Full Chisel Blog

June 29, 2008

Beating a Saw with a Hammer

Filed under: Sawing,Techniques,Uncategorized — Stephen Shepherd @ 5:47 pm

No this is not when you take the hammer away from a child because they are making noise by pounding on a hand saw but the purposeful hammering of a saw blade in order to straighten it out.

Hammering was done on new (19th century new) saws once they were made and prepared for sharpening.  Hammering was done to ‘tension’ the saw blades during the manufacturing process and work hardens the blade but allows it to remain flexible.  And of course hammering is done to old saws that are bent to straighten out a bent or kinked blade.  Badly bent or kinked blades that have cracked or fractured the steel of course the hammering will not repair that condition (I am not sure it is possible to repair).

Old blades should be cleaned of rust and brought to as ‘bright’ a condition as possible in order to have a reflective and smooth surface on the saw.  Bright tools tend to resist rust.  Once it is cleaned, it is time to examine the blade, which I do by sighting down the back of the blade.  This also applies to backed saws, sight down the back to see if there are any kinks, bends, bows, cups or curves.

 Some hammer saws on a metal anvil, and small saw anvils do exist, I however choose to use a wooden anvil with a slightly different technique.  I do have a metal plate that I use occasionally but for the most part it is on top of my chopping stump, a large squared timber cut off at about 34 inches tall, which I use for a variety of uses.

I always straighten out the back of the saw first, as this will usually straighten out the entire blade, depending on the bends and kinks.  I hold the saw with the blade straight up and down (the handle) with the back of the saw roughly parallel with the earth.  Thus by sighting down the back you should be able to see any problems with the blade.  Note where the bends are and place the blade down on the ‘anvil’ with the convex side of the blade up.  In other words the bump or kink is bulging up from the anvil.

S&J Bent

Planishing hammers with a slightly rounded face is the best, I use a small tinners hammer with a slight belly to its face.  It should be a round face to prevent denting, I however use a square faced hammer that I like and it has a slightly convex face (belly).  I am comfortable using these smaller hammers and have straightened dozens and dozens of hand saws.  I can never resist an old saw for $2.00, and rarely pay more than $10.00.  I am careful in my hammering not to go too fast and that my blows are deliberate and controlled so as not to dent the saw while attempting to remove dents.

A few ‘well placed’ blows with the hammer on the bulge, then I check the back of the saw again by picking it up holding it the same and sighting down the back again to check the progress.  I do this after every 3 to 6 strikes with the hammer.  Later I will check the saw after each blow when things are getting straight.  Now what is a ‘well placed’ strike?  That might be difficult to explain as I think it comes from experience.  Now different saws will behave differently, some straighten easily while others tend to need more attention.  I avoid striking on the very back edge of the saw (and of course I avoid hitting the teeth when working near them).

Hammering pattern

This illustration shows the pattern in which I strike the saw with the hammer, on a wooden anvil.

I hammer in a square pattern of a couple of blows next to each other near the back edge of the saw, then a couple just below.  I avoid striking the saw in the same place (unless I am working on a dimple or small bulge) but spread the blows around to prevent ‘popping’ of the blade.  That is when a dent reverses direction when struck with a hammer.  These can be tricky and may require more attention to get things flat.

S&J Straight

This saw was difficult to photograph, but the blade is now straight on both the back and the business side.  It had a significant kink and other minor bends, but it took less than 40 hammer strikes to straighten out the blade.  When striking the saw make sure it is laying as flat as possible on the anvil, this can be difficult with a bad bend.  I can tell when a blow from the hammer does something as it ‘sounds’ more solid when struck.  If it makes a bit of a ‘ringing sound’ then the blow probably didn’t do anything, it is when the blow ‘sounds solid’, that work is done.

I should mention that there are other methods for straightening metal by considering that the hammer blow will expand the metal on one side, so the work is done on the reverse side of the above mentioned technique.  This particular method requires a metal anvil as the idea is to expand the metal by hammering.  I have had some success with this method on other bent metal items, but I do not use it for saws.

I think that a craftsman should be familiar with their tools and know as much as possible about all things involved.  The study of metallurgy, of geometry, of physics, of chemistry, of morphology, the macroscopic and even microscopic study can add to the arsenal of knowledge and understanding.    Expose yourself to as much learning as you can, question experts and continue the quest, don’t limit yourself.  Like Mark Twain said, ‘Don’t let Schooling interfere with your Education”



  1. Hope you had a good weekend, Stephen.

    “Hammering was done to ‘tension’ the saw blades during the manufacturing process…”

    I think hammering–and subsequent tension–was simply a necessary step due to the warpage of the hardening/tempering processes of Western saw manufacture. Saw plate, following the hardening/tempering process was buckled and twisted despite heavy weights placed upon stacks of saw plate following the proceedure.

    Not trying to be argumentative.

    Take care, Mike

    Comment by Mike — June 29, 2008 @ 7:39 pm

  2. Mike,

    I appreciate your comments and I acquiesce to your knowledge on this subject. Perhaps putting the ‘tension’ in the blades during manufacturing is a polite way of saying straightening the plates. It would be interesting to know more about early nineteenth saw manufacturing, Joel’s post on his blog over at toolsforworkingwood on saw handle manufacturer is interesting.

    Now who is defensive? And yes I am having a Great weekend.


    Comment by Stephen Shepherd — June 29, 2008 @ 7:53 pm

  3. Tension, IMO is a bit subjective, and perhaps a bit difficult to measure. Poor sawyers would likely want more tension to prevent having the blade shaking around at the end of the downstroke on an unbacked saw. Persons with proper sawing technique might not care as much. As I mentioned in my last blog post, I have a saw that I don’t think has enough tension, but as I am learning to use it, I really don’t care. Who knows, I may even learn to like it that way. In any event, I don’t see myself trying to tension it unless it is just as an experiment to see what happens. Perhaps if I were cutting gnarly woods, I would change my tune. I really don’t know at what point you could say that you had too much tension unless you were starting to deform the saw blade.

    Tensioning and straightening are two different things and one can exist without the other although they can also overlap somewhat since a similar technique is used to produce both. It should be possible to produce a properly tensioned semicirle or just about any other smooth shape. It would also be possible to have something that was straight, but with very little tension.

    One of the banes of sawmaking is that the raw starting material is never straight enough for high quality saws. It can be cupped, warped, bubbled, kinked, and have any other number of maladies. Most modern sawmakers solve this by adding extra kerf and using relatively heavy blades.

    Comment by Luke Townsley — June 29, 2008 @ 8:57 pm

  4. Hi ya, Stephen!

    I think part and parcel with straightening is the tensioning. The main question to me is, would the sawmakers of yesteryear beat saw blanks with hammers if the plate remained straight post heat treatment? Not something I can answer pre 1890 or so. By the time Disston was able to do this it can be argued the beginning of the end of quality saws was at hand. I think that would be stretching things a bit, but someone will stand up and say it.

    Modern saw makers use plate of approximate thickness to the saws they are copying. Of course, for long saws, that would be me mostly. So we use a variety of saw plate thickness when doing repros. For the Kenyons, that is .042″ plate to start. For certain Disston styles, it is thinner.

    Loved the blog at Joel’s site. Nice Christmassy type setting. Clean ol’ man and workshop, working with a bow saw and a dozen or more tools that would have never been used for handle production. There’s a wee chance it never happened thatta way. Would make a neat card. Maybe I’ll have some printed. Sure makes it a romantic job.

    Take care, Mike

    Comment by Mike — June 29, 2008 @ 10:13 pm

  5. Luke and Mike,

    Moxon speaks of tensioning ‘It cannot be too stiff, because they are but Hammer-hardened…” and also mentions the taper of saw blades from the teeth to the back. Too little historic information for my liking on this and most every subject to which we are interested. These proceedures must have been so common that no one thought to write them down as everybody just knew them, so why bother.

    Newer books (20th Century), seldom mention hammering and rarely mention whetting.

    And that picture on Joels Blog is definitely romanticised.


    Comment by Stephen Shepherd — June 30, 2008 @ 6:23 am

  6. Moxon was speaking of hand forged out blades (well, in a commercial environment, “powered” hammers) in the making of them. Which itself, along with great variability in thickness and impurities, had to have created stresses in the plate. That in addition to warpage due to heat treatment.

    Even when rolled plate came into being, at first it wasn’t as even as later periods as regards even thickness–look at the mic’d plates of the back saws in the Seaton chest where in theory parallel plate was used. And when rolling first became the norm, the same impurities were in the steel.

    I don’t know what the cut-off point for “modern books” is, but all my references up to “recent” FWW articles mention whetting, or stoning, the sides. But then, I don’t get out much.

    At best, whetting gets the tips of the teeth with a few thousandths of their respective neighbors. It also changes the cutting dynamic by creating zero clearance at the sides of the tips–which increases friction in the cut. Worse if the whetting is taken too far.

    The picture of Joel’s–yeah, but it ws a wonderful purchase on his part. Love the thing. I really might do a woodcut and print some Christmas cards this year. Unless I cheat and use a linoleum block…

    Take care, Mike

    Comment by Mike — June 30, 2008 @ 7:11 am

  7. My lack of metal knowledge not-with-standing, I’m guessing that the term Tensioning as used by the old guys was a generic term used to describe the outcome of hammering metal: compaction. The plate has to be ductile to withstand use, sharpening, setting, etc. But the body of the plate has to have resistance to bending as well as heat from friction. If my memory serves, we now know that the crystalline structure of steel changes as it is hammered. And that is about as far as my memory goes.

    Comment by Gary Roberts — June 30, 2008 @ 8:38 am

  8. On a parallel track, but not truly on-topic, is circular saws. I have read that large circular saws for sawmills were often hammer tensioned into a very flat cone so that when spun at their particular RPM the plate would flatten out and have less tendency to wander than a flat, untensioned saw would under the same conditions. However, a lot of this kind of thing fails to pass scrutiny.


    Comment by Metalworker Mike — June 30, 2008 @ 8:59 am

  9. Yeah, I have wondered about the circular saw thing. I worked in an old saw mill in the late 1970s that the millwright would tweak on the saws because under bind they could straighten out one portion or another, create a wobble and need fixed.

    But again, I think of what they do as shaping, not necessarily hammer tensioning. Maybe it is a semantics thing. I don’t know. I do know that mere shaping of the dish creates tension via compression of the steel. Then again, this tension equals hardening and there were blades that cracked over time as well. I would equate that with work-hardening in localized areas.

    Take care, Mike

    Comment by Mike — June 30, 2008 @ 10:11 am

  10. I am looking for a good saw hammerer in North/East PA. This is for a 56 in. saw.

    Comment by george hunter — August 23, 2009 @ 9:07 am

  11. I have a saw blade with the problem that Stephen has described here. I actually may have made it worse by using to heavy of blows. The blade, a 22″ panel saw blade from a no. 7 Disston, clearly had suffered a severe dent at a spot just forward of the handle, and another just north of that one.
    The quote and problem I am referring to is ‘popping’ of the blade. That is when a dent reverses direction when struck with a hammer. These can be tricky and may require more attention to get things flat.” Can anyone give advice for how to remove the problem. It is so serious now that one can hardly tell which side of the blade was originally dented. Thank you.
    I would really like to be able to get this saw up and running.

    Comment by Acharya Kumarswami — September 14, 2009 @ 2:44 pm

  12. Acharya,

    If you hit the dent in the middle, it will just pop to the other side. You need to work around the dent, on all sides especially above and below the dent (near the teeth and near the back). The metal is stretched enough at the dent, so it needs to be worked on the edge of the dent.


    Comment by Stephen Shepherd — September 15, 2009 @ 6:11 am

  13. Hi Stephen,

    Can you upload a couple of pics of the hammers you use? I want to get an idea for size and weights.


    Comment by Brad — May 16, 2012 @ 2:28 pm

  14. Brad,

    I just use any hammer with a slight belly to the face. Mostly round hammers, but some have octagonal faces and some are square with fine edge chamfers. The weights range from 6 ounces to 16 ounces depending on the problem. The round face with no sharp edges is important.

    Also sometimes I use a steel anvil and sometimes a piece of wood.


    Comment by Stephen Shepherd — May 16, 2012 @ 5:46 pm

  15. […] The hardest part was getting a feel for how hard to hit. You don’t have to smack it with the fervor you use at the carnival “whack-a-mole” booth. There’s a Goldilocks’ knack to it—not too hard, not too soft. I also found a good tip here that details what pattern to use when hammering. […]

    Pingback by Learning to Remove a Bow: My new, old favorite crosscut saw | Hand Tool Journey-A woodworking show of hands — June 13, 2012 @ 4:03 pm

  16. Hi Stephen,
    Great piece, I found your experience invaluable with my saws. I just wanted to share some of my findings when learning about saws. I have several great older (pre -1900)saws, both new and slightly used. Steel making in these times was an art and great variability exists between the metals these workers produced. The steel was cast into plates that were cut to the shape of a saw. The blank was then ‘taper ground’ along the back. Leaving full thickness at the teeth and up the heel and handle. Teeth were then machine cut at the required TPI, setting and sharpening were also largely machine processes.
    These saws had a nodular metal structure more like cast iron than the crystaline structure of modern steel. When new, these saws were very floppy and very prone to kinking in a cut. Users of saws would, before use, have a saw doctor tune their saws, this included tensioning and hand sharpening. A ‘saw makers hammer’ was tear dropped in shape with a round face as you describe on the thin end, a short curved handle and a weight of a few pounds. This heavy hammer was used on an anvil to compress the cast steel and slightly deform the blade. The compression was concentrated in the center and executed in an almost identical manner to your method described above. Working both sides to keep the blade straight. If a new blade is bent in a curve it should remain straight across its width. A tensioned saw has a distinct curve across the blade when it is flexed to a curve. This process of tensioning also intentionally deformed the blade into a very subtle wavy (think egg carton) pattern. increasing in depth toward the back and toe of the blade. This ripple is noticeable when an old tensioned saw is rubbed with a flat stone of a decent size. This process was of importance to keep the saws vertical in the kerf (i also find these saws stay waxed much longer). A final note, modern saws (like my Lie-Nielsen panel saws) seem floppy like an untensioned saw in use, the steel is soft enough to file for sharpening and i was able to remove a kink comparable to the one you show above using your method, they do not seem to responded to the tensioning process in the same way as some of the Disston, Atkins and Simmonds saws I have had a go at tensioning. Thanks for your great piece. I have learnt a lot from the old timers I have worked with (learnt under) in the past but as time passes these fountains of knowledge seem to be fewer and further between. Although I have never posted on the interweb before I wanted to share what I think I know. Hope some of this was helpful. thanks Michael.

    Comment by Michael E. — July 13, 2014 @ 8:56 pm

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by WordPress