20/8/2014 I realised over the past week that this post is becoming far too big and therefore a major mission to download. That is why I broke it up into lots of smaller posts that should be easier to download and follow. Simply click on the “Building my benches” category and follow the numbers. I have also changed the name of the post to “My 18th Century Workbench in progress”.
I am sure you are as sick of lamination discussions as I am of perpetrating them. You will therefore be relieved to know that we are moving onto the next phase of this epic journey, entitled “Joinery and Armament”. It will be the first time in my woodworking career that I take on such ginormous joinery and as you might expect by now, it will be an exercise in overdoing it.
As much as I want to move on to this phase, I have to warn you (and myself, mind you) that there are quite a bit of squaring up left to do, on the laminated parts (painstakingly by means of hand planes predominantly), which I will add to the previous post as I get it done.
Sunday evening after packing everything away in the shop I sat down with a glass of vino and played around with some ideas for this bench’s sliding deadman. The bench is very traditional in most aspects, so I wanted to give it at least some fresh twist (pun intended). Otherwise known as je ne sais quoi. This is what I came up with.
A few weeks ago, my main man, Aldan, at Windhoek Tool Centre sourced this 175 mm Irwin Record quick-release vise for me. It is meant to become the end vise of my bench. After all the chopping and changing between different options for the tail vise, the economy prevailed and I settled on a quick-release vise. Unfortunately I only made friends with Patrick Leach a few weeks after I bought it, because I am sure he could have found me a much better made vintage model. You have to live and learn, I guess. I am however considering to ‘Paint it Black’ , (à la the Rolling Stones), as I really do not like the blue. It is a je ne sais quoi thing, I think.
Today I received the long-awaited Lie-Nielsen hardware for my bench and two new tools that will make it so much easier to build it. As you will see the shipment includes a large closed throat router plane, a large tongue and grove plane, a single screw vise (for the leg vise) and a chain drive twin screw vise (for the twin screw face vise with 24″ between screws). As you might remember, I am planning to put a twin screw vise in one of the face vise positions (as per the classic Holtzapffel design) and a leg vise in the other, as my bench will be accessible from both sides (rather than being braced against a wall). You will have to be patient in order to see the hardware liberated from the safety of it’s packaging. For now it will stay snug in it’s LN travel gear.
During the latter stages of last week I consulted Deneb Puchalski who is my go-to man when I need solid advise. I realised over the past few weeks, since reading up on the characteristics of the PVA wood glue that is readily available to me (a company by the name of Alcolin), that it will be impossible to assemble and clamp the whole bench in one go with the short open-time of these products. I therefore started thinking of using slow-setting epoxy for this task. As far as I can gather it will be at least as strong as the PVA type white and yellow glue (which is anyway quite a bit stronger than the wood), and might have an edge as far as shock resistance. Deneb advised me to consider Tite Bond III (which is a real mission to access in Namibia) or slow-setting epoxy.
The next step was to try and find a good quality epoxy for the job. I spoke to my friend Sigmund Mengerssen who told me about IBS. They stock ABE products. I spoke to a gentlemen by the name of Wolfgang who seemed to know what he was talking about. I explained the whole situation and he recommended OBE’s Epidermix 372 Epoxy Adhesive. It turns out it is the exact same product as what I have been using in small amounts for the past few years, especially for the hand tools I’ve built. This time I bought a bigger supply (1 liter) than the 90 ml I used before. Wolfgang advised me that I will have around 2 hours of assembly time with this product and I gather from the data sheet that it takes 6 hours to become ‘touch dry’, 24 hours to set for ‘practical use’ and a week until it is ‘fully cured’.
I also sent an e-mail to Attila Hoth of Southern Wood Trading requesting two beech boards in the order of 50 x 130 x 2700 that is quarter sawn and not too recent immigrants to this fair land. This should hopefully become the leg vise chop, twin screw chop and two sliding deadmans (or should that be deadmen??).
Finally I got stuck into the joinery phase during the Easter weekend. I took the age-old advise of “measure twice, cut once” quite serious and therefore spent a good bit of time on this particular task before cutting the male parts of the sliding dovetail and mortise at the top of the legs. You can see the carpenters triangle which will help me to locate the legs to their correct positions after they inevitably get mixed up during all the joinery processes.
I used the exact measurements for the sliding dovetail and mortise, which Chris Schwarz deducted from the plate in Andre Roubo’s landmark five-volume 18th century book “L’Art Du Menuisier”. As I mentioned earlier in this opus, the two books on bench building by Chris is definitely worth buying if you are contemplating a project like this. The plate below I found on the Lost Art Press blog site. You will notice the leg with the mentioned joinery in the bottom left corner (first picture).
Speaking of Chris Schwarz, he advocates that one should, and I quote:
“You don’t need to invent anything, patent anything or manufacture anything to create a workbench that’s better than what’s lurking in the aisles of many stores today. Workbench designs evolved into their highest form more than 300 years ago. And they are just waiting to be rediscovered by anyone who can under- stand that though things are always changing, that doesn’t mean they are always improving.”
Despite his advice, I designed my assembly table with the quick release vise in the middle of the end of it with a removable pipe clamp on either side. The pictures below will hopefully show that this design has at least some merit, during the few years that the assembly table doubled up as a workbench.
Anyway, after a few test/warm-up cuts in a scrap piece of Rhodesian Teak, I found a very first task for my 16″ Lie-Nielsen tenon saw. The effective cutting depth of the saw was found wanting, as you can see, but I guess it is unlikely that I would ever cut tenons of this size again.
To deal to the untouched waste left over on the sliding dovetail tenon, I devised the setup below. A combination of bench dog’s, clamps and handy beams created a very stable slot for the leg to rest while presenting it’s edge for me to go mental on it with my Lie-Nielsen carcass saw. Two kerfs in each of the triangular pieces of waste would make it much easier to later chop it away with a chisel.
Finally, some proper woodwork.
Next I had to set up my bandsaw to make accurate square cuts. As you can see in the second picture, it somehow found it’s way significantly out of square since the previous tune-up. After sorting that out I cut the necessary kerfs, to create the tenon and the back of the dovetail.
In order to cut the shoulder at the back of the leg, I found a very first job for the set of bench hooks I made more than a year ago. You can see how they performed admirably in fixing the leg while I used the carcass saw to cut the mentioned shoulder.
I just realised that there are a lot of firsts (in terms of tools being used for the first time) in this phase so far. Here you can see how I used my shop made so called Brace Bit Birdcage Awl
for the first time to make holes were the Irwin Auger bit can take hold. I decided to use the brace and bit to drill out waste before attacking the area between the two tenons with a mortise chisel. I first sharpened the bit carefully, as it probably last saw action during the Great War. It was actually surprising to see how enthusiastically it munched away the mentioned waste.
Speaking of firsts, this next task also became the first ever performed on my bench-to-be’s top. I clamped the legs to the top where it was resting on my Darwinian saw horses. The pictures show the sequence of destruction I followed to get rid of the waste.
A quick change in clamping position and I could go ape-shit on the triangular section of waste left over at the bottom of the sliding dovetail.
… x 4 and so far so good …
After removing the bulk of the waste from the male parts of the joinery, I used my shop made marking knife to mark out the exact lines to work to. In the pictures you can see how I used a whole range of tools to do the precise cleanup work.
In these pictures you can see how I used a very simple jig to get the shoulders of the tenon exactly 90º.
Again the assembly table came in handy to set up the leg for some precision pairing.
I then used the setup as illustrated below (referencing off the shoulder lines of all the legs) to identify the final length of the legs. The bench will be 82,5cm high (32.48″). Replaced the carpenter’s triangle at the bottom of the the legs after chopping off the waste with the bandsaw.
Here I took on another arduous planing task by preparing the two reference surfaces of the two sets of stretchers. After lots of deliberation, rereading Chris Schwarz’s book and even the translation of the first chapter Andre Roubo’s 18th century book “L’Art Du Menuisier” (on the Lost Art website), I decided that my stretchers might be a bit of an overkill at 170 mm (about 6½”). Therefore I shaved off 40 mm to bring it back to 130 mm. The final product will in the end be back at 150 mm after adding some Witpeer to make the stretcher appear as if it consists of only Witpeer, but we will get back to that later.
Attila Hoth dropped off the three beech boards I requested tonight. As we are about to abandon Windhoek for a 10 day migration to our Fishing Camp on the mighty Okavango River for a spot of Tiger fishing, I simply clamped them to my assembly table to stay honest while settling into my shop’s particular ambiance.
Finally I was able to progress with the joinery this weekend after our African adventure. First on the to-do list was to use the actual leg to mark out the shoulder lines of the short stretchers and two aprons. The work-holding ability of my shop made assembly table made it possible to set up a frame in order to mark out each set exactly the same. You will notice that I referenced the vertical position of the legs off the shoulders of the through mortises. For the width of the bench I finally settled on a fraction over 25″.
In order to locate the short stretchers, I made two batons of 3 mm plywood that are exactly the same length and used that to reference off the bottom of aprons, which were in turn set up flush with the tenon shoulders. With all the parts firmly clamped into position as you can see in the last picture, I used a 0.3 mm mechanical pencil to mark out the exact shoulder lines on the apron and stretcher. Then it was simply a question of repeating the process on the second set of parts.
The pencil lines generated during the above activity was then used to mark out the rest of the shoulder lines with a marking knife, while taking care to always reference off a face side or face edge. You will see that I have started using the technique/scribble taught by Robert Wearing to indicate the two face sides.
At this point I realised that I really should add the final part of the stretcher before cutting the shoulders. I decided a while ago already to laminate yet another strip to make it appear as if the stretcher is solid Witpeer. Therefore I had to mill another feral board into sophisticated strips for the top of all four stretchers and laminated it onto the already ominous looking (3″ x 5½ before the strips were added) short stretchers.
In the meantime I started sawing the shoulders of the aprons and made a point of trying to saw to the line. I did not want to remove as much material with the chisel as what I had to do on the through mortises at the top of the legs. Again I used one of Robert Wearings tricks to get the saw started perfectly. He describes how you used a chisel to pair a shallow ditch (pictured) on the waste side of the shoulder line to guide you saw. Once the shoulders were cut it was piece of pie to chop the waste away with a chisel.
The big frustration of the weekend was when I realised that I still needed to cobble together a fence for the #78 Stanley rabbet plane (that I am in the process of restoring into a working tool) before I can use it to hand-cut mouldings with it. That forced me to design and fashion this Ferrari-esque contrivance …
… which despite it’s dilapidated appearance, performed admirably in tandem with a small block plane and a card scraper in producing these hand-cut mouldings . You will notice that these mouldings were designed to hide the lamination. Together with a good colour match it makes it impossible to notice the lamination, once the stretcher’s shoulders are glued into position between the legs.
However, that took most of the day so I ended up feeling a bit despondent about the weekends progress.
I managed to chip away at preparing the aprons a little bit each night during the week. Here you can see how I squared up the shoulders first by chopping down using the prop to improve accuracy and finally some horizontal pairing to get it perfect.
Then I moved on to a plan I dreamt up to deal with wood movement. Windhoek has some of the most challenging seasonal changes in humidity of any location in this part of the world. In my shop the ambient humidity changed from 20-25% in winter to 75-80% during the rainy season last year. This means that the Equilibrium Moisture Content of the wood (which average around 6%) can fluctuate between 1-11%. This represents a major challenge when building a bench that needs to stay as flat as possible. This is the main reason why I designed my bench with a split top to allow the two parts (which will be fixed to the legs by means of the Roubo through tenon and sliding dovetail joints) to move into the space between them, rather than trying to pull the leg-apron joints apart. Please note the elongated hole for the 20 mm threaded rod that will fix the top to the apron, which is meant to allow for the mentioned horizontal wood movement.
In order to deal with the vertical movement of the top, I use the following strategies: 1) chose quarter sawn stock for the aprons, 2) cut relief gaps (for a lack of a better term) into the aprons. The holes drilled into the aprons in the pictures below, was the start of that process. Hopefully the pictures will do a better job of explaining.
Here I did a quick check to make sure the 20 mm threaded rod is able to move freely in the elongated slot.
I then grabbed the router to remove an area to allow the bolt to be countersunk below the bottom surface of the apron. This is important to ensure that a clamp can still find a nice flat surface in future, then using F-style clamps through the split top.
The router also assisted with the initial 20 mm (depth) or so, of this 6 mm (width) relieve groove, before I took to it with the drill press, a chisel and finally a bed float.
I used the bandsaw to cut two 3 mm wide relieve grooves in each tenon which extends and overlaps with the aforementioned relieve groove. This will hopefully mean that the tenon (which is about 115 mm (about 4½”) wide) will be able to expand without destructing the joint and for the apron to have minimal effect on the top when it inevitably moves throughout the year.
The three ‘fingers of each tenon were then cut to length. You will notice that there are a short haunch at the top (which stops short of the main tenon at the top of the leg in order not to weaken it) a through tenon in the middle (which will be wedged at the front of the leg) and a normal tenon at the bottom (which stops about 1″ short of the front of the leg). Both middle and bottom tenons will be pinned with a custom made 10 mm Assegaai dowel.
Then I turned my attention to the end (short) stretchers. I followed the same process as with the aprons. I only difference was that I found it a bit tough going with the carcass saw cutting the shoulders so I tried the tenon saw and it worked like a charm. The teeth of the tenon saw are set for ripping but I can honestly say that it worked as well as the carcass saw, only better because of the bigger teeth and extra weight and length it brought to the party.
The waste was removed by chisel.
The shoulders were squared up by chopping followed by horisontal pairing.
Seeing that it is insignificant whether the stretchers move slightly, I only cut the relieve grooves in the tenons to protect the integrity of the join.
As with the aprons the three fingers of each tenon were cut to different lengths. In this case the two outside tenons were shortened and the one in the centre were left long to become a through tenon to be wedged. Both the outside tenons will be pinned too. I plan to do the opposite with the side stretchers where the outside fingers will be through tenons and the central one shortened.
On to the female parts of these joints. I used the same setup as before to mark out exactly where to cut the mortises using the actual aprons and stretchers. Again the position of the legs were referenced off the inside shoulders of the through tenon at the top of the legs (see first picture).
Once again I took the measuring out quite serious spending most of a Saturday morning to measure out exactly were the mortises should be cut in the legs. I made the effort to measure the shoulder of each specific tenon at the top and bottom (of the in and outside) with the Veritas marking gauge (pictured) and transferring that to the leg to ensure that I get as close as possible to perfect. The four reference points generated in this way were then used to cut the long sides of the rectangle by means of a marking knife and straight edge. The short sides were cut using a square guided by the pencil marks of the actual tenons on the leg as demonstrated in the previous set of photographs. You will notice that each mortise have a entrance side and a exit side.
Since I recently realised how useful dividers can be in marking and laying out all sorts of measurements in woodwork, I used it to establish which size drill bits would work best for the removal of the waste from these mortise. It also came in very handy in marking out where to drill the holes. You will notice how I marked out the location of each assault on the mortise, as well as writing the diameter (in millimeters) of the spade/Forstner bit to be used in pencil.
All the tenons will be pinned in two places. Here you can see how I marked out where to drill the holes for the 10 mm custom made Assegaai dowels. These dowels are still to be made. Please note that the holes for the long and short stretchers will be in the same plane, which is contrary to what is usually done. I decided to break this very sensible rule as the design of my tenons (for reasons of wood movement) does not allow enough room to have them in different planes while still being imbedded in a robust hunk of timber. You will see how I plan to overcome this particular challenge as we progress.
Here I started removing the bulk of the waste from the mortises on the drill press. It looks like a piece of pie, but it took ages, trust me.
Then came the real elbow grease part of the job, removing the rest by means of a mortise chisel and a shop made mallet. Whack, whack, whack, whack, whack, whack, whack, whack, whack, whack, ……………………………………. (weeks later)………………………… whack, whack,whack,whack ………………….. (even more weeks later) …………. whack …… OK I am sure you get the picture.
Some evidence that I really did do all that drilling and especially all that whacking.
While all the whacking was going on I milled down the swarthout (which was laminated six weeks ago or so) destined for the cleat under the shelve between the stretchers, in order to take a well deserved break.
….. and whack whack whack etc etc ……
At this point in time I took a break from all the mortising, to build a monster Panel Gauge. Look out for a post entitled “Makeshift Panel Gauge”, which is soon to be released. I need one for the next major hand planing war that is about to break out in the previously peaceful surroundings of my workshop. I need to flatten the edges of the two parts of the twin-top. They are too big to fit through my planer, so after flattening one edge I will need to use a panel gauge to mark out the opposing side and hand plane that parallel to the reference edge. The mortises below is only one step away from being finished.
In the meantime I realised that I need to start squaring up the two beams of the twin top. As they are too wide to fit in my planer upright I will have to do that (planing to width) by hand planing. Once I get one edge square to the face side I will need a panel gauge to mark out the opposite side. Problem is, I do not have a panel gauge of that size. The idea was to build a makeshift panel gauge as I wanted to get onto the next step with the bench, but my tool design-OCD kicked in and it led to a protracted design and manufacturing hiatus. The end product of which is featured in the pictures below. I wrote a post on the process entitled “Makeshift Panel Gauge”.
This past weekend the process of handplaning these monster beams started with a vengeance. I like starting off by hollowing out the face side ever so slightly. I used the short straight edge to check all along the length of the face side. If it does not pivot on the edges, it is not hollow yet.
Next I check for wind using my truly makeshift winding sticks. This is why I start by hollowing out the face side from side to side, otherwise you get the wrong reading when using the winding sticks. If there is a bump at any point across the width, the winding stick can adopt one of two different positions. Once the winding sticks make contact with the face side’s edges only you get the true reading and can therefore proceed to fix the twist. In this case I had minimal work to do as I removed (by hand planing) most of the twist prior to feeding the beams to my planer about 6 weeks ago.
Once the face side was slightly hollow and twist free, I could square the face edge using the face side as reference. You can see how my shop made jointer came in very handy.
Then I used my newly produced panel gauge to mark out the opposite edge to be parallel to the face edge.
I went through the same motions with the other side of the top. You can see in some of the pictures what beautiful full length shavings I produced with my shop made jointer.
In order to remove the bulk of the material on the opposite edge, I first used my shop made scrub plane as a very aggressive fore plane (with the grain). You can see the shavings produced by the heavily cambered blade (3″ radius), measured 0.35 mm (0.014″) in thickness. It made short work of the waste that needed to be removed.
Again, just some evidence of all the heavy hand planing that I have been engaged in for the past few weeks.
I finally returned to the mortises and tenons of the aprons and stretchers. It all needed some TLC to fit together perfectly. I used a float and a few chisels to do the fine tuning. I have to say that for a first attempt at such massive joints it all came together splendidly. You will notice the gaps at the top and bottom of the through tenon/mortise created to except wedges.
This weekend I finally managed to do the last massaging work needed to fit the short aprons and stretchers to their respective legs. You can see how well it all came together. Thank you David Charlesworth for all the tips in those DVDs that made it possible for a novice to get it done like this. As you can see I added quite a few layers of leather to one face of my heavy dead blow mallet during the previous week in order not to damage the work while tapping these gargantuan tenons home. Once assembled I positioned the legs on the two parts of the twin-top for marking out the through tenons and sliding dovetails.
I consulted the installation instructions of the Lie-Nielsen twin screw vise to make sure that the position of the legs does not interfere the positioning of the vise. Of coarse I did not keep track of the fact that the table was now upside down and that I needed to consider this when placing the legs and twin screw vise. After I did all the careful marking out and stood back, it dawned upon me that my twin screw face vise would end up on the right hand side of the bench once it is flip over onto it’s legs. Bugger!! So I had to start all over, but still prefer this to not realising and stuffing up the bench completely.
Once the various areas were marked out for the second time, I carefully transferred it to the opposite side and clearly indicated the waste to be removed.
In order to have a slightly more stable top beam to work on I clamped the two part of the top (first picture) together with two pipe clamps. Then I used my Lie-Nielsen tenon saw to cut the sides of the sliding dovetail dado (not sure if this is the correct terminology, but I am sure the pictures will make it clear). The pictures show the sequence of cuts I made in the waste area.
The chisel did the rest as you can see.
I thought I could use my Lie-Nielsen router plane to clean out the bottom of the sliding dovetail dados but it did not have the required reach.
Drilling out the waste from the mortises had to be done by hand as a 1″ drill in a handheld drill would surely destroy the motor in no time. This was thus the first job ever for my new purchased no 923 10″ Stanley brace. It is such hard going that I had to engage the ratchet mechanism to allow me to pull only, rather than trying to go through the full 360º, which is simply beyond my strength.
This weekend I started doing further work on the long stretchers, which included final hand planing of the swarthout cleats you’ve met before. In the pictures below you can see how my shop made winding sticks came in handy.
Between working on the stretchers I got so tired of drilling out waste with the brace and bit that I decided to giving the electric drill a go using a spade bit. I first set the drill to it’s slowest (and therefore strongest) speed and realised that it actually made short work of the waste removal. The drill did not even heat up at all, but never the less I did the drilling in many short bursts between other work. Of coarse the downside was that I realised how much religious baggage I still carry, as a peculiar guilt fell over me when the electric drill completed the job I started with the brace and bit. Luckily it did not manage to get much of a hold on me before moving on to other hand tool work.
My shop made panel gauge again came in handy to mark out the area on the inside of the long stretchers where I had to remove some Witpeer to accommodate the cleat. I removed the bulk of it with my Festool router and tidied up the mess left with a Stanley no 78 rabbet plane and my shop made shoulder plane. These tools leave a much better finish and enables you to dial in slowly to the absolute exact size of the rabbit needed.
The next rabbit is meant to accept the endgrain of Kaapse Swarthout boards sitting diagonally on top of the cleat to form the shelve. I used the same sequence of tools used.
While all this was going on I set up the second part of the twin-top in order to saw and chisel out the sliding dovetail dados. You will notice that I used the cleats temporarily as braces for the Darwinian saw horse as they were a bit wobbly when I did the same to the other part of the twin top. It made a massive difference.
Once again the Festool Domino enabled me to line the additional strip of Witpeer up perfectly with during the lamination process. As per usual, I used the Proletarian sanding contrivances to get rid of machine marks on the glue surfaces.
I started cutting the tongue and groove joints on these Kaapse Swarthout boards destined to become the shelve below the twin-top using my Lie-Nielsen tongue and groove plane. The first task however was to arrange the boards so that the edges would fit best in terms of colour and grain pattern. Once I felt please with the arrangement I marked the order using a carpenters triangle.
Once I started cutting the joinery, I realised that the boards were obviously thicker than ¾” (the thickness where this no. 48 plane centers on) which meant that another “tongue were left on the male edge of the joint. After taking one shaving I first drew a picture on a piece of paper (pictured) to work out how to fix this problem and realised that I simply had the remove the extra tongue with my #78 rabbet plane and all should be sweet.
As you can see here.
While all this were going on I liberated the Lie-Nielsen vise hardware from the safety of it’s packaging. Here you can see what it looks like.
On Saturday I started my next onslaught on the female parts of the leg-to-top joinery.
Luckily I had another look in Chris Schwarz’s book on building benches and saw that he actually make reference to the fact that one needs to remove the blade adjuster from our router plane to reach this depth. It meant that I could use this very useful tool to clean out the majority of the through dovetail dado (for lack of a better term), before using that flat surface as a reference to guide my chisel while removing the rest by horizontal pairing.
Here are two closeups of the router plane with the depth stop and blade adjuster removed.
I then took to the long stretchers with my shop made scrub plane to remove the excess timber from the strips I laminated on last week.
I am very happy to report that I managed to fit two legs to the one part of the twin-top this weekend. The first was quite a mission, but the second not so much. One really needs a healthy dose of patience for this work and it does not help much that I am starting to feel pressured to get the bench assembled before the changes in ambient humidity (probably some time in November).
The second leg. You might notice the gaps on either side of the through tenon’s mortise that was created to accept wedges.
I inserted the apron and took these photos to give you an idea of the base structure of the bench.
A week ago I started to take the Irwin quick release vise (destined to become the end vise) apart in order to “paint it black”, as I am a Stones man.