The interior of the cabinet is divided into three sections. The bottom of the interior has a curved front drawer for storage. The middle of the interior has a shelf, which is the top of the drawer box, for bottles of your most precious liquids. The top of the interior has another shelf for storing six old fashioned glasses. The top shelf has a false back which keeps the glasses arranged neatly at the front of the cabinet for easy access, and to avoid having a deep dark cavern at the top of the cabinet when it is empty.
Making the cores for the interior partitions is pretty much the same as the process I used while making the doors and top/bottom panels back in the early weeks of the project. Most of the edgebands are made of poplar and are hidden from view. The curved drawer front is a laminate of birch veneers. Below is a combined plan view of the interior sections of the cabinet. The left side of the drawing shows the curved drawer front, the curved front of the drawer box, and the straight interior wall of the drawer box. The right side of the drawing shows the curved wall that provides the false back to the old fashioned glass display shelf. The top of the drawing shows the edgeband of the flat panel shelves that is bricked together from pieces of poplar.
Next week I hope to get the veneers applied to all the interior core pieces. The show is rapidly approaching and I am beginning to feel a slight sense of panic.
Early in the project one of the instructors was looking at the top surface of the mock-up and said it might look pretty cool with a little spoiler on the back edge. "What is a spoiler?", I asked and I got a response similar to the following.
A spoiler is a low lip that is sometimes put on the back edge. It kinda swoops up and blends in with the rest of the edge banding. The trick is to get it installed just a little bit proud and then use a custom scraper to bring it down to perfectly meet the top. You don't want to blow it and be too low or gouge the surface with the scraper. It's the top of the cabinet and everyone is looking at it.
I have never put a spoiler on a cabinet and a Google search for "cabinet spoiler" or any other combination just resulted in links to movie trailers. But the description seemed enough to go on and why not add a thing that kinda swoops to the top of the cabinet that could ruin it. Sounds like a great thing to try.
Last week I worked on getting the larger edgeband that wraps the front edge of the top and bottom attached. This week I will be working on the other part of the edgebands at the rear of the cabinet. The top and bottom of the cabinet both have separate edgebands that are attached to the rear of the cabinet. The top has the spoiler edgeband that swoops up from the surface and the bottom has a flush edgeband.
The first step is to glue-up the parts in a bending form. I decided to use the vacuum press again. I was worried about the press being able to squeeze the straight laminate strips around the inside and outside of the curved form so I pre-bent the strips on a hot pipe. Hot pipe bending is pretty much like it sounds. You take a propane torch and blow it through a steel pipe and gently heat the wood to soften the lignum and slowly bend the wood around the pipe. You can see in the photo below that the wood strips are pre-bent before they go into the vacuum press.
The glue-up for the spoiler edgeband went great. The glue-up for the flush edgeband on the bottom of the cabinet did not go great. The result was slightly twisted and I didn't want to straighten it by shaving off wood for fear of exposing a glue line on the final piece. So I took a different approach the second time and bent it using a male/female bending form and clamps.
The next step in the process is to shape the spoiler before gluing it onto the top panel. I routed a groove into the edge of the spoiler using a router bit shaped like a donut. I haven't figured out what it is officially called yet, but I found it on the router bit rack and it seemed like it would do the job. I built a point fence that would allow me to follow the profile of the curved edge band and route the groove just below the top edge. Later I used a spokeshave, carving knife, and sandpaper to give the spoiler it's near final shape.
The final step in the process is to fit and glue the edgebands onto the rear of the top and bottom panels. I previously routed a slot in both the panels and the edgebands to make sure that the edgebands would have the proper vertical alignment as I proceeded with the fitting. My goal is to make the edgeband precisely mate with the curves at the rear of the panel and the mitered edges of the edgebands that are already installed on the panels. I start off with the edgebands being too long for the opening and slowly plane the ends to try and get the angle of the miter on the installed edgeband to match the angle of the miter on the edgeband I am trying to install. It is a tricky and nervous operation to try and get the angles to match before I shave the edgeband too short and end up with a gap.
I slowly plane the ends of the edgeband, checking after each few plane strokes to make sure that miter still matches and adjusting the shooting board using blue tape and shims. The angle keeps changing as I get closer because the profile of the edgeband keeps changing as I am able to clamp the edgeband closer and closer to the profile of the panel.
After a few setbacks, a stupid mistake, and a recovery involving some woodshavings and enough clamping force to potentially start a fusion reaction in the panel I was able to get both the top and bottom edgebands fitted and glued. Not perfect, but certainly good enough.
I have mentioned edgebanding in some of my previous posts. Those were captured edgebands which are glued to the side of a plywood substrate and then veneer is applied to the top and bottom capturing the edgeband. There are two reasons that we do this at The Krenov School. First, if this is an exposed edge then a strip of solid wood is definitely finer than looking at the edge of a piece of plywood. Second, and sometimes debated, is to provide a strip of solid wood that can be more easily hand tooled to a smooth surface and seamless joint. Try making the edge of a piece of plywood perfectly smooth. Go ahead, I'll wait.
This week I've been working on the applied edgebands for the top and bottom of the cabinet. An applied edgeband is a piece of wood that is often glued to the edge of a veneered panel to add profile detail or hide the view of the raw edge which may look like a sandwich of veneers and captured edgeband (or plywood if you don't do the captured edgeband thing).
For this cabinet, the applied edgeband wraps the kidney bean shaped top with a strip of wood that swoops up into a spoiler detail at the rear of the cabinet. The applied edgeband on this cabinet provides the overhang of the top and bottom that visually captures the vertical sides of the cabinet.
The edgeband is about a 1/4 inch thick and is made up of 4 thin strips of wood that are flexible enough to be bent around the profile of the kidney bean. I divided the edgeband into two parts: the smooth concave curve of the front edge wrapping around the sides of the cabinet, and the rear edge where the curve has both concave and convex sections (i.e. a squiggle). The two parts of the edgeband are made separately and then applied to the edge of the panel.
The longer edgeband which wraps most of the panel is bent around a large form. The form made of plywood and routed to match the profile of the top/bottom panel. The inside of the plywood is removed to make a ring that will allow clamps to be used to apply even pressure to the laminate strips while it is being glued. In the picture below you can see a few long plywood pieces radiating from the ring. These pieces are used to align the bottom edge of the strips during glue up. You can also see a few extra holes drilled into the form around the tight radius corners that allows me to double up on clamps as the strips bend around that curve. After making up the thin strips of wood, I apply some Unibond 800 glue and wrap the strips around the form along with a bending caul made from cardboard and veneer and other bendy things. And then I use a few clamps to apply even pressure along the entire length of the edgeband.
After the glue cures, I remove the clamps and clean up the edgeband surfaces. I made a sample edgeband from poplar, which is shown below, to test out the form and the process. Jumping ahead a bit, you can also see the prototype spoiler that will be mated to the edgeband that will complete the applied edgeband for the panel.
I need to make sure that the ends of the edgeband have miters that are perfectly smooth and square so that the edgeband/spoiler at the rear of the panel can mate seamlessly. After sawing the end of the edgeband to the proper length and approximate angle I use my shooting board with a wedge to hold the curve at the proper angle (along with some blue tape) to plane the end of the edgeband square and smooth.
The final step is to glue the edgeband to the panel without any gaps or glue lines. This turned out to be much more difficult than I had predicted. After a day or two of trying to find a clamping scheme that produced the desired results and chasing down the tiniest imperfections that I had introduced on the inside edge of the edgeband and outside edge of the panels, I came up with a solution that seemed to work consistently. First I wrapped the edge band and panel with one of the most powerful strap clamps that I had ever seen. Second I encased the entire setup with a set of hard cauls and bar clamps. Finally, by tightening the bar clamps in the right order I could slowly move the pressure around the panel closing the gap along the way until I ended up with a seamless joint.
This week the project went from 2D to 3D when I added the joinery for the top, bottom, and rear panels that I have been working on the past 8 weeks. I am still many weeks away from gluing things together, so you will still need to use your imagination to picture what the piece will finally look like. On the bright side, when the time comes the project is going to transform from a pile of parts to a nearly completed piece with a few quick glue ups.
The top and bottom of the cabinet are joined to the rear panels with a series of dowels. The dowels provide additional strength to the joint as well as alignment of the curved panels in relation to the doors and other cabinet parts. The top and bottom panels are 3/4" and 5/8" thick respectively and the dowel holes are carefully (and with trepidation) drilled to within 1/8" of breaking through the veneer on the other side.
The key to doweling is to make sure you drill the holes for the dowels are in the right spots on the two parts that you are joining together and that the holes that you drill are pretty darn close to being straight and true. If the holes do not align then the parts will not align, and if the holes are not straight and true then you will not be able to bring the joint together. The Krenov School way of accomplishing this is to build a doweling jig out of hardwood and use it to guide a drill bit as you drill the two sides of the joint.
The doweling jig for these joints is made out of a piece of maple. I used a template to cut the jig to precisely match the profile of the rear panels which also matches a portion of the curves on the top and bottom panels. I glued a series of small strips of wood to the edge of the jig that are used to register the jig against the edge of the panels. The jig is attached to the piece being drilled with a few screws and a drill press or an electric drill (with the help of a drill block) are used to transfer the holes from the jig to the workpiece. The photographs show the jig in place on the flat panel and the resulting holes on the curved panel. The drill block is also pictured in the lower left corner of the curved panel photograph.
After the holes are drilled, a couple of the dowels are inserted and the pieces are put together to find out if the whole operation was a success or if you need to plug the holes and try again. In this case it was relatively successful!
Step A complete! I now have two panels and a top and bottom that mostly fit together but I have not yet figured out how I want to join the left and right rear panels together and what kind of detail I want running down the spine of the cabinet. Since I am in the habit of kicking most decisions down the road on this project I decided to build a mating piece with two groves in it that I can use later to stick in my desired detail (in software engineering terms, I just slipped some feature creep into the project and sold it to the project manager as an extensible feature).
The spine is held into position by a few locating dowels that run through the spine and into the rear edge of each panel. I used the spine as the doweling jig to transfer the holes between the two panels. Next challenge to consider was clamping. The only feasible option that I could come up with was to glue temporary clamping blocks on to the interior edges of the panels and use those to pinch the edges together and bring the spine joint home. I really did not want to add a bunch of glue blocks to my beautifully prepared birch veneers on the interior of the cabinet and I do not know yet how painful it will be to remove them. I am sure it will all be just fine.
It will still be a very long time before things get glued together permanently as I have come to accept that I need to build and fit the interior shelves and drawers while the carcass of the cabinet is still in pieces and I can drill holes and trace curves.
To top off a challenging but productive week, I roughed out some hinges for the doors. More about those in another blog post but here are pictures.
Now that the doors and rear panels of the cabinet are done, I need a top and bottom so that I can get to work on assembling things.The top and bottom are flat panels which are a lot easier to build than the curved panels of the doors and rear panels. I have chosen Sipo (very similar to Mahogany) as the veneer for the top and bottom panels.
The process for flat panels is generally the same as the curved panels I have been making in the vacuum press.I cut the substrates from off-the-self furniture grade plywood and shape them to match a template of the cabinet.I apply an edge band of solid wood.I cut and process a plank of Sipo into veneers and cut them to match the size of the substrates.Then the substrate gets covered with glue and everything gets sandwiched together in a mechanical press.
It still took me the entire week to make the top and panels because I’ve been working on the volleyball court, a memorial bench for Matt Mecaro (class of 2016/2017), and have a quick trip back to Ohio to attend to some family matters.
For years the woodworkers have looked up from their workbenches at the end of a long day and shuffled their hunched bodies outside in response to the daily plea to form two volleyball teams that will engage in a few rounds of “prison rules” volleyball.For years the woodworkers have lamented playing in the the dirt and the dust and the weeds.For years the woodworkers have talked about getting sand for the volleyball court.Perhaps things are just fine the way they are and there is no need to change these traditions, perhaps these are not traditions at all but just a lack of action.
This year I decided I would help organize a fundraiser and we would see if we could raise enough money to buy some sand and improve the volleyball court.We set a goal of $1000 in order to get 26 tons of beach sand delivered (2-3 inches).We set up a GoFundMe and Todd emailed the alumni with a call to action.We exceeded our goal and raised over $1700!No turning back now.Perhaps this is an improvement to the volleyball pastime or a detriment to it.Only time will be able to judge the actions of the class of 2018.
Making the rear panels for the cabinet is not much different than making the door panels.I am a little bit wiser now, that is about the only difference.This time around I made some extra long cauls that I hoped would put additional pressure on the edges of the rear panels to make sure that the veneers were pressed firmly against the substrates.These seemed to work out well and by the end of the week I had two rear panels to match my two doors. Pictures are worth a thousand words.
Sycamore veneer waiting to be applied
Parts of the rear panel waiting to be glued and vacuum pressed
The guts of the doors were completed last week and this week I wanted to try and get the veneered skins on the doors.Todd led me to a plank of curly birch in the wood room that had an awesome bubbly lava lamp like character to it.I thought this would be great for the inside.I also thought that it would be great to use an often overlooked wood.
I unfortunately did not take any pictures of the process used prepare the veneers from a plank of wood, but suffice it to say that you slice off some rather thin (3/32” thick) slices of wood as wide as the plank (8” – 10”), plane/sand the slices smooth and thin enough to bend around the forms, and finally glue the slices together edgewise to form a sheet large enough to cover the inside face or outside face of the door.
I made the rather bold assumption that gluing the veneers onto the substrates would go as smoothly as gluing up the plywood substrates.This did not turn out to be the case. The vacuum press was not quite able to pull the veneers tight against the substrate at the very edge of the door.After some discussion and conjecture with the instructors we modified the glue up with some additional cauls to put extra pressure on the edges of the door, but even then the second door also had a few gaps between the substrate and the veneer.If you look closely in the photo below you can see a thin gap between the outer veneer and the edge band.
“It’s not a mistake until you can’t fix it” – Jim Budlong
It took about a day, but using a Zona saw and every other thin kerf saw I could find in the tool cabinet I was able to slice down about half an inch between the veneer and the substrate to remove the glue which had hardened within the gap and allow the veneer to be pressed tight against the substrate with a small amount of force.I worked some glue into the tiny joint and applied as many clamps as I could along the length of the gap.After the glue had set and the joint was cleaned up again the gaps were gone and no one was the wiser.
This was a challenging but great week in the project.Not much happened in terms of woodworking but some major design decisions occurred. The original mockup of the project had a set of swiveling drawers at the top of the cabinet and two doors at the bottom.There was much debate about to keep the grain patterns properly aligned between the door and the drawer and how much effort and risk that involved.There was also debate about how to keep the curvature of the door and the drawer consistent, as there could be slight changes in the shape of the door fronts versus the drawer fronts over time as the drawer fronts would be captured by the drawer sides and completely unable to change while the doors might change.I seemed to have two options.Option 1 requires me to perfectly align the exterior veneers of the doors and drawers during each glue up with little trimming or waste along with a lot of luck that the doors and drawers would maintain consistent shapes as they were being constructed.Option 2 would expose the partition used to support the drawers and use it to provide a transition between the drawers and doors that I could use to hide any disruption in the grain pattern or disguise any difference in the shape of the doors and drawers.Neither of these options was appealing to me.I didn’t want to rely on good luck for the success of the project and no matter how I mocked up the divider between the top drawer area and the bottom door area I could not stop the project from looking like a layer cake or ice cream cone.While lamenting my conundrum with the ever helpful and wise Todd Sorenson, he me if I had considered just having two doors and get rid of the drawers all together, then I would not have any of these problems.I thought for a minute and decided I would rather trade these problem in for a new and different set of problems.The swing-out drawers are gone!
With that monumental design change out of the way, I was ready to start making two doors.I decided where I wanted the doors to break from the cabinet and cut the curved plywood cores that I made last week.I glued the curved sycamore edge bands to the top and bottom of the door cores and trimmed those flush.I then glued straight sycamore edge bands to the front and back edges of the door cores and trimmed those flush as well.I spent the next day squaring up the doors so that the faces ran plumb and the orientation of the curve was true and identical on both doors.
Last week I finished making the form that I would use to create the curved panels for the carcass of the cabinet. The next step is to create a curved substrate that I will eventually veneer with the woods that I choose for the project. The substrate is create by gluing together four layers of bending poplar or wiggle board. The bending poplar is a special plywood about 1/8 thick that is very flexible in one axis but stiff in the other. The reason that a plywood is used as the substrate is that it will not expand and contract like a solid wood substrate. This means the door is unlikely to change shape or size as the weather changes.
Creating the substrate is straight forward. Cut a few sheets of bending poplar to size. Cover them with glue. Set them on the bending form and place them into a vacuum bag. Suck all the air out of the bag and let the weight of the world press the layers against the form to create the desired shape.
Once the glue has cured I slide the plywood core off of the form. I will need to cover the edges of the new plywood with something pretty so that the plywood does not show when you look at the top/bottom edge or side edges of the doors. These are called edge bands and they can be straight pieces of solid wood glued to the vertical edges of the door panels, or they can be curved pieces of wood glued to the top and bottom curved edges of the door. To make the curved edge bands I use a similar process but instead of bending poplar I use thin and narrow strips of solid wood and bend them around the form. If you look closely at the photo you can see the top and bottom edge bands at the far edges of the form. I had to use a clamp in order to get a little bit of help bending the last few inches of the edge band to meet the form. Those clamps are foreshadowing of a learning opportunity I will experience next week.
If you have been following along and any of this is making sense, you are probably wondering exactly what type of solid wood I have in the vacuum press for those curved edge bands. You would be correct in the assumption that I have made some final wood selections for the project.
I spent a good part of week 3 in the wood room and the back room at The Krenov School looking at various planks. Pear was suggested by one instructor and the thought of a nice even creamy texture and tone was very appealing. Old growth Douglas Fir was the suggestion of another instructor and the thought of the tight grain accentuating the curves of the cabinet was also very appealing. There was a plank of Curly Eucalyptus that is absolutely stunning but probably a little too crazy for the outside of the cabinet. I have a plank of Eastern Swamp Ash that I picked up in Michigan while driving across the country this summer.
Suddenly and not unexpectedly I was trapped in plank purgatory overwhelmed by the endless options for the cabinet that I might build. Late in the week as I was rummaging around my workbench I came across a stack of Spalted Sycamore veneers that I had cut on the last day of school last year with fantasies of some summer project. They had quite a bit of vertical character like the Douglas Fir that would show off the curve of the cabinet and they had a whole bunch of crazy drippy graphics like the Curly Eucalyptus. I thought they looked great and I had plenty of them.
The Spalted Sycamore looked great on the carcass but was not an option for the top of the cabinet. I wanted something rich and homogenous and complemented the colors in the Sycamore. Back to the wood room and more digging. Another day in plank purgatory and in another bout of frustration I rummaged around my workbench and came across a piece of mahogany that I had used last year to test various finishes. I held that up to the Sycamore and BAM!, that looked pretty darn good.
At this point I could not discern if the beauty of Spalted Sycamore and Mahogany combination was due to serendipity or exhaustion. I decided I would head to Kinkos over the weekend and do some digital printing so that I could hotrod my mockup next week.