Sunday, July 19, 2015

Treasure chest with curved lid part 8, assembly

After I had glued up the lid, I cut the ends of so they were level with the sides. Then I started rounding over the outside of the lid using a plane.

This was when I realized yet another drawback of this design. Work holding for such a shape is difficult to say the least. It reminded me of trying to hold a fresh seed from a water melon. Not the most efficient shape if you want to retain full control over the situation.

I took a break from the free hand planing, and decided to glue up the lower part of the chest. This went conspicuously smooth. Even the diagonal measurements were spot on.

More planing and I was satisfied with the shape of the lid. I sanded the surface with some grit 60 to make it a bit more smooth.

I planed the lower part of the chest  to remove any protruding pins or tails. My solution with the floating bottom that has a lip made it impossible for me to plane the lowest part of the chest. So I had to use a chisel, and later I will follow up with some sandpaper.

The lock was mounted after making a mortise. I had to bend the upper plate just a bit, since the upper edge of the front board is not square to the front, another disadvantage of making slanted sides.

I  more or less try to follow the advice given in "The joiner and cabinet maker" from Lost Art Press, when it comes to mounting a lock. The most important thing is to keep everything centered around the pin for the key. Since all locks I have ever been able to find are full mortise locks, my biggest challenge is to not break through the sides of the board. If I succeed in making the mortise, the rest usually goes pretty smooth.

Next task was to fit a striking plate in the lid. I marked out the position from the lock, and I even remembered to make the mortise below the striking plate wider that the square hole itself so the lock would actually work.

The hinges were mounted their own width from the ends, which is pretty common.
I started mounting them on the lid, because I could use the lower case for holding the lid while I performed the job. This was for once in this project a nice work holding solution.

Since the lid did indeed end up being 1/4" too narrow, I decided to mount the hinges so this would be visible on the back of the chest. That way the front will look as nice as possible. I thought about removing about 1/8 from the front and the back, but discarded that idea, because it would make the lock visible from the front. Further more I was afraid that it would lead to that the mortise for the striking plate could penetrate the front of the lid effectively ruining the project.

I still need to make some feet and an escutcheon for the chest, and I am considering making some lifts for it as well.

View from the front.

Work holding of the lid.

An escutcheon will add to the look.

Interior of the chest.

Tuesday, July 14, 2015

Treasure chest with curved lid part 7, the lid

After the glue had dried I decided that it didn't look right if I'd led the inside stay as it was with flat boards forming the surface.
I adjusted the scrub iron in the plane, so it was able to take a bit of a cut even when only the sides of the plane were riding on high spots. It was far from ideal, but I managed to get the inside to form some sort of a curve instead.
There were a few places I wasn't able to reach, so I took the iron out of the plane and used it as a scraper. Again it is not an efficient trick that will make your friends gaze in awe, but it helped me git a bit closer to a decent shape.

The surface looked like crap after my exploits with the scrub plane and the scraping experiments, but I found some 60 grit sand paper in the deck workshop, and then I just started sanding away.

After quite some time I was satisfied with the way things were looking, and I was ready to move on to attaching the ends.

Originally my plan was to make the ends straight down, but for some reason I decided that it would probably be too easy, and that I should challenge myself a bit more.
Therefore I decided to make slanted ends.

I immediately sawed the ends of the lid so they were slanting. Then I "squared" things up the best I could using the smoothing iron in the plane.

I am getting used to making angled dovetail joints on flat boards, so I guessed that this wouldn't be
much different.
Guess I was wrong.
It really is a lot more difficult to make dovetails that will connect a semicircular flat board at an angle to a less than perfect half cylinder shaped object.

At first I marked out the tails on the end boards and cut them, I even made sure to take into account that the cut had to be deeper on one side than the other. It looked fine until I tried to put them on the lid itself for marking out the pins.
Even I could see that there were no way I would ever be able to make some pins that would allow that tail board to go on. The tails had to have different angles depending on if they were placed on the side part of the ends or on the top part of the ends. I realized that this slanted end challenge was going to be a lot more difficult than I had imagined.

I spent some time trying to figure out how I was supposed to mark out the dovetails, and ended up making the pins first.
I could use a compass to step them of, but since the reference side was curved I didn't have a system for marking out the angle of the pin. My solution was to cut a small dovetail marker that looked like a triangle, and then I placed it so it looked fair to me. At least I knew that the slopes would be the same.

Marking the depth of the tail was done using my marking gauge. Next came the problem concerning how to mark out the direction of what would be the sides of the pins. Normally I use a small square in form of a piece of cardboard. But a square is best if it has got some sort of reference edge that you can measure from.
I tried various ideas before settling on simply eye balling the lines. For that I used the back of the panel saw that we have on board. I positioned it at the pin markings and tried to make it follow the length of the lid as well as I could.

Once all the markings had been taken care of, I sawed and chiseled out the waste between the pins.
Normally I like to undercut my waste area, but I figured that it would end up showing when I get to the point where I can plane the outside of the lid to the final shape.
So the waste area was finally cleaned up using a coarse file (we haven't got any rasps out here).

I made a couple of new end boards instead of the ones with the faulty tails on them, and after a lot of fiddling I managed to mark out the tails and cut them too.
This involved some guessing regarding where to stop the cut on one side, and then just hoping that it was the correct place when I placed the chisel to remove the waste.

I dry tested the end boards, and they seemed nice and tight. When I tried to gently remove them to be able to add some glue, both top pieces broke off.
I simply glued the pieces back on and then trimmed the pins for those areas a bit more.

Finally I glued on the ends. And I pretty much immediately regretted not undercutting between the pins because there were gaps on the outside of the chest. I am afraid that these gaps will be more visible than any undercutting I could have dished up with.

The end with faulty dovetails.

Marking out for the slanted cut.

Sawing, the line from the bucket wasn't completely straight after all.
The glued up lid.

Thursday, July 9, 2015

Treasure chest with curved lid part 6, the lid

The lid is going to be almost semicircular from the front going to the back. Length wise it will be straight.

I started by drawing an arc that would correspond to my plans for the lid. 
On a board I marked a distance of 11" near one edge and a center between those. I then placed one leg of the compass something like 1.5" below the center and adjusted the other leg to touch the points 11" from each other.
Then I scribed the piece of a circle using that setting.

I measured the curved distance ( I know you can calculate it as well, but I don't remember how to since it is only a fraction of a circle).
I got the distance to be 16" (41 cm) more or less, so I made 6 boards with a width of 2.75" (7 cm). Each board was intentionally cut about 4" too long.
Using my scrub iron in the plane, I planed a bevel on each side of all the boards. I just eyeballed the angle.
The boards for the sides of the lid each got a bevel that was twice the angle of the inner ones. In hindsight it would have been smarter to wait with these two bevels, but I know that for the next time I am going to make such a lid.

I flattened what was going to be the inside of each board, and then I laid out all of the boards and numbered them. I had added an arrow showing the direction of the grain, so I didn't make it more difficult for myself than I had to.

Each pair of bevelled edges were then jointed together. Pretty much like how you do with any glue joint. The only difference is that here I had to turn one board the opposite direction, to get the bevels to match up. A few swipes with the smoothing iron in the plane, and the seam seemed tight enough.

When all the joints were completed, I again lined the pieces up. I added some blue masking tape to hold the pieces, and then carefully folded the lid. I placed the assembly on the scribed arc, to see if it looked fine.
It turned out that I had been a little too aggressive on the bevels, so I was about an inch short.
I made a seventh stick about an inch wide to rectify the width of the assembly.
This stick was made with out any bevels (at least that is what I tried to, but I didn't check it with a square).

The now seven pieces were lined up, taped and flipped over.
Now the bevels were open so I could add some hide glue to both sides of each joint. I used a small brush to get a nice layer of glue all over the surface.

Once all the glue was applied, I gently lifted the two sides of the lid and when the shape was obtained, I put the assembly upright on the work table.

I adjusted the individual boards so the inside edges of the joints lined up the best I could. I then added some more masking tape to help close the inside seams by putting some tension on the opening of the lid.

After the glue had started to set, I carried the piece out to a nice and warm spot so the glue can dry.
I then measured the width of the assembly and found that I am still about 1/2" short..

It doesn't matter much, because one of the joints of the centre stick is not very tight on the outside.
The plan for tomorrow is to remove the centre stick, and glue in another one which is maybe 3/4" wider. That would give me a bit of surplus wood that I can plane away to make the lid fit nicely.

The scribed arc that represents the ideal shape of the lid.

Initial testing of the bevels.

Flattened on the inside

Jointing the bevels

Ready for the first test.

Checking the shape of the assembly.

About an inch too narrow.

Now there are seven pieces in the lid.

Glued up, view from the inside.

Glued up, view from the outside.

Wednesday, July 8, 2015

Treasure chest with curved lid part 5, completing the bottom.

Today I managed to make the groove for the interlocking of the bottom. Actually it was a bit easier than I anticipated. Easy workability and a nice smell are definitely positive features of spruce.

I could ride the body of the grooving plane on the outer edge of the bottom, and then adjust the iron so it was protruding something like 5/16" (8mm). The finish of the groove isn't superb, but it doesn't have to be.

After making the groove I dry assembled my parts: Sides, ends and the bottom.
Inside it looks just like I had envisioned. But on the outside there is a bit of a gap. I still haven't made up my mind if I should add a small piece of wood to act as a moulding, or if I should just leave it the way it is.

My problem with adding some moulding is that it will disturb my design idea of the rounded outer edge of the bottom to look like a piece of moulding in itself.

I didn't plan for movement on purpose, but the bottom is able to move just a bit in the groove which is good news.
Wood movement is mostly limited to expansion after being worked out here, so being a little bit loose is not a problem.

My plan was to continue on the lid, but I had to make a slight alteration in those plans, as our refrigeration compressor started acting up.
It has been giving us problems for the last couple of years, and we haven't really been able to pinpoint the fault.

After fiddling with the compressor, I just managed to saw some stock for the lid, before it was time to make an evening round and stop for today.

Test assembly showing the small gap.

Close up of the interlocking parts. 

Inside view.

Monday, July 6, 2015

Treasure chest with curved lid part 4, working on the bottom and finding a couple of design flaws

The glue had dried and it did indeed still work even though I passed the "best before" date written on the container.

The panel was not flat, but I guess it comes pretty natural after glueing up to cupped boards. At first I traversed it with the scrub plane, and then when it was flattish I started going with the grain.
After a bit of workout, it was nice and flat.

I dry assembled the carcase of the chest, and after checking that it was square, I traced the outline of the bottom onto the bottom panel.

According to my sketch it will look good if the bottom is protruding on all sides, just something like 1/4". This could technically be achieved by adding a small piece of moulding, but I didn't bring any moulding planes with me, so I had to think of another way to make this small detail.

The plan was to make a groove near the bottom of the carcase, to house the bottom, That is pretty traditional.
In order for the lower part of the bottom to protrude I had to make a rabbet along all the sides of the bottom.
Then on the remaining full thickness part of the bottom I had to make a groove, so I would end up with a thin piece of bottom that would fit into the grooves of the carcase sides.
In my mind this worked very well.

Design flaw No 1
I normally stick to the grooving before dovetailing, but I didn't do it this time because my original idea was to simply make a rabbet and then nail the bottom on.
Due to my interlocking groove idea, I had to make a set of grooves without messing up the dovetails, so I wouldn't get a groove that would be visible from the outside of the chest.
On the sides it worked really well, since I just cleared the tails.
On the end boards I found myself needing to make a stopped groove - as in stopped in both ends. I tried using my grooving plane, but it pretty quickly became evident, that this tool was not intended for making that kind of grooves.
My solution was to use a utility knife for defining the sides of the groove. I adjusted the iron of the grooving plane so it protruded a little less than 1/4" (5 mm to be exact). That made it possible for me to use the grooving plane as sort of a router plane.
It took a bit more time than I had expected, but the grooves turned out well.

Design flaw No 2
If you need to make a groove on the side of your rabbet, be sure not to make the rabbet wider than your grooving plane can handle..
This flaw was discovered after I had finished planing the almost 3/4" wide rabbet all the way around the bottom.
I think that I might be able to tackle it sort of the same way I did with the stopped grooves. But I decided to call it a day and not risking to mess up things by keeping on working too late.

After imposing those additional challenges on myself, I would like to explain why I changed the design in the first place:
I wanted to make sure that there would be no crack between the sides and the bottom if it should move with the seasons. Therefore I got the idea of housing the bottom in a groove.
And making one bottom, and then glueing on a decorative bottom seemed like an idea that required too much work. Guess I was wrong about that.

Flattening the bottom.

A stopped groove.
Note the protrusion of the iron on the plane.

The bottom with rabbets.

Sunday, July 5, 2015

Treasure chest with curved lid part 3 dovetails and glueing up the bottom.

Today I chopped the dovetails of the four corners of the chest. A benefit of using spruce is that you can make the dovetails really tight since the wood compresses very well.
This time I made the tails first like at home, and I do find that I am able to make nicer dovetails that way compared to when I make them pins first.

Last year during the chairbuilding extravagnza, Brian Eve brought some liquid hide glue for us to use.
It stayed at my place, and I found out that it actually passed its "best before date" a little while ago.
For some reason, I have never tried using liquid hide glue for dovetails. Out here I am always stressed during glue ups because my normal white glue will dry very quickly because of the temperatures in the workshop especially during the summer months.

While at home, I decided to bring the liquid hide glue with me on board this time, so I could use it before it gets way too old.

My plan is to try using the liquid hide glue as the only glue on this project. I don't know why I feel all excited and insecure about that, since it is a pretty time tested glue type. The only thing is that it could be too old, but I kind of doubt that the "best before date" means that the glue will not stick to anything as soon as you pass it.

I found some more spruce that could be glued up to form a bottom.
First I jointed the mating sides, and then I discovered that there was a crack in the wide board.
I opened the crack by bending the board a bit, and squeezed some glue into it. I then applied hide glue to both surfaces and pressed the joint together with a couple of clamps.

Adding glue to both sides of a joint is a habit of mine from working with white glue. I have no idea if it is required or even encouraged when using hide glue, but I figured that it wouldn't hurt.

The joint went together as it should the first time, so I didn't have any reason to use the slow setting time to shift the joint around anyway.

After some time the glue had already dried, but I have still left the clamps on, because I won't be working on the bottom until tomorrow anyway.

Dry testing the dovetails.

Saturday, July 4, 2015

Treasure chest with curved lid part 2

Yesterday I found a set of pallet sides for half a Euro pallet.
It was a set that I had saved from some time ago, when we received some spare parts. the wood has been stored in the engine room, so it is bone dry.
Not surprisingly the quality is not furniture grade, but that has never stopped me from attempting to make something out of it anyway.

The wood is flat sawn spruce, so it is no wonder that it has cupped quite a bit. The two sides of the chest had cupped the least, so I managed to flatten them with out sacrificing too much of the thickness. The end boards were a different story. They ended up being roughly 1/8" thinner than the sides, but as long as they are uniform in thickness, I'll probably be able to make it look all right somehow.

After sketching a square sided chest with a curved look, I decided that it looked a bit dull. Next I tried to draw a canted chest with a smaller footprint than top, and that looked better.
When I added an exterior bottom and some feet it actually looked quite fine. So that is going to be my main plan for the time being.

I decided that the size of the chest on the upper part should be 16 x 11" (40 x 28 cm).
My idea is that the canted sides should not have too bold an angle, as this might make the chest a bit unstable. So I have just shortened the lower part of the sides and ends with 3/4" (2 cm) in each end.
It will give each end a ratio of approximately 10:1.
By looking at my sketch, it will look pretty good if I make the exterior bottom a bit wider than the bottom of the carcase, and the feet a bit wider yet. So the feet might actually end up being directly under the upper corners of the chest.

With the sides and ends flattened and jointed on the long edges, I sawed them all to the desired length.
I wanted to make use of my moving fillister plane, so contrary to my normal procedure at sea, I am going to make dovetails with the tails first.
So I planed a shallow rabbet for the back side of the tails. Next I used a set of dividers to lay out the tails and a small piece of cardboard to mark out the sides (just as usual).

I sawed the tails and decided that I didn't want to push my luck by starting to chisel out the waste tonight.
The parts were quickly taped together to give me an idea of the size and shape of the project.
It looked a bit too deep, but the depth will be reduced once I make the dovetails.

The lower part of the treasure chest.

Friday, July 3, 2015

Treasure chest with curved lid

I actually managed to make a functioning porch before going back to sea, although I didn't finish it completely.
But there is enough porch so the family at home can use it until I come home and can work on it again.
As usual, I was too busy building the thing to remember blogging about it and documenting it with some pictures, but I plan on taking some pictures once I am home again.

This time I have considered making a spice chest or a Milkmans' workbench.
The spice chest will probably just end up half way completed like the Gerstner inspired tool chest that is still waiting for me to complete it.
The workbench would require me to first make a tap and die set to be able to make wooden threads for the work holding.

The electrician on board was so kind as to bring a couple of chest locks with him from Buenos Aires in Argentina.
So my mind immediately started going down the slippery path of making a new chest of some sort.
I have made sea chests and tool chests before, but I have never made a treasure chest with a curved lid - So I guess that is going to be the project this time.

The first step will be to find some stock and start preparing it.

An Argentinian chest lock.