VM chair project

UKworkshop.co.uk

Help Support UKworkshop.co.uk:

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

Mr T

Established Member
Joined
25 Feb 2008
Messages
1,111
Reaction score
8
Location
Ilkley, West Yorkshire
IMG_7527 72.jpg

For some time I have thought about offering a different sort of a chair making course. Most chair making courses are for turned chairs and not everyone likes turning (I know some people find this hard to believe!). So I was thinking of a course making a jointed chair. About a year ago a student showed me a picture of this chair which is occasionally made by students at the Krenov style colleges in the US. It’s a copy of a chair made by a friend of Krenov’s, Vidar Malmsten. I liked the mix of Chippendale with Scandinavian minimalism. So I thought if they can make it in California why not in Ilkley!

I am indebted to Craig Johnson at Studio Tupla in Minnesota, his blog on making the chair was very helpful. Woodbrains of this forum also helped in trying to track down some drawings, unfortunately this was unsuccessful.

In the absence of drawings I drew up my own workshop drawings using photographs from the internet. I then made a full size mock up in poplar. This was not fully jointed, only held together with screws, it was made just so I could check out the ergonomics. The design of the back is critical, I wanted the chair to be very comfortable so the angle of the back and the curvature of the splat were considered closely. I decided that back the was swept back too far making the sitter a bit too reclined, the seat was also a tad too deep. To adjust the back I just packed the crest rail forward to find the new angle.
IMG_7172_opt (2).jpg

Once I was happy with the comfort I drew the workshop drawing in detail. The drawing was very important in the making, being used as a reference throughout the project. In most projects time spent on the workshop drawing is well spent.
I thought about using dominos for the jointing but decided not to for three reasons:

1.The compound angles and curves would have made using the domino difficult, I thought mortice and tenons would be easier in this situation. Especially the way I had decided to do them.

2.The chair was a prototype for a course and I know that most students don’t have a domino, so using a domino would not have been appropriate

3.St James would have spun in his grave if he had known I was making the chair with dominos!

After drawing up a cutting list and preparing the parts I started by marking out the mortice positions on the legs.
Morticing the back legs was going to be tricky! I wanted to use a router as this is the most likely piece of kit for my students. I decided that this was best done before ny shaping of the legs. So when cutting out the back leg blanks I made sure there was a straight reference edge on the front that could be aligned on a vertical reference on the workshop drawing, this vertical reference was used as the datum for marking all the mortice positions on the back legs. I would then route mortices through the waste into the eventual positions of the legs within the blank. Marking and morticing the front legs was easier as they were not going to be curved, just tapered.
IMG_7170_opt.jpg

Morticing was done with an 8mm pocket hole bit from Wealden.
IMG_7188_opt.jpg

The long reach of the pocket hole enabled me to rout through the waste between the front reference edge of the back legs to the eventual leg position within the blank.
IMG_7168_opt.jpg

To cut the mortices for the cross rails in the back legs the fence ran against angled packing, the angle being picked from the workshop drawing.
IMG_7190_opt.jpg

A back leg blank with mortices cut.
IMG_7194_opt.jpg

After the mortices were cut the leg was trimmed to length on the table saw, the angle of the top cut being picked from the drawing.
IMG_7196_opt.jpg

A 10mm dowel hole was drilled at 90 degrees to the top end cut, this would eventually be for fitting the crest rail.
IMG_7197_opt.jpg

With all the jointing work done on the back legs they could be shaped. They were cut approximately to shape on the band saw, leaving about 2mm spare, then finally trimmed on the router table using a 50mm Easitip replaceable blade bearing guided cutter from Wealden. The template for this had previously been cut using a router and trammel.

The next job was to form the tenons using the table saw. More on this later.......
 

Attachments

  • IMG_7527 72.jpg
    IMG_7527 72.jpg
    86.2 KB
  • IMG_7172_opt (2).jpg
    IMG_7172_opt (2).jpg
    26.8 KB
  • IMG_7170_opt.jpg
    IMG_7170_opt.jpg
    25.4 KB
  • IMG_7168_opt.jpg
    IMG_7168_opt.jpg
    23.3 KB
  • IMG_7188_opt.jpg
    IMG_7188_opt.jpg
    20 KB
  • IMG_7190_opt.jpg
    IMG_7190_opt.jpg
    19.9 KB
  • IMG_7194_opt.jpg
    IMG_7194_opt.jpg
    11.7 KB
  • IMG_7196_opt.jpg
    IMG_7196_opt.jpg
    23.1 KB
  • IMG_7197_opt.jpg
    IMG_7197_opt.jpg
    22.5 KB
Another lovely WIP. Very interesting and well explained, thank you.
Nice workshop!
Regards Rodders
 
I agree with Rodders!
Very interesting to read how to deal with all the angles. I too shall enjoy following this build.
 
Cutting the tenons

Having cut all the mortices I started to work on the tenons. The compound angles involved on the
side rails led to considerable head scratching. Craig Johnson in his blog on making the chair used
the table saw to cut the tenons, I’m not sure if that is the Krenovian way or just how they all do it
in the US. I decided that I would try doing it that way.

To achieve the compound angle I set the saw angle to that of the splay between the front and back
legs, this was picked from the workshop drawing as was the length of the rails. If I had adjusted the
fence on the sliding table to achieve the “vertical” angle I would have had to reset it from one shoulder
to the other on a tenon. This may have introduced inaccuracy, instead I made up a set of tapered
packing pieces which could be used to angle the piece from the table fence. But cutting the other
shoulder was not just a matter of flipping the packer as this would have meant the saw angle would
be wrong. To cut the other shoulder I needed to cut from the other side of the blade. So I clamped a
cross piece to the fence which allowed me to take the packer to the other side to cut the other shoulder.
The packer was stuck to the cross piece with double sided tape so the saw would trim the end, the
cut end could then be used to align the previously cut shoulder with the line of cut for the new shoulder.
You have to have been there to really understand this!!

Once the shoulders were cut I cut the tenon cheeks on the band saw using another angled packer to
get the cutting angle. To cut the thickness of the tenon I first took a cut with the angled packer against
the fence then used an additional straight packer the width of the mortice plus the saw kerf to shift the
piece over the required amount.

The tenons on the cross rails were more straight forward as most of them, except the bottom back rail
had square shoulders. These I rough cut on the band saw then trimmed on the router table using a tenoning
cutter from Wealden tools which gives a very clean cut.

When you use a router for morticing you often leave yourself with the choice of squaring the mortice or
rounding the tenon. Mostly I go for squaring the mortice, I think this is easier as it can be difficult rounding
the tenon when the grain direction is unhelpful. However on this occasion I fancied a change and rounded
the tenons, it seemed to work OK, although some were a bit ragged.

IMG_7204_opt.jpg

IMG_7200_opt.jpg

IMG_7208_opt.jpg

IMG_7206_opt.jpg

IMG_7211_opt.jpg

IMG_7213_opt.jpg


Before I could fully fit the tenons I had bit more shaping to do on the legs, especially the front legs.
I think the design of this chair works partly because of the subtle shaping and tapering of the parts.
I think one would need to make a number to fully explore this.

Before shaping the front legs I drilled for 10mm dowels in the tops, these would be for fitting the arms.

The front legs have a number of tapered faces at the bottom. Also the rounded top is not turned as it is
off set to the outside, so it is better to create the rounded taper with a spoke shave and block plane. The
offset is noticeable in the positioning of the dowel hole in the earlier picture.The inner edges of the
back legs were tapered at the bottom using a jack plane.

Once all the tapering was complete I did an initial fitting of the tenons, if the tapering was true to the
drawing the shoulders should have sat down neatly onto the faces. They were OK but would need a
little tweaking before glue up, specially where the face was curved.

IMG_7245_opt.jpg

IMG_7233_opt.jpg

IMG_7246_opt.jpg

IMG_7236_opt.jpg


The next job will be shaping and fitting the lumbar and crest rails.
 

Attachments

  • IMG_7204_opt.jpg
    IMG_7204_opt.jpg
    33 KB
  • IMG_7200_opt.jpg
    IMG_7200_opt.jpg
    30.7 KB
  • IMG_7208_opt.jpg
    IMG_7208_opt.jpg
    33.1 KB
  • IMG_7206_opt.jpg
    IMG_7206_opt.jpg
    31.8 KB
  • IMG_7211_opt.jpg
    IMG_7211_opt.jpg
    33.8 KB
  • IMG_7213_opt.jpg
    IMG_7213_opt.jpg
    30.5 KB
  • IMG_7245_opt.jpg
    IMG_7245_opt.jpg
    55.5 KB
  • IMG_7233_opt.jpg
    IMG_7233_opt.jpg
    35.9 KB
  • IMG_7246_opt.jpg
    IMG_7246_opt.jpg
    28.9 KB
  • IMG_7236_opt.jpg
    IMG_7236_opt.jpg
    22.9 KB
The crest and lumbar rails.
The crest and lumbar rails both curve in two dimensions. The lumbar rail was easier to form because the curvature was not very pronounced. The crest rail required a bit more thought. The rail curves in two dimensions and is tilted off vertical, plus the grain pattern would be important in this, one of the focal points of the chair. It would have required quite a chunky piece of ash to cut the shape from a square section piece. So a 45mm ash section was sandwiched between soft wood packing pieces and the corners removed to form a new rectangular section so that the ash was held diagonally in the section. This allowed the rail to be cut from the fabricated piece with the ash grain orientated to get the best effect.
IMG_7275_opt.jpg

To determine the angle and size of the piece in the sandwich I drew out the positions of the highest and lowest sections of the rail using sketchup, I could then draw in the position of the ash piece to ensure the rail was all ash with no bits of softwood included. Even after this it took me a couple of goes to get it right! Determining how the grain pattern will appear in the cut rail is “tricky”. I need to practice cutting a few more to experiment with different grain orientations.
vidar crest angle.jpg


The shape of the rail was traced onto the blank on the top surface and the front surface using two templates created from the drawing. Before cutting the rail the position of the dowel holes for locating the rail on the top of the back legs was determined using dowel points in the tops of the legs. The holes being drilled on the drill press. The rail was cut out on the bandsaw, requiring four cuts in all, two in one dimension and two in the other. The parts being taped back together each time to maintain the square section and the tracing for the cut. I hope the series of photos clarifies this process (note that the rail in the pictures is not the one I used finally).
IMG_7276_opt.jpg

IMG_7279_opt (1).jpg

IMG_7280_opt.jpg

IMG_7283_opt.jpg

IMG_7284_opt.jpg

After band sawing the shape was refined using a spoke shave, inside ground curved paring chisel and scraper. Holding such an awkward shape could have been tricky, but I was helped by the dowel holes previously drilled. Similar holes drilled in an off cut allowed me to locate the rail on the off cut held in a vice. I have a Veritas low angle spoke shave which I have never got on with I still prefer my old Stanleys
IMG_7294_opt.jpg

IMG_7295_opt.jpg


With the crest and lumbar rails completed I could move on to the back splats and the arms. More on that later.
 

Attachments

  • vidar crest angle.jpg
    vidar crest angle.jpg
    39.4 KB
  • IMG_7275_opt.jpg
    IMG_7275_opt.jpg
    32.8 KB
  • IMG_7276_opt.jpg
    IMG_7276_opt.jpg
    31.4 KB
  • IMG_7279_opt (1).jpg
    IMG_7279_opt (1).jpg
    28.8 KB
  • IMG_7280_opt.jpg
    IMG_7280_opt.jpg
    27.7 KB
  • IMG_7282_opt.jpg
    IMG_7282_opt.jpg
    45.2 KB
  • IMG_7283_opt.jpg
    IMG_7283_opt.jpg
    30.5 KB
  • IMG_7284_opt.jpg
    IMG_7284_opt.jpg
    33.4 KB
  • IMG_7294_opt.jpg
    IMG_7294_opt.jpg
    30.1 KB
  • IMG_7295_opt.jpg
    IMG_7295_opt.jpg
    32.6 KB
That's a really clever approach, packing out with softwood. Something I've not seen before.
 
Excellent, I do love it when WIP pictures are shown as there are always tips that can be picked when doing your own projects, not that I could aspire to a project like yours.

Alan.
 
The back splats
The profile of the back splats is obviously important for the comfort of the chair. I had approximated a curve with the prototype but needed to finalise it. The chair was assembled then a steel rule was flexed into position to establish the curve, I had decided the curve needed to be tighter at the bottom than the top. The curve was then transferred to a piece of thin MDF, cut out and used to create former members on the router table.
The 2mm ash laminates were cut on the band saw. The blank, thick enough for ten laminates 100mm wide was planed up, the first laminate cut, then the face was replaned before cutting the next laminate and so on. They were then thicknessed to 2mm, running them through the P/T on an MDF board. This was only possible because the grain was fairly straight and clean, otherwise most of the pieces would have ended up in the extractor! The splats were composed of five pieces each so the pack of ten laminates was opened up to give two sets of five book matched. They were then bent over the former in a bag press using cascamite adhesive, this could have been done with male and female former and G clamps.

The shape of the splats would be very important to the aesthetics of the chair so I spent a lot of time with pieces of card deciding which shape was best. The shape was transferred to the splats, cut on the band saw then cleaned up with a spoke shave.
IMG_7303_opt (1).jpg

IMG_7305_opt (1).jpg

IMG_7306_opt.jpg

IMG_7318_opt.jpg

IMG_7319_opt.jpg

IMG_7321_opt.jpg


The splats fit into mortices in 10mm mortices in the crest and lumbar rails. A jig was used to hold the curved rails in the correct position for routing. The jig was designer to hold the rails at the correct angle (the mortices are at right angles to the rail surface) and also provide a running surface for the fence a long 10mm pocket hole cutter was used.

IMG_7308_opt.jpg

IMG_7326_opt.jpg

IMG_7328_opt.jpg


Next is assembly and fitting the arms.
 

Attachments

  • IMG_7303_opt (1).jpg
    IMG_7303_opt (1).jpg
    37.6 KB
  • IMG_7305_opt (1).jpg
    IMG_7305_opt (1).jpg
    29.2 KB
  • IMG_7306_opt.jpg
    IMG_7306_opt.jpg
    31.3 KB
  • IMG_7318_opt.jpg
    IMG_7318_opt.jpg
    29.4 KB
  • IMG_7319_opt.jpg
    IMG_7319_opt.jpg
    60.7 KB
  • IMG_7321_opt.jpg
    IMG_7321_opt.jpg
    34.4 KB
  • IMG_7308_opt.jpg
    IMG_7308_opt.jpg
    48.5 KB
  • IMG_7326_opt.jpg
    IMG_7326_opt.jpg
    44.1 KB
  • IMG_7328_opt.jpg
    IMG_7328_opt.jpg
    35.2 KB
Assembly and glue up.

The seat rails were given a gentle curve to the outside and the corners rounded ready for the cord seating. They receive a strong twisting force from the Danish cord seat, to counter this I drilled for 8mm dowels on the inside of the tenon using a dowel point to aid location of the second hole. The inside edge was also drilled for the lines of “L” shaped nails used for the Danish cord. I will talk about the spacing and positioning of the nails when I post about weaving the seat.
IMG_7333_opt.jpg


After a final clean up and sand the chair could be assembled. This was done in stages. First the front legs and rails, the back legs and rails without the crest rail and splats, then the two assemblies together with the side rails, finally the crest rail and splats. In the earlier stages the unglued joints were kept assemble dry to ensure correct alignment of the parts. I was careful when gluing up the splats and crest rail as over clamping could distort the assembly the pictures show how this was dealt with.

IMG_7338_opt.jpg

IMG_7340_opt.jpg

IMG_7344_opt.jpg

IMG_7347_opt.jpg


Fitting the arms.
The mortice for the arm had already been cut in the back leg when the other mortices were cut, I made it 20mm wide to take a domino. The end of the arm has to be profiled to fit the shape of the back, this is easier with a loose tenon so I used the domino jointer to cut the slot in the end of the arm. The arm is angled in two dimensions but the tongue (domino) is horizontal, so I cut the angle of the arms splay outwards first then cut the slot with the domino jointer and finally cut the angle for the back curve. This gave a rough fit which then had to be refined to the exact profile of the back. This was tricky! The back is curved in both directions. I mainly used a curved paring chisel for this. To help with fitting I applied chalk to the back and offered up the arm then removed the chalked areas on the arm end. When I make the chair again or offer it as a course I would simplify this process by having the back flat across it’s width in this area, the only shaping needed then would be for the vertical curve of the back.

Once the ends were fitted the 10mm dowel hole for fitting to the front leg was drilled. This was located using a dowel point. Finally the shape of the arm was formed using band saw and spoke shave and the arm glued up.

IMG_7288_opt.jpg

IMG_7289_opt (1).jpg

Soap finish.
This is the bit that Mr Mosian has been looking forward to!

I decided to use a soap finish as it is a traditional Scandinavian method in keeping with the design. It’s the first time I have used it and I must admit I am quite impressed with it.

As the finish is water based I dampened the surface and cut back the raised grain with 180 grit.

It is important to use pure soap flakes with no additives, I used Boots soap flakes ordered online. These were mixed in equal parts by volumewith boiling water to dissolve into a translucent liquid. This was rag applied. The soap solidifies as it cools on the surface so it’s a bit of a balancing act between too little and too much. I tried to keep the soap liquid in the jar by standing it in a tin of hot water. When the soap dried it dried white in areas where it is spread thick, but this came away when the finish was buffed with a dry rag. I did two applications but I think one would have been OK.

The finish left is interesting. There is little colour change and little sense of coating but it feels very smooth and friction free quite pleasant. It remains to be seen how resilient it is, but I would guess that it could be replenished very easily.

The last task was weaving the Danish cord seat, that will be in the next post.

IMG_7355_opt.jpg

IMG_7357_opt.jpg
 

Attachments

  • IMG_7333_opt.jpg
    IMG_7333_opt.jpg
    23.9 KB
  • IMG_7338_opt.jpg
    IMG_7338_opt.jpg
    70.9 KB
  • IMG_7340_opt.jpg
    IMG_7340_opt.jpg
    74.7 KB
  • IMG_7344_opt.jpg
    IMG_7344_opt.jpg
    41.4 KB
  • IMG_7347_opt.jpg
    IMG_7347_opt.jpg
    40.4 KB
  • IMG_7288_opt.jpg
    IMG_7288_opt.jpg
    24.6 KB
  • IMG_7289_opt (1).jpg
    IMG_7289_opt (1).jpg
    27.6 KB
  • IMG_7355_opt.jpg
    IMG_7355_opt.jpg
    44.4 KB
  • IMG_7357_opt.jpg
    IMG_7357_opt.jpg
    30.1 KB
Mr T":3mnf04t3 said:
This is the bit that Mr Mosian has been looking forward to!

I decided to use a soap finish as it is a traditional Scandinavian method in keeping with the design. It’s the first time I have used it and I must admit I am quite impressed with it.

As the finish is water based I dampened the surface and cut back the raised grain with 180 grit.

It is important to use pure soap flakes with no additives, I used Boots soap flakes ordered online. These were mixed in equal parts by volumewith boiling water to dissolve into a translucent liquid. This was rag applied. The soap solidifies as it cools on the surface so it’s a bit of a balancing act between too little and too much. I tried to keep the soap liquid in the jar by standing it in a tin of hot water. When the soap dried it dried white in areas where it is spread thick, but this came away when the finish was buffed with a dry rag. I did two applications but I think one would have been OK.

The finish left is interesting. There is little colour change and little sense of coating but it feels very smooth and friction free quite pleasant. It remains to be seen how resilient it is, but I would guess that it could be replenished very easily.


You are absolutely right Mr T. I thank you for the good explanation of the process. In fact the whole WIP is well written and a delight to follow.
I guess that the soap finish is best used on a tightish grained wood, as opposed to something like Oak. Have you any thoughts on that?
I am already looking forward to the seating installment. That Danish cord work looks like something that could be carried out in the home workshop without too much trouble.

xy mosian
 
Danish cord seating.

That Danish cord work looks like something that could be carried out in the home workshop without too much trouble.

You are quite right Mr Mosian. I have not done much seat weaving but know that with material such as seagrass and tape etc it can be a right pain having to drag all the material through for each weave. This system overcomes this, making it a lot easier.

Danish cord is made from twisted and plaited brown paper, making a surprisingly strong and hard wearing cord. The system I adopted used “L” shaped nails hammered into pre drilled holes on the inside edge of the seat rail.

This was the first time I have done this type of seat so I had to do a bit of research on the web. There is a bit about how to reseat a chair but not much about doing a new one, so I could not find anything on the spacing of the “L” shaped nails. Eventually calculated the spacing based on the thickness of the cord and the number of turns. This came out as 7mm staggered between two rows and 19 and 14mm on the front and back rails respectively, the spacing is different from front to back because the rails are different lengths but have the same number of strands. The nails were driven in to give about 6mm protruding (I might make this 8mm next time).

The weaving is started by winding the warp from front to back. The start of the cord is wound above and below adjacent nails on the front rail then a loop pulled up and over the rail to the back rail. This is hooked round the first hook on the back rail and tensioned by pulling the strand from the front and hooking it round the next hook at the front and so on. The warp strands should be as tight as possible. The warp is finished by winding the cord above and below adjacent hooks on the side rail.

The gaps between the warp strands are filled in by winding cord round the front and back rails, starting from the middle outwards.

The weft is started at the back left hand corner by winding the cord over and under adjacent hooks. A loop of cord is then woven above and below the pairs of weft cords and over the opposite side rail and hooked round the first hook, the cord being tensioned by pulling on the end of the loop from the other end. The weft should not be tensioned as much as the warp otherwise it is difficult to push the stands tightly together. As the weft progresses it should be checked occasionally for straightness. A blunt screw drive is good for moving the strands for tightness and straightness.

The final loops at the front can be a little tricky as there is not much room to thread them, again the blunt screw driver is useful for moving strands around to make room. The weaving is finished by tacking the cut end to the seat rail, this is the only time tacks are used. After a little tidying up of the strand spacing the “L” shaped are tapped home to tighten their hold.
I finished the cord with two coats of bleached shellac applied with a polishing mop to counter dirt and staining.

It has not been possible to explain the weaving process fully. For a little more detail please check out my blog on this http://www.christribefurniturecourses.com/index.php/vm-chair-project-vi/

It was now time to try out the chair. Firstly it is very light, but also very firm, the back provides good support, especially in the lumbar area.

Making this chair allowed me to think through many of the constructional problems involved in a complex design. I now have a full set of jigs and formers which future students can copy sothey can repeat the chair they make on the course in their own workshop.




IMG_7358_opt.jpg

IMG_7360_opt.jpg

IMG_7363_opt.jpg

IMG_7366_opt.jpg

IMG_7368_opt.jpg

IMG_7369_opt.jpg

IMG_7372_opt.jpg

IMG_7373_opt.jpg

IMG_7371_opt.jpg

IMG_7374_opt.jpg
 

Attachments

  • IMG_7358_opt.jpg
    IMG_7358_opt.jpg
    32.8 KB
  • IMG_7360_opt.jpg
    IMG_7360_opt.jpg
    29.7 KB
  • IMG_7363_opt.jpg
    IMG_7363_opt.jpg
    30.2 KB
  • IMG_7366_opt.jpg
    IMG_7366_opt.jpg
    41 KB
  • IMG_7368_opt.jpg
    IMG_7368_opt.jpg
    44.3 KB
  • IMG_7369_opt.jpg
    IMG_7369_opt.jpg
    44.8 KB
  • IMG_7373_opt.jpg
    IMG_7373_opt.jpg
    45.4 KB
  • IMG_7372_opt.jpg
    IMG_7372_opt.jpg
    42.1 KB
  • IMG_7371_opt.jpg
    IMG_7371_opt.jpg
    72.8 KB
  • IMG_7374_opt.jpg
    IMG_7374_opt.jpg
    50.8 KB
I guess that the soap finish is best used on a tightish grained wood, as opposed to something like Oak.

Actually I think it may work better on open grain. Ash is quite open grained and it worked a treat on that and much Danish furniture is in Oak with a soap finish. The open grain would pick up the soap well.

Chris
 
Back
Top