Workbench Wednesday – #7 (2011) Roubo Sawing Bench

In great part due to the rapidly forming manuscript for To Make As Perfectly As Possible – Roubo On Marquetry and the number of my essays describing the seemingly arcane practices of 18th Parisian workshops, in 2011 I built a slightly diminutive (2/3 scale) version of Roubo’s sawing bench as illustrated in Plate 278, Figures 10 and 11.  The dimensions for my version were determined by the space in my basement workshop; I now wish I had made it full sized.  Doing so would have doubled the mass of the bench, and in this instance mass is really the only important thing.  The construction was real meatball woodworking, I simply fit and fastened together 4×6 tulip poplar stock then drilled holes for the vise screws all the way through the top horizontally.  For the female threads I simply used the wooden vise nuts that came with the threaded screws.

This workbench has only one purpose and function, to hold a work piece firmly while it is being resawn.  What we found immediately in battle was that the bench vise did a fine job of holding the work piece in its jaws tightly.  Unfortunately the bench was simply too light to perform well in action as the whole thing danced around the shop with every saw stroke.  The only way we could get it to work was placing anti-skid pads underneath each leg and then loading it up with as much weight as was handy.  Currently this sawing bench is the storage home for several hundred pounds of fire bricks I keep handy.

Roubo alludes to this problem himself, extolling the virtues of massive weights being stored on or under the bench to hold it steady, or even more likely bolting the entire unit to floor.  In my old basement shop this made no sense on the concrete floor in a tiny space, it makes more sense now that I have wooden floors and lots of them.

Most recently I used the saw bench during my demo at Working Wood in the 18th Century at Colonial Williamsburg, employing an attendee as my stabilizing weight.  In preparation for that demo, and in response to my having cannibalized the unit to use the original vise screws on other benches, I made new screws and screw nuts with my Bealle threading unit.

I cannot say I have used this bench enough to become facile at resawing veneer, the best I can do is about eight leaves per inch.  Only time will tell if I ever get to the point where I can saw a dozen leaves to the inch like the old timers, but if I do this bench or one like it will be part of the equation.

Workbench Wednesday – A Detour

Before I move forward to discuss the next workbench in my inventory let me be diverted to discuss the retro-fitting of a previous bench, my Smithsonian Roubo, such that its location, role and function in the studio are completely new and immensely more valued.  Over time the bench had come to occupy the end of the classroom space, primarily because it was the only bench I had that could fit there.  It was not really large enough to suffice as a student bench for workshops so instead I employed it primarily for metal-working type projects including saw making and sharpening, hardware mounting, parts fabrication, etc.  (sorry for the lousy picture; I had already removed the leg vise for another bench, replacing it here with a Record 53)

When I recently removed the generic end vise and mounted instead the ~125 lb.  Emmert Universal Vise in its place, one piece of a convoluted equation began to take shape.  I knew the vise needed a robust platform and this little-used bench performs the function perfectly.

A second element in this equation was expanding the work space on the side of the barn housing my shop; I reorganized it so that my own shop would extend an additional nine feet to include the full footprint of the 14′ x 36′ bay in the timber frame.  (Of course that meant that I needed more workbenches there.  Stay tuned on that one.)

A third component in the equation was a beloved niece-in-law had expressed an interest in learning woodworking (actually I have four beloved nieces-in-law, but this is one in particular).  The odds are pretty good the second of the petite Roubos I built originally for my Handworks booth would eventually end up in their apartment.  So, I removed it from the critical space it occupied adjacent to my third child before it became too disruptive to do so.  I moved that little bench down into the newly opened space, for the time being.

Since nature abhors a vacuum something needed to go into that space previously occupied by the petite Roubo.  Hmm, I really did like having a metalworking-ish bench in the middle of my herd of woodworking benches…  Palm, meet forehead.  Soon I had the old, almost extraneous Roubo bench relocated, revived and recommissioned, sitting where it will be used daily.  I removed the second vise and stocked the space underneath with a lot of my mechanicky tools.

I have additional plans for this bench which I will chronicle when they unfold.

Here is a gallery of the Emmert Universal Vise showing off its moves.

Workbench Wednesday — #6 (2011) A Vise-less Roubo

Almost coincident with my previous Roubo this much more successful bench was built in the barn, using surplus SYP timbers from the barn itself.  The barn floor plan had been reconfigured from its days as a dairy barn so there were several pieces of unused timber available for new projects like this bench.

From the outset I made the objective that this bench would be the simplest possible, big slab, big legs, no vises relying instead on holdfasts.

Since we were not yet living in the Highlands the schedule of the bench project depended on my periodic visits to work on and in the barn.

The almost 6-inch-thick slab was glued up on an unseasonably warm January day when we were able to get the shop space up to almost 50 degrees with my two kerosene heaters.

The next session of working the slab was less than a month later when the outside temperatures were near single digits, and we were barely able to crack freezing in the space.

In another three weeks it was mild enough to work in shirt sleeves.  Ahh, weather in the mountains.

Once the top was dimensioned and the mortises cut and chopped, the leg tenons were worked with the pieces being held in my Emmert K1.  The legs were tulip poplar timbers left over from who knows what project, measuring 6″ x 8″.

Soon it was time to sledge home the leg tenons into the top mortises.

I added oversized stretchers and left the bench pretty much like this, in use, for another couple of years.

Eventually I spent the time on the final flattening of the top and filled its voids to make it a first-class working surface.  It was not smooth but I didn’t really care about that as long as it was flat.

Immediately after that I sealed the whole bench with thinned varnish/tung oil mix.  That really emphasized the cross-hatch pattern employed during the flattening process.

Once the sealant had hardened I toothed the entire top and pushed it up against the window and it remained there until last summer, when it was relocated to the other end of the shop where it still serves as my primary finishing station.

This was my first truly massive bench, so heavy I can barely move one end of it at a time.  I love the thick slab top but the legs are a little too large and the stretchers are just ridiculous.  With a re-do I would reduce the legs to more like 6″x 4″ and the stretchers to 3″ or 4″ wide at the most.

And, a bench without clamping vises works just fine, thank you very much.




Workbench Wednesday – Opus #5 (2011) my first “Roubo”

In the title Roubo is in quotation marks because this bench was not strictly a Roubo, it was more Roubo-lite.

2011 was a turning point in my bench mania.  In the preceding quarter century I had built or salvaged only four benches.  From 2011 forward, inclusive, my inventory grew by more than a dozen.  To be sure, going from a home basement workshop footprint of slightly over 200 sq.ft. to one closer to 7,000 s.f. in The Barn probably had something to do with it, but the truth is I was simply becoming increasingly fascinated with workbenches and vises, and now had the time and space to indulge that fascination.

My first personal encounter with a Roubo bench was back in the 80s when I got to test Rob Taruleh’s bench at an event where we were both speaking.  It was intriguing but I was not then in a place to give it a go.

By around 2004/5 my successes and seniority at the SI allowed for my daily activities to become almost entirely self-directed; as long as I was productive within a broad framework of organizational priorities, did not ask for too much money to spend on my projects, and was not a trouble maker my daily activities were in great part at my own discretion.  I made certain not to abuse this freedom, I remained productive in my scholarship, projects and publishing, I obtained almost all of my discretionary funding via external collaborators, and bit my tongue on a regular basis.  (This last one was my biggest hurdle — Mrs. Barn says that one of my greatest challenges is that I am generally unsuccessful in hiding my contempt for knuckleheads and grifters).

Despite a sufficient number (4) of top-of-the-line German workbenches populating the furniture conservation studio I occupied for almost three decades at the Smithsonian Institution, my growing involvement with L’Art du Menuisier combined with the incessant evangelizing of Chris Schwarz compelled me to give the Roubo workbench a try in my own daily work space.

This is a fairly long-winded exposition as to why I made this Roubo-ish bench, it was because I wanted to.  Even then I did not cause any waves using only surplus materials laying around the storeroom or my conservation studio, or in the case of the legs from my pile at home.

By 2011 with the Roubo franchise building a head of steam I felt it was time to experiment with the form, but not really go whole hog.  Instead I took a couple days and built a Roubo-“ish” bench that was a so-so success.  My starting point was a five-foot-long slab of laboratory counter top (missing one corner) that was about to be sent to the dumpster.  Since it was only 2″ thick I backed it with a piece of 3/4″ plywood (yes, I knew even then this could be problematic, but thought I could get by with it being inside a tightly climate controlled space; little did I know that within two years I would go from climate control to climate, and lots of it) and grafted on another piece to fill the corner.

I also made the mistake of cutting the dovetailed tenons through the top at 45-degrees.  In more recent work I have stayed with 60-degrees.

Once the unit was assembled I flattened the undulating top with a fore plane and a jointer, before finally surfacing it with a toother.

For the leg vise I used a vintage vise screw I had in my collection, but made the movable jaw out of some oak that was laying around.

For the next almost-two-years this was my everyday workbench and I liked it a fair bit, and it provided the validation I needed to develop my experience even further. It was a little too small and lightweight, issues that have been addresses in subsequent models.

When I departed the Smithsonian they had no interest in keeping this workbench for the furniture conservation studio so I just loaded it up with all my stuff and took it to The Barn.  Soon enough the unregulated climate there wreaked its havoc on the composite top, crowning it almost a half inch.  With a little time and a scrub plane this was resolved.  It has remained pretty flat over the past three years.

The bench is too small and unrefined for my work now, although it is in the inventory.  It’s at the end of the classroom, and at the moment I have it set up with a Moxon vise on top for saw sharpening and making.  The leg vise was removed, being both in the way and needed for another project.


Workbench Wednesday — Bench #4 (2008), Tom’s Bench

More than a dozen years ago I crossed paths with Tom, a retired police detective and rustic woodworker (that’s a description of the style in which he generally worked, not a commentary on his considerable skills.)  Tom had a booth at a community yard sale selling some of his tools and, well, you know how the rest of that story goes.

Shortly thereafter we started spending Wednesday evenings in his shop making sawdust, sharpening tools, making and repurposing tools, and on occasion solving some global problem.  Or more likely, just working on projects while we enjoyed each others’ silent company.

When we started that routine that lasted for a decade until we moved to The Fortress of Solitude, Tom’s shop was set up for him to work pretty much alone so one of our first projects together was to build another good workbench for me to use.  Projecting from my own Old Faithful I designed and built a larger torsion-box-top bench on a petty standard base.  Over a period of several Wednesday nights the bench took form until it was ready for battle.

Tom had never built a torsion box and was fascinated by the whole concept and process, especially when all the parts of the 1/2 baltic birch box were glued together into a 60″ x 24″ x 5″ thick whole.  Once that was done I affixed the 60-inch twin screw vise to one of the bench faces (I had obviously already prepared the holes inside the box grid before glue-up) and suddenly he had another unique high performance tool added to his shop.  I think we had it sitting on a pair of sawhorse for a couple weeks while I used it to fabricate the base, but before long it was all together and sitting on the floor for me to use.

Well, he used it a fair bit as well.  Eventually Tom added a pair of rising stops to either end of the bench and I added a shelf underneath and it was done and put to work immediately.

For several years it served me well in Tom’s shop and I fully intended to leave it behind and for it to become part of his inventory when we moved away.  (My weekly evening in Tom’s workshop was one of a very few things in Mordor I left behind with regret).  Tom promised to visit The Barn and in fact he came a fair bit early on, and he had come several times to work on the barn with me.

On his first trip to work in the barn he surprised me by bringing the bench with him.  He said he wanted to make sure he would always have a workbench whenever he came for a visit, which has not been as frequent as I would like.

But the bench is sitting in the barn, just waiting for him to join me.  Maybe a little cluttered, but ready and waiting.

Workbench Wednesday — Bench #3, (2005?) A Sjoberg Salvage Operation

The bench today

Sometime in the mid-aughts I came into possession of a raggedy c. 1970s Sjoberg workbench that was slated for the dumpster.  It was a rickety old thing that needed to be pretty much disassembled and reconstructed to make work properly.  Before, it wobbled so bad it was probably an injury magnet, afterwards it was stout and sturdy.  I eventually added another four inches to its height by simply bolting on some mondo skids.  Since it was still unusably light I then bolted the skids to the floor and now it is one of my high-value work stations for when I am doing conservation projects for smaller decorative objects or gunsmithing, engraving, marquetry or carving.

Some of the accouterments I have added to it include a tilting fret saw table that fits perfectly into the end vise, and a swiveling ball vise made from a pair of toilet flanges, a duckpin bowling ball (it’s a Maryland thing) and a hot melt glue gun and wood scraps.  This resides on the floor underneath the bench when not in use.  I’ve also got a stereo-microscope sitting there for those times when I need it, and a pair of Gerstner tool boxes filled with the tools I need for these projects on a shelf above it and a small drawer unit with carving and engraving tools sitting on the shoulder vise end.  It is light enough to move aside easily whenever I need that vise.

The bench was one of the very first things I brought to the barn, probably within the first year after it was enclosed, in ’09 or ’10.   It was the closest thing I had to a functioning and available workbench and it arrived even before there was the final flooring underfoot.


Being pretty light it was moved around frequently until the spaces took shape.  It finally wound up in a perfectly fitted niche just inside the entry door, adjacent to the propane wall furnace.

It is not the perfect bench, but the purchase price was perfect and with a bit of finessing it has turned into a valuable contributor to the shop’s functionality.  Were my life situation different I could probably make a decent living with this as my only bench.

Workbench Wednesday — Bench #2 (1990) Bench Auxiliaries

I was mighty pleased with myself when the bench was finished. It was easy to move, easy to set up and take down, good and sturdy with great clamping. Clearly, I had solved all the problems which led me down this path in the first place. However, a little use of the bench showed me that reality was slightly less idyllic.

Unfortunately, its lightness (55 pounds) which was such an asset in my master plan for the “perfect portable bench” was also a big liability once the bench ceased being mobile and was set up as a work station. The bench was so light I couldn’t really work it hard without moving it or even knocking it over. I had to figure out some way of weighting the bench while in use. My “no loose parts” vision was about to bite the dust.

Fortunately, the solution was as simple as building two thin (approx. 1 1/8″) torsion boxes to serve as shelves resting on the crossbar of the end/leg units on either side of the folding bracket, and which could be attached to the underside of the benchtop when not in use. By putting all my tools and supplies on the shelves, the bench now had enough mass for my use. It still wasn’t heavy enough for general cabinetmaking, but it was more than adequate for restoration.

Then a second problem cropped up. The working height for flat-ish or small objects was great, but setting chairs or case pieces on the bench raised them too high. For these taller pieces I needed a lower work surface which would still fit my ideal of lightness and strength. I could have built a lower version of the bench, but instead tried something different. This time I fabricated a torsion box about the same size as the bench top which would fit on a pair of small, low trestle horses. These trestle horses were made from lighter than normal elements, for example the main components were 1″x 1″ and the crosspieces 3/8″ thick.

Some minor modifications to generally employed designs yielded another light, strong and stable unit. By making the top bar of the horses removable, each post of the trestle could then become a tenon. Constructing the torsion box in such a manner that the bottom side had openings to function as extensions to the mortises already incorporated into the grid fitting the tenons of the sawhorse posts, the pieces fit together as a small worktable suited perfectly for holding taller pieces at a more comfortable height. While this new unit was not as “neat” as the workbench, it solved the problem. It also provided more flexibility than simply building another, lower table/bench.

This saga does not end here as I continued working on newer iterations of the concept, but those episodes will be recounted in future Workbench Wednesdays in a few months.

Next week – A derelict salvaged from the trash heap that was transformed into a little jewel.

Workbench Wednesday — Bench #2 , 1990, “Final” Version

Building on what I learned from the initial prototype of the portable restoration workbench I charged ahead with a “final” version.  I say “final” in quotes because this concept is one I have continued to tinker with even to this day, and a late entry in this series four or five months from now will focus on the most recent one.

For this first final version I used the same structural strategy for the top, modifying the stock weight being the only real modification.  On this second iteration I used 3/8″ A/C plywood for both the faces and ribs of the torsion box top rather than 1/4″ and 3/4″ luan for the same purpose on the prototype. Making the box approximately 3″ thick and placing the ribs in a 6″ x 6″ grid achieved a satisfactory result with essentially no change in weight while yielding a stout structure. In fact, I could do some pretty serious joinery and carving on the bench, which I couldn’t do with the prototype.


I wanted a large-capacity vise on my bench, but there was no point in defeating my original purpose by building a lightweight bench and then installing a heavy vise on it. The vise(s) I built opens about 12″ and employs aluminum threadstock for the screws, tapped holes in the endpiece of the top, and 2″ x 3″ x 24″ wood jaws. The screws pass through same-sized holes in the movable jaw, and terminate in simple wooden handles passing through holes drilled in the ends.

One pretty dramatic change in this function was to switch from standard steel threadstock to 1-inch aluminum rod stock that I had cut at a local machine shop (I already had the aluminum rod stock in my scrap barrel).  It’s been a long time since this project but I recall paying $25 for the job.   All it took was to set up the pieces in the lathe and cut the threads with a single pass, along with a single groove above the threads for some retaining collars.  I’m guessing it took ten minutes maximum, so $25 for a quarter hour of machine shop time sounds about right.  It might have been $25 for each pair of screws, but either way it seemed pretty reasonable.

My procedure for incorporating the vise into the bench top involved drilling oversized holes in the internal grid but not the added fixed jaw, so I couldn’t assemble the box all at once. I cut my grid pieces and glued them to one face of the torsion box using 315 gram hot hide glue, which was the adhesive employed throughout the project. After laying out for the vise screws, I drilled holes through the fixed jaw (the torsion box end pieces) to tap-out and received the threaded screws. I was unsure whether holes the same size as the aluminum rod stock used for the vise screws would be large enough to allow the screws to go through the grid members, given the inevitable wobble in the screws. To be on the safe side, I drilled larger holes through the grid to a slightly longer distance the desired vise opening dimension. The vise screws would then pass through these openings as the vise was closed. After I was sure the vise was fully operational, I glued the second face on the grid and the top was complete.

By fabricating simple leg end units with long folding diagonal braces, the problem of too much shimmying parallel to the long axis was overcome on the final bench. The leg units were fabricated from the clearest 2x construction lumber I could find, and the folding braces from 1/4″ x 1″ aluminum bar stock with 1″ x 1″ aluminum angle stock for the mounting brackets. The slightly heavier face plywood made it easy to screw all these elements to the underside.

The leg units were not installed symmetrical relative to the short axis, but rather were off-set by half the leg width so they could fold up next to each other rather than on top of each other. This way, the table could become a remarkably compact unit which needed absolutely no assembly or disassembly; it could just be folded up. Following final assembly I added a snap-on strap for carrying the bench like a large suitcase. By throwing the strap over my shoulder, I could easily carry the bench for long periods of time and maneuver it through pretty tight quarters.

Next week, some accessories.


Workbench Wednesday — Bench #2 , 1990, Prototype(s)

Because of the rigidity and (light)weight requirements for the portable restoration workbench, the obvious choice for the top was a torsion box, the construction of which was virtually identical to those described in the woodworking literature. My first attempt used 1/4″ faces and 1/2″ ribs of luan plywood. The weight and rigidity were good, but the working surface of luan was a little too fragile.  Nevertheless it was a confirmational “proof of concept” exercise.

My first step was to make a simple torsion box as a test run.  It was near-perfect as the top and working surface.  I faced this prototype with cork on one side and plastic laminate on the other, and it remains in use in the shop to this day almost thirty years after being made.

Concept One – check.

My first inclination was to make a trestle base for the bench, but considering my master scheme this idea had some significant drawbacks. While easy to build and assemble, the knock-down/assemble/disassemble trestle requires multiple parts, and one of my stated intentions was to not build anything that required keeping track of lots of pieces. Instead, on my first attempt I tried using a pair of commercially available tubular-steel folding-table legs. These worked fine along the short axis of the bench but weren’t stiff enough along the long axis because of short diagonal braces.  Though a good concept, this shortcoming and the unnecessary weight of the steel legs made it a non-starter as a final option.  Even with those hefty legs the unit weighed in at only 55 pounds, well within the limit I wanted.

Concept Two – check.

As for the vises I simply used pieces of 2x stock cut to match the dimensions of the ends of the top combined with 3/8″ steel threadstock from the hardware store and some wood handles epoxied onto the ends.  Again the concept was successful but the fine-ness of the thread made the vises tedious to use.

Concept Three – check.

Finally I threw on a piece of upholstery webbing to serve as a shoulder strap so the folded bench could be carried and maneuvered easily.  That concept was an absolute winner.  Throwing the strap over my shoulder it was a breeze to navigate the twisting rabbit trail to a work site.

As a final test run I took the prototype to a job and it worked out just fine within the confines of the limitations enumerated above.  With what I learned from this it was time to move forward with a final iteration.

Check, check, check, and check.

By the way, this bench is still in use but as a food preparation station for the barbecue.

Workbench Wednesday — Bench #2 , 1990, Part 1

This is the next of an occasional series on the workbenches I’ve built over the years. — DCW


By the late 80s not only was my young career at the Smithsonian humming along but my outside work for private clients was keeping me as busy as I wanted to be.  An awful lot of those projects were “on site.”  Whether the client was a private collector or an institution, with increasing frequency I was working at their location. There are many reasons for this; cramped quarters at home, the legal liability of transporting very valuable objects, the cost of renting a truck and hiring someone to help out (I usually work alone), etc. Regardless of the cause I often found myself working in unfamiliar, and usually unequipped, surroundings.

Thus, several times a year I would move lock, stock and “workbench” (saw horses and plywood or door) to a new location, a truly onerous activity.  One project in particular, in a museum’s sub-basement down three flights of a spiral staircase, provoked me to build a high-performance lightweight folding workbench that remained in use for many years before retiring to the deck o serve as a serving table when I cut back on my on-site work.  I have improved on both the design and execution of the bench in recent years but for now I will stick to the 1989 prototype and the 1990 finished bench.

This growing commitment to on-site work justified a little investment of my time to work out the problems inherent in occupying space that was not my own. In the end, that process of finding a “better way” resulted in the design and fabrication of a new workbench and some companion accessories to make the task of working in a portable studio more manageable and productive.

What did I want?

The only thing I was sure of was that my sawhorse and plywood routine had to go. But what arrangement was to take its place? Being a lazy fellow, my first actions were to look around at the market to see if any of the available “portable” workbenches were suitable.

I discovered only two real options; a small, pseudo knock-down version of the European-style butcher-block-top bench, or a Workmate. I looked at a couple of the former, and own one of the latter. I found the portable Eurobenches to be too small and unsteady for my use (and quite frankly, too “cheezy”). In addition, they still weighed-in at around 100 lbs. Since I would have to carry it by myself, it was simply too heavy. I tried my Workmate on a couple of projects, but it wasn’t exactly what I really wanted because it was too top-heavy and the work surface was too small. My search for a manufactured bench to suit my needs wasn’t exhaustive, but nevertheless, I decided to design and build my own portable workbench.

The process of procuring a new, high performance portable workbench began with the question of exactly what I wanted out of the bench, regardless of its source. When I decided to make my own, I had only to review those requirements and build to fit them. Back to the original question, what were my specifications for the bench? The answer was simple; 1) the top had to be perfectly flat and about 2’x 4′ (any smaller and I might as well stick to my Workmate, much larger and it would have been unwieldy for some of the pathways I had to traverse), 2) it had to have a vise sturdy enough to take a modest beating, 3) the bench had to be very light, compact, and easy to set up and take down because I didn’t want to have to assemble a kit each time I moved, and 4) it couldn’t cost a fortune in time or money to acquire. It was also important to remember that the bench wouldn’t have to stand up to immense weight or stress, since the pounding necessary during general joinery was rarely required in a conservation project. Any particularly heavy work dictated by a specific treatment would still have to be done at home.


The bench I ended up with was not an example of exquisite handworked joinery, but it did require precise machine woodworking techniques. The bench shown here is my second version and still not perfect, but it is completely functional and does satisfy the requirements listed above. It might be bragging but I think it remains the finest high performance portable bench design I have encountered to date.

The top is the right size for my needs at the time (although it could be made any size), and contains a pair of 24″ twin-screw face vises at the ends which opened about 12″. The bench folds up into a compact unit weighing only about 50 lbs., making it extremely easy to move about.  In fact, the lightness presented another problem which I will describe later on. The long diagonal braces, which fold with the legs, provide plenty of stability to the bench allowing even more aggressive cutting and pounding than I had hoped for. In addition, I built several accessories which enhanced my capabilities for restoration on-site. Finally, in terms of both fabrication time and materials cost, the bench was very inexpensive. It took about six well-planned hours and $75.00 worth of materials to build, plus about $25 of machine-shop time (in 1990 dollars). You can fancy it up as much as you want, but you can have the basic unit quickly and inexpensively.

Next time — the “proof of concept” prototype