Exampe of how a small thing can send a design in a different direction

[krtwood]

A jewelry box design that I have reused several times is to have a rectangular box suspended between two vertical natural edge slabs.  Add an interesting board for the drawer fronts and you end up with something pretty cool but with pretty simple techniques to build it.  I like to be able to offer some products that have some wow factor but are still reasonably affordable. Here's a couple examples from the past  https://www.etsy.com/transaction/23190764  https://www.etsy.com/transaction/38370782  

 

I started out to make a couple more of them and I'll start with the slabs for the sides which will define the height and depth of the center box and then board for the drawer fronts whcih will define the width and work from there.  Here is one of the boards I picked for drawer fronts.  Let's pretend it hasn't already been sliced and diced for the moment.

 

 

Two knots with some figure around them.  It's a pretty funky board that I think most people would end up tossing because what the heck can you do with it?  Again it's something that keeps the cost down because I'm not just throwing expensive highly figured boards at the project.  I'm taking something that's been cut out of a normal board and seeing what I can make out of it.  But pay special notice to the little bit of wane in the bottom right corner.  That is the small thing that has up-ended my plans, but more on that in a bit.

 

 

If I do a straight bookmatch on this we get something pretty cool, but it looks off balance with the two small knots on the same side.  This would maybe work if it was turned 90 degrees, but it doesn't work as drawer fronts to my eye.

 

 

Flipping one set over gets me what I want.  It's got balance and it's dynamic with the grain crossing across the middle the way that it does.  So great let's get started making that simple box.  But wait a second, remember that little bit of wane?  If I cut the boards off square short enough to get rid of it then I am going to lose a lot of the figure that is next to the large knots.  I don't want to do that.  If I had ripped the board down I'd be intruding into the knot and I don't want that.  But I can't leave it like this either.  The only other alternative I can think of is to make the sides of the box not be straight.  They would have to curve inwards in the middle, making sort of an hourglass shape.  When you are making something for the purpose of selling it you can't make this decision based solely on what you want to do.  You have to think about what this is going to end up doing to the price and whether you think you're going to end up with something that is actually going to sell.  A change that is going to add a lot labor and brings up the cost now takes the project into territory where the expectations are higher.  Now maybe some other choices in the design that were also to keep the cost down don't make sense anymore.  But if I was just going to chop the suckers off I wouldn't be making this post about it, now would I?  So here is the template I came up with for the sides.

 

 

What would have been a simple glue up (maybe not even a glue up since it's only 8" wide) from 4/4 stock with some dados for the drawer guides and then cut the dovetails now has to be made from 8/4 stock and then rough cut out on the band saw.  The off cuts cupped immediately so I knew I was in trouble with the workpiece cupping.  Which it did and I let it sit for a few days to make sure it wasn't going to keep moving.  I did plan for this and left myself an 1/8th to play with.   Fortunately I have a cnc so it wasn't that big of deal to flatten the outside, flip it over and do the inside and also take care of the dados for the drawer guides.  Normally I would just do through dados on the table saw if it were a flat panel, but with the cnc it was just as easy to make them stopped which is one of those design choices that I have to make because of the higher expectations of a higher price.  Now the drawer guides are going to have to get a rabbet, because I am not squaring up 16 dados.  It's not like making the rabbets is a big deal, but it's just an example of the ripple effect. 

 

 

A bigger deal is the rabbet on the back of the sides for the back panel.  I want this to follow the outside of the sides but since the inside doesn't follow the curve I can't just do this with a rabbeting bit.  I was thinking I would have to make a template for the router to follow, but I came up with a different approach.  The curvature of the table saw blade was pretty darn close to the shape I was looking for so I did it with stopped cuts on the TS.  It probably ended up taking just as long as if I'd done the template since I had to fuss around with it a fair bit, but I don't like to waste plywood on templates if I don't really need them.

 

 

Through sheer luck intense planning and foresight the straight sections of the sides were just long enough to still fit in my Akeda dovetail jig, so that part was still straightforward and I was able to get onto the glue up

 

 

And now I can work on creating a curve in the top and bottom.  I think the curve in the sides follows the grain of the drawer fronts so well that it really needs to also follow it on the top and bottom.  It didn't really make sense to try to cut this curve before assembly where I would have to rig up some way to keep the boards vertical when once they are assembled the box will just do that for me.  I just did a rough cut on the band saw and then used a template and top bearing bit to work from both the front and back.  My bit is only 1" cutting length so I still have 6" in the middle to deal with.  I might do some with hand planes if I can figure out a way to hold the box, my bench is not the ideal hand tool bench.  But I'll get it down to end on the edge sander.

 

 

That gets us caught up to where I'm at now, right as I hear thunder rolling in.

[Tpt life]

Thanks for talking through the design process. That is very enlightening for me.

[TRBaker]

Outstanding, krt. I'm looking forward to seeing the finished product.

[wdwerker]

It is looking pretty great so far. The curve in the top is going to look awesome.

Love that Spirit of the Woods box!

Nice to see a CNC used to solve practical functions instead of cutting pretty patterns.

[krtwood]

Thanks guys, got the curves on the top and bottom pretty close to where they need to be.  I did the middle 2/3s on the edge sander but had to go to the spindle sander towards the edges where the curve turns concave.  Had to come in from both sides and still left a mess in the middle so it was a bit of work with a file and then scraping (my first time with a scraper, I don't know why I waited so long to try it out).  Also got the back fitted.  There's a bit of a gap in the curves, but it's the back and I'll live with it.  I cut it REALLY close with the curve almost coming into the back panel.

 

 

I had planned to use some hickory slabs with this piece.  I really love the combination of hickory and walnut, but the way this project is going now hose slabs are nice but they aren't going to cut it.  So I checked out ebay and bought this, along with another slab from the same log.

 

http://www.ebay.com/itm/271226020566

 

I'm thinking about the rails and I think I also may go with something nicer here, maybe wenge.  I think they need to do more than just follow the vertical curve of the top and bottom of the box.  There is that curving grain in the top that I was able to get centered and at both the front and back so I think I need to follow that.  I don't want them to be too chunky though and they could get fragile if there's too much of a curve and they are thin.  So I have some thinking to do there.  If I can get it to fit within a 1.5 or 2" turning blank that makes life easier.

 

[krtwood]

Drawers made.  Six are 1.75" high and the middle two are 1.5".  On the Akeda jig the minimum spacing for two tails out of the box is 2".  With some minor modifications you can do 1.75" without any trouble.  1.5" takes some swapping in and out of guides in different places and not going all the way through with the router, then finishing it up with a chisel.  Getting the drawer fronts on is an important milestone.  I've learned to give the glue 5 minutes and then check it in the case to make sure the gaps are what they should be.  More than once I've returned to the shop the next day to find one of the gaps is huge and I'm up the creek.  With it all together is the next best "So this is what it's going to look like" moment to the first coat of finish.  The drawer fronts are just rough cut at this point.  I am tempted to break out the flush trim bit but the little voice in the back of my head has some concerns. 

 

 

My next task was to figure out what to do with the rails.  I was doing this in the CAD software that came with my cnc (Vcarve Pro) which is not 3D software so it was a little hard to visualize what it was going to look like from front and top views.  I wasn't sure about my initial design that curved upwards so I also did a flatter design.  I almost never make prototypes but given I also had some questions about making it so out came the poplar.

 

 

I'm leaning heavily towards my initial design.  I was also on the fence about whether to go light or dark and now it's definitely going to be dark.