Update March ‘21 Coffee Table WIP : Complete!

[Fitzroy]

Won’t be up to the standard of many of the projects on here but hopefully it’ll entertain.

Shamelessly copying a table we saw in JL.

I’ve a few last wide planks of sycamore that are c. 650mm wide that I plan to use. I want the top as a continuous board which at 1000mm will reduce the width a little. Some measuring up and the boards were cut to rough size. Easier to hand saw than try to manhandle over any machines.

The slab for the top had some serious twist and thinned out over its length, the sawmanship was a little lacking when these boards were cut. The board was c. 45mm thick and I figured I could get a 30-35mm finished flat top from it. I was up for the challenger although I knew ripping it in half/thirds and rejoining was probably the more sensible move.

First thoughts about a router flattening set-up.

Some refinement later. Using a 1 1/2” router bit by Yonico about £20 and cuts very cleanly. Dust mask was essential and wife made me derobe before being allowed in doors, much mess was made in the shed.

The board was roughly flattened and then cut to size on the table saw, it’d lost enough weight to be manhandled with enough control. Once to size it had a final flattening. I pulled a couple of chunks of metal out the board before starting, one of which turned out to be a tooth from the saw which had made a darned mess of the board as it dug in before snapping off. The black stain is where the saw tooth sat in the wet wood after sawing.

I took another couple of mm off the board and the damage was still there. Using a chisel I trimmed back to discover the damage was at least another 4mm deep; thickness I couldn’t afford to loose. I’d have to have a crack at a patch.

Selected patch was cut and glued in about 5mm deep and planed flush. Not perfect but it’ll suffice.

I decided to plane and sand the top flat and smooth to make sure the boss liked the look and to make sure i could get a finish I was happy with before I get into the truing up the remaining boards and joinery.

I was suffering with edge damage on my plane and couldn’t work out why. Once I’d sanded smooth something glinted, and a magnet revealed the culprit. I though it was just a little knot.

On the back of the board is a significant cavity that’ll I’ll need to epoxy fill. Another first and skill to learn, that’s for next time. Also the original has box joints that I’d planned to router cut with a jig and pattern bit, but the wife has mooted dovetails. I’m just rather nervous about my first DTs being 600mm wide and 35mm thick. We’ll see!

Fitz

 

[MikeG.]

Excellent, Fitz! That's a bit of bad luck to come across not one but two chunks of steel in one piece of wood.

Before you experiment with dovetails on a showy joint on a showy piece of furniture, grab a few scraps and do a practise run, or two. The key skill with dovetails is marking out, and having a sharp pointy knife is critical. Have a go and see how you get on before making any decisions for your coffee table.

 

[woodbloke66]

Before you experiment with dovetails on a showy joint on a showy piece of furniture, grab a few scraps and do a practise run, or two. The key skill with dovetails is marking out, and having a sharp pointy knife is critical. Have a go and see how you get on before making any decisions for your coffee table.

Having just finished the book 'The Why and How of Woodworking' by Michael Pekovich, I can recommend the use of blue tape for marking out joints, especially dovetails. He's a particularly big fan and having tried out his method(s) the other day, it does work!
That wasn't the main reason for reading the book though as I was interested in the excellent section on making a Japanese 'kumiko' panel which I shortly want to build into a project - Rob

 

[Fitzroy]

Got some time in the shed and made a lot of shavings turning very twisty, cuppy, bowy, boards into flat sized and ready to be glued into ends and base.

I also started to contemplate if I could manage the dovetail ends. My attempt is below. The offcut was not totally flat which likely didn’t help. They came out ok, opinions on if you think glue would fill the gaps and look ok?

Main concern is these two pins and sockets took me about an hour. The whole table would have 60, that’s 15hours work. I’ll knock up a box joint router template next weekend and see how they look at the time involved.

Onwards and upwards!

Fitz.

PS. Yes I realised afterward you can’t knife line across the whole set of pins/tails

 

[MikeG.]

That's a great start. I'm sure you've learned a lot from doing those. Take them apart and find out why those little gaps occurred, and you'll learn even more. Not having accurately dimensioned boards is just about the biggest complication for good dovetails, and the other big deal is avoiding breakout. Chiseling in to half way then flipping over is key to that (mind the debris on the bench when you do that). I'll repeat for the umpteenth time Mike's Rule of Halves: place you chisel half way between the edge of the waste and the gauge line, every time, until the amount left to remove is too small to be able to do another half. This means the first time you chisel touches the gauge line it is only having to remove a sliver.

I'm not sure what you mean by "you can't knife line across the whole set of pins and tails".

 

[Fitzroy]

That's a great start. I'm sure you've learned a lot from doing those. Take them apart and find out why those little gaps occurred, and you'll learn even more. Not having accurately dimensioned boards is just about the biggest complication for good dovetails, and the other big deal is avoiding breakout. Chiseling in to half way then flipping over is key to that (mind the debris on the bench when you do that). I'll repeat for the umpteenth time Mike's Rule of Halves: place you chisel half way between the edge of the waste and the gauge line, every time, until the amount left to remove is too small to be able to do another half. This means the first time you chisel touches the gauge line it is only having to remove a sliver.

I'm not sure what you mean by "you can't knife line across the whole set of pins and tails".

Thanks Mike for the comments, and yup trying to implement the rule of halves. My comment regards the knife line, I marked across the width of the whole piece at the base of the tails. Which means I have a nasty great knife line at the base between the tails!

Fitz

 

[MikeG.]

That not a problem. Many people leave it on as a statement that the dovetails are handmade. Otherwise, just plane it off (or use a belt sander). Obviously, if you want to get rid of it you won't score it very deeply in the first place. I often keep them, and if it is part-removed by cleaning up the dovetails after the glue-up, I've even been known to reinstate the line.

 

[Fitzroy]

After a few busy weekends doing family stuff I got a few hours back in the workshop to practice the joints for the corners of the coffee table.

It’s safe to say dovetails are out the window, need to practice on a more manageable scale before leaping to 30mm thick stock on 600mm wide pieces.

So it was onto box joints using a template and router. There are a few different approaches on YouTube but the one that I liked, read thought I had a chance of replicating, was this one. The jig is a router template that clamps onto the board ends.

Before charging into making the whole jig I planed up some fingers and screwed them to a piece of wood, clamped on a board and had at it with my router. Lesson one; a router bit extended 35mm in a big ass router, 3 1/4hp, is terrifying and tries to bang the wood all over the place. I figured out I needed to mark out the teeth and cut away 95% of the waste on the bandsaw. Proof of concept was it felt possible.

The jig utilises a set of fingers that are held in place with T-track and nuts. For cost sake I found I could use 2020 aluminium extrusion, and parts. My final jig for the coffee table will be 600mm wide and have c.20 fingers. For £25 I got 1m of extrusion and 25 nuts and bolts. Giving me plenty to make up this prototype jig.

Learning two was that I needed more clampage. The first joint I cut on the jig the board shifted slightly during the cut and it ended up gappy. This was resolved with a block of wood screwed in place to prevent lateral movement of board in the jig.

I was concerned over the weight of the router deflecting the jig. However, with a 50mm thick piece of wood for the main body and 20mm thick fingers it was solid when all tightened up and any movement was at the interface between jig and bench so the board position was maintained relative to the router.

Also rather obviously I figured out that routers don’t cut into square corners, yes I’m a noob! 12mm MDF was used as a backing board that gave a nice clean finish and prevented tear out. I’d tried some offcuts of real wood but the router tended to jump as it encountered the board. The mdf was much kinder, just needed to ensue the dust mask was well fitted!

The final joint off the jig, having incorporated all my learnings, was really pleasing. The joint went together with minor clamping pressure, but I did need to squeeze in each set of teeth. I’m concerned I need lot more clamps for the glue up.

Learning is a process that often end in failure on the path to success!

So proof of concept complete now I need to remake the jig in a 600mm wide version. This piece of oak has been awaiting a project for a while. It’s a sleeper offcut that’s dried in the shed, it’ll have too many defects for anything nice but it’ll do nicely for a jig.

I’ll take a few more pics when I make the main jig.

Fitz.

 

[MikeG.]

I made a box joint jig for the router many many years ago, but unlike yours it sat on the end of the wood, not the face. This meant that the corners it cut were all square. You do need a sacrificial backing board, though, to prevent blowing out the back edges. The other advantage it had was that I could do multiple boards at the same time. Before you hack that lovely big lump of oak around why not have a little think about the alternative.

 

[Fitzroy]

There’s one of those in the failure pile. The problem I found is you end up with an internal round corner that you have to chisel square. With 160 to square up in total my conclusion is I’m going with the end mount jig where the only chisel work is minor tweaks.

Fitz.

 

[MikeG.]

We're talking at cross purposes. I am actually describing what you have made. My mistake.

 

[Fitzroy]

So, back on this project after a couple of months working on other stuff. I’ve joined the boards for the bottom panel, and made a pretty poor effort of shooting the edges so the glue lines are very visible. Then I had a number of defects to fill including a large cavity that I decided I would epoxy fill.

I ordered some GlassCast 50 from easy composites. Sourced a clean empty pot and some old kitchen scales and it was on with the mixing.

The defects go all the way through the wood so I taped up the show face with masking tape, flipped the board and part filled from the back.

I’d not really read much about what tape to use and figured the viscosity of the epoxy would prevent most leaks. Hmmm, the deep fill epoxy takes hours to start to thicken and pushed past the masking tape in a few places. My fix was to wipe the drips off and seal the leaks with CA glue, which worked ok.

The epoxy recommends a maximum fill of 25mm in wood. My pour was small but decided to do it in two layers in case it went pear shaped. So 24hrs after the first pour I filled the defects to the top. I was amazed how clear the resin was.

So the following morning, 18hrs later, it’s all set up and I’m itching to take the masking tape off. I flip the piece, remove the tape and sand back. There are a few minor holes but it’s looking good.

I left the top flipped over so I could come back and fill the holes. It turns out it really is 48hrs till a full cure. This evening i found the material from the second pour had made a break for it under the effects of gravity, a very pretty pattern but oh balls!

It’s on the back and the majority of the defect is still full so I’m going to leave it be and see what I have when it’s all sanded back.

Definitely a school day today!

Fitz

 

[Trainee neophyte]

This is brave - a huge piece of wood and perfection required. I have just made some slabs which are begging for a similar treatment, but luckily it will be a couple of years before they are dry enough to force my hand. In the meantime, I had better practice some joints.

Oddly, with this enormous lump, it looks like a hand saw and chisel would be the simplest option, but there is no way I would lose my dovetail virginity to such a project, either! This is precisely why God invented electricity.

 

[Fitzroy]

Wow, what a year, the ‘rona, working from home, home schooling, house rearranging and new furniture making, gate building, tree chopping, and secret Santa’ing. 7 months later and I’m back on the coffee table.

I’ve claimed the project languished due to other priorities, but to be honest it’s as much been an excuse to hide behind those other projects as a necessity. The idea of getting the end joints to a suitable level of accuracy that everything goes together, and any gaps are tolerable was so daunting it put me off starting!

So where did I stop? I’d discounted dovetails as a step to far, and box joint jig version one had shown promise. Version 1 was 20cm wide so needed scaling up.

A piece of oak, a sleeper off cut, was planed down to make a beam 100x60x700. An end plate and base plate were also made.

A groove was cut in the beam using the table saw and finished to depth with a router plane, the 2020 aluminium extrusion is screwed into the jig.

I’d learnt on jig V1 that clamping the workpiece to the jig was key, and the beam has through holes to enable clamping. The base plate and end plate are screwed and glued in place. The end of the main beam is finished as square as possible using a shooting board and plane.

The end plate forms a reference to get the fingers perpendicular to the jig, so getting the end plate as square as you can is critical. Fingers are slid in place one at a time and each checked for square.

With the jig set up I marked up two boards, made the cuts and fitted them, or tried to fit them........The joint was too tight, and out of alignment in a couple of places. I went back to the drawing board, should I say original article on the jig. I had failed to use a shim between each finger to give some joint space, the original jig used a cigarette paper. How thick is one of those! Turns out about 0.05mmm, which is equivalent to three thicknesses of my kids tracing paper.

I took all the fingers off the jig, and rebuilt it with shims between each finger. Additionally I used extra bits of tracing paper to shim every finger as close to square as I could. More fettling took place after joint two, and joints three and four were more successful. But with a piece as large as this would all four corners actually go together?

The two boards to be joined are fitted in the jig, with mdf backing boards for the router to eat into. The box teeth locations are marked on the boards and most of the waste cut away on the bandsaw.

TBC

 

[Rorton]

Brilliant work so far. I really like the glossiness of the wood when the epoxy is on it. Cover the whole thing in epoxy after !?

 

[Fitzroy]

With a little help from the coping saw on the boards too big to do on the bandsaw.

I’d learnt on jig V1 that trying to route out the entire joint was a recipe for disaster, obvious really with a 30mm deep joint, and a router cutter extended about 60mm! I tended to cut a few mm inside of the lines, easy to do and left an easy cut on the router.

The end of each board was not completely square. To remedy this shoulder lines were marked up with considerable care to get them correct. The jig was then spaced/shimmed such that the shoulder line was at a constant depth.

Two joints came together ok. But needed lots of sanding due to earlier issues discussed.

The last two joints were much better and by some miracle it’s worked out swimmingly.

The was a mess up on one joint where the router cutter dropped in depth, balls, but I noticed as it happened and it’s only effected one cutout so should be fixable.

Now that I have joinery that I’m happy with I can think about finishing. More to follow.

Fitz.

 

[Fitzroy]

I took a wee dram to the shed and spent a nice couple of hours considering edge details. This is the bit of the process I really enjoy, but that also takes forever. The touches that make a piece your own.

I’d glued up the earlier test joint. I’m using 30min epoxy to give me maximum open time, there is lots of surface to apply glue to before getting it together square. Even with the test joint and only 5 teeth it took me a good five mins to get glue everywhere!

I want to round over the edges and am thinking about a 5mm or 12mm radius round over. So on the test joint I marked the round over and the planed to it, then a bit of sanding and it’s done. It was a nice and rewarding hand tool process. Whether I switch to the router for the actual piece we’ll see.

I like both the round overs but am leaning towards the 12mm look.
5mm

12mm

It’s hard to tell in the photo and the wood pattern on the 5mm doesn’t help.

The other opinion is do I round over the inside edges also. That’ll have to be done with the router so will need to wait until tomorrow when I can make noise.

Fitz.

 

[RobTy]

12mm for me. Looks great.

 

[billw]

I've been looking at it for about ten minutes on and off and still can't decide for sure but my preference seems to be leaning towards 5mm.

 

[jcassidy]

Fantastic work, inspiring. I too found out how runny resin is on a wee project for the kids, I built a frame to keep it in but neglected to make sure it was completely tight and half the resin ended up on the gravel.

Anyway, I'd plump for the 5mm, I think if small details are key to a great look, keep the details small!