Question about resawing

I have a question for you all. A few years back I built a Roubo-esque frame saw in an attempt to conquer hand tool resawing. Frankly, I’ve not had great success with it. It’s 4′ long, and sports an .032″ steel blade sharpened at 2 ppi. Rake angle is fairly low which makes the blade difficult to start. Once in the kerf, this thing eats wood fast. The problem with it is that it’s difficult to control.

I’ve made the easy fixes. And this wasn’t my first rodeo with frame saws. There are 2 pins attaching the blade on each end. The pins are strategically placed so that the tooth line sees a fair bit of the tension.

The problem I seem to be encountering is that the blade twists, seeking the path of least resistance. You can hold the line on your side, but you may loose it on the far side.

To help reduce the twist, I removed the upper pin on the far end, driving all the tension through the teeth (just above the gullets really). This helped somewhat, but not enough. I also tried extra tension. But I may be too mechanically sympathetic. The tensioning bolt is designed to put a stick thru to produce GOBS of tension. I could try that.

Other problem I’ve encountered is that when sawing veneers, the saw naturally gravitates to the weak side. Roubo’s illustration seems to indicate they began the cut in the sap and moved toward the heart with each successive cut (don’t have the image handy. If anyone’s got it and can email it to me, I’ll post it here.) Not sure how you stop this. Is it possible the saw in the image is set on one side only?

If you have experience with one of these saws, I’d like to hear it. The only other saw I’ve seen like mine is the one in the Hay Shop. Maybe Ed Wright has been through this and they’ll blog about it. If you’ve got any advice for me that you don’t mind sharing here, I’d appreciate it.

Adam

30 thoughts on “Question about resawing

  1. gavinmac

    This is a topic which was a near obsession with me a couple years ago. I thought that if the problem of resawing could be solved, an entire panorama of pure hand tool woodworking would open up. Jointing and planing stock to thickness is one thing… resawing another thing drenched in sweat. Anyway, what I was using was a stumpy version of the saw pictured, with a blank blade cut from a standard panel saw(27″ I think). It was very slow, up to 45 minutes a cut to resaw a 5′ long, 12″ board. The pace slows dramatically after the sawdust can’t entirely exit the cut. The blade was 2 tpi, with a neutral rake angle.

    Several experiments followed:

    I attempted to cut my own teeth in a loggers crosscut saw I bought at a flea market, but even after tempering the steel several times, the metal was just too tough. Probably just as well, as the blade was much too thick.

    Secondly, I obtained a length of bandsaw blade from the local sharpening guy. The blade was for an old Hitachi resaw bandsaw, and was very thin, about and inch and a half wide, and stellite tipped. This seemed very promising as every other tooth was tipped, and were rakers, not set to one side or another. I think that might be part of the equation: teeth ‘upset’, and not set with fleam. Anyway, the blade cut amazingly fast, but sadly, was impossible to control.

    Lastly, I did some reading on the subject, and formed some unscientific theories… 20-22 guage steel is about the thickness desireable, with 3 tpi. Up to roughly 6 degree forward tooth angle, with a ‘wolf’s tooth’ pattern.
    The wolf’s tooth pattern coupled with the upset teeth, could yield results… I think. I know, I’m ranting. It’s been a long time since I thought about the subject, and I can’t remember my logic regarding the tooth geometry, nor could I defend it.

    One technique I did get good at with the saw I used was to twist the saw to regulate the rear of the cut, and move the saw left or right to regulate the front of the cut. Sort of a compound angle to keep it tracking straight.

    Also, I drafted, and made prototypes of a gibb used to tension the blade. It was similar to the type of gibb used to secure sawblades to scandinavian log mills which use multiple reciprocating blades. The blade is essentially bolted to a plate (the gibb) with two or three bolts. The other end of the gibb, which attaches to the frame, is basically a large square cut-out (this is made of 3/16 mild steel). The end of the cut-out fits over a kerf cut on the outside face of the saw frame member facing away from you while sawing. The gibb on the operators side is similar to the one on the opposite side, except instead of a closed box, it has the end removed, and the sides of the cut-out slide in kerfs cut on the top and bottom of the saw frame member facing the sawyer. A bolt attached to the front gibb with a pin goes through a hole in the front of the same piece, with a large wing nut to tension. Clear as mud? This type of attachment was very strong, and enabled high tension.

    Anyway, In the end I made several pieces of furniture entirely without electricity, and learned more in those pieces than anything before or after. Plus I gained a huge appreciation for craft before the industrial age – and electricity in general!

    If you ever conquer the mysteries of the frame resaw saw, please post! Thanks.

  2. Ryan Mails

    I do all of my resawing alone with a 4′ framed veneer saw. I saw mostly pine and walnut, so I have the 2 pt blade filed at a pretty savage rake angle. I tend to agree that twist in the blade is inevitable.

    My blade is attached with single pins through posts in the frame ends, in the same manner as bow saw. So, of course, the blade twists, but I try to use this to my advantage. In sawing, I adjust the degree of twist to compensate for the tendency of the blade to wander. Basically, if the blade is drifting left on the back side of the cut, I twist the rear post a bit to the right.

    Its crude, but it works for me.

  3. GregM

    If it’s Roubo we’re trying to emulate, then to Roubo we must go …

    I don’t have access to a real copy, and I don’t speak French either, but a scan of the plate in question is available here: http://toolemera.com/bkpdf/roubomichaudBK.pdf (pl. 278, page 12 of the PDF). Figs 6 and 7 show detail of the saw blade, which has no set at all. Some of the other figures seem to show detail of how the blade is fixed/tensioned, but the image isn’t that great and there’s no explanatory text either. I can’t quite make it out.

    Still think this may be frustrating as as a one-man operation though …

  4. Bruce Jackson

    Adam,

    I was looking at Anthony Guidice’s Seven Essentials of Woodworking in the middle of my spring cleaning (I know, my calendar’s weird, but that’s what you get for living well south of Georgia and about in line with the Bahamas). Guidice seems to think that you don’t need set if you’re cutting dry, as opposed to green, wood. Come to think of it, I wish my band saw blades I use for ripping (3 tpi hook) don’t have any set – it would make my life a lot easier if the blade track my straight way instead its own whimsical path which I have to plane to straighten out. In any case, Guidice also suggests, and I now agree, if you have set, one side has to be dead-on even with the other. That way, you don’t have to call the ex-Governator to come tighten your bow / frame saw for you. Hope that helps.

  5. Adam Cherubini

    Gruss Gott Pedder, vielen dank. That saw looks remarkably like Roubo’s in style. I’m having trouble making out the fixings. Looks like some sort of plate and pin arrangement?

    Honestly, I’ve tried both ways, by myself and with various unskilled helpers (sorry guys).

    I suspect that the end fittings may be a red herring. Even firmly clamped the blade can (and does) still twist between them. I’m guessing that blade width, thickness and fixings aren’t the answer. It may well be that the blade is simply too agressive (too much rake) and that in a single stroke, it moves so much wood that it finds its own way. Keep in mind that when we saw thick stock, we have very little control over what the blade does inside the wood. You can move the saw around on either side, but that doesn’t change what is happening inside.

    If I haven’t mentioned already, I believe the Hay shop saw is 6ppi (mine is 2). When I saw it go last, I thought it was slow moving. It will be interesting to hear what Ed and Bill say about it. They’ve promised us a blog on that subject.

  6. Pedder

    Hi Adam,

    do you saw with another woodworker or do you try alone? These things are made for two man.

    This is the URl of a picture I took in Skansen Museum (Stockholm Sweden) http://hw.roesch.de/Bilder/B2577.jpg

    You can see the fixing of the blade at the right end. This will ceep the blade from turning.

    Best regards
    Pedder

  7. Adam Cherubini

    Hi Nick,

    No. I think I know where you are going tho. I think that would happen with only 2 pins both located far from the neutral axis. I think I could stretch the lower edge and end up with the top edge shorter and "doinking" or "oil caning" back and forth. I’ve had this happen to rip saw plates I was making.

    All the old saws seem to have 2 pins on at least the near side. Some have end fittings that really spread these out (like the one at the Hay Shop).

  8. Adam Cherubini

    Nick,
    When you tension a wide blade with a single pin say, the blade rotates about the neutral axis. Adding tension doesn’t help may even make the situation worse. There are other ways I can react the moment and I have tried a couple of them. The part that holds the pin(s) has a narrow slit in it to help stop the blade from rotating.

    And yes Bill, I have to shim that support piece or it will rock, just like yours. I tried to make a tighter fit there, but that piece needs to slide on that arm and I guess the arm shrunk a little. My plan was to fit it with bras shim stock, but I haven’t got ’round toit.

    I won’t give up guys. There’s a good deal left for me to learn and your comments have been helpful and encouraging.

  9. Nick Webb

    Adam,

    If you put the tension anywhere other than along the neutral axis of the blade then it will bow. You can see this if you take a wide rubber band and stretch it by pulling along one of the edges. That edge will curve outwards, putting the other edge into compression. But rather than compress the far edge will bow one way or the other. So I think the answer may be to move the pins nearer to the middle of the blade, rather than further away.

  10. Rick HIll

    Hi Adam
    I think you are using Too course a pitch, and looking at the length of the
    frame in the picture I bet if you shortened the frame and blade the a lot
    of your resawing problems will go away. It would be like lowering the
    blade support on a bandsaw. All a frame saw is, is kind of a bandsaw. You
    did’t mention the width of blade? This would make a big difference in resawing by adding stiffness.
    Rick

  11. Adam Cherubini

    Thanks Greg.

    Warren, I’ll have to spend more time with this saw. I could try a less aggressive blade. It could that because this saw cuts so quickly, you can ruin a cut with a single stroke.

    Still, I feel as though it would be very tricky to saw veneers as shown in the picture above.

  12. Shannon

    I’m hesitant to say there is an engineering problem with your saw Adam as it sounds similar to the ones we use at the Steppingstone museum and I remember your post about it and thinking it was a well built saw. It has been said above, but I’ll second it that the only time I have ever done this with a frame saw it was a two man operation to keep on the line on both sides of the board. The only way I have found to do it well by myself was using a regular hand saw and of course flipping the board. Good luck

  13. W Mickley

    Adam, many of the things mentioned are helpful, but I suspect you are too much in a hurry to correct the cut when it goes off. If you lean to much to one side, the blade actually drifts the opposite direction inside the cut. It is like a vibrating string that is dampened at two nodes. Then when you think you are straightened out, the wave inside is still pulling the blade. I suggest using a very gentle leaning to correct the cut and think about making the correction over the next six inches instead of the next two. Back in the 20th century, people used to suggest cutting down just one side at a time, but I find straight across yields a better board and this also seems to agree with historical evidence.
    Warren

  14. James Watriss

    A few short anecdotes…

    We resawed a huge (4" thick) plank of mahogany back when I was at North Bennet. There was a severe amount of blade drift… enough so that when we flipped the thing end for end, we sawed about a foot past the end of the previous kerf and couldn’t figure out what was up. The blade was still coming through the kerf on each side, but inside the wood, it was drifting to one side on each cut, so that when we finally pried the whole thing apart, it looked like a Z… the blade had gone around the middle of the plank.

    Re: veneer sawing… If the saw is drifting towards the thin side, my guess is that because that side is more prone to vibration than the thick side, those vibrations are causing the wood on that side to vibrate against the set teeth. So the added cutting going on from the increased vibration against the side of the teeth could (my theory) cause that side of the cut to widen, and provide a path of less resistance. When I saw thin slices (1/32"-ish) from boards with my ryoba, I don’t have the same problem… but (from what I understand) the set on those is minimal: the saw plate is scraped in the middle to provide clearance instead. So… zero set, or stone the sides of the teeth enough that the points are completely smoothed. (ie, the teeth still have the chisel shape, are set wide enough to provide clearance, but can slide through the wood without scraping the sides of the kerf.)

  15. Mark Schreiber

    I usually keep flipping my work around, sawing at about 30-45 deg angle. When the saw begins to cut on the far side, time to flip. Also, I assume you have no set to the teeth. If there is set or a slight set more to one side than the other can lead a saw to drift.

  16. GregM

    Doesn’t Roubo’s illustration show *two* people operating the saw? Maybe that’s part of the secret.

  17. Josh B

    Bill,

    I happily stand corrected. I always assumed that flipping the board like that was kind of cheat to make up for a less than optimal saw or technique. Nice to know it’s pretty much how it’s always been done in a one man operation. I still find it cumbersome actually flip the board though. When you resaw how do you hold the board? I’m just clamping it in my leg vise at the moment.

    Adam,

    I’ve always sawed my tenons by flipping like this – it’s worth the slight interruption of flipping it in the vise for accuracy gained. Guess I need to start thinking the same way about my resawing. Thanks for starting this discussion, I’m learning a lot.

    Cheers,

    Josh

  18. Bob Rozaieski

    The only ones I’ve ever used had blades made from 1-1/2" bandsaw blade. They had the same problem as yours. Way too easy to twist and get off line. So I would saw out the corners with a tenon saw, then finish the cut with the frame saw, occasionally switching sides to attempt to keep a consistent cut line. Sometimes I got lucky, but most of the time it went way off line. I’m sure it had way too much set, and not enough tension. The user probably contributed just as much to the problem as the saw though. Guys like Frank Klausz and Mike Dunbar can saw really well with continental style frame saws. They may not be resawing, but I’ve seen them rip some pretty thick stock. So I’m sure there’s quite a bit of technique to it as well. Apparently I don’t have it :).

  19. Adam Cherubini

    About the double pin connection- All the period examples I have seen have double pins on at least one end. In Jay Gaynors book there’s a saw with a tapered blade that has a single pin on the far end (IIRC).

    Since the pins just rest against the channel section that holds them, I’m unsure how much moment there is. I can say that a single pin in each end centered in the blade will allow the blade to easily rotate about the line between to the two pins. You’ll get very little tension on the toothed edge. The idea of adding a moment isn’t necessarily a bad one, however. Not sure if bandsaws do this.

    I agree with flipping the stock, Bill, and I sometimes flip tenons too. Not ashamed of either. IIRC the Hay shop’s saw like mine is fairly fine- maybe 6ppi? My blade is 3" deep (measured from Roubo). I don’t believe a deeper section would be better, guide the saw etc etc. It would be just harder to tension. I think blade depth has to do with bending front to back.

    This may be an engineering problem or it could be a user error. I’m happy to entertain either!

    Adam

  20. Bill Pavlak

    Adam, we’re looking forward to doing several blogs on resawing and on our large frame saw in general, but it might be a month or so until we get a chance to post on that topic.
    As for our saw, twist is certainly an issue. Generally, we put a fair amount of tension into the blade, but not too much. I’ll usually rely on a stick to put the last bit of tension into the blade. Of course, the act of tensioning usually results in blade twist. We check for this with winding sticks and then adjust the iron mechanisms that hold the blade in place (I have no idea what the proper name for those is?). On our saw there is some play between this mechanism and the frame (is it that way on yours?). We usually remove the twist with a few gentle hammer taps on the mechanism and then lock it in place with wooden shims/wedges. We’ll typically check for twist again periodically as the sawing progresses. (I’ll try to e-mail you some photos of the above in the morning since that’s a clumsy explanation.)
    We’ll typically use the frame saw on stock over 10" wide (I’ve done 20" successfully a few times and I’m certainly not the most experienced resawyer in the shop). Generally we make this a two person tool, though occasionaly we’ll use it solo. Contrary to Josh’s comment above I don’t think that switching sides down the length of the board is something to "resort" to, but a sound practice. All of us in the Hay shop do it this way and there are plenty of 18th century saw marks to suggest they did the same. Perhaps switching sides could be judged inefficient in cutting tenons, but I think its the best aproach in resawing. It allows you to only advance the saw along the line you can actually see and it balances out any irregularities in the saw and in the motions of the sawyer. Oh and it should also result in a smoother cut which will reduce the time spent planing after sawing.

    We have two pins on either end of our saw and I’ve never noticed that to cause any problems. As for set, I prefer as little as possible for resawing, which allows the blade to really start to guide itself in the cut. We tend to view resawing as a race that the tortoise wins: slow and steady (also a really light grip on the saw). By slow and steady, I don’t mean inefficient work though.

    Sorry for the rambling comment! I’ll make sure the others guys see your blog in the morning and see what they can add.

  21. Steve

    One thing that can lead to blade drift is the gullets filling up with sawdust, which obviously becomes more and more of an issue the wider the board you’re trying to resaw. Aside from increasing the gullet volume by modifying the tooth shape to resemble a hook- or skip-tooth bandsaw blade, I think about the only thing you can do is let up on the pressure so that the blade doesn’t "eat wood" so fast.

  22. Josh B

    I have a similar frame saw but smaller, 24", and I’m using a 3 1/2 TPI, 3/4" bandsaw blade. The blade is captured and pinned into slotted bolts on either end of the saw and requires a terrifying amount of tension to work halfway well.

    I wish I could say that resawing is all beer and skittles with my saw and technique but like you I find resawing with it to be a frustrating chore most of the time. I have to resort to the trick of sawing down from the corners and flipping the stock in the vise every few inches so I can saw from the other side. Otherwise the far side of the cut starts to meander off on it’s own personal voyage of discovery blissfully ignoring my intentions. The one bright spot is that I’m at least able to get good cuts made this way, but it’s slow and annoying enough that I don’t do it very often. I’m curious to see what tips your readers will have, hopefully someone knows the secret for making hand resawing awesome 🙂

    -Josh

  23. David Cockey

    2 pins on each end? Is that copied from other saws known to work or you innovation> I wonder if the 2 pins are part of the problem. You mentioned removing the pin from one end and that helped some. Try removing the pin from the other end also. Two pins on an end will result in a moment on the blade. That moment could result in an instability of the blade which manifests as the blade trying to twist. So try one pin per end.

    Also, my guess is the pin at each end should be close to centered in the width of the blade, not offset towards the teeth for the same reason.

  24. Bob Easton

    My frame saw isn’t as long as yours and is 5 tpi. I have used it to resaw a lot of long boat lumber, white cedar. I have not used it with any hardwoods.

    Early on, the drift was very pronounced. Stoning the set on the side it drifted toward helped a lot. Stoning helped me correct more than higher tension did.

    It still drifts gradually, and my answer is to saw a bit (usually 10-12 inches), then flip the lumber and saw a bit more. The results were exactly what I needed. BTW, 12 – 14 inch boards that are 12 – 15 feet long use up about one Snickers bar each.

    I have a tutorial about my boat lumber resawing which you can find at the link provide by my name. (or: http://www.bob-easton.com/blog/?p=475)

  25. Brian

    Probably tension is your main issue here–I found that I needed an awful lot for my frame saw to work. The blade would bow very slightly, and start tracking the grain. On flatsawn stock, this was a big problem; there’d sometimes be a little hump that would track the shape of the rings.

    The pitch of your teeth sounds about correct; mine is finer and it isn’t as fast.

    Also, how deep is the blade? The deeper the blade, the straighter it will track in the kerf. As an extreme example, Japanese sawyers used enormous blades to resaw (see Odate’s book). In the absence of that, you may need to alternate the side you’re sawing from.

  26. Dave Anderson

    Hi Adam,

    While I don’t have personal experience in resawing with a frame saw, I spoke with Marcus about this a few years ago. He and Ed always did the sawing as a 2 man task and tended to work corner to corner to keep the saw on track. Their saw was much finer toothed (sorry, can’t remember the ppi) and it was slow work even in the walnut I saw them cutting.

    I realize that this doesn’t help you much, but maybe you are going to have to go with a finer PPI and just suffer that the work will be much slower and there will be more cleanup work to do on the veneer slabs yu cut off.

    Best regards,

    Dave Anderson NH

  27. Mike Witteveen

    Adam,
    Does the saw consistently pull to one side only? Would twisting (or winding) the blade very slightly opposite this counter that reaction? I don’t think it wold take much twist to affect a cut, after all,that saw if huge!
    I don’t think tension alone will account for it, like a catenary in a long wire, you can’t get it perfectly straight no matter the tension. I think your idea of saw set is on the right path.

    Mike

Comments are closed.