Workbenches that are powered by wedges, friction and stops have been fascinating to me since I first started looking into Roman workbenches. My interest and research into these benches eventually became “Ingenious Mechanicks.”
And now an old Norwegian Sloid (Sloyd/Slojd) manual has shed some new light (for me, at least) on these wedge-based benches – thanks to some drawings and text.
Eivind Reed of Breim, Norway, sent along these drawings plus a translation he made from the first Norwegian textbooks on school Sloid, “Sløidlære for skole og hjem” (Craftsmanship for School and Home), which were written by H.K. Kjennerud and Karl Løvdal. Here is the translation:
No. 244. Wedgebench. Pine in the benchtop, birch in the front vise, the board at the front end, planing stop and wedges. The bench can be made larger or smaller according to the intended use. By the drawings you see that the planing stop will sit slightly within the edge of the front board. The fit in the mortise must be so tight that it stays without slipping down. To avoid having to remove the planing stop when using the birdsmouth, we make two recesses for the teeth, so that it can be flush with the benchtop.
The bench can be put on a box, kitchen counter or similar. It will of course not measure up to a regular workbench, but when you get used to it, it does good service. To plane the face of a board, put it on the benchtop, thrust it into the planing stop so that the teeth sink in and it rests toward the front board. The teeth stop the board from moving backward when withdrawing the plane. To plane the edge of a board, we put it in the front vise [a crochet]. If it will not stand securely, we use the wedge. If the board is narrower than the thickness of the benchtop, we drive down the planing stop and put the piece in the birdsmouth. To shoot the ends of the board, we wedge it in the front vise, back vise or we use the front board as a shooting board. To rip boards, we use the front or rear vise as we see fit. To crosscut, we lay the piece on the bench as usual. To avoid losing the wedges, and to keep them always at hand, we hang them by string on eye hooks on a fitting place on the bench.
No. 245. Wedgebench to attach to the wall. See No. 139. The bench is attached with hinges and can be put up when not in use.
There are some clever aspects to both of these benches that are not covered by the text.
Both benches are on the small size, like the Milkman’s Workbench. The sizes are in metric. (My mind defaulted to American customary units when I first looked at the drawings. I saw a benchtop that was 6” x 20” x 80”. Dumb American.) The second wedgebench (No. 245) is longer, but that’s mostly to make room so your handplanes don’t poke you a new window in your wall.
One of the bits of cleverness are the rotating toggles below the benchtop that allow you to hold work on edge. I’ve seen sliding bars, but not toggles. These are much simpler to make and install.
The best stuff is the wedges. The opening in the benchtop has one straight side and one angled side. The angled side is 6° off vertical. The wedges shown below the notch both have faces that are angled at 6°. One wedge for small work; one for larger. The angles on the vise and the wedges keep the clamping pressure square to the workpiece.
In investigating early benches, all the notches that I recall encountering had square-sided notches (the dovetailed notches in the Saalburg bench are one exception). Clamping work in those proved less-than-spectacular until I tried using softwood wedges with almost no slope on them. The softwood compressed when struck and then held the work like crazy. These angled notches are another excellent solution to the problem.
The crochet is called out in the text, and it’s nice that this one also comes with a complementary wedge. Both the crochet opening and the wedge are at about 11.3°.
Also interesting is the “front board” – basically a full-time wide planing stop. It’s only about 3/8” thick (9mm). Combining that with the toothed planing stop is pretty clever.
When I first looked at figure No. 244, I assumed the holes in the benchtop were for holdfasts. That is, of course, silly for many reasons. It’s likely because of the way they are drawn that they are the way you fasten the benchtop to a table or box.
If I make another Roman bench, I will definitely incorporate the angled notches and wedges into the design. Thanks so much to Eivind for the image and the translation.
With the last large batch of dividers, it became clear that the demand for the heat-treatable tool steel version is much greater than anticipated. So, for the latest batch, half of the sets in inventory are O1 tool steel, and the other half 12L14 steel.
At the moment, there are about 50 sets total available, with another few dozen sets milled and awaiting hardware. Unfortunately, as with so many things this year, backlogs in production and delivery have made it increasingly difficult to predict timing – but I believe I have found a solution to the pivot hardware issue using off-the-shelf hardware developed for the custom knifemaking world.
So, over the past week I’ve been developing a simple approach to modifying pivots manufactured specifically for folding knives. This hardware is specifically designed to hold very tight tolerances, which is perfect for the needs of the dividers. Because they are intended as a sort of DIY-customizable solution, the bolt-and sleeve (aka “Chicago bolts,” or “sexbolts”) system only needs to be modified to take the #10 spanner drive I use for adjustment, and some light shaping to make them nearly identical to the custom manufactured hardware I’ve always used to date.
While this is no guarantee against production delays – these bolts are occasionally prone to stock shortages themselves – having another, much larger channel to source this hardware makes it much easier to avoid long-term stock issues.
So, if you’ve been wanting a set of dividers, this is a good time; but as always the goal is to have these available on a continuing basis, and this development puts the dividers much closer to that goal.
O1 and 12L14 dividers as well as a few remaining hammer kits are available at daedtoolworks.com.
We are happy to let you know that Texas Heritage Woodworks is now making handmade slipcases that keep “The Woodworker’s Pocket Book” handy in your tool chest or tool cabinet.
Made with 14 oz. waxed cotton canvas and #69 bonded nylon thread, the slipcases install easily on any wooden surface thanks to the four brass eyelets and included #6-5/8” slotted screws. The slipcase is designed specifically for the “Pocket Book,” which slips in and out easily, and holds it securely on a tool chest’s lid or cabinet door.
Like everything from Texas Heritage Woodworks, the slipcase is made by hand in their shop in Menard, Texas. It will easily last as long as the book and your tool chest.
We love them. I installed one on the lid of my Anarchist’s Tool Chest. Megan installed one on the lid of her tool chest and also installed one on the lid of a Dutch Tool Chest she is constructing for her forthcoming book.
Note that all proceeds from the slipcase (which is $25) go to Jason Thigpen and his family at Texas Heritage. Lost Art Press isn’t taking any royalties. We think the world of Jason and his family and the work that they produce. Even if you don’t need a slipcase, we encourage you to check out the company’s tool rolls, aprons and leather goods.
We are currently sold out of “The Woodworker’s Pocket Book” in our store. Another press run is in the works, but we don’t expect to be restocked until mid-April. However, several of our retailers have stock, and we encourage you to support them. All of these retailers had stock or were taking pre-orders as of today.
An image from a Latin version of “Laws of Hywel Dda,” a Welsh law book. This version dates from the late 12th century or the middle 13th century. Note the tapered legs and the sticks beneath the arm.
Editor’s note: The following is a draft chapter from “The Stick Chair Book,” due out later this year. I just wanted to give Peter Galbert a heads up that we’re changing every reference to “Windsors” in “Chairmaker’s Notebook” (just kidding).This piece has been updated to reflect ongoing changes in the manuscript.
When people see a stick chair for the first time, a typical response is to call it a “primitive Windsor.” Unfortunately, every syllable of that expression is incorrect.
And that’s OK. We live in a world where the term “Windsor” has expanded like a gas to mean almost any piece of furniture where stick-y components are mortised into a plank – Windsor table, Windsor stool, Windsor bench, Windsor printer stand.
It does make you wonder: Where did this furniture come from? A place called Windsor?
Perhaps.
As furniture historians point out, the origin of the word “Windsor” to describe a class of chairs is complicated and has yet to be definitively settled.
So let’s start at the beginning. Furniture where legs are tenoned into a plank – what is sometimes called “staked furniture” – goes back at least to the ancient Egyptians. Three-legged staked stools with beautifully curved legs and a saddled seat have been found at Thebes (1400 BCE). And the National Museums of Scotland has a similar one from the same time period.
A low stool from Thebes, Egypt. Found at the tomb of Kha, Eighteenth Dynasty. Note the curved legs and saddled seat.
Staked furniture of all kinds shows up in Western paintings and drawings through most of human history. Stools, benches and tables are the most common forms. So, the idea of putting sticks into a slab of wood is at least 3,400 years old.
What I’m interested in, of course, is this: When did people start making chairs this way?
The simple question is complicated a bit by language. The term “stool” can sometimes mean a “backstool,” which is a stool with a backrest that is a solid board or an array of sticks. Some people consider a backstool a “chair” and not a stool. So that clouds the timeline. Old writings that mention “stools” might actually mean “backstools” and those might be chair-like.
The earliest stick chair – legs, seat, arms and backrest – that I know of is from a Welsh book of laws that dates from the late 12th century or the middle 13th century. The book is the “Laws of Hywel Dda”; the chairs are drawn in a particular copy that was written in Latin instead of Welsh (this copy is referred to as the “Peniarth MS 28”).
The book is illustrated and has two images of important men sitting in chairs (one is at the beginning of this chapter). Both appear to be armchairs. Both chairs have tapered legs below the seat. One has sticks under its arms, and the other has four shapes below the arm. The shapes could be cut-outs in a solid plank. Or the shapes could be objects holding up the arm.
Also from “Laws of Hywel Dda,” this chair is slightly different. The structures under the arms might be cut-outs.
“Welsh Windsor chairs sounds to me like saying Welsh Scottish oatcakes, or Welsh Wexford glass” he wrote. “The chairs I am writing about are very definitely Welsh, and they are called stick chairs in Wales. They do, however, fulfil exactly the definition of what has come to be known, in Britain and the United States, as Windsor chairs. My judgement is to stay true to my original thoughts; only time will tell if I am mistaken.”
So if early stick chairs aren’t Windsors, where did Windsor chairs come from?
First, let’s dispense with the myth about the origins of Windsor chairs that gets repeated in popular culture.
“The most popular meaning stems from the story which describes how George III was caught in a rainstorm near Windsor,” writes Ivan G. Sparkes in “The English Country Chair” (1973). “Taking refuge in a cottage, His Highness sat on the best chair in the room and being well pleased with its comfort, required similar ones to be made for Windsor Castle. Unfortunately for this theory, the style existed and was so called long before the Georges came to the throne of England!”
Another (slightly more plausible) theory appears in “Popular Technology; or Professions and Trades. Hazen’s Panorama” (1846) by Edward Hazen.
“The Windsor chair seems to have been first used for a rural seat in the grounds about Windsor castle, England; whence its name. It was originally constructed of round wood, with the bark on; but the chair-makers soon began to make them of turned wood, for the common purposes of house-keeping.”
I do like that this theory hints that bark-on sticks played a part in the history of the Windsor and they were originally outdoor chairs.
In the last decade or so, historians have used probate inventories and paintings to present a clearer picture of the origin of the term. The best synopsis of the current thinking was published in Regional Furniture, Vol. XXIV, by Robert F. Parrott in 2010.
The most interesting part of the evidence are two inventories taken two years apart of the same household, one in 1721 and the other in 1723. The first inventory was for the husband who died of a stroke; in the listing of the equipment for the garden are “Forty eight Forrest Chairs.” Two years later there is another inventory, and in the section on garden equipment are listed 60 “Windsor” chairs. Presumably these are the same chairs, but the household has bought another dozen.
“Presumably therefore, the type of seat originally described as a ‘Forrest’ chair sometimes went under the alternative name of a ‘Windsor’ chair,” Parrott writes. “This, then, may be another reason why the early history of the Windsor has been so difficult to ascertain.”
A very early English chair from a private collection in the United States. Could these early chairs, which have no backsplat, be the “Forrest” chairs referred to in early 18th-century inventories?
Forrest Chairs We don’t know exactly what these early chairs looked like, but we have some clues. Since the 1970s, several early chairs have shown up at auction houses, at the Victoria & Albert Museum and through some sleuthing. These chairs are far simpler than the typical later English Windsor and could be a stylistic link between stick chairs, Windsor chairs and American Windsor chairs.
These early chairs share many characteristics with stick chairs. There are no stretchers – the strut legs are simple turnings. There is no backsplat – a very common feature on English Windsors. And the ornamentation is incredibly restrained compared to later English Windsors. There is a simple scratched groove around the seat and the comb. The front posts under the arm have a little shape. But that’s about it for decoration.
As a maker of stick chairs, I contend these are the prettiest English Windsors I’ve ever seen. I am also struck by how much these early chairs resemble American comb-back Windsor chairs. It’s rare to see an American Windsor chair with a backsplat. And the rake and splay of the legs looks far more American than English.
It makes me wonder – and this is a bit of conjecture – if these early chairs inspired American makers.
John Brown also had some thoughts on this matter. He came to a slightly different conclusion.
“The oft repeated statement that American Windsors derive from the English chair could be in error,” Brown wrote. “For historical reasons, and because of similarities in design, there seems to be a more direct link between the Welsh chair and the American Windsor. Perhaps the English version is the cousin, and the Welsh chair is the father!”
So About that Name, ‘Windsor’
Once you know these chairs may have been called “Forrest” chairs, you have to wonder, why did the name switch to “Windsor?” Was it because the chairs were first made in a place named Windsor?
William Sergeant found evidence of the earliest-known maker of Windsor chairs in a village in Lincolnshire, which he discussed in a 2018 article in Regional Furniture. That maker, Joseph Newton of Fenton, placed an ad for “New-fashioned” Windsor chairs in July 1725.
Newton’s ad also mentions there are makers of these chairs in London. What’s important to know is that Newton’s shop was nowhere near Windsor Castle (it’s about 140 miles away).
Parrott and other historians have found connections between chairmaking activity near Windsor and where those articles went to London. But Parrott admits the link is still tenuous.
One possible theory for changing the name is that the term “Windsor” gave the form a royal flavor and is in line with the French naming furniture styles after kings (i.e. Louis XIV).
Or perhaps the name “Windsor” could have become popular first as an insult to the chairs, as Sparkes wrote in 1973.
“In the end I find myself agreeing with those writers who connect the origin of the name with the manufacture and sale of these chairs to the London dealers at the Windsor Market and along the main road from Windsor to London. For one can imagine the London chair dealers, used as they were to the finer mahogany and walnut products of the London workshops, referring in a derogatory way to the latest batch of beech chairs ‘up from Windsor’.”
Today the term “Windsor” gets applied to broad classes of furniture that have no connection to Windsor Castle. Or pieces that have nothing to do with the House of Windsor, which was founded in 1917, or the town of Windsor. It can be confusing. At times I fantasize about a world that has switched back to the earlier and more evocative name for this distinctly English chair: Forrest Chair.
The term “Forrest” is far more descriptive of how the chairs were initially were used: as a seat for the outdoors. And, unlike the word “Windsor,” the term “Forrest” describes without a doubt where the chair came from.
And so, in this book – as a bit of a lark – I will refer to “Windsor” chairs as “Forrest” chairs.
I am certain this will catch on everywhere – just like Esperanto.
Editor’s note: The following is an excerpt from “The Anarchist’s Workbench,” which you can download for free here. This excerpt discusses one way to choose wood for your workbench: its cost per pound.
Yellow pine, Hot Springs, Ark. After years of working with yellow pine on our Arkansas farm, I convinced myself it was an ideal wood for a workbench.
Let’s Talk About Weight When you compare the weights of species, you need to make sure the comparisons are all at the same moisture content (12 percent is the typical comparison unit). You can compare the density of a species by comparing its “specific gravity,” which is a method that compares the weight to a cubic meter of water. Or you can look at the average dried weight of a cubic foot of the wood (also at 12 percent moisture content).
These are useful, but I think you can also make some important comparisons by factoring in the local price of a species. It’s like buying meat at the butcher. Is rib eye ritzier than hamburger? The price per pound helps us answer that question (and yes, it is).
For example, a cubic foot of hard maple consists of 12 board feet of maple. If maple is $4.73 a board foot, then a cubic foot of maple costs $56.76. That cubic foot weighs 44 lbs. Or $1.29/pound.
Longleaf pine (a yellow pine) is 78 cents a board foot (for No. 1 grade), so a cubic foot costs $9.40. That cubic foot weighs 41 lbs. Or a remarkably cheap 23 cents per pound.
Because I live to poke fun at Ipe, let’s run those numbers. Ipe costs $17 a board foot, so a cubic foot costs $204. That cubic foot weighs 69 lbs. So Ipe is $2.96/pound. Not a great deal at the wood butcher’s.
The chart compares some of the common U.S. hardwoods and softwoods using typical Midwestern retail prices circa 2020 (this is not wholesale or trade pricing). This cost-per-pound calculation is simple to do yourself using your local prices.
Here’s how: Take your cost per board foot (use 8/4 prices) and multiply that by 12. That’s the cost for a cubic foot. Now divide that number by the weight of a cubic foot of that species (a statistic that is easily found in books and online). The result is the cost per pound.
$175 workbench after 20 years of use, 2020 The yellow pine in this workbench is now about as hard as maple and moves little with the seasons.
Do the Math From the chart, ash looks like a good choice among the hardwoods. The problem with that assessment is that by the time you are reading this book, white ash might be almost extinct. The emerald ash borer has devastated the ash forests in the United States. So, you might not be able to buy it at any price. And if you do find it, you want to ensure it hasn’t rotted. We have been plagued by punky ash for the last few years as the sawyers have milled up trees that have been standing dead.
Aside from ash, poplar and the maples are a great bang for the buck. Both are easy to work, readily available and fairly cheap by the pound. I’ve made workbenches using all three species and think they’re fine. Neither is considered a noble species for a workbench, like European beech. But as long as you aren’t out to impress anyone, go for it. You’ll have no problem finding those species at almost any lumberyard in America.
But if you want to go full redneck, read on.
Softwoods that are used for structural members in home construction – the yellow pines, Douglas fir, hemlock and some spruces – are an outstanding value. They are heavy, cheap and readily available at any lumberyard. After working with them most of my life in residential construction and workbench building, they remain my No. 1 choice for workbenches.
Here’s why: Anyone can buy it. You don’t have to search out a specialty lumberyard or set up a commercial account. Just go to the home center if you want (though I always prefer family lumberyards). They have plenty.
Also important: They have plenty. A typical home center or family lumberyard will have hundreds of planks of 2x material in the racks on any given day – everything from 2x6s to 2x12s – with lengths from 8′ to 16′. At a home center, you can spend hours sifting through the racks to find the best boards – the employees don’t care. At a family lumberyard it pays to ask permission (they will sometimes be happy to help you). Either way, just be sure to restack the lumber nicer than you found it.
Here’s another buying tip: Some lumberyard chains carry No. 2 yellow pine, others carry No. 1. The price difference is minimal, but the quality isn’t. No. 1 is worth the extra nickels. If you find a yard that deals in No. 1, you might be able to buy all the wood for your bench in one swoop. If you buy No. 2, you might have to hit all the yards in your town, county or region.
Yellow pine is easy to work. I’ve built yellow pine workbenches using only hand tools, and using a full-on machine shop. It’s friendly stuff. Yes, there can be some knots, but if you pick your boards with care, you’ll have almost none of those to deal with.
So there must be disadvantages. Yes, but they are slight. Construction lumber is sold in a wetter state than hardwood lumber.
While hardwoods are typically sold at about 12 percent moisture content (or at equilibrium with some environment) that is not the case with construction lumber. It is wetter.
How wet? In the Midwest it might be 15-20 percent moisture content. On the West Coast, it might be even wetter (as in wet enough to ooze and squirt water). So, you need to gather up what you need to build your bench, cut it to rough length, stack it and wait a bit.
It might also be “case hardened” because it was kiln dried too quickly. When lumber is rushed through a kiln it can develop tension that is released when you cut it. It’s particularly obvious when you rip a board. Sometimes the wood will pinch so hard on a blade it will stop a 3-horsepower table saw like pinching out a candle.
How do you deal with this? It’s not difficult.
Plan to cut things a bit over-wide. And have some wooden wedges handy to keep the kerf open when you rip the wood. After that first rip, a case-hardened board will usually lose all of its fight.
Make your rip cut to a shallow depth at first – less than half the thickness of the board. Then raise the blade, flip the board end-over-end and finish the rip.
The final disadvantage: Softwoods are uber-redneck. No one is going to “ooh and ahh” over your choice of yellow pine. It’s the mullet of the forest.
The True Cost of Yellow Pine per Pound I’m not a trusting soul. After I calculated the cost of yellow pine per pound (23 cents) based on published statistics, I decided to see if that worked in the real world. So, I weighed several 2x12x8s and came up with an average weight of 30.4 pounds each.
These were boards I’d had in my shop for months, so they had likely lost some of their water weight (as all softwoods do). Plus, the boards in this particular pile were fairly average – not full of sap or with lots of heavy summerwood. In other words, they were a bit on the lightweight side.
Each of these boards cost $8.81 each, so that’s 29 cents per pound – about 6 cents per pound more expensive than the published weight tables indicate. But still a great deal.
I wondered, how did that work out after surfacing the boards and gluing them up? What was the cost per pound of “finished” yellow pine?
Here’s how I calculated that. The benchtop for the workbench at the end of this book is made from nine 2x12s, ripped in half, glued up and planed so the top is 5″ thick. Nine 2×12 x 8s cost $79.29. After gluing up the top, I managed to weigh it on a heavy-duty scale we use for shipping crates. The top weighed 240 pounds. That’s 33 cents per pound. Still a bargain (if you ask me).
