Techniques

Tenon Details

Mortise and Tenon joint is the first woodworking joint I learned to cut.  At first, it was tenons by hand and mortises by plunge router. Closed mortises, roughly half the material deep, and stub tenons, with four shoulders. Everything co-planar and nothing fancy. Words like “reveal” and “haunch” were unknown to me.

Even the picture is poor quality.

One of the first joints I ever cut.  Even the picture is poor quality.

After a few months of voraciously consuming woodworking scholarship (mostly via YouTube) and much trial and error, I knew enough to plan a specific reveal for a joint. A nice roundover leading to a slight reveal can make a world of aesthetic difference and can be accomplished by simply varying tenon shoulder depth.

This is a vanity sink I built for my brother's first house.

The vanity sink I built for my brother’s first house.  That’s the matching footstool on the right.

Later, I learned to always cut mortise and tenon joints to align the long-grain gluing surfaces (and maybe even do a double mortise and tenon) for maximum strength. This allows for different board orientations (and more delicate-looking frames).

Later still, I discovered the haunched tenon, and I’ve been using it as much as I can ever since.

They all basically look like this.

This is the angled-leg bedside table.

I understand the true purpose of the haunch. It permits maximum tenon height (for maximum resistance to twisting force) while still keeping the mortise closed at the top (and thereby stronger than an open mortise, such as a bridal joint). But I also appreciate the economy of the joint. The haunch means I have to chop less mortise overall without compromising joint strength. On a rectangular table with eight mortises (of 4 inches or so each), a half inch haunch saves me a full mortise-worth of chopping.

I also rarely do full length tenons, preferring a bottom shoulder of some depth (usually the width of the mortise, because a chisel is a handy straightedge) to hide any roughness of the mortise hole. Assuming a bottom shoulder of another half inch, that’s a whole extra mortise worth of chopping saved in a single project. And with the haunch on the top of the tenon, I’m only cutting three shoulders – maybe even only two shoulders, if the material is thin enough.

All this time savings means I get more done in the shop with less effort.  And best of all, none of these “shortcuts” ever show in the finished piece.

JPG

An Obvious Downside

I’ve mentioned a couple times that my main source of cheap pine is off-cuts from tongue-and-groove pine siding.  I’m running a bit low, but I’ve managed to make a three-foot tall stack of 36″ to 48″ long boards last for most of a year.  While the material is fluffy and easy to work, it isn’t exactly stable, and most of the time, is significantly twisted and cupped (thankfully, not often bowed).  This presents an obvious problem: final board thickness.

Before preparation, any given piece is only about 11/16″ thick, and the “raw” board can have almost 1/8″ of cup/twist or more.  This means that after flattening and straightening to S2S, I’m often a barely a shade over 1/2″ on some parts of the board.  Passing the piece goes through the thicknesser takes it down to a hair over 1/2″ all around at S4S.

P1000072

This pile of shavings is from S2S’ing only two 19.5″ x 4″ boards.

This is why I’ve gotten in the habit of only S3S’ing these boards when the project can allow for it.  For example, the bottom shelf of the toy workbench was made of four such boards.  The underside of that shelf will never be visible, so I left the extra mass in the boards by only S3S’ing them.

P1000066

That’s a machinist granite slab adding clamping pressure.

I also use this trick when I need an inside tenon shoulder or when a board is too wide for my thicknesser, in each case where the inside face will otherwise be concealable.

This trick isn’t available in some applications, however.  For instance, I can’t attach drawer runners to the twisted inside face of a side rail (like for the angled leg side table) and the underside of a tabletop should be trued to mate well to the table frame.  I guess you could technically use rabbets to solve each problem, but that seems like even more work than just thicknessing properly.

I haven’t disassembled much antique furniture, but I find it difficult to believe I’m the first person to cut this corner when possible.  I think this shortcut thing is becoming a running theme.

JPG

Shortcuts That Make Sense

I think I first heard the expression when I read the Fellowship of the Ring for the first time as a teenager:  “Shortcuts make long delays”.  I always took that expression to heart and I try not to cut corners in my woodworking (other than breaking the arrises). Sometimes, though, shortcuts make sense.

For instance, I rarely cut through tenons.  Unless they are for a specific design element, or the wood is particularly thin in the first place, through tenons just aren’t necessary  They look cool, and I guess a through tenon might add strength in the right situation, but why do the extra work if you don’t have to?  For that matter, why cut a full length stub tenon when a shorter stub tenon will do just fine?

Side rails to keep the legs rigid and square.

Test-fitting the side rails for tight fit between the shoulders and the legs.

On the mini workbench, I had cut some 1.125 inch tenons on the side rails (see above).  Then I looked at the thickness of the legs themselves (1.5 inches) and the position of the rails (flush to the underside of the bench) and determined that a .75 inch stub tenon would be more than sufficient.  A .75 inch tenon would translate into a mortise that I could chop in a single pass back and forth, saving time overall.  So I hacked half an inch off each of the tenons.

So I hacked half an inch off the tenons.

It’s not lazy; it’s efficient.

The side rails will still do their job  with the shorter tenon (i.e, keeping the legs rigid and square front to back), but that’s more a function of the tight fit between the tenon shoulders and the leg than a function of the dept of the mortise.

Stub tenon or not, that's a tight fit.

Stub tenon or not, that’s a tight, square fit.  And yes, I keep my miter box on the floor.

Speaking of shortcuts, I have also decided I’m not going to mortise the front and back rails into the legs.  Instead, I will half-lap and screw them into the insides of the legs, mainly because I’d like the ability to remove or re-position the slatted shelf if necessary in the future.  Mortising those rails into the legs would make that much harder.  Plus, it will save me a bunch of time (and I really enjoy the ease of cutting half-lap joints).

So next time you’re planning out a woodworking project, think long and hard about any available efficiencies.  I’m not saying scrap the corner dovetails on your tool chest in favor of nailed rabbet joints (although that would be perfectly fine if done correctly).  But I am saying don’t go overboard if you don’t need to do so.

JPG

Through the Gauntlet

Sticking with the theme from Michael’s guest spot last week, let’s talk about preparing rough stock for joinery.

One of my most favorite things to do in the world is plane a piece of wood S2S (i.e., Surfaced Two Sides) by hand.  Going from rough stock to two perfectly straight, perfectly square sides entirely by hand is such a joy.  A face mark, then a square mark and some arrows to indicate grain direction.  Onto the next board!

The pile of wood grows almost as quickly as the pile of shavings.

And the pile of boards grows almost as quickly as the pile of shavings.

One of my least favorite things to do in the world is plane a piece of wood S4S by hand.  Going from two sides perfectly straight, perfectly square to four square entirely by hand is such a pain.  And that is why I still own and often use a thickness planer (it lives at my parents’ house), even though I do mostly hand tool woodworking.

Ready to load into the car.

Ready to load into the car.

On thicker stock like the above Douglas Fir which is for a miniature workbench for my niece and nephew (more on that another time), it is so convenient to send the S2S boards through the thicknesser to clean up the remaining sides.  Then just a quick smoothing plane to take out any plane marks and I’m good to go.  On thinner stock, I’ll actually thickness just one face (making it S3S?) and use the traditional method to square the final side.

Top piece is now S4S; bottom piece is now S3S.

Top piece is S4S straight out of the thicknesser; bottom piece is S3S(?) and ready for ripping/final squaring down do the gauge line.

I quickly came to realize that when thicknessing it’s imperative to know grain direction.  Tear-out from power planing is so much worse than tear-out from all but the heaviest set hand plane. That’s why I’ve gotten into the habit of indicating grain direction on all my S2S boards.  Keeping track of which way to pass a board through the thicknesser can mean the difference between a light pass just to square up a side and having to mill down another 1/4 inch to get to clear grain (or, even worse, having to mill another board entirely).

My thickness planer may be an hour and a half round trip drive away (plus the time for milling), but it ultimately saves me time and frustration.  And that is good enough for me.

JPG

A Rectilinear Corner

Editor’s Note: today, we’re doing something a bit different.  Michael, a friend, fellow-woodworker and fellow-lawyer, whom I have known for literal decades, has graciously agreed (after much prodding by me) to write a guest spot for TheApartmentWoodworker.com.  I think Michael brings up a prescient point about the modern woodworker.  In the past, many artisans would have contributed to a single piece of furniture.  Nowadays, hobbyist woodworkers are convinced they must be jack-of-all-trades craftspeople.  This is certainly not true, and there is great value and efficiency in outsourcing certain project parts. I also learned a new word (rectilinear). Thank you, Michael!
JPG
– – –
Like James, I woodwork in a small shop in my spare time. I use mostly handtools: not only because I enjoy them, but because, for me, they’re simply more convenient. I believe in using the most efficient tool for the job, and when the job involves cleaning up after yourself, and not tracking dust through the rest of the house (you know, the one your partner just cleaned), then handtools quickly score points on the efficiency scale.
But when the snow melts, and I can pull out a contractor’s table saw and rip down some boards in the yard, the efficiency scale leans the other way. Could I have done the job with handtools? Sure, but who cares? This is a hobby and I only have a limited time to do the things I enjoy, one of which is creating beautiful pieces of furniture.
I love the subtle camber on the front front (I hope that not just camera bend).

Editor’s Note: I love the subtle hollow on the front frame (at least, I hope that’s hollow and not just camera distortion).

That’s why I am a firm believer in outsourcing some woodworking tasks when it makes sense to get the project done. For example, I don’t turn my own legs. I don’t have a lathe, or have the room for one.  I’ve never even tried turning. Maybe I will someday, but for now, when a project calls for a turned leg, I go to places like tablelegs.com. There are other excellent websites as well, where a skilled turner will make you any style leg you want – from cabriole legs to bun feet. I can usually find what I’m looking for, or at least something close that I can modify to fit my project. The quality is great, and consistent.  The point is, my lack of shop and skills shouldn’t pin me into a rectilinear corner.
Bun feet!

Editor’s Note: Bun feet!

Sure, I could take a class at the local woodworking club and eventually develop the turning skill to do it myself.  But I want to actually finish my project this decade. Again, I say who cares?  And by outsourcing I am supporting other woodworkers.
But please don’t order your legs pre-mortised.
Don't be lazy: haunch your mortises.

Editor’s Note: Don’t be lazy: cut (and haunch) your mortises yourself!

The Trusty, Underappreciated Rub Joint

I own lots of clamps, but an unfortunately low number reside in my apartment woodworking shop. As a result, as often as I can, I turn to the rub joint.

The Rub Joint: for when literally any amount of clamping pressure will do.

It’s a simple joint, really. A thin film of glue is applied to two jointed boards, which are rubbed together back and forth until the glue becomes grabby. The glued-up boards are then left to dry under gravity and atmospheric pressure only. It works well with other types of joints, but edge-jointing smaller tabletops (the above is for a footstool) is where the rub joint really shines.

There are three tricks to ensuring a proper rub joint, though:

First, the jointing must be crisp and square. Gaps in the mating surfaces are the bane of a good rub joint, even when using gap-filling adhesives, such as epoxy.

Two, there is such a thing as too much glue. Use only a thin film of glue, because without true clamping pressure to force squeeze out, the boards will float on a thick film of glue and the edges won’t meet in a strong, straight joint.

Three, leave the glued-up boards to dry on a flat, level surface. You’re using gravity and atmospheric pressure as your clamping force, so a flat, level surface to support the work will ensure the glue joint is perpendicular to gravity and atmospheric pressure as it dries.

A tight, strong joint with no clamps.

A tight, strong joint achieved without clamps.

A well-executed rub joint is just as strong and gap-free as a clamped joint. Once you master the rub joint, you too may find that sometimes, very occasionally, you can have too many clamps.

Or, you know, just can just use pinch dogs.

JPG

Nesting Saw Bench

Been ripping a significant number of longer boards lately for the planing slab project, which motivated me to finally make a smaller, companion saw bench to nest inside my main saw bench (both are based on the Christopher Schwarz 2009 version).

wpid-20141220_133010.jpg

I’m psyched to use something other than a dovetail saw for joinery, for once.  Lap joints are the best.

All stock preparation and joinery only took a few hours and I am very happy with the result.  I just need to let the glue dry before I fit the stretchers and level the seams and feet.  Should definitely make ripping easier, as I can now support both ends of a long board won’t need to stop halfway through and flip it around anymore.

Also, I started using the wagon vise on my bench like a front vise for thicker stock that wont fit in the actual front vise.  Works well enough; still need to make an actual moxon vise at some point, though.  I guess I’ll add it to the to-do list.

wpid-20141220_122804.jpg

Small-space woodworking is all about improvising! 

Hope everyone has their holiday shopping wrapped up.  I’m one home center gift certificate away.

JPG

A Planing Slab In Process

Ever since the LED light arbor project (linked here), I have been searching for a reliable way to support 5-6 foot boards for face planing on my 4 foot worktable. For thicker stock, the overhang didn’t really matter; the work was rigid enough to support itself. On thinner pieces (less than 6/4 or so), although edge planing was usually fine, flex at the ends while face planing caused me to unknowingly plane humps into the work, no matter how sound my technique.

wpid-20141008_083040.jpg

Just not enough tabletop real estate for face planing longer, thinner stock.

After much debate (I even consulted Christopher Schwarz on the issue, whom I thank greatly for his quick response and prescient input and whose Lost Art Press I once again plug), I decided to laminate a planing slab (like a Japanese workbench). Made from inexpensive materials, the slab would be rigid and flat and could be clamped to my worktable with angle irons or leaned against the wall when not in use. It would also have a Veritas inset vise and dog holes for Veritas planing stops and holdfasts.

Rough sketch in hand, I went to the home center on a lazy Sunday and dug through the 2x10s for an hour or so (the 2x12s were so saturated they were almost too heavy to move on my own) and found 36 feet of the driest, straightest, clearest Douglas Fir 2x10s available (all from the bottom of the pile, sadly). In the parking lot, I used two lumber carts as saw benches and piled the six 6 foot lengths in the car.

Quite proud of myself for this one.

Quite proud of myself for this one.

Ever since, in my spare hours, I have been hand ripping 3 inch wide lengths, face planing the gluing surfaces and laminating the pieces, one at a time (Paul Sellers’ workbench playlist on YouTube is a great how-to for preparing dimensional softwoods for laminating into a bench top).  The whole lamination should be approximately 18 inches wide (which equates to about thirteen boards) , 72 inches long and 2.75 inches thick when finished. I think, though, that when the slab gets to be about 10 inches wide (seven boards or so), I will flatten the bottom by hand and skip plane it with my thickness planer. That way, I can more easily finish up the second half of the lamination, using the first half as an (albeit smaller) planing slab.

Only four more laminations to go and I'll have half a planing slab!

Only four more laminations to go and I’ll have half a planing slab!

Once I have the space, I will probably re-purpose the planing slab as the bench top for a proper workbench for lighter work (and maybe add a face vise). Until then, though, I am looking forward to having a real planing surface, so I will keep laminating.

JPG

Shavings Piled Like Snowdrifts

Sawdust, wood shavings, waste wood. Hand tools or not, going from rough-sawn lumber to finished furniture makes a mess.

The main benefit of hand tool woodworking is the significant reduction in breathable particulate matter (i.e., wood dust) in comparison to power tools. Not a total elimination, of course, but with hand tools you don’t need breathing protection just to cut a joint. Ditto for hearing protection (although chopping mortises in hardwood in an apartment sometimes calls for earmuffs and sound dampening technology [more on that in a few months]).

That having been said, whether by scrub plane or power jointer, by tenon saw or table saw, by mortise chisel or plunge router, you’re still removing stock and the waste has to go somewhere. And that somewhere tends to be the floor around the workbench.

In my shop, I never let the shavings pile like snowdrifts (as romantic as the image may be). After each task, the waste is swept and sorted. Chips in the trash, shavings in the drum liner [seen in the background of most pictures], dust in the shop vacuum. Shavings and wood dust are slippery and there are few things worse than extricating embedded splinters from my living room rug because I tracked chips all around the house.

A clean shop is a safe, happy shop. When your shop is also your home, that is doubly true.

JPG

A Dovetailed Box

Over the last couple weekends (in what little free time there was), I have been making a simple dovetailed storage container for my car.

Not winning any awards, but definitely sturdy enough to hold a pair of aquasocks, a multi-tool, a portable air compressor, an emergency water bottle and some miscellaneous supplies.

Not winning any awards, but definitely sturdy enough to hold a pair of aquasocks, a multi-tool, a portable air compressor, an emergency water bottle and some miscellaneous supplies.

I’ve been meaning to replace with something more permanent the heavy duty cardboard file box I use in my car to maintain organization of odds and ends. Finding myself with some leftover 3/4 birch plywood, I decided it was time to dovetail a box.  I don’t cut dovetails often and I like these little, non-critical projects to keep my skills up.  Mostly, though, this project was good opportunity to test the capabilities of the Milkman’s Workbench in a situation where perfection is non-essential. Spoiler alert: it performed admirably, as advertised.

Just to be clear, I’m not here to teach you how to dovetail.  If you are looking for instructional videos, there are wonderful dovetailing tutorials all over teh interwebs (I highly recommend Paul Sellers or Rob Cosman).

Dimensioning:

I want to say from the get-go that plywood is not ideal for dovetailing (I rarely use the stuff anyway) and dimensioning sheet goods with hand tools is almost more trouble than it’s worth.

I typically rough cut sheet goods with a jigsaw and use a flush trim bit and a straightedge clamp to clean up and square the edges, but I am inside again and plywood dust is gross.  Instead, I rough cut four pieces by hand (2 sides and 2 ends) and planed straight and square with a #4 bench plane one end piece and one side piece to use as patterns.  I then lined up the factory edges and used double-sided tape to attach the pattern piece to the second rough cut piece (show-side to show-side, which will be important later).  Using the affixed pattern piece as a guide, I planed down most of the overhang on the rough piece to approximately 1/32″ (to have as little waste as possible) and cleaned up the final edge all around with a flush trim router bit.

wpid-20141026_202006.jpg

Less waste to remove means less gross dust to clean up.

The bottom piece, which is 1/4″ plywood from the craft store, floats in housing joints in the assembled box and is just hand ripped reasonably square and straight and fine tuned with a block plane.

Joinery:

There shouldn’t ever be a ton of stress on the glued up joints, so at a little under 12 inches high overall, I went with three tails per corner.  I’m a “tails first” kind of guy and typically I prefer to hog out most of the recess waste with a coping saw (then chisel down to the line), but go with what works for you.

wpid-20141101_184513.jpg

Layout is simple with a decent dovetail marking guide. I use the Veritas version.

I did not separate the now-identical side pieces prior to sawing the tails.  I had a good reason, however: supporting the show-side to prevent blowout is critical when dovetailing plywood.  By leaving the pieces affixed, the show-side of each piece acts as a backer board for the show-side of the other.  As an added bonus, the tails you saw are (hopefully) identical.   Plus, I find working on the wider stock makes sawing perpendicular easier (on plywood, you should also eyeball perpendicular against the layer lines).  All in all, it saved some time and, again, not a museum piece here.  I note that this trick does not work for cutting pins when you cut the tails first; you will need to separate the pin boards.  It may work if you cut pins first (I think the pins may be asymmetrical, which may be a neat design element), but I haven’t tried it.

Tails first.

Tails first.

A coping saw and a chisel made quick work of the first set of recesses between the tails.  All of a sudden, just after I finished sawing the second set of tails, the double sided tape gave way and I had to finish up the tail boards individually.  I also decided to put down the coping saw and chiseled out the remaining pin recesses (more on that later).  Beware cracking the veneer when chopping out the waste with a chisel.

I then scribed and sawed the pins in the usual way, also using a chisel to cut the tail recesses.

Not bad for plywood

Not bad for plywood and being out of practice.

I routed the 3/8″ deep housing joints for the bottom piece with a 1/4″ up-spiral bit and a fence (dust extraction works much better in this router orientation) and tweaked the fit of the bottom piece with a block plane.

Assembly, Gluing and Finish:

A backer board also comes in handy when seating the joints during final gluing.  I always use a piece of wood to cushion against the mallet blows when seating joints and the support for the tails guards against chipping and blowout if/when the joint is a little on the tight side.  Unlike solid wood, plywood de-laminates and cracks instead of compressing, even when lubricated with glue, so the extra support goes a long way.  There is a chance the box might get wet, so I opted for overkill with Titebond III.

wpid-20141115_133305.jpg

Mechanical joints mean single-direction clamping pressure.

After the glue had dried, I flushed any protruding pins or tails and trued the base of the box.  Then I softened all corners with a radius plane and plugged any large gaps in the exposed plywood “end grain” with some water-based wood filler.  No finish required.  It’s not pretty, but it’s not meant to be pretty.  Everything then got a quick 220-grit orbital sanding to remove the pencil marks.

wpid-20141115_202712.jpg

Sitting on my saw bench, “clamping” the work with my feet.

I had the idea of threading some 7/16″ nylon line from the boating supply store to make a rope handle.  I tested a couple different lengths and finally settled on a single strap that I could sling over my shoulder (as opposed to individual handles on the ends).  I hope I don’t regret the shortcut.

Conclusions:

I’m pretty satisfied with the box; it is solid and it holds stuff.  Truth be told, it’s been done for about a week.  I am just staggering posts.

wpid-20141124_082621.jpg

There are gaps. It happens. I’m not looking for a critique of my dovetailing.

Next time I dovetail plywood (I still owe my father some marine plywood drawers for his Frers 33), I am definitely skipping the coping saw. It works fine for solid wood, but is tough to control in plywood (with the multi-directional grain of the layers and all) and the veneer can tear easily.  Not to mention the awful screeching sound.

The Milkman’s Workbench worked quite well for this project.  Early on, it became clear the grip of the front vise screws wasn’t great until turned so tightly they mar the work, so I added leather pads to the ends of the front screws.  This improved holding power at much softer clamping pressures and work slippage nearly disappeared (although with increased grip, the work rotates a bit when the screw is engaged).  Front vise flex decreased dramatically as a by-product of lower clamping pressures as well, and I recommend anyone using a Milkman’s workbench take the time for this easy upgrade.  After a few more months of use, I will do a full retrospective on the bench’s total functionality.

Much gripper.

Much gripper.

As I was working, though, I started fantasizing about making a significantly larger version of the Milkman’s workbench.  Something with at at least 36″ of maximum clearance between the widest dog and the wagon vise and not less than 30″ of front vise inside clearance.  This would mean 40″ plus of total workbench (and probably a third screw in the front vise); my workshop table is slightly over 50″ long and should easily handle a clamp-on bench of this size. I would probably make the main bench deeper too (for a larger chopping surface) and maybe abandon current front vise configuration for something more like a moxon vise…

Several hours with pencil and graph paper later, it became abundantly apparent that what I really need is not a larger Milkman’s Workbench, but rather a planing slab (like a Japanese workbench, maybe with some dog holes and a Veritas inset vise). Something around 18″ x 60″, made of laminated pine framing timbers.  I guess I know what my next project will be after I finish those footstools for my brother’s new house.

Also, I can’t say it enough: plywood dust is gross.

JPG