A Hand Built Roubo, and the Education of a Woodworker

[David Keller NC]

Mike,

Thanks for the feedback and for the guide block recommendation. I'll definitely try it tomorrow.

I've been using chisels, too. I was wondering if you(or anyone) has any experience with Lie-Nielsen's joinery floats. Do you think they'd help a novice mortiser out? These are the ones I mean.

Thanks,

David

David - Yes, a float is WAY easier than a chisel when it comes to making adjustments to either mortises or tenons. The problem with using a chisel to do this is that it's almost impossible to telegraph your intent to the chisel - it will want to follow the path of least resistance, which in this case is along the grain. A float works more like a file or a rasp, and will largely ignore grain variations.

If you decide to get a joinery float, remember that you must sharpen them - they are not "ready to be used". Sharpening is quite easy with a 3-corner saw file (I would recommend the 7-inch regular taper for this job, though you can use smaller and thinner ones).

And to reinforce your instincts - when using hand tools, there is almost no advantage to making any joinery element uniform and identical to one another. That's a recipe for uncontrolled variations that result in sloppy joints, and is an attitude/way of working that's a product of power-tool woodworking. When doing these operations by hand, there is very little labor savings trying to make parts uniform, and in many cases, it takes more labor than fitting one piece to its mating joinery element.

[Jim Koepke]

I was wondering if you(or anyone) has any experience with Lie-Nielsen's joinery floats. Do you think they'd help a novice mortiser out?

Not sure, but I might try convincing SWMBO that it was needed just to get a new tool.

when using hand tools, there is almost no advantage to making any joinery element uniform and identical to one another.

I think of joint uniformity as the impossible dream.

jtk

[David Keller NC]

I think of joint uniformity as the impossible dream.

jtk

Hmm - well, personal opinion here, but I don't see uniformity as a dream (or a goal) - I see it as the voice of the inexperienced. Sort of like the huge number of threads on just about every WW forum that have a title something like "I used a set of feeler gauges and my Jointer fence is out 0.001" over it's length - HELP!"

[Bill Moser]

Hmm - well, personal opinion here, but I don't see uniformity as a dream (or a goal) - I see it as the voice of the inexperienced. Sort of like the huge number of threads on just about every WW forum that have a title something like "I used a set of feeler gauges and my Jointer fence is out 0.001" over it's length - HELP!"

Yup. When I built my workbench, I leveled it with a 6 ft starrett straightedge, and feeler gauges (only down to .0015, I'm such a slacker). That was a great feeling. My bench was really, really flat. For, like, 5 seconds. Mostly, the problem with us (newbies) is that we don't know exactly how much accuracy is needed for a particular job, so we generally tend (once we've f***d up a joint or two) to over compensate. I'm getting better at that, slowly. Two pieces of advice for other newbies:
1) mark the joints that go together (paying attention to the reference face & edge)
2) don't sand/plane/erase these marks before glue-up. Not that I've ever done that.

[Frank Drew]

And to reinforce your instincts - when using hand tools, there is almost no advantage to making any joinery element uniform and identical to one another. That's a recipe for uncontrolled variations that result in sloppy joints, and is an attitude/way of working that's a product of power-tool woodworking. When doing these operations by hand, there is very little labor savings trying to make parts uniform, and in many cases, it takes more labor than fitting one piece to its mating joinery element.

David,

Perhaps I haven't understood your point, but I think I don't agree with you here, unless you simply mean that each individual joints might need some small amount of final fitting.

[David Keller NC]

David,

Perhaps I haven't understood your point, but I think I don't agree with you here, unless you simply mean that each individual joints might need some small amount of final fitting.

To say this in more detail:

What power-tool WWs typically do (and I used to do) is strive for very accurate machine setups that produce very uniform results, without "cutting to the line". In fact, if you're producing multiples of something by machine, it is considerably more efficient to use stop blocks, jigs and the like to avoid marking out your pieces.

One sees this often on The New Yankee Workshop, which is a classic machine approach. Norm will cut his mortises with a mortiser, and then spend a few moments fine-tuning his tenoning jig on the tablesaw with a test piece to achieve a good fit. He will then run all the pieces. While this does include a bit of "fitting the piece to its mate", the difficulty with this approach is that every piece of wood is different, and machines and jigs are inherently less accurate than a skilled hand-tool worker.

But there is a great advantage to this approach when making a small production run - you wind up with many rails and stiles that are reasonably identical to each other, and are interchangeable with each other.

Now the opposite approach - when cutting parts using hand tools, you must mark each piece. Many of us have had the unpleasant experience of accidentally erasing these knifed lines and the extreme difficulty in adjusting the fit of these pieces once those marks are gone.

Given that you have to mark each piece of both halves of a joint, it makes much, much better sense to mark the first halves of the joints (possibly by measurement, there's no harm in using the same measurement for the first halves of the joint), and then mark the second half of the joint from the first half, and given the slight variations in the particular piece of wood, mark these pieces from a specific first half (i.e., you have to number both halves of multiple joints to avoid confusing them). This approach does two things: it gives you a better than reasonable shot of having your second half of the joint precisely fit the first half with no adjustment, and if adjustment is required, those adjustments narrow in on the precise mate to the part, instead of a "global mean" of the other, similar parts.

Overall, using the hand-tool approach of fitting a specific piece of wood to another specific piece of wood results in tighter joints and, in my opinion, better craftsmanship than the make-identical-parts power tool approach. However, it has the downside of being much less efficient if you're making multiples.

But - it's much less mind-numbing than freaking out over the couple of thousandths inaccuracy of your table saw fence, jointer bed, or cross-cutting jig.

[David Wadstrup]

Hello,

This is a little off-topic, but I am a little confused by the performance of my saws, and would like to hear your thoughts. There is a science to saw sharpening which is way beyond my general woodworking knowledge. Earlier in this thread I mentioned that my rip sawing seemed to go more slowly than I would have expected. That I am ripping 8/4 ash partly explains things. Also, my 24" rip saw is 8 PPI, which, I guess, is suited more for 4/4 stock(what I will normally be working with). So, the slow pace didn't really trouble me too much. Today, however, I used my 24" cross cut saw for the first time. It is 9 PPI. I was trimming the 3.5"x5.5" legs to final length, and the saw cut through them in no time at all. I was surprised because I had assumed that cross cutting would be a slower process vs. rip cutting. Is this a mistaken assumption? Out of curiosity, I tried a little experiment. I used the cross cut to rip some of the same wood that I had been working on earlier. To compare, I also did a short length of rip with the true rip saw. Perhaps it was me, but it seemed like the cross cut ripped faster than the rip saw. It also started and sawed as smoothly. Of course, I am now thoroughly confused. Does this sound odd to any of you? Are there any saw experts out there that can explain this performance?

Thanks for your help,

David

[John Coloccia]

Hello,

This is a little off-topic, but I am a little confused by the performance of my saws, and would like to hear your thoughts. There is a science to saw sharpening which is way beyond my general woodworking knowledge. Earlier in this thread I mentioned that my rip sawing seemed to go more slowly than I would have expected. That I am ripping 8/4 ash partly explains things. Also, my 24" rip saw is 8 PPI, which, I guess, is suited more for 4/4 stock(what I will normally be working with). So, the slow pace didn't really trouble me too much. Today, however, I used my 24" cross cut saw for the first time. It is 9 PPI. I was trimming the 3.5"x5.5" legs to final length, and the saw cut through them in no time at all. I was surprised because I had assumed that cross cutting would be a slower process vs. rip cutting. Is this a mistaken assumption? Out of curiosity, I tried a little experiment. I used the cross cut to rip some of the same wood that I had been working on earlier. To compare, I also did a short length of rip with the true rip saw. Perhaps it was me, but it seemed like the cross cut ripped faster than the rip saw. It also started and sawed as smoothly. Of course, I am now thoroughly confused. Does this sound odd to any of you? Are there any saw experts out there that can explain this performance?

Thanks for your help,

David

The crosscut saw is filed like a bunch of little knives so it CAN start easier sometimes, though I don't really notice a difference anymore and I'm not sure I ever have....it's maybe less grabby. The problem using it to rip is that it tends to push the wood fibers (like lots of little straws) out of the way as opposed to cleanly shearing them off. Now, you have a lot more surface area to push through with a rip saw, that is cut like a chisel, so it will tend to work "harder", if you get my meaning, but you should get a cleaner and straighter cut. I don't know about faster. Splitting wood is faster than sawing it, for example, but we saw for precision.

My guess is that your rip saw is not sharpened very well, but even if it is I'm not sure that I would judge it based on speed. I would judge it based on accuracy. If you think about how a crosscut saw works, there will be wood fibers that are practically untouched and in the way when you use it to rip. Those fibers can easily push things off line and even hang up the saw (you will eventually abrade or break the fibers off but that's not the goal).

Anyhow, that may not be a satisfying answer, especially since you seem happy with the performance of the crosscut saw for ripping. The other question is if your crosscut saw is actually filed crosscut...filed well, anyway. It can be very poorly filed in a hybrid pattern, giving mediocre but similar behavior for crosscutting and ripping. Given a choice in most wood, I'd rather crosscut with a rip saw than rip with a crosscut saw so I have to think that your rip saw is very dull and that your crosscut saw is possibly not sharpened in a true crosscut pattern (or poorly so). A symptom of a poorly filed crosscut saw will be tearout on the end grain. This is from the wide teeth (as opposed to the sharp, knife-like crosscut teeth) grabbing the wood fibers and yanking them as opposed to cleanly severing them.

[Hans Braul]

I felt guilty for about a second, but decided it would be ok to use a power drill in this case if I promised myself that I would dig the other 8 with just the chisel. They will be 1/2" wide, and thus much easier.

Thanks for reading,

David

I guess this is where I get confused. I get (sorta) why you would take on a project and limit yourself to jus hand tools. It's the guilt I don't get. If it's easier with a Forstner bit, then why not use a Forstner bit??? Seems like going out of your way to make things difficult. But never mind - as I said before, I admire your determination to make this project work and look forward to watching your progress.

Regards
Hans

[Tom Vanzant]

Generations ago, the cabinetmaker would have his apprentice do the rough work that we might use a machine to do since we don't have an apprentice. Where's the difference in the real world?

[David Keller NC]

I was surprised because I had assumed that cross cutting would be a slower process vs. rip cutting. Is this a mistaken assumption?

David - To expand on what John C. said, the speed at which a handsaw will cut is influenced by a bunch of differnt factors, not just whether it's filed cross-cut or rip. Obviously, the strength and skill of the sawyer is the biggest influencing factor, but there are many other, less obvious contributors. Some of those are: sharpness, degree of set (bigger set means removing more wood), rake, fleam and shape of the teeth, eveness of the teeth, sawplate thickness, hang angle of the tote (this is the angle formed between a line drawn parallel with the grip and a line perpendicular with the toothline), breasting of the saw tooth line, and probably a few others that I haven't thought of.

As John notes, the principal reason for choosing a rip-filed saw to rip wood is accuracy - cross-cut saws tend to wander excessively when ripping, but that tendency also depends on how well-behaved the grain is in the wood.

[David Wadstrup]

David,

Thanks to you and John for the saw tutorial. What you said makes a lot of sense, and I now feel like I have a better understanding of how saws work. I'm now of the opinion that my saws are performing just as they should be performing. The cross cut is lightening fast, even through my 3.5"x5.5" legs, and there is only minimal tearout. And the rip tracks incredibly well. I cut a few 8' lines, and I was able to cut straight on the line the whole way though. I thought I was just a natural, but now I'm thinking it maybe had more to do with the saw than my abilities.

Thanks for sharing your knowledge and opinions,

David

[David Wadstrup]

I've spent the time since my last post waiting for a set of Lie-Nielsen's joinery floats to arrive. It turned out that they were the perfect tools to help me clean up and square the mortises that I cut. They looked professional and precise once I was done with them, and I'm glad I made the purchase.

Today I cut my first tenon. It was slow going, but very satisfying work. I'm discovering that I'd much rather spend my time in the shop joining wood instead of milling it. Anyway, I have a question, and hope you can help... The tenon fits into the mortise ALMOST perfectly. The rail meets the leg exactly where it needs to(surfaces flush), and they sit perfectly square to one another. I was actually pretty surprised that this joint came out so well. The one problem, though, is that I'm afraid that the fit might be too loose. How loose is too loose? I read somewhere that each piece should be able to hold onto the weight of the other by friction alone. Is this true in all cases? My leg, for instance, weighs about 20lbs, and the rail's tenon would pull right out if it tried to lift it without supporting the leg. Holding the leg, however, is a different matter. The rail stays joined. It will work itself loose and fall out if I shake the pieces around, though.

What do you think? I definitely will shoot for a more snug fit next time, but am wondering if you think I should try to remedy the looseness of this one? Does it sound too loose? Are there any tricks to getting a tighter fit back? I'd really appreciate your advice.

Thanks,

David

ps If it makes a difference... I'll be securing this joint with both glue and drawbore pins. But keep in mind that this will be one weighty bench and will likely see a lot of stress put on its foundation.

[Salem Ganzhorn]

If they are going to be drawbored they are probably ok a little loose. But you could also glue some veneer on the tenon to take up the slop. By the way i am nowhere near an expert but i did use drawbored tenons for all my leg joints when i buit my benhcrafted style split Roubo.
Salem

[David Keller NC]

David - If you draw-bore the joints, and use an appropriate wood for the draw-bore pins, it will make little to no difference if there's a bit of slop in the mortise. The ideal is a flexible, tough and dense. White Oak, Hickory would be preferred, but red oak, maple, etc... will also work.