Saw Filing - Theory

Introduction
The Jig
Saw Setting
Saw Filing
Some Theoretical Issues
Sloping Gullets
Copyright (c) 2002-15, Brent Beach

A couple of topics here that do not fit well into any other page.

File Direction - Do you file with the tip of the file pointing toward the saw handle or the saw toe?
Metallurgy considerations - Modern metallography has discovered that files are among the worst ways to shape steel. Do we have any alternatives.

File Direction

The Options

There are two different ways to file a saw. Method I, with the file tip toward the saw toe. Method II, with the file tip toward the saw handle. Different authorities prefer one method over the other, without proving their way is superior. Here are two with exactly opposite opinions.

From Charles Johnson's Saw Dictionary - work with the file tip toward the saw toe.

Always begin at the "heel" of the cutoff saw, filing on two teeth, running the file level and straight, giving the bevel and pitch most suitable for the work and timber. File with the front of the tooth, giving it the feather edge, and against the back of the tooth, giving it the smooth inferior cutting edge.

Then file back on the other side of the saw from the "toe" to the "heel," taking care not to file too little or too much, filing the large teeth as much and the small teeth as little us necessary, to make the teeth and spacing even sized, bringing out most of the teeth sharp and to an even size.

From the Disston Lumberman Handbook - work with the file tip toward the saw handle.

When the saws are manufactured the sharpening is usually done by filing every other tooth from point to butt or handle end. The filer, standing at the point of saw holding the point of the file inclined toward the handle, works in that direction and against the front or cutting-edge of the tooth set toward him. After filing every alternate tooth, maintaining a uniform angle and bevel, the saw is reversed in the clamp, the filer changing his position accordingly and proceeding to file the alternate teeth on this side, also beginning with the first tooth set toward him.

By this method the operator is in better position to see if the fronts of the teeth are being brought up keen. It also does away with the "feather-edge" thrown up on the cutting-edge of tooth which is done by holding point of file toward point of saw, or in other words filing against the set of the back of the tooth.

The important distinction is whether you file with or against the set on the front of the tooth.

With and Against

In the following this distinction is important, so I will introduce a bit of terminology to make the distinction clear.

The file sits in a gullet with one tooth on either side. Since teeth are set in alternate directions, the tooth on one side of the file is set away from the filer and the file is moving in the direction of the set on that tooth. I will call this the with-tooth. The tooth on the other side of the file is set toward the filer and the file is moving against the direction of the set on that tooth. I will call this the against-tooth.

So, every gullet you file there is both a with-tooth and an against-tooth. The difference is whether you are filing the front of the with-tooth or the against-tooth.

The Burr Argument

The first argument is based on the burr left by filing. As the file scrapes metal off the tooth, it leaves a small burr on the far side of the tooth. This burr is on the inside of the tooth if the tooth is set toward the filer - the against-tooth. This burr is on the outside of the tooth if the tooth is set away from the filer - the with-tooth.

This is important because the inside of the tooth is not involved in the cutting action. Only the outside of the tooth is, and even there only the tip of the outside of the tooth.

If you use Method II, the burr appears on the inside of the front of the tooth, on the outside of the back of the tooth. If you use Method I, the burr appears on the outside of the front of the tooth, on the inside of the back of the tooth.

Why does it matter where the burr is? It matters because these burrs are broken off by contact with the wood as you saw. The breaking off could produce a degraded cutting edge. If this is true, then it makes sense to have the degraded edge on the back of the tooth rather than the front of the tooth (for saws that cut on the push stroke). This suggests that Method II would provide a longer lasting sharpening.

I will take some microscope pictures of teeth filed each way and include them here.

The Visibility Argument

Disston suggests that by filing from toe to heel against the front of the tooth, the filer is always in a better position to see the tooth fronts.

Charles Johnson says: "[a right handed filer, working with the file handle toward the saw handle, gets] ... the better view of the teeth points on the second filing [saw handle on the right], and is best adapted for rapid work, while the [saw handle on the left during the second pass] gives the better view for accurate filing of the teeth.

This is important - the filer must be able to see the teeth clearly, in particular the points and the flats on those points created by jointing, with visibility most important on the second pass.

There are 4 permutations of the two saw positions (handle on the left or right) and file positions (file tip pointing toward saw heel or toe). A third factor is whether you are right or left handed (left handers please reverse all terms in this discussion.) Each complete saw filing involves having the saw handle on the left during one pass, on the right during the other.

In these drawings of a saw as viewed from above, the zig zag is the gullets, the lines going to the edge mark the cutting tips of the teeth (the most important part when filing). The position of the saw handle and the filer are labelled and the file position shown by the dark line. The bevel on the front of the tooth is marked in green, the bevel on the back of the tooth in blue.

	Filing with the set on the tooth front, looking at back of next tooth. As bevel increases, filer crammed into blade. Filing even numbered gullets.
	Filing with the set on the tooth front, looking at the front of that tooth. Filing odd numbered gullets.
	Filing against the set on the tooth front, looking at the back of the next tooth. Filing odd numbered gullets.
	Filing against the set on the tooth front, looking at the front of that tooth. As bevel increases, filer crammed into blade. Filing even numbered gullets.

Using Method I (drawings 1 and 2), file handle toward saw handle, the tooth fronts are away from the filer on the first pass (saw handle on the left), toward the filer on the second pass.

Using Method II (drawings 3 and 4), file tip toward saw handle, the filer is still on the wrong side of the file for one pass down the saw.

With my jig, I can rotate the file away from the tooth for a quick visual check of both sides of the gullet, then rotate it back into position with no loss of accuracy. Since this is easy no matter which direction the file points, I don't think the visibility argument favours one file direction over the other.

The Surface Quality Argument

I think you get a better tooth surface when you file with the set.

When you file with the set, the file moves easily and smoothly across the face making little noise. When you file against the set, the file skips and bounces as the teeth contact the leading edge of the tooth, making quite a bit of noise. It is my belief that the surface on the with tooth is much better than the surface on the against tooth. I believe that this effect is as great as the burr effect, if not greater.

I will take some microscope pictures of teeth to see if any effect is detectable.

The Sloping Gullets Argument

A hand filer has the option of sloping the gullets to achieve tooth shapes not achievable by machine sharpening.

The sliders determine the slope of the file. If you want a slope, make sliders with angled saw kerfs.

When filing sloping gullets, it is usually easier to tilt the saw in the saw vice than to tilt the file. My saw vice only tilts away from me - the geometry of the vice and the bench clamp prevent tilting toward me. This means I can only file up the gullets, not down the gullets. Since the gullets always slope down toward the handle, the file handle must slope back toward the handle as well.

So, with standard saw vices, it is difficult to use Method II if you want sloping gullets.

Summary

Convenience during filing, durability of the resulting file - how to decide. This could well be a personal decision. Try it one way, keep track of how easy it was to do the sharpening and how long the saw was sharp. Next sharpening, try it the other way.

In my experience, comfort during sharpening is more important than a possible slight durability gain.

Metallurgy considerations

Metallography is concerned with how abrasives, including files, affect the crytalline structure of metals. The general result is that the coarser the abrasive, the greater the damage to the metal. With increasing damage we always get decreasing durability.

A file is about the coarsest abrasive we can possible use.

Are there any alternatives?

The graphic on the right shows the cross-section of a triangular files used in saw filing. The file has teeth on all size sides - the corners as well as the faces. This means that when you file a saw, the file cuts the bottom and the sides of the gullet.

Plane Iron Sharpening

I demonstrated the effect of using coarser abrasives in the durability of plane irons. I tested using a single plane iron, always with the same edge geometry. The first test, final abrasive 15 micron. Second test, final abrasive 5 micron. Third test, final abrasive 0.5 micron. The third test produced a more durable edge.

We cannot simulate abrasives graded at 15, 5, or 0.5 microns with ordinary metal files. Are there other types of files?

Remember, the files we use must have the rounded corners shown in the above model.

Other File Types

There is another type of file that looks triangular, but the corners end in a point, not in a flat. If used exclusively, these files would soon stop working. The corner, since it is not removing metal, would not cut down into the saw plate. The file would also produce a very sharp gullet bottom.

I have several triangular files that are made of bonded abrasives. These files are quite short - perhaps because a long bonded abrasive file would snap. They are made of finer abrasives than any file would correspond to, but are still not in the 5 micron of 0.5 micron range. These would be hard to use because they are so short. I have not tested them with saw tooth durability as the test goal.

I have seen small triangular files coated with 10 micron diamonds advertized. These files have sharp corners with no abrasives on the corners. If you sharpened regularly with these files, you would occasionally have to file down the base of the gullet with a standard metal file.

In suggesting that a saw filed with diamond coated files would produce a better result - an equally sharp that that is more durable - I am in no way criticizing anyone who uses regular files to sharpen saws. First, the needed files are not available. Second, the benefits are not proven.

However, if the needed files did become available and benefits were proven, also assuming reasonable cost for the files, I think most filers would gravitate to the diamond files.

Questions? Comments?

You can email me here.