Tree Table Project

UKworkshop.co.uk

Help Support UKworkshop.co.uk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

yetloh

Established Member
UKW Supporter
Joined
1 Dec 2008
Messages
1,590
Reaction score
227
Location
Sussex
I’m not a great thread starter but thought one or two of you might be interested in a major project I completed a while back. Because this is a completed project I can start off with a photo of the finished article so anyone who is not interested in the “how” can just look at it and decide whether or not they like it and not have to plough through to the end.
tree table 2s.jpg


A bit about the inspiration may be useful for starters. The idea for the table started germinating when my daughter was living in a small Edwardian cottage. At the time she had a small round pine dining table, bought second-hand for very little, with the top secured to the pedestal by a single large screw through the centre into the turned pedestal column – utilitarian but certainly not pretty. I thought it would be good to make something better. Space was restricted so a pedestal table seemed best, with no legs to get in the way. I got to thinking that a pedestal table is a bit like a tree, with a trunk, roots for stability and branches to support the top, so why not make it more literally tree like?

Although a lot of thought went into the design, it evolved quite a bit during the making, partly to solve unforeseen problems but also to make it a better resolved design. Much of the making was done on 5-day courses at West Dean College where I had the help of Bernard Allen, the tutor and very good friend, but also that of my fellow students who were always ready to lend a hand. Courses like these, where each student could make whatever they wanted were fantastic learning experiences and always hugely enjoyable but sadly, I fear, a thing of the past. Bernard has now retired and I can’t imagine anyone else wanting to take on the extra pre-course work of mentoring students, each with different project plans, let alone the stress of running six to eight separate projects in parallel.

Making the Tree Table took three years of very much part-time work.



My original plan was to make the legs from steam bent sycamore – an expensive lesson because every attempt at bending it failed; I didn’t know then that serious bending requires a ring porous wood like the English ash I eventually used.

The Pedestal

Steam bending, particularly when the component is relatively thick as here (18mm) is a lot less predictable than veneer laminating, not least because different amounts of bending are required across the thickness, leading to considerable stresses. When bent round a former, the inside of the bend will tend to compress the wood along the grain, while on the outside the forces will try to stretch the wood. While compression is perfectly possible and manifests itself in a way a little like the folded skin on the inside of a bent finger, the scope for stretching the wood on the outside of the curve is very limited indeed. This isn’t a problem with veneer lamination because each veneer behaves separately and they can slide against each other. To mitigate this problem with solid wood bending I used a steel strap bending jig to hold the work during the bending process. This aims to limit the external stretching and force more internal compression. Here is CAD image of the design I used. Apologies for the broken image, but the drawing and annotations give a better idea of how it works than I could give with words alone.

View attachment TT bending jig.jpg
And here is the actual (much longer) jig with wood wedged in it preparatory to actual bending.

tree table 006.jpg


And here the wedge is more clearly visible

tree table 008.jpg


Here is the wedging arrangement. Note that the strap wraps right round the wedging block and is bolted through the handle to cope with the considerable forces involved in a achieving the bend.

tree table 009.jpg


This photo. is of a practice run for the initial attempts – hence the sycamore blank rather than ash.

tree table 017.jpg


An early bending experiment with the West Dean student/tutor team - that’s Bernard Allen in the blue shirt giving directions. It was a bit like a bunch of headless chickens at first, with people getting in each other’s way in our efforts to get the bend done and clamped before the blank cooled too much, but things improved as we all got used to working systematically.

More to follow, hope you find it of interest. If anything isn't clear, please say so.

Jim
 

Attachments

  • TT bending jig.jpg
    1.4 MB
Looks good!
The steel bending strap was more or less what Thonet invented for the making of his bentwood furniture. The idea being that the outside of the bend shouldn't stretch and delaminate but instead the inside should compress. In fact you can often see a crumpled look on the compressed inner sides of Thonet's bent bits
 
Thanks everyone for your interest and feedback, here is the next instalment.

Clamping the hot leg blanks was something that evolved with experience and proved the value of the age old “keep it simple” principle.

tree table 016.jpg


This was our starting point, the former was supported at its sides on softwood bearers so that three I beam sash cramps could run underneath and a big block cushioned the straight central section and spread the pressure evenly. Note the two holes in the former to take big F or G clamps near the centre of each bend.

tree table 023.jpg


Some of the clamping was rather makeshift – who clamps onto a a clamp? But this may have been a temporary measure, too long ago to remember!

tree table 025.jpg


Along the way we realised just how well the big hole/F or G clamp combination worked but the clamping was still over-complicated . Achieving that bend took a lot of pushing and shoving and the work/former tended to fall off those side bearers and, when speed is of the essence, stability and the fewest clamps possible are the best solution.

tree table 036l.jpg


This was the definitive arrangement. Simple, sequential and fast to put into action with a couple of “clampers” to back up a couple of “benders”. Our student/tutor team quickly became pretty fast and efficient at it.



tree table 034A.jpg


Once the blanks had cooled and come off the former, there was significant spring-back so they were clamped around single layer 12mm MDF keepers with support blocks – I had to order six big F clamps to be delivered to West Dean from Axminster. The blanks were transported home in and kept in these keepers. Once in my workshop they were placed in a makeshift plastic sheet enclosure with a dehumidifier to dry them.

At the end of this process the spring-back was quite a bit less but there was still significant variation in its amount across the eight pieces that were bent. In the end I discarded the extreme variations and picked the four that were closest in size and the most visually harmonious. Even so, there were still significant variations - a problem that come back to later.

All this sounds like a very profligate use of wood but one of the great things about West Dean is the huge range of craftspeople and artists you get the opportunity to meet. On a later course I met a brilliant letter carving tutor who was running a course at the same time as my furniture course; he was delighted to have my spare legs which he cut into sections, carved with lettering and sold through galleries and his website.

Before moving on to the next stage, a look at the West Dean steam box may be of interest.

tree table 029l.jpg


A ply box with a door at each end houses a length of stainless steel flue liner supported at each end by ply bulkheads. Maintaining as much heat as possible is vital to effective steaming, so the space around the liner is packed with fibreglass loft insulation. Improvised rubber seals on the liner ends and doors are quite effective at containing steam and heat. The box is arranged to slope down towards the back door which has a pipe fitting to take the pipe from the wallpaper stripper that can be seen on the floor. The back end of is propped up to slope to incline the box the front , where a drain tube can be seen to carry away the condensate which drips into a bucket.

On the top of the box is a digital thermometer with remote probe. There is a nearby hole in the box and corresponding hole in the liner through which the probe passes. Once warmed up, the thermometer indicated a consistent 95 - 98°C, which is pretty good. Note the piece of wood that is visible a little way along the liner – there were three of these along the length, to hold the workpiece clear of the liner and allow steam to reach every surface.

Jim
 
Hi jim, great writeup 🙂 glad you found the time to share, things like this give people inspiration and seeing what can be done gives a bit of confidence to get stuck in and try 👍
To the rest of you, wait till you see the chairs 😉
 
tree table 043.jpg


The first time the legs were put together – in the workshop at West Dean. This was a moment of minor revelation because it was when I realised just how beautiful the view of the pedestal from above could be with its flower-like form. This was too good to conceal, so it was easy to decide that the top would have a central glass panel rather than be solid, as originally intended. Coincidentally, it also served to avoid the tricky technical problem of bringing multiple radial veneer segments to meet perfectly at the centre of the top.

tree table 046.jpg


Shaping the mitre on the trunk section of a leg, using a block plane and a 45°guide block at West Dean. You may be surprised to see that I am doing this by hand rather than with a router, but that would have involved devising a way of securely holding both the work and the router in the correct relationship whilst minimising the risk of something going irretrievably wrong in the blink of an eye. I doubt that the router would have saved much time; besides, the old fashioned simple answers can be so much more satisfying. What does surprise me about this photo is that I have never owned one of these LN block planes – I guess I must have borrowed it.

tree table 059.jpg


A simple hardboard template helped keep the leg profile consistent – I learnt this from watching Karl Holtey form a plane handle for a No 98 smoother. Unlike most other planes - the part you grip is oval in section rather than flat and simply rounded front and back, and all the more comfortable for it.

But I digress, these were the main tools I used for shaping.

tree table 066.jpg


These American Woodjoy spokeshaves really are wonderful tools and well named. I have three of them – flat based, curved base and one with a cigar shaped brass centre section for tight internal curves. Sadly, I understand the maker is now retired so they are no longer available. I got mine from Classic Hand Tools so there must be a few around in the UK. If you see them advertised, I strongly recommend you snap them up.

Tree table -481.jpg


This Sorby drawknife was OK for rough work but too clumsy for much of the finer stuff. I bought it new and it was very coarsely ground so needed a lot of work to get a decent back and bevel.

tree table 070.jpg


Spokeshaves apart, this little Veritas 4” carvers drawknife was my main workhorse. It is very manoeuvrable; altogether a joy to use. Mine is A2 steel so is fairly average in its edge holding ability so I wrote to Veritas asking why they didn’t make it in their PMVII steel and got a message back saying they would pass on the comment. Having just looked it up I see that the latest version is PMVII, so will need sharpening much less often. Pricey but worth every penny in my opinion.

Working in three dimensions with these hand tools was hugely instructive. It taught me a great deal about reading grain direction which years of working square stuff had failed. to do. The tactile feedback through the tools is a vital and very satisfying part of it. An Arbortech would have been much quicker, but the noise, dust and vibration…. No doubt it is, for many, a matter of priorities, but I am sure most of those who automatically reach for the power tools have no idea what they are missing.

tree table 060.jpg


Curved cabinet scraper. Perfect for smoothing out curves.

tree table 079.jpg


It takes a lot of work to generate this many shavings without using a plane!

Next time I will be getting to the knotty problem of how to join the legs securely.

Jim
 
Thanks everyone for your interest and feedback, here is the next instalment.

Clamping the hot leg blanks was something that evolved with experience and proved the value of the age old “keep it simple” principle.

View attachment 151773

This was our starting point, the former was supported at its sides on softwood bearers so that three I beam sash cramps could run underneath and a big block cushioned the straight central section and spread the pressure evenly. Note the two holes in the former to take big F or G clamps near the centre of each bend.

View attachment 151774

Some of the clamping was rather makeshift – who clamps onto a a clamp? But this may have been a temporary measure, too long ago to remember!

View attachment 151775

Along the way we realised just how well the big hole/F or G clamp combination worked but the clamping was still over-complicated . Achieving that bend took a lot of pushing and shoving and the work/former tended to fall off those side bearers and, when speed is of the essence, stability and the fewest clamps possible are the best solution.

View attachment 151776

This was the definitive arrangement. Simple, sequential and fast to put into action with a couple of “clampers” to back up a couple of “benders”. Our student/tutor team quickly became pretty fast and efficient at it.



View attachment 151777

Once the blanks had cooled and come off the former, there was significant spring-back so they were clamped around single layer 12mm MDF keepers with support blocks – I had to order six big F clamps to be delivered to West Dean from Axminster. The blanks were transported home in and kept in these keepers. Once in my workshop they were placed in a makeshift plastic sheet enclosure with a dehumidifier to dry them.

At the end of this process the spring-back was quite a bit less but there was still significant variation in its amount across the eight pieces that were bent. In the end I discarded the extreme variations and picked the four that were closest in size and the most visually harmonious. Even so, there were still significant variations - a problem that come back to later.

All this sounds like a very profligate use of wood but one of the great things about West Dean is the huge range of craftspeople and artists you get the opportunity to meet. On a later course I met a brilliant letter carving tutor who was running a course at the same time as my furniture course; he was delighted to have my spare legs which he cut into sections, carved with lettering and sold through galleries and his website.

Before moving on to the next stage, a look at the West Dean steam box may be of interest.

View attachment 151778

A ply box with a door at each end houses a length of stainless steel flue liner supported at each end by ply bulkheads. Maintaining as much heat as possible is vital to effective steaming, so the space around the liner is packed with fibreglass loft insulation. Improvised rubber seals on the liner ends and doors are quite effective at containing steam and heat. The box is arranged to slope down towards the back door which has a pipe fitting to take the pipe from the wallpaper stripper that can be seen on the floor. The back end of is propped up to slope to incline the box the front , where a drain tube can be seen to carry away the condensate which drips into a bucket.

On the top of the box is a digital thermometer with remote probe. There is a nearby hole in the box and corresponding hole in the liner through which the probe passes. Once warmed up, the thermometer indicated a consistent 95 - 98°C, which is pretty good. Note the piece of wood that is visible a little way along the liner – there were three of these along the length, to hold the workpiece clear of the liner and allow steam to reach every surface.

Jim
You want to be building a 40' cutter with that steam box!👍🤔👍!
 
@yetloh
Beautiful design and craftsmanship. Reminds me of the work by John Makepiece. Stunning. Looking forward to watching this thread. Thanks for sharing.
Fred
 
You want to be building a 40' cutter with that steam box!👍🤔👍!
Now there's an idea, but on reflection I think I might be a bit old for that, not to mention persuading the wife it might be a good idea, Speking of boats, I do like some of this guy's stuff https://pettersouthall.com/ . I particularly like his slatted table which uses some boat building techniques.

Jim
 
@yetloh
Beautiful design and craftsmanship. Reminds me of the work by John Makepiece. Stunning. Looking forward to watching this thread. Thanks for sharing.
Fred
Thanks Fred. Although the design came out of my head, we are all influenced by people we admire and I'm quite sure that he has influenced/inspired me.

Jim
 
an inspiration to us all........
I'd be proud to own it let alone the skills to build it....
Thanks Clogs. I think my daughter is pleased with it. What skills I have I must credit to my good friend and West Dean tutor, Bernard Allen. He has always encouraged me and I have never once seen him suggest or imply to a student that something might be beyond them. I am sure that this is part of the reason why I have never been put off doing something becayse I have,'t done it before. I have had plenty of disasters but always just got on and started again if necessary. The downside is that I am painfully slow so could never have made a living from furniture.

Jim
 
Thanks also to Martin and COWS for your kind words. There is quite a long way to go yet and I hope everyone will enjoy the ride.

Lastly, particular thanks to Baldkev for his encouragement to start this thread which I don't think I would otherwise have done, it really is good to know that the effort is appreciated.

Jim
 
Thanks also to Martin and COWS for your kind words. There is quite a long way to go yet and I hope everyone will enjoy the ride.

Lastly, particular thanks to Baldkev for his encouragement to start this thread which I don't think I would otherwise have done, it really is good to know that the effort is appreciated.

Jim
Im glad you shared, we all get to learn a bit along the way from reading these sort of project threads and they give inspiration to future readers too.
You clearly have a lot of skill and patience and you should be incredibly proud of your efforts. Just imagine what it might cost to commission a furniture maker to make this.... a lot!

👍 😎
 
With most of the shaping done, it was time make some decisions as to how the to fix the four legs together to form a really strong cylindrical trunk. Clearly, just gluing them together at their mitred edges would not provide a structure of guaranteed long-term integrity. I considered several options including a piece of square section steel tube with threaded inserts in each leg and set screws tightened from inside the tube, but eventually settled on the simplicity of a block of maple through the straight section of “trunk”. I could screw through this to fix two adjacent legs, but I needed a means to fix the other two legs without the fixings being visible.

My tutor/mentor, Bernard, suggested keyplates might be the answer. One advantage of sharing the south coast with lots of boaty people is that there are small businesses locally that service their boaty needs, not least stainless steel fabricators. Over the years I have built up a good relationship with one such craftsman from whom I can occasionally scrounge the odd scrap of stainless, which is what I did so I could make four of these. Working in metal is not my favourite activity – cold, dirty, scratchy stuff – but it makes a change.

TT block-2.jpg


Here is the maple block with two plates installed. Two adjacent faces have these exit holes for the screws fixing them directly to a leg.

Tree table 157.jpg


While, on the entry side, the hole is slightly counter-bored to take the screw head within the recess in which the screw head for the opposite leg slides while the leg is being fitted - I hope that is clear! Basically the screw head for the leg that is directly screwed to the block must not interfere with the head of the screw in the opposing leg which relies on the keyplate to secure it to the block.

Tree table 161.jpg


The finished component. This method of fixing requires very careful layout in order that each leg ends up in exactly the right relationship to its neighbour and that the maple block ends up in the correct position inside the central ”trunk”. The beauty of it is that it can easily be taken apart as many times as necessary to adjust the screws that locate in the keyplates to achieve the perfect tight sliding fit. As a final step, each leg to block face was glued with Titebond liquid hide glue which has a long open time and is, at least theoretically, reversible should disassembly ever be necessary, that is of course, if the person trying to do it can work out how it all works.

Tree table -2301.jpg


One possible problem I had anticipated at the design stage was the possibility that the pressure of people leaning on the tabletop might put excessive strain on the bottom of the leg to block joint. At the top, splaying stresses would be limited by the fixed tabletop but no such constraints would apply at the bottom, so I had my stainless fabricator make these stays – shown here with the locking nuts yet to be tightened. Having the legs firmly but not yet permanently fixed in a trial assembly proved the viability of the screw/keyplate/stay combination, but it also provided the opportunity to pull up a chair with a temporary mdf tabletop clamped in place and see how it all felt. The answer was, far too bendy; something I had not anticipated. Simulation of a diner resting elbows on the tabletop flexed the top section of the legs so much that there would be serious risk of a “wine in your lap” event. This new problem was the downside of the decision to use ash for the legs to achieve the bend I wanted, solving one problem but creating another. Had I succeeded in bending the sycamore which had been my original it might not have arisen, but we have to start from where we are rather than where we might have been. Some serious stiffening measures were needed, which I will come to in my next post.

Jim
 
Back
Top