The Roubo-Derived Luthierie Workbench

[Prostheta]

This is a work in progress....my main posts will be formed with the final intention of this becoming more of a permanent article on the main site. Wider user input is encouraged as it would be fantastic to broaden the article scope to include the experiences and knowledge of other PG members. Talk about your work area, please!

If any of the thread seems somewhat unformed or unfinished, the chances are it is in the process of being edited :-)

This article is being written with the intention of documenting my personal build of a heavy-duty workbench and to explore the subject of productive working areas as a wider subject. Fundamentally my bench will be configured for general woodworking use with certain aspects taking into account the demands of luthiery. Although almost all of the design decisions have been laboured over prior to any of this being written I decided not to present the bench as a simple "build project" and instead to explore the design pressures and describe the wider subject matter of workbenches.

The estimated cost for my the project itself is low compared to many off-the-shelf workbenches however it does not skimp on the factors that make it more than an "expensive wood table"! The major factors in keeping cost down are purely related to the choice and availability of materials in the quantities, dimensions and quality required. Although the rationale will be outlined later, it is just as justifiable to use inexpensive constructional Pine as it is to use Hard Maple or other similarly costly woods. Equally relevant, the choice of workholding hardware can affect the final figure more than the wood itself....much the same as an electric guitar build!

--------

I realised that I needed my own workbench because my current working environment is not a permanent one and changes depending on daily needs and usage. The quality and consistency of personal projects is definitely affected by this as every time I need to move workpieces and set up my working area it invites accidents with workpieces, stock and tools. Not good! Pretty much all of the requirements a workbench would need were right there in front of me. It can't take up massive amounts of space but needs to be large enough to handle my work confidently whilst maintaining a degree of portability instead of being a permanent edifice. These seem like completely incompatible requirements unless careful consideration is taken into balancing them out correctly. Oh, the fun!

My initial introduction to the field of fine woodworking benches was through the publications, writings and musings of one Christopher Schwarz. Researching further into projects Christopher's writings have inspired (or have had reference made to) revealed a polarised set of opinions, some unfairly devaluing his work as overly fundamentalist and dogmatic (isn't this a positive definition for those who preserve valuable traditions?), many who formally copy his work to the letter and others who just vaunt the (largely similar) works of other authors.

So anyway. The dogmatically fundamentalist work that Christopher HAS done regardless of overcritical online piddlings produced a fantastic pair of books on the subject with much additional materials in his blog posts and other online resources. The first of the books purely describes the detailed construction of an established and very useful 17th-century design and the second more of the rationale behind workbench design along with many other specific bench configurations.

The second book is the one I will refer to more than as not. In "The Workbench Design Book" (Google book preview available that link), Christopher goes to some lengths in underlining how the needs of the one outweigh the needs of the many (hah) and how defining pressures are paramount to decisions made in a personal bench design. Are you a hand planer? Do you cut lots of dovetails? Do you need to knock your bench down for mobility? Are you a powertool user? Are you working on something as small as a nut or lighthouse walls? Is the bench going to be shared with other people? If anything, this blows accusations of narrow-minded dogmatism out of water....

Not all of us are gifted with the luxury of an excess of space. Some of us - myself included at present - work in shared space where items are often moved into temporary locations which change as the demands do. I am sure that we would all love the opportunity of having a beautiful 12ft length of beautifully-planed French Oak (or five) sat in the best naturally-lit space(s) in a shop with wooden floors for our sole attention. Perhaps some of us do! In that respect I guess I should point out that I am not intended to identify the "ideal" workbench for all luthiers. Surely such a thing does not exist and any attempt to assert to the contrary is bound to fail. Hopefully by the time this article matures, the broad majority of circumstances and setups amateur and regular luthiers will benefit from will be discussed and explored. The overall point being to describe what works, why and for whom certain ideas will be of the most benefit....not just those for my own bench!

I warmly encourage others to fill in the blanks or correct me where my knowledge fails me.

So let's sit down with an Irish Coffee and look at where this journey is going to take me, why specific decisions were made (or yet to be made given that at the time of writing the project is yet to make-wood-smaller) and how this can be extended into making or upgrading your own workspace....

[Prostheta]

A little on my current circumstance might shed a little light on where I am coming from when it comes to my own project. I decided to change my career a while back and am studying for a woodworker's degree here in Finland. The workshop itself consists of 9000ft²/850m² of semi-industrial machinery from semi-automatic lumber profiling machinery to pneumatic clamps, CNC machinery to fine hand-carving tools. Three years worth of classes study at any one time putting space at a premium. My own personal working demands and desire for stability in the working environment necessitate making my own bench that can be moved if necessary but is large enough to cope with my work. Whilst a few workbenches already exist in the hand tool room, these are flimsy light affairs scored and spotted with dried glue from years of (ab)use. I certainly wouldn't trust chucking up a meticulously-planed body between dogs and a tail vice on any of those for fear of the surface marring the workpiece!

Ultimately this bench will leave with me when I graduate so it will not be possessed of the temporary nature its surroundings are. The design pressures will be defined purely by existing circumstance which is a mixed bag of luthiery and general furniture making. Perhaps both will dilute each other to a degree however I hope that I can find a happy medium that works for me without sacrificing too much of what I am aiming for in each discipline....

[Prostheta]

This post will probably get a little long and is likely to meander all over the place in a contradictory manner. If you want to skip to the bench-y bits instead of reading about my vacillation and poor decision making skills, go to the next post.

After having pored over many online articles, blog posts, books, participations-in and voyeuristic-skimming-of discussions on the merits of anything from holdfasts to top wood choices I realised that there is perhaps too much information on the subject of workbenchs. Whilst the availability of knowledge can only be a good thing (information wants to be free!) it becomes a time-consuming chore to reduce it all down to what is useful for one's own use. It just goes to highlight how our widely differing needs (and perceptions of them) vary.

Meanwhile wood is not making itself smaller....

In his first workbench-related publication "Workbenches", Christopher Schwarz described the detailed process of building a workbench based on a 17th-century design André Roubo described and illustrated in his master body of work, L'Art du Menuisier (The art of the Joiner). This somewhat iconic style of bench has proved to hold several surprises which people still mass debate over at great length. Whilst not describing much of the design rationale or having room to explore comparative pros/cons of the bench, Christopher's book makes an otherwise large project easy to consume and produce with a fantastically useful bench popping out of the other end. Great! I would have named it "Workbench" however....

Christopher's second book revisits this posterboy bench whilst opening out the subject with several other designs including variations in storage, workholding, budget and work usage, all of which fundamentally form the final design. A great read. For the people who just wants one book to produce one bench rather than going full rocket-scientist like myself it certainly hits the mark. Whilst I am a great advocate for reading as much opposing material as possible in order to make up ones own mind on any subject, this book does an excellent job of imparting the most usable information in a logical, parsimonious and sufficiently complete manner.

Identifying my own needs was the first step. These turned out to be a confusing list of desires which conflicted with each other! Such is me. The extent of my workholding strategies and fundamental working methods appear to be the defining factors in narrowing down the field.

My heavier work is unvariably carried out on free-standing machinery and there are assembly tables available when it comes to manipulating larger pieces. This pleasingly rules out the need for a deeper bench and happily marries up with my gut feeling that a shallow bench would be less burdensome. A large squarish footprint does not store or transport happily and tends to be a set of compromises. Since I guess that everything I need for my course project work exists elsewhere I should concentrate on not duplicating that....unless of course it is to my advantage. The length of the bench is only needed in my imagination, realistically. Instruments don't require enormous length and neither does my coursework. Anything longer doesn't need to come under the remit of my own bench.

More than anything this satisfies my personal insistence on making a smaller bench than as not. Luthiery doesn't generally require large amounts of space (just large amounts of stuff!) plus I would like the resulting bench to be moveable when needs require it to be. A long bench would become a monolithic nightmare.

More often than not when sat carrying out what would normally be regarded as detail or fine work, I find myself easily frustrated at the lack of immediate reliable workholding. What a pain! Certainly, good in-situ workholding methodology is going to be of paramount importance.

Despite my best efforts to talk myself into a different design I decided - and guess that I had long wanted - to build a Roubo style bench. Shallow in depth (arms reach, in fact) long-ish in length and heavy in countenance. Perfect! Again, why overcomplicate matters? The style has been documented by many people with the pros and cons detailed at length.

The pure mass of the bench confers stability and durability whether the bench is constructed entirely from hefty Oak, a combination of heavy top/light structure or even completely from simple airy Pine. The structure is deceptively simple, yet it yields several important advantages which may not be immediately obvious. The most important to me is the slab top which allows clamping around all of the edges unlike commercial benches which merely have an "apron". The number of times I've hunted for a flat spot under the table of a pillar drill, morticer, sander, etc. and found nothing close where I needed it is just silly now.

Despite being thick and potentially quite long the top is comparatively not that deep which allows quick access to tools stashed at the rear or on the wall. The legs are stout and joined through the top for maximum structural connection and minimum flex. Pairs of stretchers at the sides, front and rear connected using chunky mortice and tenon construction through the legs complete a simple but physically awesome structure. The legs are perfectly flush with the front edge of the top allowing the integration of a front leg vice. Although this is perhaps not as useful for luthiery as some vices, this is where my vanity project habit kicks in. Nothing looks more impressive on a Roubo style bench than a full-depth chop leg vice, and in any case these can be removed should they become a hindrance. Definitely a must-have for when I do hand-cut dovetail joinery. Totally impossible to do well using anything else!

The top alone on a modestly-sized Roubo bench commonly weighs in more than the average person; the weight almost doubles when factoring in the supporting structure. Now, I certainly don't need anywhere near the kind of lengths demonstrated. Indeed, at the size which this design is useful to me, the weight could be perceived as overkill. Perhaps so, however I would rather have that weight behind the bench than too little. Additionally it means that whichever way a workpiece is secured to the bench it has an enormous amount of mass behind it ensuring stable safe work.

[Prostheta]

At this stage it does not seem obvious as to why a Roubo bench is the best choice for luthiers. In most ways it is not. The specific workholding configuration of Roubo benches best suits work that requires access to manually working all three sides of a potentially large piece of wood. Whilst this is still something that I will require from my bench from time to time, luthiers work more in the small with workpieces that demand far greater flexibility and delicacy in access and workholding.

The majority of strung instruments - other than pianos, harps, double basses, etc. - are closer to human-sized, generally never flat in any dimension for very long, and certainly luthiery requires more up close and personal work that takes time in a quiet bubble of unsociable OwnSpace. It would seem that the bench of a patternmaker or woodcarver would be a little more appropriate. The single biggest difference in the broadest sense between all of these is the workholding. I'm sure I could be shot down for saying that, however this is the internets and people say all kinds of crazy stuff these days.

Ask any patternmaker what the central workholding method on their bench is and the answer one Metric 100% of the time will be "an Emmert-style vice". Unlike the standard edge/face/end workholding methodology found on say a Roubo or a Holtzapffel, the bench found in a pattern shop benefits more from unhindered 360° access to the workpiece in more weirdy bendy stretchy ways than an exotic dancer could dream of in their career. These things are seriously cool however they are also seriously expensive. If I were to build my dream bench on a whatever-budget, this would be the jewel in the crown. Unfortunately I do not have that luxury....

Go drool

http://www.highlandwoodworking.com/patternmakerswoodworkingvise.aspx

http://www.leevalley.com/us/wood/page.aspx?c=&cat=1,41659&p=31143

Perhaps one day I might find myself in a position to fit such a monster (40KG!!!!) however at present this is right out of my current project bounds. Nothing wrong with mentioning it of course....and if you want to send me one as a weekend toy then cool.

A far more viable alternative to this behemoth is the "Guitar Repair Vice" as sold by Stewart McDonald:

http://www.stewmac.com/shop/Tools/Vises/Guitar_Repair_Vise.html

This is actually quite affordable (other than maybe shipping!) and is easily retrofitted to benches through the dogholes. Perhaps when this becomes available I could incorporate this into my project....

Thankfully the workholding methods I am not yet able to incorporate into the initial design are gamechangers, allowing me to build a more-or-less original Roubo now and hotrodding it later!

The only significant alteration would be in the bench top's height for fretting, electronics or other in-your-face work. Attempting this work for protracted periods on a planing bench will likely result in back and shoulder fatigue due their lower heights. Equally, a higher bench set for detail work will be awkward for planing. Since I intend to use machinery for my planing operations more than as not, setting my own bench marginally higher from the outset is the best compromise.

So! Roubo with longer legs it is. Sexy.

[Prostheta]

As luck turns out, the dimensions as given in Christopher's book work out perfectly for seated work and match the dimensions of the desk I am writing on right now. Extended working periods feel comfortable and in the correct ergonomic position, everything is more or less within arm's reach and the top isn't excessively large in footprint.

My rough sketch is not much changed from the dimensions of that in Christopher's book, other than being in Metric :-)

The leg vice absorbs improvements introduced by Jameel over at the Khalaf Oud workshop....

http://oudluthier.blogspot.fi/2009/09/roubo-bench-one-year-later.html

....and certain ideas Jim Kirkpatrick added to his own Roubo....

http://www.talkfestool.com/vb/woodworking-projects/3068-benchcrafted-split-top-roubo-bench-build.html

Specifically, the introduction of balancing bearings on the lower glide bar take the weight bearing off the wooden screw so it can concentrate purely on opening and closing the vice. Pretty much how I thing FWD cars are a little silly since the same wheels are both steering and applying power to the road!

Additionally I have decided to fabricate the vice nut and hub garter from Acetal holopolymer instead of wood. This should improve the vice's durability and ease of use. I am still in the process of fine-tuning the leading bearing wheel design primarily so it does not interfere with the mortice and tenons in the leg, however it would be perfect if the vice were able to be closed up flush. This may require a shortening of the leg vice's chop however we'll see where this takes us.

Also illustrated in the drawing is a more traditional end vice style which I have decided will eventually be a sliding tail vice:

http://benchcrafted.com/TailVise.html

This is where blackmail comes in handy so I can get my son to mill me a vice instead! Unfortunately it does mean that a certain degree of commitment to this will be required from the outset. It is what it is. I believe that it should not be too problematic given the method for milling the square dog holes meshes in nicely with the prep work for a vice of this style.

[Prostheta]

I visited a boat building school today where they have about 6-7 workbenches made in the Scandinavian style. Metal dogs (ouch) and tail vices that showed signs of sagging rather than being balanced and/or supported well on their runners. Equally, the mechanisms utilised Acme threads and metal hardware. I found one bench which used wooden threads for the vices which worked far more solidly than the small diameter metal screw threads which made up my mind in one respect. Not to use a metal thread for the sliding tail vice!

All of the benches used a knock-down method of through mortice and tenons with tusks holding the tenons tight at the other end of the open mortice. These invariably loosen with time and use, especially when the bench structure flexed as readily as they did. If the bench can't take light pushing from one or other end, something is certainly amiss.

[ScottR]

What kind of wood is normally used for wooden threads? I have seen them used this way too, but never gave it much thought.....

What kind of care must they get to keep the threads from cracking and breaking off? They are basically thin slivers of end grain taking the load, aren't they. I expect something with a lot of interlocking in the grain would be good.

Lignum vitae has a history of use as bearing material for marine prpeller shafts and is interlocked typically.....might be ideal. On the other hand a metal rod with threads as large and coarse as the wood shaft might be the best of both worlds.

SR

[Prostheta]

You can use any hardwood really. It is accepted that the threads will chip to a degree over their lifetime however unless you use a very small number of mating turn counts in the threads they fit into this is not really an issue. The contact over the rest of the thread allows it to run freely.

Your suggestion on an interlocked grain rings true. I was looking around for some Lignum Vitae and Hickory the other day and hit no good leads. Hickory is like glass on the endgrain and Lignum Vitae is just so waxy it would be a wonderfully running screw. I wasn't aware of its interlocked grain so that's a fantastic bonus.

I would rather do this all the way rather than any half measures, so wooden screws it is. Metal threads are too slow-acting for vices unless you do metalwork and need torque. Wood vises are generally wood and/or leather-lined so the clamping pressure is spread over a larger area. The torque just isn't needed. I've seen demonstrations on how much holding power leg vices have even with light pressure so the adventure of making all of the parts from scratch is too good to miss.

In the absence of any woods with properties advantageous to making wooden screws I will just use Birch. I have a lot of it on tap (no joke intended). I'll be keeping the jigs used to produce the wood screws so if I get the opportunity to upgrade to Lignum Vitae, etc. then the option is there.

In other news, I purchased two cylinders of Acetal (Delrin) holopolymer to serve as the vice screw "nuts". One will be fitted to the rear face of the leg vice's leg and the other to the sliding tail vice block. I may purchase more to serve as running bearings for the vice wheels also. Going to play with some decorative ideas for the leg vice chop and garter rings....

[Prostheta]

Excellent. My Acetal arrived today.

The legs are sat in the clamps back at the shop. 4x laminated blocks destined to be cut down to 100mm x 140mm x 900mm are curing so perhaps these'll get sorted on Monday. I made up the 100mm from our stocks of low-grade 50mm and 25mm Birch, much of which contains pith and faults. After planing and jointing the pieces are around 42mm and 22mm leaving me a potential laminate of (2x 42mm, 1x 22mm) 106mm. Additionally I chose reasonably attractive "show" faces for the legs which contained character such as the occasional birdseyes, etc. The "faulty" nature of the wood means I get it for a song and it doesn't look plain and boring! Structurally the wood is fine, especially when laminated up the way this project requires.

Over the weekend I will do a little work on the bench design to add in upper short stretchers underneath the top. These will allow me to "split" the top into two and leave a specifically sized gap of 40mm. An flush-fitting slotted insert lives in this gap which can be used as a tool stash, or can be lifted out to sit proud as a workholding stop.

It also allows me to reduce the complexity involved in the top's laminating strategy. Instead of a whacking huge 1800mm x 500mm x 120mm laminated top (probably over 70kg on its own!) being worked, I can reduce this down to 2x 1800mm x 230mm x 120mm. Much easier to manipulate!

The downside of this is that both halves of the top will be far harder to make co-planar as they are mounted independently of each other despite being on the same structure. Unfortunately this means I will have a lot of hand-planing to carry out after final assembly....right when we are usually supposed to be celebrating project completion....

I only realised as I typed this that our thickness sander's maximum thickness capacity is 100mm. Perhaps enough to sneak through the legs in one dimension but certainly not the other, and definitely not the top! I guess I will have to spend some quality time sharpening the jointer blades....

The vice screws will be made from Birch. I don't have money in the kitty to invest in expensive woods for specific components at this time so perhaps these could be left as "future upgrades" or perhaps for a second bench build. I kind of like the idea that all of the wood was grown and milled within half and hour's drive of here anyway. That of itself is pretty cool.

[ScottR]

I'm pretty sure I'm going to be green and drooling by the time you get this done.

SR

[Prostheta]

Maybe not. A workbench is nothing more than a glorified table with suitable working conditions such as workholding and ergonomics. Even then the word "ergonomics" can be broken down to "decent chair" and "decent surface height". The whole workholding thing is the crucial difference. I just wish I happened to have an Emmert vice or similar to install on it. They really would be the perfect front vice for a luthier. My whole leg vice thing is just because I want to work on boards for dovetailing furniture. In that respect it is a compromise, however I hope that illustrating why it is might make this whole writing thing valid.

[Prostheta]

Check this out:

http://www.mprime.com/Emmert/how_to.htm

[Prostheta]

I made a few modifications to the design today both in a practical and a cosmetic sense. The legs will be located using rising dovetails (more on that when we get to them) plus the overall length is shorter by about 30cm. I just don't need 1,80m!

I would love to spring for one of these:

http://www.axminster.co.uk/axminster-axminster-pattern-makers-vice-prod21294/

Perhaps some day I can install this on the rear side of the bench to make it a proper "walk around".

[Prostheta]

Okay - a little more on the idea of the sliding dovetail and my decision for employing it. The basic design properties of a Roubo style bench is the usage of a solid top (not some weedy thin affair with an "apron" making it look thicker than it is) with legs that are mounted flush to the edge of and through it.

The methods of fitting the legs as advocated by Christopher Schwarz are drawn directly from Roubo's drawings and are a basic pair of mortice and tenon joints for each leg. One is a dovetail exposed on the edge whilst the other is a fully enclosed through tenon:

This method is valid however it is also fundamentally work intensive due to the need a substantial enclosed mortice through ~4-3/4" thick top! Really no fun to be had there. If there is one thing I dislike, it is getting bogged down in a design point which had no real bearing on the finished item and eats up work time for no return whatsoever. Instantly then, this one is straight out of the window or left on the doorstep as a recycle/reuse donation for a dogged traditionalist to collect ;-)

Roy Underhill (of "The Woodwright's Show" fame) employs a novel form of mortice and tenon joint leg mount called a "rising dovetail". Being a dovetail tenon this joint adds significent stability within the mortice and locks the leg securely in place. On second inspection however it looks impossible!

[Prostheta]

The secret behind the rising dovetail is in what you don't see (dovetail afficianados love to do this to you). The hidden rear face of the dovetail is in fact not parallel to the show face but is angled backwards....

From a mechanical standpoint this makes a lot of sense. Thanks to the theory of gravity, the weight of the bench top is always pushing this joint securely together over the shoulders. The only difficulty is the strategy used to mark up this beast and ensure that it actually works. One thing I have not managed to find is a guide on how to design these rising dovetails. Apparently Roy wrote about these in one of his publications however as a mental exercise I'm going to take the concept and work it backwards. Unfortunately there is probably no practical use for this joint in luthiery!

The most important aspect of this joint to me is that it is exposed on the edge of the bench. Of itself that makes hogging out the mortice a hundred times easier.

[Prostheta]

I decided to make some small modifications to the leg vice design to facilitate working on the face of say, a body blank. One of the most common tasks a luthier will have is working on the faces of a large lumps of wood like these, whether we are routing out pickup cavities or finish sanding a face.

Holding a flat and relatively thin workpiece securely can be pretty difficult without impeding your access across it. The foot of your clamp is ALWAYS in the way when you are navigating your router base around a template! Well, perhaps it isn't but it certainly feels that way....

The Roubo commonly sees more work from hand tools as it stands, whether they be planes, drawknives, chisels, etc. These tools share a common human push/pull action which requires specific workholding in that plane of movement. Planing a body blank flat for example could be carried out by clamping it up between the sliding tail vice and a bench dog.

Luthiers - unlike cabinet makers - rarely work on items that are so conveniently flat and square! Trying to hold the body blank for a Strat after cutting the outline makes this workholding method far less useful.

Enter the revamped tail vice design!

By adding two reversible brass dogs into the top edge of the leg vice's chop (the larger holes underneath are for finger access to push them up) the leg vice can now hold work across the top of the bench! These dogs significantly increase the number of workholding options available, especially given that cauls conforming to the usually-odd shapes of instruments can be made.

As for the corresponding dogs on the other side, I am yet to make a specific decision. I would like to think that I could drill in a double line of 19mm holes for fitting Veritas Surface Clamps into:

http://www.leevalley...1,41637&p=57059

....the obvious downside of course is the cost of adding these in. Certainly useful and items like these are what will decide the sizes and locations of dog holes around the bench.

[ScottR]

I think you are developing a vice fetish.....

Be sure to show us how to assemple that rising dovetail if you get it worked out. I'm still not getting my head around it.

SR

[Prostheta]

Okay Scott, this one is just for you :-)

I worked out on the bus that a rising dovetail is merely a normal dovetail (extruded parallelogram) but set at an angle. Where this angled dovetail emerges across the top and front face of the bench top you get two more parallelograms which resemble a traditional dovetail. It is just a trick which works because it isn't easy to think in angled cross sections and why would we? Everything else is set at 90°. It's just a neat trick with the benefits of a dovetail and some smugness built in. Here's how I planned one up:

This is a leg with a dovetail on one edge. I skewed the top face forward to make it a bit quicker to model.

This is how it looks from top/right/front with an intersection added to show the front face of the leg and the part of the angled dovetail that would be missing.

Closeup of how the full rising dovetail intersects with the bench top.

Finished item.

Clear as mud, Scott? It might help to think about how it fits together. The "full" parallelogram on the top of the dovetail corresponds with the shape on the bottom, but shifted backwards. This means the leg is inserted from slightly behind and moves forward as it slides upwards into the mortice.

[Prostheta]

Not really a hugely enlightening photo of the progress so far, however this is the majority of today's work. The morticer can't handle 14cm thick workpieces so the four front stretcher mortices had to be hogged out using a long Forstner bit (it is 90mm/3,5" deep, 30mm/1-1/4" wide and 100mm/4" long). Tiiiiiring work. All of my marking out was done using my new laser-cut Incra T-square which I intend to review. Fantastic bit of equipment which I heartily recommend.

These mortice and tenon joints will be drawbored in addition to being epoxied up (unless I run out of epoxy in which case Titebond). That is to say, Oaks dowels are driven through the leg and the tenon, however the corresponding holes in the tenons are offset by 2-3mm towards the shoulders. This causes the Oak to pull the tenon in tight to the mortice and around the shoulders. Having practiced this a lot recently I look forward to the results.

Each leg is 861mm long (1mm to be planed off the end tenons when fitted), 100mm thick and 140mm wide. Severely weighty legs! The lower stretchers are currently in the clamps being laminated up. A dry fit may be on the cards tomorrow however I should probably prioritise the two halves of the top so I can get the rising dovetails sorted.

More, I promise.

[ScottR]

Clear as mud, Scott? It might help to think about how it fits together. The "full" parallelogram on the top of the dovetail corresponds with the shape on the bottom, but shifted backwards. This means the leg is inserted from slightly behind and moves forward as it slides upwards into the mortice.

I confess I scrolled down and peeked at your first image before reading. Instant enlightenment! Brilliant!

You've got to love the old designs and tricks employed by the craftsmen of old. Not a computer in sight when those were developed.....nor electricity either.

That is a very flat, square, crisply cornered leg you've got there.

SR

[Prostheta]

Thanks. I'll be adding either a light bevel or a heavier stopped bevel later. Not sure yet.

Old style craftsmen did things like this intuitively rather than reading hence why I thought it a positive exercise to work this one out for myself. I've seen people make these but nobody explaining how to design them.

[Prostheta]

Quite a bit of progress today although not much to look at I am afraid. Most of it was spent cleaning out the hooking great mortices for the tenons! Still, all four legs have been morticed and the stretchers made to fit tightly and flush. Nowhere near the stage where I can glue and drawbore them all up yet though....

The tenon shoulders were cut on all four faces using the table saw before cutting to size on the bandsaw. The sizes were roughly 1mm over the mortice size leaving a quick shaving with a wide chisel for fit.

Although not all hammered in flush for the photo, all of the stretchers sit snugly in the mortices. Perfect! As you can see, I abandoned common wisdom and incorporated heartwood into the legs. This Birch pallet is a lot cheaper and at these sizes this really won't be problematic. i think it adds character anyway....the surface checking out from the heart will more than likely get a good pasting with some pigmented epoxy if I can a. find my epoxy and b. remember to buy some pigment.

(the lower stretcher looks like it is seated at a weird angle! it truly isn't....)

[ScottR]

Nowhere near the stage where I can glue and drawbore them all up yet though....

This is another step I'm looking forward to seeing. I don't think I formed the correct mental image from your description....

SR

[Prostheta]

Try this. Christopher Schwarz demonstrates how it works nicely in this video....notice he is working on his Cherry Roubo around the assembly stage!

[ScottR]

I'm assuming the taper of the peg plays an integral part in pulling the tenon and shoulder in tight? That was the part I did not form a mental image of. Reminds me a bit of the clamping techniques that use wedges to apply pressure.

SR