the spline will shrink/swell more than the frame material, and show a fault. To eliminate this problem you would need to cut splines by ripping thin strips off a board thick enough to yield the needed width, or resaw from quartersawn material.

However, I believe the most likely reason for face spline faults is not allowing enough time for moisture to equalize during assembly. Moisture from the glue and/or any filler must have a chance to disperse, not just from itself but also from the wood that absorbed it when applied.
 
I have always ripped strips from thick enough boards to make my inserts, and then put them in the time saver machine to have a perfectly flat gluing surface.
Maybe these faulty corners have not been sawn the same way, I have never considered the difference between radial and tangential expension in the equation... Very good observation!

These cracks appeared three month after... The frames have successfully been on display in a gallery for a month without any problem...
 
Also, I would like to ask as an aside and not to jump your post, Francois: if basswood splines are used to join a hard or soft maple frame, would this lead to any
future difficulties like the one pictured?

Actually, less likely. You can see from the specs below that the Radial shrinkage of Basswood is closer to the Tangential ratio of Maple than Maple is to itself. So this means that even with large swings of RH, the Bass will move more like Maple than "conventionally cut" maple splines will. And, if the RH is stable, and the Moisture Content of both spline stock and moulding is the same and at Equilibrium MC, there would be no issue.

(from Wood as an Engineering Material USDA Forest Products Laboratory, FPL–GTR–190)
Shrinkage - Radial/Tangential
Basswood, American: 6.6 / 9.3
Maple, Sugar (hard): 4.8 / 9.9
Maple, Red (soft): 4.0 / 8.2

Again, using spline material from the same species, cut from the same grain direction will result in the most long term stable situation.
 
In my opinion the frame profile is oriented radially/tangentially while the spline is oriented longitudinal. The spline longitudinal grain doesn't shrink relative to the profile grain. In the picture you can see the top of long edge of the spline has popped out.
 
Could it also be that the miter of the corner itself has gapped under the face spline in these defective corners? The frames are large, and if the miter is stressed and moves even 1/64", it
loosens the glue bond that the back of the face spline has with the front of the frame. So then it would be much easier for the thin wood of the face spline to do its thing
(swelling and shrinking) independent of a glue bond.
 
In my opinion the frame profile is oriented radially/tangentially while the spline is oriented longitudinal. The spline longitudinal grain doesn't shrink relative to the profile grain. In the picture you can see the top of long edge of the spline has popped out.

When I first opened this thread it was on my small laptop and I couldn't see any details of the picture.

Looking at it now I see what you mean, agree in principle, but I believe that it's more likely that the opposite occurred. That is, the spline shrunk, not the profile.

While it's possible that the frame shrinking in width could produce the same result, the physics involved would make it much less likely. Because of the miter, when a moulding shrinks the outside of the corner will remain in the same location, it's the inside that would moves out, closer to the outside. The leverage required by the positioning of the spline means it would have to be a very weak joint in order for it to let go.

In the first scenario the spline can just start pulling away from the recess cut. The only force required is the glue joint at the point of contact. With the second scenario it has to push across, not just the entire width of the joint, but that force is also spread across the entire surface area of the spline.

Of course, one last scenario would be that both spline and moulding were at the same (high) MC and both shrunk. In that case, I'm really not sure what would happen but it would seem likely to end in disaster of some sort.
 
Could it also be that the miter of the corner itself has gapped under the face spline in these defective corners? The frames are large, and if the miter is stressed and moves even 1/64", it
loosens the glue bond that the back of the face spline has with the front of the frame. So then it would be much easier for the thin wood of the face spline to do its thing
(swelling and shrinking) independent of a glue bond.
I did some stress test with these assembly and it's impressive how much force it takes before breaking or cracking. A good basswood strainer is also the key to a strong joint.
But rough handling can also cause these kind of damages.. I'll never know how they handled these pieces.
It travelled 300km by truck in november (here in Quebec, it's pretty cold!!) and then it was shown in a gallery for a month without any structural problem. They discovered these cracks three month after in their storage space..

But anyway, I'll do the repair and in the future, I will pay more attention to the way the wood is splitted, his moisture content and glue drying time before applying the finish. The insert should have exactly the same structural properties as its surrounding environment.
 
Here is a picture of the last spline inserts I used for this project. Usually, I use 2" timber to make them radially.
We can see that these new inserts are oriented longitudinally and it's the main structural problem for these frames.

The second image explains very well these expansion rate differences.

IMG_0683.JPG woodscience_2.jpg

Thanks to Mr. Waldmann and other Grumblers for all the infos and feedback!
It's always a precious source of knowledge here!

(And it makes me practice my English!!!)
 
What brand of lacquer do you use? I like the finish it produce.
 
What brand of lacquer do you use? I like the finish it produce.
I use this: http://chemcraft.com/index.php/en/products/products-quebec/waterborne-qc#aqualac®-ii-topcoat
Aqualac II as a finishing coat, pre-catalysed water-based lacquer.
And Aquaprime as a primer.

These are not available in the US, but they have similar products for the US market.

The picture of my broken miter is not really good as it looks textured, but the finish is satin smooth.
I use an HVLP turbine system.
 
You are to be commended for using face splines that aren't going to be seen.
If it were me, I would probably install that last top spline conventionally, 3/16" back from the face so it's glued on 2 sides so it stays put.
As far as the seamless look is concerned, 2 coats of primer, sanding, and 2 coats of w.b. lacquer AFTER sealing and sanding the crack with a latex filler (Bondo-type filler like you stated)
would get rid of the miter line just as well, wouldn't it?
Also, I would like to ask as an aside and not to jump your post, Francois: if basswood splines are used to join a hard or soft maple frame, would this lead to any
future difficulties like the one pictured?
I give it a try for some test frames!
That way, the joint is reinforced very close to the face, and directly to the weakest point of the miter: where the glazing sits.
 

Attachments

  • IMG_0689.JPG
    IMG_0689.JPG
    42.6 KB · Views: 42
I give it a try for some test frames!
That way, the joint is reinforced very close to the face, and directly to the weakest point of the miter: where the glazing sits.

I'm not sure of the actual dimensions, but based on proportions of the spline size, we would do 2 splines on the smaller one, and three on the larger. Unless you are using some other method of reinforcing the joint.
 
Yes Francois. You use thumbnails to hold the frame together, and then a top spline to reinforce the miter up front where the glass is- the point of highest stress.
I join the frame half-way using v-nails at the bottom to hold the frame together- and I use 2 splines in the top half of the profile. Strong joint either way.

Bon chance mon ami. J'aime beaucoup Quebec. Je fais mes vacances chaque ete a Mont Laurier et Lac du Cerf. Beaucoup des grand brochets, achigan, et dore.
Tout le monde est content de voir venir l'ete de Quebec!
 

Attachments

  • IMG_1574.JPG
    IMG_1574.JPG
    50.6 KB · Views: 35
Yes Francois. You use thumbnails to hold the frame together, and then a top spline to reinforce the miter up front where the glass is- the point of highest stress.
I join the frame half-way using v-nails at the bottom to hold the frame together- and I use 2 splines in the top half of the profile. Strong joint either way.

Bon chance mon ami. J'aime beaucoup Quebec. Je fais mes vacances chaque ete a Mont Laurier et Lac du Cerf. Beaucoup des grand brochets, achigan, et dore.
Tout le monde est content de voir venir l'ete de Quebec!

Nice frames! I give it a try and will paint them to see! It's really fast to join this way, I use a biscuit joiner to make the kerf and a thumbnailer for the assembly ( I wish I have an Hoffmann... )

Fishing is also one of my great passion... Especially fly fishing!
Votre français est excellent!
 
Francois,
I seem to be having the exact same issue with my painted frames with face splines. My shop is not tempertaure controlled so I can imagine that can be part of the problem. I am wondering what worked for you in the end? Im interested in trying the basswood splines instead of the maple.
 
Francois,
I seem to be having the exact same issue with my painted frames with face splines. My shop is not tempertaure controlled so I can imagine that can be part of the problem. I am wondering what worked for you in the end? Im interested in trying the basswood splines instead of the maple.
Hi Marlon,
First, I apologize for the late answer... I’m not here very often...
I've stopped using face splines, because I live in a country that has huge temperature and humidity differences between winter and summer; It affects how the wood is reacting and it’s often unpredictable.
The best results I had were with maple splines, with the exact humidity percentage as the moulding used. It needs to be exactly the same density, weight and structure to expand and contract with the frame.
A good miter done with a Morso, combined with the Hoffmann system is for me one of the key for seamless joints...
 
Back
Top