Saturday, June 27, 2015

Handsaw tension.



Gabe has questions regarding handsaw tension, as do I.



  1. This removing the tension from the edge with the hammer, did it make a noticeable change in how the saw cut before you re-tensioned? It seems quite counter intuitive that striking along the tooth line with a cross peen would remove tension. My grandfather has an Irwin hard pointed saw, but as a full length panel saw. It cuts very well, but the saw flops around in use like a beached fish. It seems obvious to me now that it needs tensioning. Any reason you used a round faced hammer to tension? How did you know when the saw felt like it had enough tension?
    Lol, I'm full of questions today. As far as the smaller distortions, how does it go for you to sight them visually? I mean to say, are you having much luck holding the saw up in front of you and viewing the plate at a low angle? Most of the time when I am looking at a saw the distortion I can see in the plate does not readily suggest how to be corrected with the hammer. You're doing great work! Though a humble little saw, that tool now has a lot more value.



Hi Gabe!

I really wish that I had some good answers here, your questions are all good ones.....questions that I share! Tensioning of handsaws is one of those questions that a few of us are asking, but finding answers is proving difficult.

George Wilson (now retired, tool makers from colonial Williamsburg) indicated in some long ago thread on SMC that they didn't actually tension many of the full size long saws that they made on site, and they worked just fine. What the heck is tensioning anyways, and is it necessary? Here is my thinking, such as it is.....

When you cut wood, force and friction cause the blade of the saw to heat up, not like HOT, but the heat is definitely there. The heat effect is greatest at the tooth line, and causes the saw blade to expand slightly, effectively lengthening the saw a tiny amount. The effect being greatest at the tooth line, if there isn't some way to accommodate the extra length, you get wandering cuts or binding. Striking the body of the saw to "tension" it, is essentially making the body of the blade longer than at the tooth edge, so that as the blade heats in use, all things are evened out. To me this would mean that, rather than the edge being "tensioned" in its normal state, what we are actually doing is putting it in a state "compression". Tensioning would then be a misnomer. I'm not sure if this is the case, but that's how it works in my mind, haha.

I doubt that this saw was tensioned originally as I saw no marks on the plate that indicate that it was run through an English wheel....lines of distortion running longitudinal to the body. I think that's how they do it nowadays. The saw plate on this guy is just relatively thick, and very stiff. When I struck the tooth line at the hash marks, to remove or neutralize the tension, it was mostly for forms sake. The idea is that I would be "lengthening" the tooth line to be equal in length to the body, and I used my faux cross-pein hammer to give more directional force, a stronger lengthening effect.

In re-tensioning the blade, if we look back to Hasluck Handyman's Guide,


he calls for the use of two different hammers to fix bends, the cross-pein hammer for specifics and a larger dogs-head hammer for the general area around the bend. The larger hammer is used for tensioning, I am assuming because we need to input more force. Bob Smalser calls for using a 2# hammer for tensioning, so he has something similar in mind.

It would've been interesting to have tested the saw prior to tensioning, but it didn't occur to me. This was one of those cases where I just got pissed, grabbed the hammer and started whacking things. My direction and intent were there, but the pictures came after the effect. Let's all try to do better in the future, because this is helpful information, haha.

As I spend more and more hours staring at these different saw blades, I guess that I am training myself in various ways to see deviations. When I bend a saw blade and look at the way the light reflects, I now see much more than I did a few months ago. I would have to say that my eye is looking differently now, somewhat akin to the way an artist sees the space around an object as much as the object itself, seeing the "negative space". Mark Grable refers to this as right brain/twilight brain thinking, and that would be correct as well. I don't look at the blade itself, but more how the light gets distorted by the bad spots.....kind of. And to be clear, I claim no facility here either, I'm just starting to see this stuff. When I was tensioning this blade, how did I determine the areas that needed extra tension? You could call it a guess and be just as correct as any excuse that I could devise, haha. It just felt like it needed a few extra whacks to bring things back into line.

When staring at the blade side on, if you just bend the blade into an arc there might not be anything to see, but if you squeeze that blade between your hands as you bend (inducing a compressive force) you might see much more.

Somewhat similar, I have two ryoba, one cheap and the other much higher quality. The cheap one bends smoothly and evenly, also rings loud and bright. The higher quality saw bends smoothly as well but as I bend it, it feels like the saw plate close to the edge is stiffer, while the main body is more flexible. It doesn't bend evenly exactly. Smooth, but not even. When rung, the sound it makes it clean, but also lower in pitch or perhaps richer/more complex.

I was looking at a handful of old, rusty saws yesterday, and spent a minute bending them back and forth, sighting the length, just kinda checking them out. One of the saws was an old 1920's Disston, one was a "guaranteed superior", and a couple were just.....crap. The Disston was trashed, but the GS saw was stiff, bent in a way that felt nice to me, but more than just stiff. I don't know yet how to describe it. Most interestingly I thought, were the crap saws. Though they bent clean, smooth, and easy, they just felt if you were sawing something, they would be flopping all over, wandering from one side of the line to the other. It just feels like something was lacking.

I have been looking at some pictures of a nice Maebiki-oga that was on eBay not too long ago.

Of the great variety of hammer marks that you see on these handmade saws, the maebiki have some of the most distinctive.

I'm looking at the upper corner, and the wealth of impact strikes that are common here. Is this to thin out the metal in that entire corner, reducing drag in use? Is it greatly thinner there, to allow a degree of correct-ability when your cut runs away from the line?

Is it to drive mass into the greater body of the saw, as part of the tensioning process?

When I look at this photo, I see hammer strikes that are pushing (fullering, in blacksmithing terms) metal from the tip of the saw into the main body. Then, that mass gets pushed /spread along the length of the saw, as evidenced by the little line marks from a completely different type of hammer.

The opposite side of the saw shows a similar (though not exact) system in play. Obviously this is done for a specific reason and I would love to know.

Gabe, you've got two maebiki right there....what are your thoughts?










  1. This really clarifies things up...I thought originally that when people were 'tensioning' a saw, they were compressing the plate, causing the metal to start to be drawn towards the hammer hit, making it more tense and rigid. Kind of like a guitar string drawn towards the tuning peg, becoming tighter.

    Your theory makes a whole lot more sense, I had forgotten about the heat produced by sawing. People always think the old ways are primitive and easy, and sometimes they can be, but hand tools seem to be incredibly complex under the surface...Complex in the work and in the human part of it all.

    1. Hey Steven!

      I hesitate to suggest that things might be completely the opposite, because what you are thinking is exactly what I have read, time and again. I really want to be clear here.....this is just me thinking out loud. I am just trying to figure this stuff out, and I'm asking all of you folk to help me, haha.

      The tensioning process is essentially the same thing as what is known as "hammer shrinking" in the automotive bodywork trade. You strike the metal, it gets drawn towards the point of impact,actually making the metal slightly thicker and more stiff. It is the surface profile of the metal that is changed. You can use a waffle face hammer ( a shrinking hammer) or a slapper file and things really get working, tearing up the metal so much so that you need to grind again before finishing.

      The traditional description of handsaw tensioning......doesn't it just sound like the saw would get looser during use, get all floppy? Maybe what is happening is that the traditional description is correct, and it is the metal raised around the hammer impact that creates a band of tension, a stiffer region where you want it, right next to and supporting the tooth edge? It is definitely true that a deformed flat panel is stiffer that one that is perfectly flat.

      It seems most likely that the traditional description is correct, and that my understanding of it is still to immature. Like you say, these processes seems so simple, but it is often that complex things are actually very subtle and sophisticated, only appearing simple to the glancing eye. Doesn't this make you want to get a bunch of saws, try tensioning different ways, explore how simply whacking a saw blade with a hammer might make it perform better? Maybe it doesn't make much difference?

      What do you think? Help me figure this out!

  2. Wow! I read this a couple of hours ago and needed the time to think about your description. This really is one of those things that we must try on a bunch of saws. I have a stack of old crap panel saws waiting to be worked on, but in truth they don't excite me enough to put the time into them. I need saws that beg to be used, like the maebiki-oga!
    Part of this dynamic that I find interesting is that a saw looses its tensioning from use. I've also hear of saws loosing their tempering (over much longer time periods) as well. In simple terms we think of steel as hard and immutable, not the reality of this crystal lattice that is constantly dynamic and interactive with its environment.

    Those photos of the maebiki-oga are fascinating. Have you seen:

    The photos are not great but it give a few more examples of working saws used by one of the only living master kobiki. Its worthy to note how rough some of these saws are by comparison to the refinement of a carefully scraped ryoba. Not being a blacksmith I can't make good judgements about the tools and techniques that result in the hammer marks we see. Some of this stuff looks like the signature of a particular smith, or style of saw. The maebiki-oga that I have used the most with the ground plate of .085"-.090" has a gradual taper from the hammering at the tip of the spine edge that thins out to .065". However, the other saw that is an entirely hammered plate varied greatly, .060"-.080" throughout the whole thing, with no taper at the tip of the spine edge. Can you imagine hammering a sheet that large at a power hammer to within .020", let alone by hand with a team of strikers? Of course, my little micrometer can only measure along the edge of the saw. I've been thinking of ways to make a reliable deep throated caliper for measuring the mid plate, so I can't say about fullering. Those other hammer marks in the pictures, what kind of hammer was that? It had some serious sharp edges, really strange. Thanks for posting these photos! I'm gathering a collection, trying to notice commonalities and derive a bit of understanding

    1. Thanks for the reference to your saws Gabe!

      I have heard mention of saws loosing their tension. Interesting, no? Doesn't that sound as though......So we have a saw blade, just the slightest bit elongated at the center. This stretching action along the middle of the saw body is trying to pull the toothed edge along with it, putting the metal in tension. That meshes with the tissue paper analogy shown in the illustration. Because the steel of the blade is actually an EXTREMELY slow moving plastic state solid, over time the toothed edge would "catch up" with the body, and would no longer be in tension. That makes sense to me.

      The "age hardening" of steel (precipitation hardening of a supersaturated solid solution), a blade loosing it's temper..... fascinating stuff!

      Let us know how you like your Japanese sawyer book, I'm thinking of getting my own copy but I'm hoping that there are lot of pictures, haha. The link you posted was the same guy, right?


Like all of us, I am figuring this out as I go, so when you see something that is incorrect or flat out wrong (and you will!), let me know. This is a learning process. Real people and names, please. Constructive comments and questions are very welcome, but hate speak/politics are not! Life (get one!) is too short.

Thanks, Jason