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.


Understanding Saw File Sizes

Files are sized by length and by their thickness (shorter files are naturally thinner to start with). Then, there is a variety of sizes available in each length... This is what follows after the length, e.g. a 6" 'Slim Taper'.

The common size tapered saw files you will find are 4" through 8" long, and are listed as (in order from widest to narrowest):

Heavy Taper
Regular Taper
Slim Taper
Extra Slim Taper (aka Xtra slim or X slim taper)
Double Extra Slim Taper (aka XXtra Slim or 2x slim taper)
The "taper" is just that, a slight taper to the thickness of the file at the end.  It aids in starting the file to cut properly and evenly.
Here's a quick visual on some of the different sizes - these are essentially the file sizes I keep on hand for my own use:
Smaller files (and the smaller as listed above) have a sharper corner, therefore leave a longer tooth (and a deeper gullet) when you sharpen. This is important in fine-toothed saws, especially those 12 ppi and finer, as the gullet carries away the sawdust... Too round of a corner can also make sharpening more difficult, as the file is just too big to fit in the slot for the tooth. However - too sharp of a gullet can be a problem too, making it easier for the metal to tear at the base of the gullet... That's usually not a big issue, but it's something that Grimshaw thought important enough to mention, so I won't question his authority - and I have seen it, though admittedly only on saws with poor quality steel.

Finding a balance is what it's all about. The file needs to be properly sized for the tooth you are working on - too small and you risk wasting your file quickly. Too big and it' also a waste:

Note that I did not illustrate a rounded corner in the section through the file in the illustration above - it comes to a sharp point, which is not the way it is in reality. It is slightly rounded over, and that is the roundness referred to above. Larger files will have a larger radius at the corners.

Not all of the classic texts nor current web sites agree on the exact size you should use for a given PPI... This may have been because of what was available or perhaps even what a particular company sold or had on hand... Here's the table I put together from a plethora of different sources:

Simonds (a well known maker of saw files) recommends these sizes:

Tools for Working Wood recommends the following sizes:
      Needle File for 15ppi and finer saws ( file is 7 3/4" long x 2nd cut)
      4" Double Extra Slim (12-15 ppi - short stroke)
      4" Extra Slim (11-13 ppi)
      5" Extra Slim (11-12 ppi)
      6" Double Extra Slim (10-11 ppi)
      6" Extra Slim (9-10 ppi)
      7" Extra Slim (8-9 ppi)
      7" Slim (5-7ppi)
You can see it varies from what I put together, and it's just more evidence there is no hard fast rule.  There is no reason to doubt their recommendations - they have a pretty good grasp of what saws require, and following their advice won't steer you wrong.
For reference - the difference between TPI (Teeth Per Inch) and PPI (Points Per Inch) is just in where you measure the tooth - from point to point or from gullet to gullet. PPI is point of tooth to point of tooth. Some text refer to TPI, while others refer to PPI, so I've included both in the table above.

I hope that helps clear things up a bit... I'm sure it's way more than you wanted to hear...

The Art of Saw-Filing: Scientifically Treated and Explained on Philosophical Principles By Henry Wells Holly
First published in 1864, still in publication
ISBN-10: 1408667150
ISBN-13: 978-1408667156
Available at Google books here:
You can download a PDF copy from Norse Woodsmith here:
Grimshaw on Saws
ISBN-10: 096180887X
ISBN-13: 978-0961808877
Available on Google books here:




Hi Leif,

Yet another simple distillation and compilation useful information!
Very much appreciated- keep it up!


I have been around the saws of various types but the teeth per inch never entered into the equation I used a fine tooth on paneling or plywood,A rip blade for ripping abd a cross cut for you guessed it.thehandtoolco[dot]com


Hi Leif and thank you very much for these informations.

I would add one: the correct method to measure the pitch is to measure 10 teeth (or 10 points) and then calculate 10 divided by the measured length.
P.S. The PDF version do not work.


Thanks, Andrea... 

The pdf version should work now...

"the correct method to measure the pitch is to measure 10 teeth (or 10 points) and then calculate 10 divided by the measured length."

I've never heard this... it's very interesting.  However - I would postulate that it's not much different than comparing six to a half-dozen.  For instance, if there are exactly 13 points in one inch, using that method the result (by measuring 10 teeth and dividing that distance by 10) the result would be 13.2.  So, in my opinion I would say either method is substantially correct and you cand and should use either method that you either believe in or that makes you feel comfortable... 

In the end,  we all should remember that it is a simple hand saw and we should not over-think it too much....

Thanks again, Andrea - just shows you can learn something new every day!



You're right, Leif.

I used the wrong words, sorry.
The method I described is not "correct" one, but the "standard" one.
And it is the same for determining the pitch of a screw or of a bolt when you have not a thread pitch gauge.
But, just for curiosity, how you calculate the value of 13.2, in your example? To me it results 12.987 or 13,004 according to the approximation. Yes, it is a wrong result, but it is much closer to 13 than your.
Best regards from Italy,


I used AutoCAD to measure the distance of 10 teeth in a 13 lines per inch diagram to come to the amount of 13.2, so I am pretty sure that is accurate...

I think I have heard of that method being applied to screw sizes... However, you might be interested in this article from L.S. Starrett on how to determine the number of threads per inch there are in a screw:

"Determining Threads per Inch (T.P.I.)
The most simple and common measurement is determining the number of threads per inch. This can be done with an ordinary steel rule. To do this, line up the one-inch line graduation of a steel rule with the crest (point) of a thread (see right) and count the number of crests (points) over that one-inch thread length. That number is the “threads per inch.”

"If the overall thread length is shorter than an inch, count the crests over only a half-inch, then multiply the number by two. For example, if you counted six threads over a half inch, then 6 x 2 = 12 -- there are twelve threads per inch. If you counted five threads over a quarter inch, then there are twenty threads per inch, and so on."

This is essentially the method I've shown above...  So while it might not be the "standard", I believe it's safe to say it is an accepted method.

Thanks for commenting!



I searched in my books, in my files and on the web a confirmation to the method I described, both for saws and for thread, but I've not found it.

So, evidently, I have dreamed it.
At this point, instead of calling it correct or standard, we have to call it a imperfectly creative approach. :-)
Best regards,


Ah, now I understand where 13.2 comes out.

It depends on the accuracy of the instrument used.
Instead, previously I made only a theoretical calculation... Sorry.


Hi, found your imformation very helpful. Is there a standard for the height of the teeth in relation to the bottom of the gullit and the tpi. If there is do you know of a formula to calculate it 



Hi Estonianed!

There is no "standard" depth of  a tooth, really - it is a function of the geometry of the file and of the number of teeth per inch there are. A saw file is a 3 sided triangle so the depth would be the result of length that is between each tooth with the teeth running down at a 30 degree angle between the points (algebra/trigonometry was never a strong point for me, but I'm sure there is a formula that would work to figure that out).  You must then also take into account how much roundness there is on the corner of the file being used, which varies on the length and slimness factore.  An extra slim file has a sharper corner than a regular slim file for example, and the same is possibly true for a 5" file over a 4" file - one must also consider the manufacturer as well, as different manufacturers (not to mention an individual saw filers methods) might vary from one another a slight amount.

If you are interested, you might want to check out Brent Beach's excellent treatise ont he subject.

Japanese saws use a much sharper angled file and the gullets are much deeper as a result.  There is an interesting post on the Giant Cypress blog that might interest you that shows some of the geometries involved in both western and Japanese saws.

Hope this helps...



Hi leif,   thanks for the prompt responce and the info. I am going to brush up on my maths and check the links you recomended after I finish this reply!