Friday, February 25, 2011

A few of my favorite non-woodworking, woodworking books

When I opened my first shop, I spent most of my time doing laps; futzing with this, fixing that, wondering where the time went, and why I hadn't gotten anything done. I compared that to my time at school, and it occurred to me that there was a lot more going on in the school environment that I hadn't really been seeing clearly. Like most well-executed details, all of the supporting elements were pretty much invisible, and streamlined enough that they did what they needed to do without calling attention to themselves. We had a machine room and bench area, with instructor supervision and tutelage, carts and machine room jigs, etc, and all I really needed to do was work.

In the pursuit of learning to be a better woodworker in the real world, I've had to reverse-engineer a lot of the supporting background details that I took so much for granted. But the process of learning how to build and run the rest of the machine takes a while to figure out. Making sure that I have what I need, where and when I need it, so that I can just focus on working, was a lot more involved than I thought, and these books have helped me to find my way.


-The Toyota Way by Jeffrey Liker

I'm not producing cars, I'm not in mass production, and I'm not doing anywhere near the kind of business that Toyota's doing. But this was still a great book that helped me become more productive. The core of Toyota's method is about slow, steady progress, and improving the end product by improving the process. Refining the process may involve procedure, shop flow, product design and design process... there's so much support work that goes behind any particular step. And I think that there's a lot for small woodworkers to learn here.

Woodworking is manufacturing. I don't think hobbyists want to see it that way, because who wants an industrial job for a hobby? But while many of them have a lot of fun shopping for new tools, I think a lot of them also end up spending a lot of time in the basement or garage futzing, and straightening up, and trying to clear off the bench, rather than actually working. I know I did. And I've seen a lot of pictures of benches... and shops... that were so clogged up with bits and pieces and half finished projects that there was no hope for any actual work to get done. It may feel counter-intuitive to apply a large-scale production mindset to a guy with basic tools in his basement. But the careful evolution of work habits, work space, and design is critical for lone woodworkers if they're going to get better. Setting up and maintaining a smoothly operating shop space is an ongoing process, and not just a series of attempts to fit 10 pounds of crap into a 5 pound bag.

This is a radical departure from the feeling I used to get from reading the glossy pages. ("Buy this machine, and you'll be able to build... THIS! In a weekend!") I really don't feel like there's a quantum leap to be made in the human ability to produce, no matter what the ads and articles are trying to sell you. Everything requires proper technique and procedure, and an appropriate amount of time. The tool store can sell you new toys, but they can't necessarily tell you how to incorporate those tools into your work flow and your shop in a way that will make sense. And that's the kind of problem that Toyota helped me to solve as things went on.

-The Back of the Napkin by Dan Roam. Back of the Napkin is a treatise on visual problem solving. A picture is worth a thousand words, and the human mind's ability to interpret a well thought out illustration can provide insights that those same thousand words still probably wouldn't inspire.

Problem solving is part of the process of gradual refinement. Sketching things out is a normal way to solve problems, especially for jigs and procedures. But it's easy to get carried away and draw things that look remarkably like the things you've seen before... even if those solutions aren't that great. When I started redesigning my crosscut sled, I found myself redrawing it again and again, and listing what I really needed, and what was really critical as I drew it. I went through 5 or 6 pages of sketches before I really started wrapping my mind around what was going on. I'm still a novice when it comes to visual problem solving. But I've learned a lot from it so far.

-The Gantt Chart, a Working Tool of Management, by Wallace Clark. Despite a technical background, I wound up with an English degree. None of my classes taught me anything about time or project management. They just made the papers longer, and assumed we'd figure it out. This book presented the most straightforward way to plan larger projects that I'd ever seen. While my engineering friends all know about Gantt charts, I'm pretty sure they generate them by computer. This was written in 1914, and everything was being done by hand. And that's generally how I use them, since I don't have a computer in the shop. I can generally compile a pretty decent procedure, or at least figure out what's next. But that doesn't really tell me much about how deep into the woods I am, or where I am in those woods. This is a tool that gives me a much more visual, intuitive way to actually see just how much more I have to go, and where I am in the context of what's left to be done.

-Getting Things Done by David Allen. Gantt charts may help me plan out a project and keep track of where I am in the process, but I still need to actually get into the groove and make things happen. This book is more about how to form more productive habits. The way he wrote it, this really is a book about how to manage paperwork in an office, so it took a while to adapt to the way I work. But he makes some very good points about storing ideas in places where you know they'll be safe, and organizing a work area so that it doesn't get too stagnant. For instance, I used to lose an hour or two a day to some side-track project that was tool or jig-oriented, because I just felt like I had to get it done, or it would be lost in the cloud of things that never get followed up on. Now I keep a notebook handy to write down those little side track tasks, so I can immediately regain focus on the task at hand. It helps me keep track of those side projects, so I know they'll get done, and in the meantime I can remain focused on what I'm doing.


This is by no means an exclusive list, but this post is already getting too long.

Sunday, February 20, 2011

Undoing past mistakes

The home front has been pretty quiet lately. Ariel had a test to study for last week, so I spent last Sunday in the shop. This week I was able to get back to the bench at home and get some work done. Some, but not much. It's amazing how slow some of this seems to go sometimes.

Since I brought the small chest of drawers home, it's been wobbly. It's solidly glued up, but it still rockes back and forth on the bench whenever I did anything. The reason why is pretty simple: it came from a big beam of poplar that I resawed a couple of years ago. Every other board from that beam had given me huge hassles. It had been kiln dried, but when it reached equilibrium, it had some wonky internal tension that all came out when I cut it up. So, the sides of this case, 2 years later, were giving me a hard time. 

This week, I started fitting the two drawers that I've dovetailed so far. I got about 10 minutes in before I realized that I had a problem. Somehow, the inside of my nice, square case was tapering in, and the drawers were getting wedged. It turned out that the side was bowing inwards from front to back. More hassles with the wonky lumber.

To take care of that, I stood the case on its side, and slid a jack plane carefully in between the drawer dividers, so plane the inside of the side flat again. Once I was done with that, I planed the bottom flat, too, figuring it was high time I got around to doing that.

So, lots of shavings produced, but it was a one step forward, two steps back kind of day. At least the case sits flat now.

Getting it together

About a month and half ago, I started keeping a log book that contains entries every 30 minutes to track where my time actually goes when I'm in the shop. And at the end of each day, ideally, I've been logging how many hours I've been putting in at the shop, where they go, when I'm slacking off, etc. The general idea was that if I can see what I'm doing on a given day, and how many hours I'm really working, it'll give me a better idea of how I can improve.

About a week ago, I looked at the accumulated list of numbers, and realized that it wasn't really telling a story. It was just raw data. So, I pulled out a book that I found a few years ago on Gantt charts. It was written circa 1930's, and published by a British publisher... and it's straightforward, to the point, and incredibly usable.

Most of my engineering friends know about Gantt charts, for planning work, and tracking progress. I'd never heard of them. But then again, I was an English major. We weren't really taught much about project planning... which is a shame. I really could have used some help in planning out some of my longer papers.

For the uninitiated, a Gantt chart is a straightforward visual way to track machine or personnel usage, track project progress, plan out projects, etc. In short, it's the solution to the problem I was having... at a glance, I can see how my work habits are improving, or not. But there's a lot more. It's also going to be a huge help in keeping things on track, because it provides something that all of my 30-minute logging doesn't do: it shows where the time is going, on what part of the project, and it will show whether or not I'm actually on schedule. If I fall behind, or something takes longer than planned, I can see just how far behind I am, which will help motivate me to work even harder. And in the future, I'll be able to look at previous charts, and use them as experimental data, to see where my time estimates matched up with my theories, and where I was way off, so I can get better at providing an accurate time estimate to myself and my clients. Small adjustments, gradual progress... good stuff.

Friday afternoon, I sat down with the book, and planned out the remaining parts of the project I'm currently working on. The project has been winding down, but there are a lot of details I've been trying to corral and pin down to get this thing out the door. I didn't realize I had about a week's worth of work left, but I do. And now I have the tools to see what's going on, and try to keep it on track to be done in a week.

The joy of being self-employed is that I get to be my own boss. But it also means yelling at the employee when he's not working. Now it also means making sure the shop foreman has a clue about what needs to get done, so he can keep on top of the employee and make sure he gets everything done on time.

At least I have other people to talk to when I'm working. 

Wednesday, February 16, 2011

Coolest thing this week

There's a tool I've been sitting on for almost 4 years, waiting for an opportunity to use it. Yesterday the opportunity presented itself. I'm sure that I didn't squeal like a little girl in excitement.

Well, mostly sure, anyway.

The tool in question is the cove cutting wheel from CMT.  I learned about cutting coves on the table saw at school, but we did it using regular table saw blades and a LOT of sandpaper for cleanup. This works MUCH better.

The wheel came with a crown molding set that I bought when I was a retail stooge. (I think you can get the wheel as a separate piece: CMT part number 235.006.07)  There's only one real problem: It's 7" in diameter, which means it isn't compatible with my hyper-intelligent SawStop. The SawStop will only run 8" dado stacks, or 10" blades. If the diameter isn't right, the saw will know, and the motor won't start. It's the only real complaint I have about the saw. But I still have the contractor saw, and it doesn't know or care what I bolt in, so the cove cutter runs just fine. And this is the part where I start gushing. It works WONDERFULLY. The rim of the thing is really heavy, to give it the inertia to keep it spinning. I knew at first sight that this thing was going to be cool.

People asked me why I kept my contractor saw when I bought the SawStop. There are a lot of reasons: It's paid for, I might need to make a quick cut, but not want to break down a setup on the primary saw, my router table is bolted to the contractor saw... they're all ok reasons, but the truth is, I kept it so I could run this cove cutter. And I have to say, It's worth it. With a dado stack, and this cutter, the contractor saw is a whole new animal that's somewhere between a table saw and a router table. I've heard of molding cutters for table saws before, and typically the feedback I got about them was "bad idea." Too many stories about knives coming loose, etc. This is a different story. It's one solid piece, with the carbide welded solidly to the disc.

Last but not least... the coves are SMOOTH. Smoother than most router cuts I've ever made. And it's quick. Table saw blades are naturally designed to cut on the edge, and they're preceded by some pretty small gullets. They can be used to cut coves, but it's a slow, tedious process. This blade has six nice, round teeth, that are sharp all the way around, and huge gullets in front of them. So it can take out a lot of wood at a time, and the gullets are huge enough to handle the bulk that's removed.

Sunday, February 13, 2011

Cutting table, chapter one: the racetrack

So, since Saturdays are designated as "My day" shop project days, I took yesterday to get going on phase one of the cutting table build. This will take a few Saturdays to get through, but will be beneficial in the long-term.

The general idea came from this post by a guy named Steve Jones over on the Festool User's Group website. He came up with a more efficient way to process plywood, that I really think makes sense. In short, the concept is that a Festool track saw is a lot lighter and easier to handle than a full sheet of plywood. Since the two need to be moved past each other in a directed fashion, why not move the lighter of the two? It makes a lot of sense to me... especially since full sized sheets are heavy and awkward, and I'm more likely to bungle the process or damage the parts in the process of dealing with them as a result.

In the case of this particular project, the top and bottom of the torsion box that I'm building for the table have full-sheet dimensions: 48"x96". Since the parts coming out of the process are basically the same size as they were when they went in, and we only have one cart for full sheets of plywood, I didn't have any other place to put each sheet after it was processed. So, I did the only thing I could: I dropped the blade on Chris' Oliver table saw, cleared everything off, and started race-tracking the parts around our huge outfeed tables. I have a pair of long trestles that are designed for supporting plywood when it's being fed into or out of a saw, so I used them to help slide panels to other surfaces, too... like the contractor saw, which is on the left in the above picture. And at the far end, near the double doors, is the plywood cart, holding incoming sheets. It was an interesting day.

For this phase of the project, the chief frustration was my dado stack. When I originally got set up, I bought a Freud Dial-a-dado. The short story is that there's a large aluminum adjusting knob that adjusts a threaded spacer between the outer plate, and then next inner one, to make minute adjustments to the width of the dado that's being cut. In theory, it's great, because it means no fussing with shims. In practice, it has proven to be a royal pain in the ass. As the steel arbor nut is tightened down against the aluminum adjusting knob to hold the stack securely in place, it grinds in, turns the knob, and throws the adjustment way off. So, the nice, precise adjustment mechanism becomes instantly imprecise, and it took me roughly an hour to get the proper width dialed in and secured. I was not happy.

The big eureka moment of the day was figuring out how to make sure that the dados had consistent depth. Since the outfeed table isn't as level as I'd like, and plywood can warp, it's hard to keep the sheet from lifting up away from the cutter. The solution was to take the insert out of the saw, round down the front edge, and adjust the leveling screws to get the front edge re-aligned with the table surface. Why does this work? The depth of cut is gauged against the top surface of the insert, and the problem comes when the plywood isn't in contact with that surface. Rounding the front end of the insert and re-aligning it had the end result of lifting the center of the insert up. So, it was always supporting the plywood, and the depth of cut was maintained.

Inspirational words of the day

"Men were not intended to work with the accuracy of tools, to be precise and perfect in all of their actions. If you will have that precision out of them you must unhumanise them.

If you will make a man out of the working creature you cannot make a tool. Let him but begin to imagine, to think, to try to do anything worth doing, and... out come all his roughness, all his incapacity... failure after failure... but out comes the whole majesty of him also."

-John Ruskin


Topped 3000 page views today.

Thanks to all of you for reading. It's humbling and exciting to think that people are interested.

It's late, I'm tired and more, but I wanted to say thanks.

Friday, February 11, 2011

Shop tips

A better glue bottle

This was originally the product of my frustration with store-bought glue bottles. I had one that was getting clogged, and I squeezed a little too hard during a glue up. The top came off, glue went everywhere, all over the project, and my temper went through the roof.

Soda bottles are rated for pressure, so I knew it wouldn't blow out on me. (If you don't believe me, put a cap on an empty one and stand on it.) And it brings its own benefits. Drilling the cap on one side gives more control over the glue stream, because a little rotation of the bottle one way or the other will re-align the stream. It doesn't have a long, pointy nozzle to get clogged up. And depending on how big a hole you drill, the bottle can dispense small, controllable beads, or deliver massive quantities in a hurry... whatever's called for. And if you decide later that you want a smaller hole... well, shucks. Where will you ever find a new cap? Oh, right. EVERYWHERE.

I don't know how the chemistry works, but titebond I doesn't seem to bond with the plastic at all. It'll build up on the cap around the hole, but it pops right off when I pick at it. And it doesn't stick to the inside of the bottle at all. I've been using this bottle for a few years now.

And the price is right. A five cent bottle deposit is much cheaper than the $2-3 you'd pay for a dedicated glue bottle at the hardware store.

Production crosscutting stop

I tried to work this into the first entry on the crosscut sled, but it just didn't fit. This is a crosscut stop design that I discovered when I had a shop in Medford. It made sense immediately when I saw it. There's about an inch of wood hooked onto the hinge. Flip it out of the way to crosscut something to rough length, and then flip the block back around when you crosscut the other end to finish length. It's a great production tool.

Thursday, February 10, 2011

Evolving a better crosscut sled: followup review

So, in the past few days I've been able to take the sled through it's paces. I'm really impressed. Having zero-clearance inserts built into the floor makes a HUGE difference. Having them made out of something as cheap as masonite is a bonus, because I can already tell that I'm probably going to go through two or three of them on so many projects. But combining the clean cuts that those inserts give me with the reliable precision of the Kreg stops gives a whole that is greater than the sum of its parts. I'm mad as hell that I didn't think of this before.

The ability to cleanly cut rabbets and dados is huge. The veneer on the plywood I'm using on my current project has already shown a tendency to blow out horribly when crosscutting. A fresh insert fixed that problem. And using the Kreg stops made rabbeting out side panels for bookcases a snap: once the dado stack is properly set up, run it through the insert. Nudge the panel up to the edge of the insert, and set the stop at the other end of the panel. Next, slide the panel over the gap in the insert, and stick a piece of plywood cutoff as a spacer between the panel and stop. This sets the panel to the right position for a rabbet that's exactly the thickness of the plywood. Re-set the stop, and run rabbets. SO easy.That's the first and most obvious method that I've figured out so far... more will come, I'm sure.

I've already had a few ideas on what I would do differently on the next version; it's just the kind of guy I am. How would I build another sled more simply and easily? What functions did I not consider when I did this one? I"m already kicking myself over the replaceable insert: It's a brilliant idea that I could have done better. If I run the blade at a 45 degree angle, I'm pretty sure I'm going to run it through the screws that are holding the insert in place. I hadn't really thought about using a beveled blade with a sled before. In school, the sled was just a tool for running 90 degree crosscuts. Nothing with a dado stack, nothing with a tilted blade. And the Kreg fence has really changed everything. Having used this sled a few times, it's very clear to me just how much more is possible.

Or will be, when I make a newer, modified-again sled.


Tuesday, February 8, 2011

Evolving a better crosscut sled

Over the past 4 years I've gone through a few versions of cross-cut sleds.

The kind we used at school was very plain, and while it worked, it was not what I would call optimal. It was basically like this one that I made for my contractor saw a few years ago. Big fences, mounted as square as possible to a huge piece of MDF. Setting up repetitive cuts meant clamping a huge L-stick to the fence, which was a huge pain in the ass. Clamping it up meant using one hand to hold up the stick in place, one hand to operate the clamp, and another hand to hold the tape measure, to make sure the stick was being clamped up in the right place. 

My first attempt at a miter bar was intended to solve most of the problem. The L-stick would have a wooden bar mounted to it that would hold the stick up in a horizontal position, and a knob would thread onto a bolt through said stick, and the bolt would engage with a T-track to clamp the thing down. In short, I could use the knob to set the position AND clamp the setting in place, while I measured with the tape measure using my other hand. Voila... I no longer need to use three hands.

This is the most recent version, and it's the first one I'm really proud of. The Kreg rails add several important advantages: Kreg's production stop is AWESOME, the adhesive scale that sticks to the rail means I don't need a tape measure anymore, all that metal adds mass, which helps the sled slide very well, but most importantly, because they bolt onto the fence, I can loosen the bolts and shim the fences to be perfectly square if I need to.

I made the floor of the sled out of 1/2" ply, ran dadoes for the rails, as well as for the fence, to make sure everything was positively square. Each side was made separately, and then joined when I mounted the fence, and the bridge in the back. I finished the whole thing with sanding sealer, so that the plywood wouldn't distort when I waxed the underside of the sled.  (Wax works wonders... the whole thing slides like it's on marbles that are rolling in a puddle of WD-40.)

I wanted to use a replaceable zero-clearance insert, and the easiest way I could think of to lay one in and have it sit flush with the rest of the floor was to lay one whole layer of the same material. So, I made the insert,  installed it, and then used contact cement to lay down the adjacent panels. With the replaceable insert, I should be able to use the sled for dados, and mitered cuts, too.

Sunday, February 6, 2011

Cutting drawers

I cut dovetails and glued up another drawer today. I still feel like I'm a little bit out of practice. But I also feel like it's coming back really quickly... more a question of dusting off existing skills than starting from scratch.

When I picked up the twin screw vise, I had dovetails on the brain. And it's great for that. But today, I had to rip a few boards to finish width, because I cut them oversize at the shop. And I'm liking the fact that the side of the vise holds just as firmly (or more so) as the span between the screws. It's a great setup for holding short-ish boards for ripping. The tall jaws cut down noise while ripping, too. The vise holds the wood firmly enough that the board can't start resonating... at least, not within the 6" height of the jaws. I won't say it's a silent process, because it's not, but I've been aware for a long time that quiet cuts are typically cleaner cuts, because of the lack of vibration. And it helps keep my wife happy, which helps keep me working at home on weekends.

I'm going to use drawer slips to hold in the bottoms, since I've never used that method before, and I think it has a few advantages over running grooves in the drawer sides... especially when it comes to working at home. If something goes wrong with a joint, ripping another board to make a new drawer side doesn't turn into a big deal... without the grooves, there's nothing 'special' about any particular board. I may decide later on that I don't like drawer slips, in which case I'll look into getting a plow plane. But that's money I'd rather not spend... especially since drawer slips are a great use for scrap wood that's sitting around the shop. Lastly, this method will make it easier to lay out joinery, since I don't have to worry about where the groove is coming through the end of the drawer front. 

I got to put my holdfasts to work again today, holding down the parts while I trimmed out waste. Hammering one of those things in is a joy... but it did bring commentary from the next room, since it was loud enough to be heard over Ariel's headphones.

Tuesday, February 1, 2011


This winter is a hassle. Living 35 minutes away from my shop has some real downfalls, like today. And, likely, tomorrow. Not having a heated shop doesn't help, either. On sunny days, it's nice and toasty in there. But on days like today, the prospect of driving through this much snow-bound traffic, in order to spend what's left of the day freezing up in the shop, just doesn't make sense. While today's/tomorrow's storm is supposed to be truly epic, it seems like there's been one day a week for the past month where the weather has gotten in the way of it making sense to go to the shop. It's a real drag.