Be sure to visit the Hand Tool Headlines section - scores of my favorite woodworking blogs in one place.  Also, take note of Norse Woodsmith's latest feature, an Online Store, which contains only products I personally recommend.  It is secure and safe, and is powered by Amazon.


Testing Five Different Back Saws

The following was taken from a conversation I had on an internet woodworking forum.  Some folks there thought it might be nice if I made it available for reading here - so I've edited it slightly to fit into an "article" better.

Talk of the different types saws got me curious, so I went out to the shop and grabbed 5 different saws - each a little (or a lot) different from each other - to try a little test with. This is completely subjective, and I have no doubts that others could come up with completely different results - so take it for what it's worth.

Here's the saws:


  • A Lee Valley standard dozuki in the center - still quite sharp with it's factory sharpening;

  • My 100+ old 12" 'warranted superior' (13 PPI crosscut) - the upper left saw - that I've used for close to 20 years without sharpening;

  • A near 100 year old 14" Jackson (11 PPI crosscut) on the upper right that's dull as a post;

  • A post WWII 10" (12 PPI crosscut) Disston #4 in the lower right with the factory sharpening still quite sharp;

  • And finally one of mine - a 7" rip (15 PPI) on the lower left of the photo above that I've used quite extensively for the last year or so without sharpening.

The test - make five deep cuts quickly - mostly to gauge the 'feel' of each saw as it cuts, but also to see how smoothly they cut, if they wander, how quickly they cut, and leave as little material between cuts as I could. Here's what I did:

The results - from top to bottom - the top is my saw, then below it the dozuki - continuing with the Disston, the warranted superior, and finally the Jackson. Here's a closer look:

Be nice, now  that's really close up, and I wasnt' trying to make perfect cuts! The wood is 1/2" thick hard maple.

Not surprisingly - the most accurate cuts were with the old warranted superior. It's the saw I've used for so long, that it's become like an extension to my hand. The second most accurate was the NWS saw - followed by the dozuki, the Disston, and finally the Jackson. The dozuki could easily have scored higher had I been more practiced with it... but I don't use it that often, so it feels a little awkward in my hands.

The fastest cutting was the NWS, followed by the Warranted Superior, the Disston, the Jackson, and the slowest was the dozuki. That surprised me - I thought the dozuki would fair better, but it was by far the slowest cutting saw of the bunch.

The smoothest cutting was the NWS, followed by the Warranted Superior, then the dozuki, with a tie for last place with the Disston and the Jackson, each of which made exceedingly rough cuts.

The easiest starting was the NWS, followed by the dozuki, the Warranted Superior, and the last two were abysmal.

My favorite? Still the Warranted Superior.  Surprised me, I thought my own saw would score higher. It seemed to perform better, and felt better in my hand - but I still preferred the old one. Old habits die hard, I guess! 

My least favorite was a toss up between the Disston and the Jackson. Note, though - I have no doubt that I could raise their status significantly simply by sharpening them. But this test was also to test different levels of sharpness, and their affect on the cut - which I would say is pretty significant. A dull or poorly sharpened saw can obviously be a big source of frustration.

What else I learned from it - I'd say familiarity with the saw is probably the most important factor. Also, just because the saw was sharpened at the factory doesn't mean that it's sharpened well - the Jackson (dull from use and abuse) scored almost evenly with the Disston, which still had the factory grind on it.

It also seemed to matter little whether the saw was sharpened rip or crosscut.

I also see that the NWS saw wants to wander a bit to the right. A little tuning - a swipe with a stone on that side - should take care of that, though.

I think I could have done better with the dozuki with practice - but since I am most comfortable with the western style...


Curious, do you think having a push saw with 22-26 TPI would have any bearing on the results?

Well - here's a couple more, done VERY quickly - a 17 PPI gents saw and a Zona saw:

The gent's saw isn't really a fair test as it's not reversible, and I have it set up for a *very* specific task - one where the width of the kerf must be *exactly* .024", where the cuts are only about 1/4" deep. I think it's a Stanley, but I'm not sure... It does pretty well, though I also have to add that I am also very biased against these types of saws - even if my hands weren't as screwed up as they are, I hate the "push a saw on a stick" type, as it just isn't natural - a dozuki feels much more natural to me. Here's a close up of the cuts:

The cuts were done probably too quickly to be very accurate, but close enough to give me an idea of how they were cutting compared to the previous 5 saws.

The zona saw is really too fine for cutting 1/2" thick stock, IMO - it feels more like you are grinding a saw kerf rather than cutting it. But for *very* fine work, especially on small stock (around 3/8" to 1/4" or so) they are indispensible.

Overall, accuracy is quite good with both of these saws, but cutting was slow.

This is by no means even close to a scientific test, just a way for me to see the differences of the saws when used side by side, right after each other - just for fun. I just thought I would throw it out there to see if it might interest anyone else.

I still think the 2 biggest suprises to me was first that the dozuki was the slowest cutting - yes, it has more teeth per inch, but then again, it also has the thinnest blade. And - the NWS saw was the fastest cutting, and is 15 PPI. The second biggest surprise was that I still preferred using the old saw over the new one. Upon further reflection, one reason I like the feel of Warranted Superior saw better than the NWS is because the blade is thicker, and doesn't flex as easily. 

Now, a bit of maintenance - those two saws really need sharpening.  So, I did - the Disston I converted to rip, and the Jackson remained crosscut. Then, I tried the test over again, this time in mahogany:

The top is the NWS, then downwards it's the dozuki, the Disston, the still unsharpened WS, and finally the Jackson.

Results - this time, the dozuki performed better all around than the NWS, though it was still the slowest cutting saw. The NWS wanted to bind a bit - I suspect there isn't enough set on it for the softer wood - and consequently, more set on the dozuki helped it out here, where it hurt it on the maple earlier.

The WS performed adequately, but it also suffered from too little set, and wanted to bind just a bit with a tendency to wander to the left in this wood.

The real surprise was the freshly sharpened Disston. Fast, and straight. Easily the best performer on this board - and it also performed well in the maple board from above as well (not shown to preserve *some* bandwidth).

The Jackson needed quite a bit of work to bring it back, and I only gave it a touch up with the file. It has a *lot* of set, and I think it might need another session with the file to get it right, and maybe give it a steeper rake angle. It didn't want to follow the line too well at all with it's new sharpening. But since its a crosscut saw, I decided to try that - and it performs pretty well at that task. It is a fairly coarse saw, at 11 PPI, so maybe it's just not destined to do fine ripping work. Here's the crosscutting:

In order from left to right in that photo - the Jackson, WS, Disston, dozuki, and the NWS.

Of course, there's another story behind:

You can tell which one was the dozuki, huh? To be fair, look at the top photo, as being a pull saw, the tearout is on the face of the board. Of course, you can easily see the rip-filed Disston tore the crap out of the wood, where the coarser crosscut Jackson did pretty well... better than the NWS.

I must say, that the dozuki really did shine on the crosscutting. I guess that shouldn't be a surpise, as that's what it was designed for - not ripping. The order of saws used was the same as above (obviously in reverse on the back side shot), and you can see the damage the rip filed Disston did. Normally, I would have either a backer board or some masking tape on the back side to reduce the tearout, but I thought I'd like to see how they worked without.

The WS wanted to wander badly to the left here, so I let it. I guess I will sharpen it. Finally.

What I got from this round - the woods you work in make a big difference in how the saw cuts, as well as how well the saw is sharpened. Both made a huge difference here.

not all dozukis have the same cutting rate

Yes - good point. The dozuki is all by itself in my shop - I don't own any other pull saws. There's a whole world of Japanese saws whose designs I'm not privy to, nor claim to know much about. As I've often said, I am a fan of the western style handsaw - probably mostly because it's what I was trained on. There wasn't too many Japanese tools available out on the prairie where I grew up. It certainly isn't because of any lack of quality on the part of the japanese tools.

On personal preference, I can say only this - whatever works for you is what works best. The idea (there was a point?) behind this exercise was as much to see the differences and similarities that exist between saws of both similar and of dissimilar design, the effects that simple maintenance can have on performance, and how familiarity (or lack of it) with the particular tool you are using can have on your work.

Besides - I needed the practice!

Thanks for reading!