Make a Kerfing Chisel

In 2011 I posted an article on a kerfing chisel, a tool used to deepen the kerf in half-blind dovetails.

The idea for this came from a method used by Tage Frid, who used either a scraper blade or, as I viewed in a video, a section of bandsaw blade. The kerfing chisel was my effort at making a specific tool, rather than buggering up a prized scraper blade.

In the years since, I am aware of two similar tools, the first was by Ron Bontz and the second by Rob Cosman. Both are shaped like saws, complete with brass backs and scraper plates. They do the same thing.

Over the years I have received many queries how to make a kerfing chisel, and this posed a problem since I had made it from a steel gardening trowel/spatula. These can be difficult to find. I have been thinking of other ways to make this easier for anyone interested in rolling their own. This is what I came up with ...

At the moment I am testing out a few different sizes ...

Top is Jarrah and bottom is Hard Maple.

A few others ...

The parts include a scraper blade (these are 0.03" thick). I was curious to see whether this would create a problem when dovetail saws have a 0.026" kerf (0.02" plate plus 0.003" set, which is considered "fine"). See the end for more information. A ferrule is made from brass tube.

Turn the handle, and then saw the slot for the blade using a bandsaw. Place the handle on a V-jig to hold it steady. Ensure that the shaft matches the depth of the ferrule (so that is bottoms out against the end of the handle). Epoxy everything together.

So why is a kerfing tool so useful that it warrants being made into a specialised tool?

Here is a half-blind pin board socket being sawn. Note the diagonal angle, which leaves half the socket ...

This is where the kerfing chisel is used: First clamp the ends of the pin board. The edge of the blade is squared (not bevelled like a chisel), but it has the potential to split or cleave a board. The clamp helps prevent this.

Tap the blade into the kerf, to full depth, moving towards the boundary line a little at a time - be especially careful with the outside kerfs, where there is less support.

I have been doing this for a decade and may have had 2 or 3 splits in all this time, mainly from being careless.

When chopping into the socket, the deepened kerf will make it easier to split out the waste ...

The blue tape makes it easier for older eyes to see lines ...

Cleaning out the waste is significantly easier ..

Removing waste leaves clean sides to the sockets ...

This is saw-to-saw cut ...

I have done a little experimenting with plate thickness for kerfing, and so far it seems that this does not make any significant difference.

Most saw plates are 0.020” thick. Then add 0.002 - 0.004” each side for the set of the saw, and you end up with something closer to 0.030” than 0.025”.

Then there is the accuracy of sawing to the line - how close can you get? Are you within 0.003 of the line, or less? In other words, is the thickness of the kerfing plate largely responsible for changing the socket size? I do not think so. The tightness of your sawing will have more of an effect.

Plate thickness will affect how easily it cuts into the wood. Thicker should require more effort. However, this is affected by how big a bite one takes. That also is determined by the hardness, and brittleness, of the wood.

I measured some of the cabinet scrapers in my collection. They were mostly greater than 0.03”. Paint scrapers from the big box were greater than this, and some have used these successfully.

Bottom line: at this time I would argue that a blade up to 0.030” is fine. Note that I am still experimenting.

UPDATE ON THE ABOVE:

There are two important features to consider if you plan to build a kerfing chisel.

1. The first is that thickness does influence the process and outcome. The thicker the blade, the more effort to pound it into the wood. More importantly, if you are one to saw to the line, using a blade wider than the saw kerf will widen the socket, and the result will be a loose fit. Of course, if you do not saw to the line, then there is simply more effort involved.

Over the past week, while building 10 drawers, it has been apparent that a thin bladed kerfing chisel is significantly easier to use. There is much less effort involved with, say, a 0.024” blade versus a 0.030” blade.

2. The second issue is that it is not enough to simply epoxy in the tang of the blade. The blade is subject to much lateral forces, and eventually will come loose. In order to prevent this, the ferrule/mortice/blade needs to be pinned. I used 2mm diameter mild steel nails. Since I have been making blades with stainless steel (see picture for source of SS scraper blades), the drill bit of choice is one made from tungsten carbide.

When striking the kerfing tool, saw or chisel style, you want to use a steel hammer. The energy from one is better focussed than a wooden mallet. I added a ferrule to the kerfing chisel I made a decade ago, and which I have been using since. I use Japanese chisels a lot, and copied this style. The kerfing chisel in this article is being used sans-ferrule. These are all hard woods: rich reddish Fiddleback Jarrah, creamy Hard Maple (USA), and some type of West Australian Cassurina, very hard and wonderful colours. I have been using a 225gm gennou, and not seen any signs of damage to date, however use has been short.

Regards from Perth

Derek

January 2021