avoiding poorly fitting joints

[tablesawtom]

Cooler made a statement in a different post that I wanted to reply to, but didn't want to sent the post in a different direction so I thought I would start a new post. I am not picking on Cooler nor to put him on the spot, I was thrilled that he wrote it. I am just trying to point out a few facts using someone else's observation and how he or they dealt with it. And hopefully a better way of dealing with it.

His statement is this,


[quote pid='7726092' dateline='1551274994']
I use a tenoning jig.  I will have to think how that would work for me.  But I've found that the 3/4" stock is not 100% uniform in thickness (and I don't run the boards through a planer).  So I wonder if that will always give a nice tight joint.  

If I buy all the lumber at the same time I can make sure it is all the same thickness.  But otherwise I have to pick and choose to make sure it is the same.
[/quote]

On most of the TV shows, it is shown that you cut a tenon by setting the height of the blade ( which can be hard to do)  and then just flip the work piece around and you come up with a perfectly fitting tenon every time. It started with The New Yankee Workshop and it continues even to this day. And sadly this is where a lot of people get their information. If you read this word for word and leave basis out of it I hope it will solve some or better yet stop some your problems from a curing in the first place. Hopefully this is for those who might like a better solution to a common problem. 

But as Cooler has stated the joint fit can very as the thickness very's  And he stated his solution, buy all the lumber at the same time.  But the fact is, just buying all the wood at the same time doesn't mean all the wood was run at the same setting or even on the same day. or even from the same tree.

 And given the fact that he doesn't plane it,  probably means that he doesn't have a planner. And it is a problem a lot of woodworkers have for various reasons.  A few  it could be that they do not have the money, no room just starting out and trying to make do as they go. And that is why I am writing. 

Even with a planner the problem of exact thickness still exists. If the head on a planner is not in perfect alignment, and they never are, close but not perfect with the bed, one will have the same problem with fit. If the blades aren't in perfect alignment or part of the blade is duller than a different part of the blade or the pressure from the feed rollers are different from side to side or the wood was from to different trees and the density is different, the thickness will very.  

And thickness effects the fit, period. and to those who will write and say they have don't or never have or had the problem  to which I  saw hog wash. 

Whether you are setting the height of a dado blade or setting the distance of a tenoning jig. the principle and process is basically the same.  Let me give you am example: you are doing frame and panel construction that is 3 or 4 feet long. and as seen on TV you want the groove perfectly centered ( why I don't know but because they say it it must be true) so you flip the board around and take the second cut . And the plywood goes in from each end okay but get stuck in the middle. Fact even if your style is perfect the plywood may or may not be.  Not to mention the pressure of the board against the fence  an various points along the way. The rails are then test cut and fit and then cut and flipped and cut. And as my luck works out, one will fit one will be tight and one will be loose.

I would like to point out one thing as an example. it you are using a style and rail cutters on a router table,. the fit is is always correct, Why? If it mismatches on both sides it is a cutter height error,  if the mismatch is on one side then it is a thickness mismatch but the fit remains correct. WHY, because the fit is being controlled by the cutter height and the distance between the two cutters. AND the datum surface of the wood is in contact with the datum surface of the  router table surface. If it is flipped over you are changing datum's and working from two different surface Datums of the wood. Now thickness is a problem and it doesn't take much to get a poorly fitting joint. 

 A datum is a surface, line, plane, or feature assumed to be perfect.  It does not carry a tolerance, but it is the surface that everything is referenced from. An example:  the height of a router bit is referenced from the top of the router table not the underside of the table. Which is what you are doing if you flip the wood.

Solution to the problem.

If the groove is cut on the table saw one cut is taken and then the fence is moved and the second cut is taken, the thickness can mismatch by an inch but is the groove width remains the same and the same distance from the fence is maintained, if the same datum surface is uses for all cuts, And as a result the thickness problem doesn't come in to play. Centered is a moot point,  it doesn't have to be but tit sounds good on TV and helps hopefully to give them some credibility, not to mention to fill up a half hour Slot.

If a tenon is cut the the distance from the face and the jig remains constant so the distance from the wood datum ( only that surface) remains constant. A thousand pieces and it will remain the same,  there is always human error for someone who wants to find fault with the statement. And if the face is then re- positioned then the proper fit will remain the same cut after cut. The one side can be run and flattened on a joiner and even if the wood was bowed before it was flattened the fit will be correct.

If you rethink flipping the work piece and working from just one surface I think your fits will improve..

Tom

[AHill]

My joinery improved quite a bit once I added a jointer and planer. Lacking those, you still need one FLAT surface to use as the reference datum. That is, one flat surface on each piece. Otherwise, no matter how well the joint fits, twisting, cupped, or warped stock will cause problems on the other end. This is why at least a jointer is important.

[Woodenfish]

You can hide poor joinery caused by various board thickness. Before or after glue-up a vee-groove or even a saw cut right before the joint can create a breakup of the two adjoining surfaces eliminating any discrepancies in board thickness or future joint movement and misalignment. If purposely done all over it then becomes a design point versus a construction error.

[tablesawtom]

My joinery improved quite a bit once I added a jointer and planer.  Lacking those, you still need one FLAT surface to use as the reference datum.  That is, one flat surface on each piece.  Otherwise, no matter how well the joint fits, twisting, cupped, or warped stock will cause problems on the other end.  This is why at least a jointer is important.

Yes a jointer is important and yes It helps to have a planner.  A jointer is great at flattening a cupped board but a board that is bowed the length is much harder to deal with. And if you are fortunate to have one great. but not every one does.

 I have an 8" and I love the long infeed bed. To be truthful,  Some say the jointer should be as wide as the planner. My jointer blades get indicated in using the out feed table as a reference surface. The blades are indicated in with no more than .002 total indicator run out.   So that every blade cuts the same and the infeed is set at .015 for a depth of cut.  .0156 is equal to 1/64. So I know how much I am taking with each cut. Downward pressure on a jointer is a no no when trying to flatten a board, and 1/64 q butt makes it a lot easier to get it flat.. I do not use carbide insert cutter heads because I think there is more cutting  pressure and to me it takes more force to take the cut so I stay with sharp High speed knives. If you push down on a bowed board when it is on the infeed side you will get a bowed board when the downward pressure is released on the out feed side. 

Please this is not the place for discussion on why carbide is better, save it for a different post. It is about doing everything I can to get a board as flat as I can possibly make it and my planner is set up the same so that I can get a board as perfect as I can, reliable as I can, and as often as I can. 

And do you know what, as hard as I try I still can get tight , good or loose joints  because a lot of times wood has a mind of its own  and I want to get a joint as good as I can  the first time so I don't flip boards and change datums. 

There is an old saying in the trade. If you don't have time to do it right the first time how do you figure you have time to do it right the second time. Or how come there is not enough time to do it right the first time but there is always time to do it right the second time. And to some  if you don't like how your tenon joints turn out, why do the things the same way time after time and then expect different results.

To enjoy this hobby you have to enjoy the steps and a good joint will stand the test of time. A poor joint will not. Just trying to pass along information that hopefully help someone to get good tenon joints repeatedly.

Tom

[jteneyck]

You can hide poor joinery caused by various board thickness.  Before or after glue-up a vee-groove or even a saw cut right before the joint can create a breakup of the two adjoining surfaces eliminating any discrepancies in board thickness or future joint movement and misalignment. If purposely done all over it then becomes a design point versus a construction error.

That doesn't fix poorly fit M&T joints.  Tom was pretty long winded but definitely on point, IMO.  Use a common reference face when cutting joints and they will fit well regardless of thickness variation in the individual pieces of stock.

[tablesawtom]

You can hide poor joinery caused by various board thickness.  Before or after glue-up a vee-groove or even a saw cut right before the joint can create a breakup of the two adjoining surfaces eliminating any discrepancies in board thickness or future joint movement and misalignment. If purposely done all over it then becomes a design point versus a construction error.

 A design point and hiding poor joinery are to different animals. Shoddy rough work bleeds through in the finish and shows. Don't be deceived you will reap what you sow. Even if you want the joint to look old and coming apart there is still no excuse not to get the fit right so the the joint will still look like it is coming a part 50 years from now and not to have come apart in 6 months.

Quality is long remembered after a cheap price is forgotten. 

Whats in your wallet?

[Woodenfish]

My comments were made in order to inform someone of a potential fix for a found error within the scope of having no planer to accurately machine raw lumber from the beginning.

John, your recommendation is the same as the original problem you just moved it from one side to the other. That may help and then again may not for a particular situation.

Tablesawtom, there was no inference to complete shoddy joinery by me. My point is to think outside the norm and find a solution to an ailment with the material already machined under listed circumstances. No one should feel limited by not having the latest, greatest, mostest tools and equipment in order to build a satisfying project.

Improvise, adapt, overcome. USMC

[Jim Waldron]

Maybe I've got no place in this conversation and should keep my big mouth shut, but I have a couple of thoughts that may be interesting:

I'm a hybrid woodworker, mostly hand tools until it's right to use machines.  

I recently built a "crib bed" for my Grand daughter.  Interesting design with 32 vertical slats (Western Red Cedar) with tenons on each end and fitted into mortices in top and bottom rails (also Western Red Cedar).  That's 64 mortices!  Not being a masochist, I cut the mortices 1" deep with a 1/4" plunge mortiser, spaced from a reference surface.  Easy peasy, no sweat, neat and proper result.  (My mortiser is not the highest quality tool and takes a bit of fiddling to achieve acceptable accuracy. Took about 3-1/2 hours to cut all 64 mortices.)

But then there were 64 tenons.  The slats were nominally 3/8" thick by 1-1/2" wide.  The tenons were designed at 1/4" by 1" meaning some very delicate work was at hand.  Don't have a router table.  Don't have a tenon jig for my table saw.  Don't trust my table saw technique to pull this off.  Very comfortable with doing it with hand tools, so off to the bench to mark out and get it on.  

Marked the shoulders with a knife wall to exact length.  Marked the width with another knife wall and sawed those bits with a dovetail saw.  The thickness was addressed last; working from a reference surface, the depth of cut was marked by setting the depth on a router plane and marked the depth on all 64 tenons.  Then I reset the router plane to a depth just 1/64" more than a 1/4" tenon thickness, measured from the SAME REFERENCE EDGE and marked the depth on all 64 tenons.  Using normal router plane technique, I was able to shave the faces of the tenons to thickness in two passes of ~1/32" each to depth; it took about 5-6 minutes for each tenon, including cutting a chamfer on the bottom edges to ease fitting into the mortices.  (I did stop half way through to sharpen the router plane blade and since it was late, I left half for the next morning.)  All told, the tenons took about 4-1/2 hours total.  Being Western Red Cedar, the very slight over size was overcome neatly by compressing the wood a tiny bit on assembly and provided an excellent fit of all 64 joints with no fiddling.  In hardwood, a tighter tolerance would be needed.  

Hand tool guys are all trained to work from a reference face (or sometimes a reference edge) for all procedures.  Tom's advice is not news to the hand tool guys.  It's just basic methodology.  Guys who move over to hand tools from tailed tools sometimes take a while to register and digest the advice they get from experienced guys, but they do get it sooner or later.  It's good advice for a lot of machine techniques as well, as Tom has pointed out.

[jteneyck]

No need to apologize for joining the conversation, Tom.  Your input shows that the method used to remove wood is not nearly as important as the methodology.  

Dear me, though, if I had to cut 64 M&T joints by hand I think I'd find a new hobby.  My hat's off to you.  

John

[FS7]

My only addition to this thread would be that it's prudent to remember that we are all woodworkers, not machinists. I often see people talking about arbor runout and thickness differences in the thousandths range. Visible gaps might bother you, the craftsman, but the end customer may not ever see a gap or think anything of it. It may not (almost certainly won't) affect the structural integrity of the piece.

Generally speaking we are working with a natural material that is highly susceptible to environmental changes. It is often impossible to cut, let alone maintain, with that level of accuracy. A machined piece of aluminum, sure. But like I said, I'm not a machinist.

I'm also a hypocrite, as I am more of a perfectionist than most.