Episode #47: Removing & Re-installing a Saw Spine

Note: All of my old podcast videos have been moved to my YouTube channel.  You can now watch this video here:


38 thoughts on “Episode #47: Removing & Re-installing a Saw Spine

  1. so if you were to make a saw with a canted blade you would want to cant the tooth line not the spine ? or could it work both ways

  2. Thanks for showing Bob, very interesting to me as I currently have two such saws that need refurbished. Only confusion on my part is, why worry if the spline isn’t parallel to the teeth?, it will continue to support the blade’s straight line, will it not?
    FYI, for any viewers that would like additional information on this topic, a Lumberjocks member recently posted a similar blog with photos and videos at this address: http://lumberjocks.com/Brit/blog/33741

    Again, thanks for posting, I really enjoy your web site.

    • Well, you really don’t have to worry about it IF the saw works OK for you AND if the spine is seated on the entire saw plate. In fact, I typically explain to my customers that if the saw works OK as is and the spine is fully installed on the plate, then there’s no harm in leaving it that way. In many cases, however, the spine will be sitting lower at the toe and not even on the plate at the heel because of the angle it is at (which was the case with this saw). This situation will make the blade less rigid at the heel and more likely to kink in use. When I get saws like this for sharpening from customers who are restoring them, I always ask if they want the spine reset the correct way. Most do because that is the way that their particular saw was intended to be.

      Problems can arise when resetting the spine though. Often the reason that the spine is seated improperly is because the saw was either dropped on its toe, or because a previous owner hammered it lower to try and fix a bend/kink in the blade (an improper and inelegant solution in my opinion). Removing a poorly aligned spine from a seemingly straight blade will often reveal that the blade has a bend/kink in. To do a proper repair and reset of the spine in these cases, the kink first needs to be hammered out of the blade before the spine can be reset.

      I always warn my customers of this possibility before I remove a spine. I don’t charge extra to remove and reset the spine when I receive these saws, I offer it as a complementary service as part of the sharpening if the spine needs to be reset. However, if removing the spine reveals a kink, I want my customers to know before hand that it will need to be hammered out before reinstalling the spine, and straightening a kinked plate is an additional charge. So I explain all of this ahead of time, and most still opt to have the spine reset and have the plate hammered if necessary because that is the way the saw was intended to be set up and will work best.

  3. Thanks for the great video Bob. One thing I guess I missed was the value of having a canted (sloped) tooth line versus a straight tooth line that is parallel to the saw back.

    • Ask several different sawyers about this and you’ll get a bunch of different answers. In my experience, I find canted saw blades easier to use with more aggressive teeth and lower handle hang angles because the canting of the saw blade seems to help ease the teeth deeper into the cut as opposed to having all of the teeth the same depth and engaged all at the same time. I think of it like breasting on a large rip saw. I also think that a canted saw blade helps to prevent over sawing on the blind (back) side of a cut in things like dovetails and tenons. Having the saw a hair shallower on the back side means that usually when my saw hits the mark in the front, I’m there or one stroke away ont he back side. Japanese saws are still made that way today, just backwards because they work on the pull stroke.

  4. Great video Bob! I’ll add that getting the plate installed in the back is made easier if you plow a groove in a piece of scrap that will hold the opening in the back upright. This especially makes getting things started easier. Question. When making a new saw how deep do you typically drive the back into the plate? On the couple saws I made I drove the plate pretty far down. Looking at your video it seems I seated mine farther down than is needed.

    • Yep, the grooved board definitely helps. Hand screw clamps work good too. I was just being lazy ;).

      When I make my saws, I drive the plate almost all the way into the spine. I base this on an old Atkins back saw I had at the time I made my first saw that was driven almost all the way in (it was done intentionally by the manufacturer on this particular saw). I’ve seen old saws that had the back installed almost all the way like my old Atkins, and I’ve also seen old saws like the Richardson in the video that only had the back installed down about a 1/4″ or so on the blade. I don’t think there was ever any kind of rule. I think it differed from one manufacturer to another and one saw to another. I think as long as the spine is installed evenly along the back of the blade and isn’t driven so deeply that is causes the blade to bend then you’re OK.

      • Thanks Bob, been wondering about this for a while. I was looking back at mine and I did indeed drive them fully down. The one disadvantage here I guess is the saw plate has less depth, but this is of no issue until years of refiling make the plate more shallow. I used parts from Wenzloff and just realized he intends not to have you drive the plate fully down. I was looking at mine and it turns out that I have less depth than I had specified for a finished depth on the order. Looking back at the site he specifically says that they ad between 1/4 and 1/2 of plate to the requested finished depth. My bad….guess I could take mine apart, reseat them, and not drive them down as far, but since it doesn’t seem to matter probably won’t bother. Thanks again!

  5. Bob, thanks for the great video. I have a follow-up question. I have an old Disston with a spine that is U-shaped–the distance from the spine to the toothline is about 1/8 greater at the toe and heel than at the middle. But, when you look down the spine, it’s quite straight and parallel to the toothline (which is also quite straight). Do you think I should take the spine off and try to bend it, or is this too big a job for a novice?

    • Double check the spine by putting a straight edge along the back to make sure it is really the spine that is “U” shaped. I have a strong suspicion that it is the toothline that is concave, not the spine. It is very unlikely that the spine is not straight as they are near impossible to bend in the plane of the saw. While the toothline may look straight to the eye, it is very likely slightly concave, which will give the measurements you reported and will also make for a saw that can be difficult to work with.

      My suspicion is that there is a minor concavity in the toothline, so check it with a good straightedge by resting the edge of the straight edge on the points of the saw with a light behind the saw so you can see small gaps between the points and the straightedge. Don’t check the gullets as these can be straight and the toothline can still be concave if the teeth aren’t all exactly the same size and shape. If the toothline is straight, i.e. the points of every tooth touch the straight edge, then check the outside edge of the spine. Again, I suspect it will be straight as I have never seen one that was bent in this direction. If it is straight, then check the bottom/inside edge of the spine. It may be possible that the spine was not milled parallel, though this is pretty unlikely too. If checking these three areas does in fact show the spine itself to have a concave bend in it, there’s pretty much nothing you can do. I know of no way to straighten this kind of bend without the aid of a large blacksmith’s anvil and a forge. Working iron that stiff and wide will require the spine to be heated to forging temperatures. However, I’d leave it alone if that’s the case. It won’t affect the use of the saw.

      • Bob, Thanks so much for taking the time to reply. The spine is definitely curved–I checked all the ways you suggested. I’m going to take your advice and just ignore it, since the toothline and sawplate are nice and straight. Thanks again!

  6. Hi Bob, nice video.
    Just a thought to ease insertion of the plate into the back, put the plate in the freezer overnight and the back on top of a home heater.

    Due to metal thermal expansion property the plate would become just slightly narrower and the back just a tiny bit bigger, not that much that you could probably measure it but enough to ease the assembly and alignment process.

    Even easier and better joint could be made using liquid nitrogen (pretty easy to find a can those days) on the plate and a torch/oven on the back (just don’t get it red hot so it stay straight). It’s more extreme technique, would work in this case cause the cutting material is inside, you never want to overheat cutting metal material cause it may loose some hardness, but refrigerating it is not a problem. It works very well on any metal/metal tight assembly used all the time in many industry including aeronautics and rockets manufacturing.

    • Thanks for the tip Sam! I’ll keep in in mind for the future in case I run into a spine & plate that give me a lot of trouble. I’ve reset dozens of these though and have never had a problem getting them started and reset. Most of the time, they’re pretty easy to do. Getting them off is typically much more of a problem than getting them back on because they are typically rusted underneath the spine before they come off.

  7. Very useful to me. Thank you for posting. If you are ever looking for ideas for podcasts, I think I would not be the only one to appreciate a demonstration of how you go about hammering a bend out of a plate that isn’t perfectly straight.

    • I did. I sold that vise though. The machinist’s vise works fine for the 3/4″ backs for small saws like dovetail and carcass saws. I still struggle with the 1″ backs, but I struggled with them using the old vise too. They’re just really tough to fold. I’ll come up with a solution some day. Until then, I buy them from Wenzloff, though he lost his supplier for folded backs now so I’ll have to come up with another solution for any new saws.

      • That hurt when Wenzloff lost his folded back supplier. I was lucky that got mine just before he stopped getting them. I keep meaning to email Mark Harrel to see if he’d sell his backs. He sells his split nuts unmarked so why not his backs. The great thing about Wenzoffs backs was that they were dirt cheap…. like the cost of the brass. Maybe we all just need to go in together on a break so we can do a proper job of folding backs on our own. I didn’t get the good results you did folding steel in a bench vise.

        • It took me quite a few attempts and some choice words. The vise was also a pretty heavy model. I don’t think a lighter weight version would work, especially for the 1″ stock. The 3/4″ stock works pretty easily but the 1″ is a bear. Without a heavy hydraulic press brake, I think forge, hammer and anvil is probably the next best thing.

          I’d be curious as to Mark’s supplier of steel backs. Brass is way easier to fold, which is why most saws use it. The steel backs are just plain tough to make. I’ve had the parts for three new tenon saws sitting idle for a long time because I don’t like my current results making backs. I’d finally finish them and offer them for sale if I could find a better way to make the backs, or a source to buy them.

  8. I’m only making wild guesses, but doesn’t McMaster Carr (or someone else) carry something in “soft” steel that would be easy to form or bend? I did a brief look on the MC site, but don’t really know what I’m looking at as far as description.

    I did see some sheets in varying thicknesses that are under “formable” steel, but you would need to take the sheets to a metal shop to have them cut it down to width on a metal shear. The thinnest sheets were 1/8” thick, and I’m guessing that would be way too thick, but maybe there’s something in their catalog that’s “just right” (or elsewhere).

    Maybe there’s a heavier gauge sheet metal (used by HVAC companies) that could be treated to give the proper surface appearance. This type of metal should be reasonably “bendable”.

    • “Soft” is a relative term. The steel that I typically use for saw backs is considered “mild” steel and it is already bent to 90 degrees, which makes the process a little less work, but not easy. It’s still steel though and durn hard. And it should be. If it was that easy to bend, it wouldn’t really stiffen the plate like it is meant to do.

  9. Not sure what happened Bob, but I clicked the “reply” within your last reply box to tack my reply onto yours, but it still posted it as a separate reply. Oh well.

  10. A few years ago I made some brass backs together with a collegue. They were made from 3mm thick brass. It was quite a job! Bending on a big hydraulic brake. Then putting them under a 20 ton press to squeeze them really flat.

  11. Would it be possible to shift the back up a bit on the toe of the saw, without completely disassembling the saw? I have a nice 16″ Sorby with the same malady as yours. But the toothline is nice and straight now, so I would prefer not to mess too much with it.

    • You have to be careful not to buckle the saw blade when doing so. Also, removing the back allows you to check the blade for kinks that may have been hidden by the back position.

      • Hmm. Well maybe I leave it just as it is now.
        Removing kinks in blades isn’t really in my reportoire yet.

  12. Hey! I know that saw! Thanks for the video, Bob – can’t wait to get the saw in the mail (hopefully today).

  13. Great video! Reset the spine on my 16″ Disston tenon saw and it was easy after watching the video. The spine wasn’t engaging the handle mortice and the handle was wiggling. After removing and re-setting the spine everything is great. Took about 45min. Thanks again!

  14. I enjoy gutting my kill as much as the next man; but is it really necessary removing the back from the saw plate? Couldn’t you just knock the back up a bit till its parallel? Thats what I just did today and it saved me a whole lot of pain on an old tenon saw. I was dreading having to remove the handle as the slotted nuts ain’t very slotty no more! – Thanks for the vid – very informative and clear, I saw on backsaw they tell you how to remove the back to check if it is an intended canted saw but dont tell you how to put the peices back together – “square to the tooth line” makes much more sense. – Thanks again – Mark -Ireland

    • Hi Mark,
      Thanks for the comment. It’s necessary to remove the spine if things aren’t all straight in order to assess if the bend is in the back, the blade, or both. If you don’t remove the spine, you can’t assess what’s causing the bend. If the saw isn’t bent then there’s no need to remove the spine.

  15. I am late comer to this post. I would like to know if you ever seen and had in your hands a saw with canted blade. I read everything that was posted on this subject in last 7 years or so and nobody promoting a canted blade yet to show a picture of a real canted saw.
    If you have pictures, I would like to see them. Thanks, Wiktor

    • Yes, I have. I have seen and sharpened a couple of Groves (English) saws, tenon and dovetail, which were canted. I’ve yet to see an American saw that was canted. Unfortunately I don’t have pictures as they were not my saws.

  16. Bob, so from your observation of these two saws you are convinced that the tooth line was canted and not that back of the plate cut at an angle?

    • I certainly can’t say that I’m 100% positive having not seen the saws when new. However, the angle between the toe and heel and back was 90 degrees while the toe and heel were not square to the toothline. This leads me to believe that the toothline was intentionally canted. In a saw that appears canted when it actually isn’t, the angle of the toe and heel with the toothline is square. So it’s pretty easy to tell when the spine is installed improperly on a rectangular blade. Similarly, if the back were filed at an angle instead if the toothline, then the angle of the back would not be square to the toe and heel of the blade while the toothline would. The saws pictured in Nicholson are also noticeably canted with a toe and heel that appear to be square to the back. The Smith’s Key saws are similarly depicted.

Comments are closed.