Panel Saw Tensioning

[Trafalgar]

Bob Smalser:

http://www.wkfinetools.com/tRestore/saw ... Blade1.asp

"When a thin blade is struck on an anvil by a convex-faced round hammer, a dimple is created; often so small it can’t be seen by the eye. Steel from the area around the dimple is pulled inwards toward the point of impact, making the steel in the circular area radiating from the dimple stiffer, or “tensioned” on its surface.

Hundreds of such hammer blows applied in certain patterns equally to both sides of a handsaw blade can make it stiffer, can true a warped circular sawblade, or can dish a large bandsaw blade to conform to its wheels while at the same time tensioning the cutting edge."

End quote.

"The UltraSTEELTM process, developed by Hadley Industries Plc (Hadleys), is a novel surface dimpling process [well, perhaps not so novel] used on steel strip prior to cold roll forming. This dimpling process increases the strength of the final rolled products and enhances other product properties such as fire test performance and screw retention."

 

[D_W]

In use the two D8's both have the annoying flapping habbit on the return stroke. The S&J and one of the Tyzacs are nice saws. The last Tyzac isn't in working order at the moment.

So, overall, I would carefully conclude that this handsaw tensioning business is more then just a myth.

Not surprised to hear that late d8s are pretty floppy. I've had maybe half a dozen early d8s, and I had (until selling them last week) two later d8s. The early saws are stiff, but the later saws are not as stiff and the rest of the details don't look as good on them, either - the handles are blocky and the polish in the steel is more coarse.

 

[Corneel]

That would make them good candidates for some tensioning trials. Now I need two things: Time and Courage. I am a little afraid that I will turn them into pretzels.

 

[Trafalgar]

Why on earth would you risk it? You know it exists. If you have a well-tensioned saw be glad of it and let the naysayers be naysayers. Nothing to be gained. It's like looking yourself up in the phone book to make sure you've been spelling your name correctly.

 

[pedder]

And a picture with the bend saw plate. Here you can see the effect of tensioning.

Hi Kees,

I see two hollows. Not more not less. The rest is interpretation.

And I don't know how it got there. I can be an result of the grinding process.
Bu I doubt it is the result of a hammering. It is too consistent for that.
Or do you say Disston rolled the tension in the blade?

Could you forward me the hole german text via email?

Cheers
Pedder

 

[swagman]

Kees. you may wish to make contact with Rob Streeper; Saw Maker from the USA.
Can be contacted via the Australian Woodwork Forum; Saws Handmade.
Rob hammer tensions his saw plates, and has posted a number of threads on this subject.

http://www.woodworkforums.com/f278

Stewie;

 

[Corneel]

Yes I know about Rob Steepoer, I think I read all his threads overthere, but thanks for the tip anyway.

Pedder, I'll send you the texts I have found.

I posted two pictures from my Spear and Jackson saw. The first is with the plate flat. You can see that the straight edge shows no hollows, it sits flat on the plate. The second picture is with the saw bowed in the length. Like a sector of a circle, radius probably 1 or 2 meter. Now I suddenly see those hollows appear. I tested on several places on the plate and found similar effect, but not all the way to the end. This behaviour shows that it is not from grinding.

This is exactly what you expect from a tensioned saw plate. The hammered or roled band along the length of the plate, through the middle, tries to elongate the sawplate. That is of course not possible, so it sets this centerline in compression, and the adjacent areas in tension. When you bow the plate, the compressed sector suddenly gets a chance to really extend the length, it simply bends a little further. That's what you see with the lightgap meassurement.

The Disston D8's didn't show this behaviour, and they behave more floppy in the cut. So that is why I carefully concluded that there might really be something in this tensioning theory.

The Disston D8's are 0.9 mm thick at the toothline, my English saws are 1 mm thick. This could also be a factor in the floppyness of the D8's, becasue a smaal decrease in thickness means a lot less stiffness.

So far it is all interpretation. It seems like a fun diversion though. I have some junker plates so maybe I'll try to find some time for experiments. Don't hold your breath though, this is going to take some effort.

 

[Cheshirechappie]

Yes I know about Rob Steepoer, I think I read all his threads overthere, but thanks for the tip anyway.

Pedder, I'll send you the texts I have found.

I posted two pictures from my Spear and Jackson saw. The first is with the plate flat. You can see that the straight edge shows no hollows, it sits flat on the plate. The second picture is with the saw bowed in the length. Like a sector of a circle, radius probably 1 or 2 meter. Now I suddenly see those hollows appear. I tested on several places on the plate and found similar effect, but not all the way to the end. This behaviour shows that it is not from grinding.

This is exactly what you expect from a tensioned saw plate. The hammered or roled band along the length of the plate, through the middle, tries to elongate the sawplate. That is of course not possible, so it sets this centerline in compression, and the adjacent areas in tension. When you bow the plate, the compressed sector suddenly gets a chance to really extend the length, it simply bends a little further. That's what you see with the lightgap meassurement.

The Disston D8's didn't show this behaviour, and they behave more floppy in the cut. So that is why I carefully concluded that there might really be something in this tensioning theory.

The Disston D8's are 0.9 mm thick at the toothline, my English saws are 1 mm thick. This could also be a factor in the floppyness of the D8's, becasue a smaal decrease in thickness means a lot less stiffness.

So far it is all interpretation. It seems like a fun diversion though. I have some junker plates so maybe I'll try to find some time for experiments. Don't hold your breath though, this is going to take some effort.

I agree that sorting it out won't be quick or easy.

It seems that there has been some manipulation of internal stresses. Whatever was done stopped short of pushing the steel to yield point through it's full thickness because the blade's shape has not changed, which points to a modification of surface (or residual) stresses. Unfortunately, those are hard to measure, and outside the scope of 'home workshop' methods. Thus, we're left with observing the effects, which seem to be subtle rather than dramatic.

One of the difficulties in trying to draw too many conclusions from vintage saws is that we have no idea how much abuse they may have suffered during their life, or what effect that might have had.

Perhaps the best way to eliminate uncertainties - as said before - is to make two new saws, tension one, and compare. The obvious problem is in working out the details how the tensioning should be done. There may be some trial and error, here.

By the way, I tried the bend-and-straightedge test on a couple of my saws. The 1980s Robert Sorby (that states 'Tensioned Taper Ground Blade' on the blade) showed the bulges Corneel illustrated, but only to a very small, almost imperceptible, degree. The Groves, which is a much thinner blade and much older (stamp on middle of blade instead of etch, split sawnuts) did display the bulges to a more noticeable extent. It's a subtle thing to detect, though. A couple of 1930s or 1950s panel saws didn't show the phenomenon. I haven't checked all my saws yet, though.

 

[Cheshirechappie]

I refurbished the two panel saws mentioned in the last post over the weekend. Both were acquired a while ago via Ebay. The first is 20" long with a 0.030" thick non-taper-ground blade, the second 22" long with a 0.035" blade, also non-taper-ground. Both had nibs at the toe end, and solid (not split-nut) sawscrews, and a Warranted Superior medallion each. Handles are of 13/16" beech, nicely shaped on the handhold and with decent though not pronounced horns, and without lamb's tongues. The 20" has no stamps or etches, the 22" has a very faint stamp, "Buck and Ryan, London NW". I suspect that both saws are of 1920s or 1930s vintage - one reference I've seen suggests that the blade nib feature was dropped during WW2 (though there may have been exceptions to prove the rule!).

Both were toothed crosscut, 8tpi on the 22", and 6tpi on the 20". Both were in poor dental order, but the blades were beautifully straight. I've recut the 20" as a 10tpi crosscut, and refiled the 22" 8tpi as rip.

Testing both saws showed that they had no humps and bumps across the plate either flat or bent into a slow curve. So - probably not tensioned.

However, both saws now work very well, having little tendency to toe-end waggle in use. That's probably because they're relatively short, but nevertheless, it's hard to imagine what improvement tensioning them would make. It did cross my mind to get the hammers out, but since the blades are so nicely straight, it seemed daft to disturb them and risk introducing distortions.

Thus, I wonder whether 'tensioning' might be something that has some positive effect with longer flappier saws, but less with shorter ones like panel saws (unless they're very thin)?

 

[Rhyolith]

Wow, this is still going

Thanks Corneel, think thats probably the most useful post so far in terms of getting to the bottom of this! Its also an example of the sort of thing this thread has been sorely lacking in in terms of proof.

Personally I think it supports my theory (well not mine, just the one I believe is true) that "tensioning" in handsaws is a means of manipulating the internal stresses of a saw plate to achieve the same thing as a bow frame does (stretching the cutting edge). As that is how I imagine a metal plate to behave when bent if such a thing is done to it, the centre will have more compression and will be push out more steeply than the stretched outer sections (the cutting edge and back). The benefits of this are obvious if its possible in a handsaw plate, just use a bow saw with low blade tension -> then increase the tension and see for yourself!

Warning! Sorta a Rant: I don't mean to offend anyone, but can people stop having a go at engineers/technical sorts, its wasting a lot of posts. Believe me I know how annoying it is when someone cannot see beyond proof and misses more instinctual aspects, but I don't think this is happening here. What is genuinely needed is just something beyond "this is what I think so there" as is demonstrated by Corneels post. Further, now I understand what the technical definition of stiffness is (its worth taking more than a few moments to think about it if your not an engineer or physicist) I agree with CC its probably nothing to do with saw tensioning.

Its seems that Lie-Nielsen still tension their brass back tenon saws, trying to get more info on this.

 

[D_W]

Warning! Sorta a Rant: I don't mean to offend anyone, but can people stop having a go at engineers/technical sorts, its wasting a lot of posts. Believe me I know how annoying it is when someone cannot see beyond proof and misses more instinctual aspects, but I don't think this is happening here. What is genuinely needed is just something beyond "this is what I think so there" as is demonstrated by Corneels post. Further, now I understand what the technical definition of stiffness is (its worth taking more than a few moments to think about it if your not an engineer or physicist) I agree with CC its probably nothing to do with saw tensioning.

That's me, and i'm not exempt from technical as I'm an applied mathematician. I started with the idea that there's probably a lot of scamminess in the past and low hanging fruit, but have learned to detach myself from the technical background and instead examine through the lens of continuous improvement over centuries.

I don't have too many notions like this thread started with, that I suspect something strongly about a specific topic I don't have much experience with. I learned to drop that because the probability that you'll learn why you're wrong is pretty high. In this case, as with many others, embedded is the assumption that you are enlightened and hundreds of thousands of others in the past were stupid and easily fooled. Not just general population (you can certainly sell a butt jiggler exercise machine to thousands of unwashed), but professional users who didn't have a lot of spare cash.

I don't see too much of this from non-engineers - it's just what I've seen in 10 years, and I started out on the other side of it. The notion that other people are responsible for providing the unwashed information for a foregone conclusion is also goofy. The burden is on them.

Plus, in this case, it has always occurred that the high end saws that I have, that have *more* taper (and less cross section because of it), have been stiffer without being harder to the file. It is an extremely easy thing to test. All one has to do is take a saw and do something demanding (cut something like beech 12/4 or 16/4 or resaw something of hardwood) with it and the wheat separates from the chaff quickly.

 

[Trafalgar]

Well said.

 

[Rhyolith]

Again, I don't me to offend you (or anyone else) this is just what I think based on my understanding of whats been written.

That's me, and i'm not exempt from technical as I'm an applied mathematician. I started with the idea that there's probably a lot of scamminess in the past and low hanging fruit, but have learned to detach myself from the technical background and instead examine through the lens of continuous improvement over centuries.

So are you saying that we should be willing to look at this through a more subjective lens?

I don't have too many notions like this thread started with, that I suspect something strongly about a specific topic I don't have much experience with. I learned to drop that because the probability that you'll learn why you're wrong is pretty high. In this case, as with many others, embedded is the assumption that you are enlightened and hundreds of thousands of others in the past were stupid and easily fooled. Not just general population (you can certainly sell a butt jiggler exercise machine to thousands of unwashed), but professional users who didn't have a lot of spare cash.

If you could magic up a early 20th century saw doctor I would take his/her word as law on this topic, but the fact is we don't have any reliable evidence from that group, technical or otherwise. Regardless of the expertise of the people of that time (which I have a high opinion of) the fact is that the information they have left (that we know of so far) is uselessly vague, to the point where even knowing for sure if saw tensioning actually exists in handsaws is very much questionable.

I don't see too much of this from non-engineers - it's just what I've seen in 10 years, and I started out on the other side of it. The notion that other people are responsible for providing the unwashed information for a foregone conclusion is also goofy. The burden is on them.

I don't really get this point.

Plus, in this case, it has always occurred that the high end saws that I have, that have *more* taper (and less cross section because of it), have been stiffer without being harder to the file. It is an extremely easy thing to test. All one has to do is take a saw and do something demanding (cut something like beech 12/4 or 16/4 or resaw something of hardwood) with it and the wheat separates from the chaff quickly.

I would gladly take the use of a tensioned saw vs a un-tensioned saw -> how they feel, operate and perform as evidence of tensionings benefit... but this relies on knowing a saw is tensioned in the first place. I have a number of old and new saws, and have no idea which are tensioned or indeed if any are. It takes a proper test or some solid non-subjective knowledge to find this out, mainly: what is saw tensioning?/how do you test for it? Which for me brings us back round to what I think CC has been after (I think!), which is proof/evidence that is to vague and subjective.

 

[Rhyolith]

Well said.

Who?

 

[D_W]

Again, I don't me to offend you (or anyone else) this is just what I think based on my understanding of whats been written.

That's me, and i'm not exempt from technical as I'm an applied mathematician. I started with the idea that there's probably a lot of scamminess in the past and low hanging fruit, but have learned to detach myself from the technical background and instead examine through the lens of continuous improvement over centuries.

So are you saying that we should be willing to look at this through a more subjective lens?

I don't have too many notions like this thread started with, that I suspect something strongly about a specific topic I don't have much experience with. I learned to drop that because the probability that you'll learn why you're wrong is pretty high. In this case, as with many others, embedded is the assumption that you are enlightened and hundreds of thousands of others in the past were stupid and easily fooled. Not just general population (you can certainly sell a butt jiggler exercise machine to thousands of unwashed), but professional users who didn't have a lot of spare cash.

If you could magic up a early 20th century saw doctor I would take his/her word as law on this topic, but the fact is we don't have any reliable evidence from that group, technical or otherwise. Regardless of the expertise of the people of that time (which I have a high opinion of) the fact is that the information they have left (that we know of so far) is uselessly vague, to the point where even knowing for sure if saw tensioning actually exists in handsaws is very much questionable.

I don't see too much of this from non-engineers - it's just what I've seen in 10 years, and I started out on the other side of it. The notion that other people are responsible for providing the unwashed information for a foregone conclusion is also goofy. The burden is on them.

I don't really get this point.

Plus, in this case, it has always occurred that the high end saws that I have, that have *more* taper (and less cross section because of it), have been stiffer without being harder to the file. It is an extremely easy thing to test. All one has to do is take a saw and do something demanding (cut something like beech 12/4 or 16/4 or resaw something of hardwood) with it and the wheat separates from the chaff quickly.

I would gladly take the use of a tensioned saw vs a un-tensioned saw -> how they feel, operate and perform as evidence of tensionings benefit... but this relies on knowing a saw is tensioned in the first place. I have a number of old and new saws, and have no idea which are tensioned or indeed if any are. It takes a proper test or some solid non-subjective knowledge to find this out, mainly: what is saw tensioning?/how do you test for it? Which for me brings us back round to what I think CC has been after (I think!), which is proof/evidence that is to vague and subjective.

re: the tensioning thing. We're all operating from some area of lack of knowledge about it here. I don't think there is a 20th century saw doctor we can go to, those folks would've been feeding saws through rollers, though maybe I'm being errant, and perhaps that was done as a matter of skilled inspection and working something until it was as needed.

With respect to subjective or not, subjective isn't really the right notion, more full research and understanding of a problem before coming to a conclusion or even significant bias. The probability that you're going to come to the right answer is increased by a whole lot when you learn more about the problem to begin with. Perhaps the right answer won't ever materialize because of lack of being able to understand the problem, and that's OK. "I don't know" is perfectly fine, or "i suspect, and that's as far as I can go". In a case like this one where someone proposing a hypothesis can't notice the difference between better saws and lesser saws (perhaps due to lack of really heavy actual use), then there's a real problem in drawing any conclusions at the outset. Especially if plenty of other people with more experience suggest a tangible difference.

I would assume that engineers are left at work to either take a problem and come up with an engineering solution, or given the assignment to find problems and then formulate solutions and come to a conclusion. I would assume that has a lot to do with this being a trait of engineers, to assert something without having a full grasp on the problem to begin with. Most other professions aren't charged with as much analysis or problem solving, which probably limits the desire to do this kind of thing.

 

[D_W]

By the way, we have a problem of sorts in guessing which saws were tensioned and which weren't, and at least Kees is giving us some clues to start with.

The problem is even if a saw that we think wasn't tensioned is floppy, we don't know that it was floppy at manufacture, or if something happened to it. It's likely it was probably floppy at manufacture if it's not been rusted and it behaves normally under a file. But for any of us to stick our neck out and put a flag in the ground and say "it is written from here forward"....I'm willing to say I'm not sure but the evidence points to there being something to it a lot more than I'm willing to not understand the circumstances fully and put that flag in the ground.

 

[Rhyolith]

I still don't think I am grasping your issue with the engineers... anyway

I think I know what saw tensioning is now, my theory is already described. Looking at this the other way round -> I cannot see why this wouldn't work, its just a case of it being possible to internally stress steel plates in the way I described before... everything falls into place for me then and my questions are answered/answerable through simple follow up logic.

If that theory is correct then stiffening is nothing to do with saw tensioning, which using what I have learnt here, makes sense.

 

[D_W]

I still don't think I am grasping your issue with the engineers... anyway

I think I know what saw tensioning is now, my theory is already described. Looking at this the other way round -> I cannot see why this wouldn't work, its just a case of it being possible to internally stress steel plates in the way I described before... everything falls into place for me then and my questions are answered/answerable through simple follow up logic.

If that theory is correct then stiffening is nothing to do with saw tensioning, which using what I have learnt here, makes sense.

It's certainly possible that the saws that are properly tensioned return to zero faster and with better frequency, and it creates the perception of stiffness rather than actual stiffness. Who knows?

There seems to be a property in the better saws that when you put them into actual heavy use, they do better at it, and it's not plate thickness (as the cheap saws have thick plates without much taper).

(It's not only engineers who do what I described, and it's not all engineers - kees does quite a bit of looking before he comes up with a conclusion and states it. I mention engineers because it seems there is a professional tie-in to wanting to define a problem and then suggest a solution and reason through it. Why engineers? No clue, it's not physicians - they never seem to have the extra time, and they're also closely related to waiting for the results in terms of modern medicine in the states. It's not folks in the arts or humanities, they don't usually have the same desire to solve problems and be analytical. It usually seems to be engineers, or in one case recently, someone who has a quasi-technical profession that led them to believe they had drawn a firm conclusion because they could put errant data into a software package and declare they had a statistically significant solution).

Other "engineered" conclusions that I've seen:
* steel was substandard
* old cap irons are inferior to new ones (most people don't even know how to use them)
* thicker iron is always better, thin ones chatter
* newer irons are harder, they're better because of that
* if craftsmen had access to lie nielsen planes, they'd have preferred them over stanley (probably not, the price would've been higher - stanley couldn't even get people to spring for bedrocks)
* new chisels are better than old chisels because of better steel alloys
* old saw steel was not as good as new saw steel
* modern sharpening stones are one of the greatest improvements in sharpening tools (never mind the fact that similar stones were available in the early 1900s at a price equivalent to a mid grade arkansas stone, but they never sold---alumina abrasive and resin binder and all.)

It's uncommon for someone to step back and say "why are older irons softer than new hocks" without knee-jerking "because newer irons are better", and the same for the rest of those. The errant conclusion that people were uninformed in the past, easily taken advantage of, and "we know better" now. Certainly that's true for something like a tobacco enema, but beautifully made old chisels were made the way they were because professional users preferred them that way. Same for most of the rest of the stuff. An educated pool of users found no preference for harder irons or thicker irons (God knows Ohio Tool tried to sell thick irons for a long time).

We have pretty far to go finding out why things were the way they were, in general, rather than concluding they were antiquated junk.

 

[Corneel]

In this video from 6:50 onwards, Rob Steeper explains his ideas about saw plate tensioning.

https://www.youtube.com/watch?v=HITnU5m1eGo&feature=youtu.be

 

[MIGNAL]

Hmm. What's he saying, that it alters the resonant frequency? we could be getting into nodes and all manner of other things. Where's a bridge engineer when you need one.