Scrapers, Part 3

Negative Rake Scrapers;  A Low-Angle Scraper; Applications

Negative Rake Scrapers

A negative rake scraper is one that has a bevel ground on the top in addition to the normal bevel. Beyond this there is little consensus or standardization relating to the geometry of the tool. The oft-quoted advantage of a negative rake scraper is that it will give a clean cut in wood so twisted, hard, and gnarly that ordinary scrapers or turning tools cannot deal with it.  

The effect of adding the top bevel is to increase the included angle at the cutting edge, assuming the normal bevel is unchanged. For example, suppose the normal bevel initially has an included angle of 76°. We then grind a top bevel with a rake angle of 15°. This gives an included angle of 91°.

What is the significance of an included angle of 90° or greater? No burr is produced when you grind the cutting edge. Therefore, adding the top bevel makes it possible to “sharpen” the scraper at the grinder without producing a burr.

Here is a rather “busy” photo of the top of a negative rake scraper made from an old file. It has a secondary bevel on the top (80 grit), a left-over from an earlier experiment. The bevels were not honed; it’s straight from the grinder. There is no burr. It will cut and make shavings.

Could it be that the ability of a negative rake scraper to deal with gnarly wood stems from the fact it has no burr? Maybe.

Some woods are so dense and tight-grained that a scraper with a burr is not effective. The burr tends to grab, which makes it uncontrollable, and that is not a good thing. The long-standing trick in such case is to remove the burr and hope for a better result.

My big adventure with grabby scrapers occurred when I was trying to clean up and refine the inside bottom of a small dogwood bowl. I was forced to use a scraper because I couldn’t reach the bottom with a bowl gouge. The gist of the story is that it was slam, bang, and knock every time I touched the wood with a scraper that had a burr.

Not long before my adventure I had heard a demonstrator mention casually that some woods do not like burrs. He named several, which I don’t remember, but dogwood was one of them. I removed the burr from my scraper and was able to complete the project. It was not great fun, but I got it done.

Can a cutting edge with an included angle greater than 90° and having no burr cut wood and make shavings? Will it cut without a burr? Yes, but how well it cuts depends upon it having a keen edge, and even then you must press the tool against the wood with considerable force to get it to cut.

It’s a bit unusual to think of an edge as being keen when the included angle is more than 90º, but the meaning of “keen” is that the cutting edge is the perfect intersection of two planes, the planes of the two bevels. Ideally, no rounding at all would be present but that is hard to achieve in practice.

If you change the bottom bevel angle you can reduce the included angle to less than 90° and return to the familiar territory of the burr. However, the only thing you gain over an ordinary scraper is that the angle of the tool handle will be different. When you hold the handle level the business end is effectively pointed downward.

Is there an advantage in having a scraper pointed downward if it has a burr? I think not. Earlier we determined that pointing the tool downward aimed the burr more directly at the wood which, in most cases, will give a more aggressive cut.

I tested two negative-rake scrapers on a bone dry piece of walnut.  One was M2 high speed steel; the other was a test scraper made from an old file, and both had been meticulously sharpened to a keen edge. Each one produced a surface smooth as glass, all the way around, end grain and all.  

However, the negative-rake ran into problems when I applied it to a sidegrain blank of kiln-dried pine which was relatively dense and hard, for pine at least.  Considerable pressure had to be applied to get it to cut, and some tearout was evident. On this piece, an ordinary scraper with a burnished burr used in the shear-scraping mode gave better results.  

Two things, I believe, come into play here. First, negative-rake scrapers do best with hard woods, the harder the better. Second, a shearing cut will almost always yield a better surface than one where the wood meets the cutting edge straight on, as it does with the negative-rake scraper. It appears that the advantage inherent in a shearing cut won out over that of the negative rake.

Scraper with 35° Included Angle

For years I had heard about using a skew as a scraper but I wouldn’t try it for fear of ruining my skew. Finally, out of curiosity, I took a little-used 3/4” spindle gouge and reground it to the shape of a round-nosed skew, if there is such a thing.

After grinding the included angle down to about 35° and honing it to a razor edge, I applied it to the spinning wood, flat on the tool rest with the handle level. I was amazed. It cut, made shavings, and left a clean surface. Further, the cutting edge seemed none the worse for having been abused in such fashion.

It couldn’t be easier to use. Flat on the tool rest is good. It doesn’t tend to grab or catch. The depth of cut is pressure sensitive which means you can fine tune a surface to whatever precision you wish. The only shortcoming I’ve found with it is that on spindles of certain fibrous woods (sycamore, for example), it tends to pull the fibers sideways out of the surface rather than to cut through them. This does not happen with tight-grained woods.

Why does it cut? How can a knife edge applied straight to the wood produce shavings? There is no invisible burr like there is on an ordinary scraper. With the cutting edge at right angles to the surface, the geometry seems to fit the textbook definition for scraping, but it cuts instead. Instead of dust and chips you get shavings. How can this be?

I have a theory, but I’ll admit in advance that I just made it up. I would be happy to hear any ideas you may have in this regard.

The simplest case to consider is a single wood fiber (cell) embedded in a semi-rigid material, which of course is the wood around it. The knife edge approaches and makes contact with the side of the fiber which gets pushed sideways and deformed. At this point the keen edge engages the side of the distorted fiber and begins cutting through it, severing the fiber.

Why cutting and not scraping?  The fiber will be pushed over in the direction of motion of the knife edge while it is passing through the body of the fiber. In this configuration the motion of the knife edge will have a component along the length of the fiber. This will reduce the rake angle to less than 90° which puts it into the realm of cutting.

OK, forget that stuff about the components and rake angles and the definition of cutting. Have you ever watched a bulldozer with a sharp blade clearing a patch of woods? When the blade makes contact with a small tree (sapling), the first thing that happens is that the blade pushes the tree over a small amount. Then the blade begins to cut through the trunk. The angle of the cut is not straight across but is at an angle along the length. Same thing.

While my theory only addresses a fiber in one orientation (which is actually end grain), I believe it applies just as well to other orientations. In any case, there will be a deformation followed by engagement of the cutting edge.

I think I can now understand why a fibrous wood like sycamore doesn’t cut cleanly in the spindle orientation. My guess is that the fibers are torn away from the surface during the deformation stage before the cutting edge has a chance to engage and cut.

There is no room for compromise with the included angle. If you make it larger than 35° it will not work as well, and if you get up to 50° you will be dealing with a completely different tool.

Also, the cutting edge needs to be very sharp, the sharper the better. A razor edge is ideal but I’ve found the tool is still quite effective with an edge less than razor sharp, one that will not remove hair from your arm. The tool is basically low maintenance – touch it up with a hone every now and then and it will last a long time.

Applications of Scrapers

The following is a short compilation of applications that come to mind for scrapers. A scraper represents a good choice for these but for most it is not the only tool that could be used. That’s typical for woodturning in general.

Form a tenon on the end of a spindle.  A scraper with a square end is well suited for this. However, if the tenon is to be fairly wide, it is best to make two or more narrow cuts rather than one big wide one. If the tenon is to be dovetailed, you can form a straight tenon with a scraper and then switch to a spindle gouge to form the angle.

Refine a cove on a spindle. I use the word “refine” because in most cases a gouge will remove wood more rapidly than a scraper. The idea is to rough out the cove or other concave region and then clean up and refine the surface with a round nose scraper.

The same idea is applicable to a groove at the rim of a bowl. If you have difficulty getting a good curve with a gouge, try a round nose scraper in the shear scraping mode.

Clean up an outside corner.  This can be done with just about any tool, but if you have a scraper in your hand, you might as well use it to knock the splinters and roughness off an exposed corner. In my view a rough corner spinning at a high RPM is close akin to a ripsaw.

Refine the contour on the inside of a bowl.  If a bowl has an inflowing rim, it may be impossible to shape the inside with a bowl gouge. The problem areas are typically where the side transitions to the bottom and the area immediately under the rim. In this case, a scraper is the go-to tool.

Special “long and strong” scrapers are made specifically for this. They are much thicker than the typical scraper. The extra strength is needed for reaching far over the tool rest, where “far” means anything beyond about 3”.

Caution: Some curved scrapers have a nice flowing curve that seems to invite you to ease that curve up against the wood to transfer the curve to the bowl. Don’t do it; it’s a trap! You must never attempt to take a cut wider than about 3/8” at the most, and this only if you are able to take a very light cut.

If you try taking a really wide cut (say 1” wide) with an aggressive burr, it will probably grab and bring the spinning bowl to an abrupt stop. Then either the lathe will bog down or the bowl will come off the lathe and go rolling across the floor. Ungraceful, at best.

By the way: if your lathe has a belt it will probably have a means for adjusting the tension. There is no reason to set the tension to the maximum your strength will allow. Let the belt run a bit loose. Then, if you have a mishap, the belt will slip and save some excitement. (The squeak of the belt on my PM 3520 used to make my dog bark.)

Goblets.  The bowls on some goblets are rather deep and narrow so that a gouge is not a viable tool. Solution: use a round-nose scraper.

Boxes.  Scrapers find many uses in making small lidded boxes. These range from forming the lip or recess for the lid to hollowing near the bottom of deeper boxes.

One instance that practically requires a square end scraper is the bottom inside corner of a rectangular-looking box (right circular cylinder). In fact the corner doesn’t have to be square, but square on the outside and round on the inside just doesn’t seem quite right.

This is where the square end scraper with offset angles is used. The offset angles lets you reach to the bottom corner while remaining clear of the wall above the corner.

I modified my box-corner-cutting scraper to make it a lot less worrisome to use. If the bottom of the box is perfectly flat, you can’t cut the corner without having the entire end of the scraper come in contact with the bottom. This is not good because that can easily lead to taking a very wide cut. So I gave the end a slight curve. Clever, yes?

Not only can the inside corners of boxes be done with a scraper, but things that look like the inside corners of boxes can be done as well. An example that comes to mind is the recess into which a tenon is to fit (think mortise and tenon) such as you might find on the underside of an elevated tray. In short, corners invite square end scrapers.

Spear point scraper.  This is definitely a special-purpose tool but it is quite handy on those occasions when you need it. It’s used to cut fine grooves into the foot of a bowl, a detail on a finial, or the handle of a magic wand. Need a groove? Think spear point.

The key concept is “fine grooves.” If you try pushing a spear point straight into a surface, you won’t get far before you bog down one way or another. Straight in, it cuts on both sides of the point and a wide cut develops very quickly. If you need a deeper, wider groove, alternate between sides and provide clearance to avoid having the tool bind up.

Spear point scrapers are available for purchase but many turners make their own by re-grinding a seldom-used tool or by shaping an appropriate piece of tool steel.

A wobbly bowl.  I once was given the task of finishing a small bowl that had warped so much after rough turning that truing it up was impossible. Further, I was asked to “keep the warp” because that “adds to the beauty of the bowl.”

The inside surface was very rough and the wall was way too thick. After all, it had only been roughed out and set aside to dry. My first thought was to use 80-grit sandpaper to whip it in line, but that would have been a tedious job with little guarantee of success.

What I did was to use a scraper with a prominent burr (burnished) to make shearing cuts on the inside surface – with my lathe running as slowly as possible, about 50 RPM.

As the piece rotated, I let the scraper follow the contour of the piece. This caused the tool to move back and forth across the tool rest (toward and away from the headstock), sometimes as much as a half inch or more. Even so, I found I could control the tool and advance it smoothly over the surface.

In short, it worked, even to the point of making finishing cuts out near the rim. The key was to have the lathe turning slowly. If I tried to speed it up, the tool would begin to bounce and things would get out of control in a hurry.

[Back to Top]