Side Sharpening
	Finest abrasives.
	Microbevels front and back.
	Use a jig.
	Copyright (c) 2002-09, Brent Beach

Introduction

I have been aware of side sharpening for some time, but decided to look into it when HS introduced a sharpening jig called the Sharp Skate which uses this approach. You can watch a video of him demonstrating the jig here.

I will use the acronym STS for side-to-side (or side sharpening), and FAB for front and back (or forward and backward) sharpening.

This page covers my experiments with side sharpening:

• Honing testing - generally positive. In this test I used STS to hone a blade following my standard sharpening protocol: begin with a 25 degree primary, hone front and back microbevels using 15 micron abrasive, hone front and back microbevels using 5 micron abrasive, hone front and back microbevels using 0.5 micron abrasive.

  There is a very strong tendency to hone more on the part of the blade at the front on the push stroke. You will have to keep this in mind if you opt for side sharpening.

• Grinding testing - generally negative. I ground the primary bevel at 25 degrees using my coarse bench stone.

  Few will grind using bench stones, but those that do should continue to use the standard front and back style.

• Durability test done - small defects at the edge that could be the result of scratches parallel to the edge.

• Re-test done - re-honed the blade using front and back style, planed 200 linear feet - honed blade showed level of camber produced during side to side sharpening, resulting blade did not get edge defects during edge durability test.

• Page editing done - removed errors and updated early material to reflect later test results.

• Summary - if the alternative is no jig, then use side sharpening. If you side sharpen with a jig, then you are better off with a jig system like mine that incorporates microbevels and back bevels and does not have the problem of uneven bevels. However, you should still be aware of the increased tendency to camber the blade and the likelihood of edge defects during use.

  Last updated July 29, 2007.

Why Side Sharpen

It appears that STS is also popular with Japanese tool users who also prefer to sharpen without a jig. From watching the video, it appears that the standard within that group is to hone along the edge, but move the blade perpendicular to the body (as you would in the normal honing motion). [If anyone has any Internet links to sites that discuss STS, I would be happy to see them and perhaps include then on this page.]

Now, while having its origins in freehand honing, HS has invented a jig that can really only work with the direction of motion along the edge - the Sharp Skate.

Which Jig to Use

Before looking at the results of STS and comparing those results to the FAB results, a short look at jigs for STS.

The Sharp Skate costs $150 and has a number of limitations.

1. Setting the Angle

   In the video, angle setting appears to be a little haphazard. HS balanced the blade on the bevel, slide the jig down, then tightened the jig on the blade. Unless you can grind an accurate primary bevel angle, you cannot use this method. Having ground the primary bevel, you are now directed to hone the primary bevel. There is no reason to hone the primary bevel away from the edge, if you can avoid it. Jigs should let you avoid unnecessary honing.

2. Microbevels

   Microbevels not only speed the honing process, they also improve the resulting edge. The full explanation is here.

   In the video, HS increases the honing angle by loosening the tension on the blade and sliding the blade a bit. People should expect quite variable results from an approach like this.

   By analogy with my honing system, you could slide a piece of very thin plastic under the wheels to get the desired effect. Since the extension is so small (less than 1") you would probably have to use plastic about 0.01" thick. Plastic this thin would probably deform quickly with use.

   Assuming wear on this shim is not a problem, this would at least give you a reliable microbevel.

3. Back bevels

   I will not repeat all the reasons for using back bevels here, but the Sharp Skate is not adaptable to back bevels. In the video he lays the blade back flat on the stone and hones very briefly to remove the burr. This does not restore the back of the blade to a honed condition.

4. Honing Effort

   The video shows these shiny primary bevels with just a little fiddling around. Smacks of Uri Geller to me.

   One of the ways to compare honing systems is to calculate the amount of abrasive the blade runs across on each honing motion - the more abrasive the faster the honing. A FAB jig that rides on an 8" stone (a jig with a roller, like the Lee Valley or Eclipse jigs) and is 4" from the roller to the blade edge, is limited to a 4" range of motion (really only about 3" or you fall off the end). The Sharp Skate is better than those jigs, because you can slide the Sharp Skate the full length of the stone (8" on an 8" stone). Each part of the blade moves the stone length less the Shark Skate width for a net of 6".

   Honing, when using a back and forth motion of any kind, has three phases: start up, middle, slow down. The best work is done in the middle phase. During start up and slow down, the speed and pressure are variable. As a result, the honing effort is variable and more subject to error. I found that the length of honing/grinding motion was an important factor in honing/grinding speed - long motions produced better bevels much more quickly. STS honing has an advantage because of this.

   One of my problems with the video is that HS shows a tool with a shiny primary bevel (I think it was a large chisel). This may have been done simply to show the honing ability of the jig. It is not good honing practice though. Only a very small area near the edge of the tool is involved in the cutting. Time, and abrasive, spent honing away from the edge is time wasted. Microbevels save time and abrasive wear, and still produce a better edge.

   In spite of this, side honing does allow you to move each part of the blade over more abrasive than a FAB jig that rides on the stone. This is a positive for STS jigs.

5. Uneven Abrasive Wear

   One of the problems with water stones is uneven wear that results in dishing of the stone. Even a FAB honing motion with a 2" wide plane blade tends to dish the stone. People flatten their stones as often as every 5 minutes of sharpening (some more often). In the video HS uses fussy little rubs.

   I favour full length motions to reduce end effects, as explained in the previous item. I suspect the variety of motions are a requirement for using this jig, not a feature. If you use long motions you will quickly wear slots in the abrasive - both with the edge and with the rollers. The short, jerky motions all over the surface are required to keep the stone reasonably flat.

   Going back to the example of honing a 2" blade and an 8" stone in the previous item. If you use FAB honing, a 2" blade moving 4" crosses 8"" ("" means square inches) of abrasive, whether you are honing the full primary bevel or the narrowest microbevel. Using STS to hone the whole primary bevel of a 2" blade, which is about 0.2"", you use 1.6"" of abrasive. Using STS to hone a microbevel which is 0.005" wide (the last microbevel using your finest abrasive), a 4" motion uses only 0.02"" of abrasive.

   The amount of abrasive involved in one 4" FAB motion is 400 times that used in one STS motion. That suggests to me that the wear on the abrasive is 400 times greater in the STS case. This concentration of wear will lead to unevenly worn abrasives.

6. Edge Strength

   FWW on Planes and Chisels includes a short item by Bruce Hoadley (of Understanding Wood fame) called "Which way to Hone?". Hoadley has a drawing of the scratches produced by the two honing methods. He admits he has no scientific evidence, but favours FAB over STS, saying that STS could produce a scratch parallel to the edge and the metal weakened by the scratch "may break right off in a long sliver of metal".

   Any tool that is ground on a wheel or a belt grinder will have deep scratches perpendicular to the edge. FAB grinding on a bench stone produces the same type of scratches. If you think of someone combing their hair, after applying gel, you can imagine the valleys and hills running back from the hairline. A freshly ground primary bevel is like this (if you grind right up to the edge, which I do not recommend). You might also think of a series of parallel valleys.

   Honing scratches produced by STS cut across the grinding scratches. The hair comb analogy fails here because scratches are cuts. What do you get when you cut a new set of valleys across an existing set of valleys? You get a grid of pyramids - standalone structures unsupported on all sides. When these pyramids collide with the wood, they could easily topple.

   Honing scratches produced by FAB refine the existing triangular hills and valleys, producing smaller hills and valleys. The leading face of each resulting hill is supported by the lengthwise hill behind it. That is, by honing in the same direction as you grind, which is the same direction in which you use the tool, you are not weakening the edge in the direction of use.

   As you will see below, my testing suggests this theoretical objection to side sharpening may be an actual objection.

Rather than buy a jig, you can use the jig you made like mine, without any adjustments to it, and solve the first 3 problems. Rather than the 2" jig I used in the test, you could make a short jig which may reduce rocking during use. The last three problems are unchanged by the switch to a different jig.

In this picture, the stone is held in the stone holder by the long sides rather than the ends. As usual the jig slides on the wide face. Here it slides sideways, rather than front and back. Otherwise, it looks the same of the setup for FAB honing.

The slips are used in exactly the same way as in FAB to allow you to get precise second and third microbevels as you move up to finer grits.

1. Setting the Angle

   Precise, repeatable, accurate to 0.1 degrees, assuming you can set the extension to within 1/32". You can probably set it much more accurately than that.

2. Microbevels

   Same slips and technique as for FAB with this jig. As many microbevels as you want, exactly the same every time.

3. Back bevels

   By design - just flip the jig over and hone.

4. Honing Effort

   Same as with the skate.

5. Abrasive Wear

   Same as with the skate.

6. Edge Strength

   Same as with the skate.

Incidentally, the setup pictured on the right is using the first bench stone holder that I made to use bench stones with my jig, in the FAB style. If you want to try this you should probably make one more like Version 4 - which is faced on top with glass to reduce friction between the jig and the stone holder.

Similarly, you can adapt my jig for use with sheet abrasives by changing the size of the glass plate you use. If you decided to use full length sheets, perhaps ripped into three, then your abrasive sheet would be approximately 3" wide by 11" long. If you put this along one side of sheet of glass 8" wide by 11" long, you would have a 5" side area on which to slide the jig. I happened to have this 9" x 12"+ sheet of glass, so used a half sheet of 15 micron abrasive.

If I were to use STS I would probably favour the 3M abrasives on glass approach, assuming that the abrasives can stand up to this honing motion. As with FAB honing, the 3M abrasives give you the finest available abrasives on an always flat surface.

The jig in this picture is 2 inches tall. To better conform to the style of the Sharp Skate, a shorter jig might be used. I have jigs as small as 1 1/4 inches tall. Using a much shorter jig may reduce problems with rocking.

Test Setup

Abrasive Wear

This increased wear rate on a very small percentage of the grits may not have any real effect on an Arkansas or India stone - stones famous for their resistance to wear. On a water stone, the problem of creating narrow slots in the stone will have to be considered.

On sheet abrasives, I thought the real problem would be tearing up the backing. Unless you are very careful, the leading edge of the blade could easily rip the paper. It turns out this is not a problem.

Edge Strength

The question is: should the ridges be along the direction of use, or across that direction? Planing using a FAB sharpened blade, the wood makes contact with the front of all the ridges across the blade, each ridge front supported by the rest of the ridge behind. Planing using a STS sharpened blade, the wood makes contact with the single ridge (line of pyramids) closest to the edge, which has no support behind it.

Put this way, in terms of the ridges, the conclusion seems pretty clear. (You are planing the wood, not sanding it!)

Abrasive grits are box shaped solids (usually a slightly skewed). The corners of the boxes do the cutting. I generally assume that the deepest scratch a particular piece of grit can leave is about half the grit size (it could be less). The depth depends on the grit density - how many grits are in contact with the surface being abraded - and the grit size.

At one extreme, the open coat Norton 3X abrasives are able to leave deep scratches and even fracture the edge. In that case there are so few pieces of grit in contact with the blade that they can easily chip the edge. At the other extreme, the 3M 0.5 micron abrasive has so many minute pieces of Chromium Oxide in contact with the tool that each grit is exerting only a very small force. In this case the scratches are very fine.

If you use STS with 15 micron abrasive, you could get scratches parallel to the edge that are as much as 7 microns deep - 0.0003" deep - and very close to the edge. If you use a hard white Arkansas stone, a scratch this size is certainly possible since the average grit size is around 14 microns, with many larger grits. Is this a serious problem?

During use, wood rubbing against a blade removes metal at the edge. By the time a plane blade is dull, it is shorter by about 0.0002". The blade does not feel dull while this wear is occurring, only getting a little more difficult to use near the end. If the sliver of metal isolated by a deep STS scratch did break off, that part of the edge would immediately feel dull.

Of course, if you don't use microbevels, then the deepest scratch is probably that left by the coarsest abrasive you used, less any metal you honed off after that. If the coarsest abrasive you use at the edge is 15 micron (600 grit), then the deepest scratch is about 0.0003" deep. Honing with very fine abrasives, if you don't use microbevels, does almost nothing. You can see that the amount of metal you hone off during use and the depth of the deepest scratch left by 15 micron paper are about the same.

Honing Test Results

My first impressions - which apply to using 3M abrasives and my jig - are pretty positive.

Abrasive Wear

Using full length motions along the 11" of abrasive, it was quickly clear that there was no problem with the narrow edge cutting any of the three abrasives. This applied to the initial microbevel on the 15 micron abrasive, as well as the first honing on slips with the 5 micron and 0.5 micron abrasives.

If is too early to tell if the first few strokes on the 5 and 0.5 micron abrasives left narrow dull areas along the abrasive. I suspect not. No matter how hard I tried to move exactly parallel to the edge I was always a few degrees off - the scratch pattern was never exactly parallel to the edge. You might want to deliberately tilt the jig and iron a degree or two on the first couple of passes to be sure you don't cut a groove in your abrasive. You might also want to start with a pull motion, rather than a push.

So, my worries about abrasive wear were misplaced. You can use STS on 3M abrasives without particular worries about abrasive wear.

Honing Speed

My first tests used an area about 2" wide and 10" long - 20 square inches. Using FAB on a 4" wide abrasive sheet with a 3" range of motion makes use of only 12 square inches. As well, since the range of motion is so much longer - 10" versus 3 or 4 inches - the quality honing time using STS is far greater. (Again, based on my belief that the middle part of any honing motion produces the best results. STS on an 11" abrasive sheet has a much longer range of quality honing.)

The increased honing distance per stroke means that you need 60% fewer honing strokes for each abrasive grit/microbevel. Under these conditions, STS could produce faster results than FAB.

If anything, the danger is doing too much honing. You have removed enough metal (produced the desired width microbevel), before you know it.

Honing Feedback

The difficulty in seeing the microbevels might be a result of how I hold the blade when trying to see the microbevels - not a problem with the honing method. The FAB scratches are perpendicular to the edge and may show up better when I hold the edge perpendicular to the light source. It could well be that scratches parallel to the edge do not reflect light the same way, making them harder to see.

It is certainly the case when scanning: scratches parallel to the scanner light show up much better than scratches perpendicular to the scanner light. In any case, more experience with illuminating the microbevels may help me see the microbevels without magnification.

The following scan was done after a honing on the 3 abrasive grits, using the increasing slip sizes. The microbevels are hard to see, even in a 4800 dpi scan.

The question - is the microbevel uniformly wide across the blade?

The image on the far right is a scan of full width of the 1 3/4" blade. It contains two dark columns, with a middle lighter column. The three honed microbevels are all in the lighter column. This middle column is 0.07" wide. To the right is the primary bevel, to the left the edge and stuff beyond the edge that does not reflect light.

If you look at the middle bright section, the right part is not as bright as the left part - particularly at the top. The right duller part of the bright column is the 15 micron microbevel. The left brighter part of the bright column is the 5 micron micro bevel. The dark area, close to the edge (particularly near the bottom) is the 0.5 micron microbevel.

Close examination reveals that I have a tendency to press the leading corner of the blade down when pushing away from me. That is, the bevel is wider on one side of the blade than the other.

The left column shows enlarged parts of the edge. The top image shows the right side of the blade (top of the right image) - as viewed looking at the primary bevel side. During honing, this is the corner of the blade nearest me. When I side hone, I push the blade away from me, then pull it back toward me. So, this corner is the trailing corner on the push motion, the leading corner on the pull motion. There is no evidence of the 0.5 micron abrasive microbevel in this image. The brighter scratches on the left are the 5 micron scratches, the duller left part is the 15 micron microbevel.
The second image corresponds to the edge about 40% of the way from the right side of the blade (40% down from the top of the image on the right). The 0.5 micron microbevel is visible here - but quite narrow - perhaps 0.004" wide. This is a little too narrow. The area of the blade in contact with the wood over the life of the edge should all be honed with the finest abrasive to minimize friction, and thus heat, and thus blade wear. The microbevel in this part of the blade is too narrow.
The third image is of the left side of the blade (the bottom of the image on the right). This is the leading corner on the push motion. The 0.5 micron microbevel is very clear here - and wide enough. It is the darker area at the edge, covering almost 40% of the total microbevel area. It is clear that my STS honing technique needs a bit of work. When honing, my left hand controls the blade and my right hand controls the jig. I will have to apply more pressure on the trailing edge of the blade on the push stroke. It should be possible to learn to hone more evenly with just a little more practise.

This final image is the left side of the three microbevels. The edge is clearer at this size, as is the 0.5 micron microbevel. It looks like the third microbevel completely honed away the second microbevel in this area. That is in itself not a problem, but indicates wasted effort - only the area near the edge that comes in contact with the wood need be honed with the finest abrasive.

In summary, honing using STS is not easy peasy, it does require some attention to technique, but it can produce very good microbevels.

Grinding Test Results

I put my coarse Silicon Carbide stone in my Version I stonevise, put the iron in my jig set for 25 degrees and gave it a try.

Grinding requires some pressure which is best provided by the pad on the outside of your hand rather than the fingers. The exact position of the pad helps balance the metal removal across the edge. This works for a plane blade - I cannot exactly imagine it working for a narrow chisel.

I didn't check soon enough and ended up with a lot more ground off on one side. Even with close attention I had a lot of trouble correcting the earlier problem and getting a straight edge. As well, the feedback I am getting from the stone suggests that I have put some grooves in it.

The camber is evident in this image. Ignoring any extra camber near the corners, the blade is cambered 0.005" with the high point closer to the left side of the image. Not flat, but not too highly cambered. Removing the camber would be easy using FAB, but I could not straighten it using STS.

At full size, this image clearly shows the burr. In fact, there is much more burr formation using STS than using FAB - I have no idea why. I flipped the jig over and ground a bit of a back bevel (about 1.6 degrees at this extension) to remove the burr. This increased level of burr formation/retention could be occur when honing as well - I did not notice on the first honing test but will look for it when I re-hone this blade.

STS Honing Test Repeated

In this image, the edge is if the bright line just above the top line of text.

The 0.5 micron microbevel shows no scratches.

The 5 micron microbevel is next - with some scratches remaining. These scratches go downward to the right (tilt blade to direction of motion during honing)

The 15 micron microbevel is next. These scratches go up to the right. The white vertical mark is a piece of lint.

The primary bevel is last - formed by the coarse Silicon Carbide benchstone. Some of the scratches in this area are quite deep - but this part never touches the wood. (Not sure why that bit of text is unreadable).

The blade still has a camber related to the grinding problems, but is not too cambered for normal use. I can run the durability test with the blade having this camber.

Edge Durability Testing

This blade is a notched rectangle Stanley blade from the laminated blade period - before about 1935. These blades perform about the same as other older High Carbon Steel blades.

The dark vertical line in the middle is a scratch added after sharpening to allow me to find the same place on the edge after each part of the test.

The three microbevels and the primary bevel are clearly visible in this picture. My scratch begins at the bottom in the primary, crosses the 15 micron microbevel, and ends at the start of the 5 micron microbevel. The faint vertical lines in the 0.5 micron microbevel are not scratches from honing - they are faint remnants of the baby oil left when I wiped the blade after honing.

The test was normal - no surprises.

Wear is about normal as well, with the possible exception of the two small edge failures, the ripple in the wear bevel on the right side of the image. Stanley blades usually do not show this kind of edge failure. I have not tested this blade before, so do not know if this would have happened with a normal FAB grinding and honing. I will rehone the blade using the normal front-and-back style and see if this pattern repeats. [It did not - test results below.]

FAB Re-test

The far column shows the full blade. The edge is to the left, then the new first microbevel, bits of the old microbevels from STS, the primary bevel from STS, then the rest of the blade. If you recall, when STS grinding the primary I inadvertently introduced a camber to the blade which had a high spot near the left side of the blade. In this image, the new 15 micron microbevel is wider in that area and narrower away from that area. This FAB honing is straightening the edge. The new 15 micron scratches have not removed as much of the old microbevel on the right side of the blade (top of the picture) so the new microbevel shows up lighter there. The effect is quite apparent in the different brightness of the enlargements. The middle column shows the top part of the blade enlarged - the top 0.3". You can see that the new microbevel has not completely reached the side or edge of the blade for most of this 0.3". STS produced a cambered edge, FAB is producing a much straighter edge. The first microbevel is wide enough now that I should redo the primary - removing all the camber without redoing the primary will take quite a bit of honing. Since the re-test only involves jointing which uses the middle of the blade, I won't regrind now. It does show that honing cannot remove defects left by a poor grinding step.
This is the bottom 0.3". Again, the new 15 micron scratches do not reach the edge here, indicating the blade was cambered before honing. In the full sized image, I can see scratches angle up and toward the edge in the top part of this image. These scratches would have been left by the 15 micron abrasive during the previous STS honing. I should grind a new primary then rehone to get rid of all of these scratches, but will delay that until I see what happens after FAB with 5 micron.

These are full sized scans, after the 5 micron on the left, after the 0.5 micron on the right. The distinction between the 5 micron and 0.5 micron microbevel is subtle.

The difference is small, but there are no defects along this edge similar to those that appeared in the STS durability test.

Conclusions

However, if you do not use a jig, and the STS idea appeals to you, then STS with a jig will improve your edges. It might even persuade you to use FAB with a jig, which is where you really should be.

Other Reviews

Do you have an online review of side sharpening or the Sharp Skate? Send me a link and I might include that link here.

Web Site Links

Honing Jig

Home again

Questions? Comments?