I have not forgotten or discontinued this project, only when it is cold I have to first heat up my workshop before doing any work – which wastes a lot of fuel and a lot of time. Therefore in winter I never manage to do as much work as I would like to. But I managed to do something in the last two months – like building the tumbler (that has run for five days straight by now btw. and it has made a very nice satin finish on the broken blade).

But I could not do much actual work on the knives themselves, because first I must focus on making the necessary improvements. I managed only one step in the process and one failed improvement in the next one.

The step that I have managed with success is shaping the outline of the blades. Last time I ended up with three stacks of blanks held together with screws. So I took them to the belt grinder and ground the outline of all three stacks. After that I disassembled the stacks and cleaned up any irregularities, burrs etc.

©Charly, all rights reserved. Click for full size.

I ended up with a stack of 12 knife blanks. They are not all identical, but in three shapes – next time I will have to think a bit more about how to get reliably reproducible results. To save my self trouble when cutting the handle scales I have marked the blanks in each group on the tang with letters T, V and X. Why these letters? Because they are easily distinguishable from each other and can be scratched with just two lines.

This step was not actually very time-consuming before – just 10 minutes per blade, or 1, 55% of the whole process. Theoretically not worth improving. But I hope that having three groups of four reasonably identical handles will save me some time when shaping the handle – which took 110 minutes per handle, or 17% of the whole process.

Nevertheless, shaping four blades at once did bring some minor time-saving in itself – I have spent only 5 minutes per blade now, so I have saved 5 minutes from my process. This has confirmed that this was indeed low hanging fruit – it was a very easy improvement.

Next step is basic grind of the blade – and this is where I have my first failed attempt at improvement to share in my next post.

I hope this will work. If not, I am determined to fiddle with it until it works.

I found an old asynchronous motor in our cellar. It is a small thing, mere 140 W, and it lacked wiring, any elements to fix it to something and cam wheel completely with wedge. But I have managed to convince my father to connect a cable and a switch to it, and it was working. So last few weeks, whenever I have got an hour or two, I was building a tumbler. I did not document the building process, because there is not much to it, really.

©Charly, all rights reserved. Click for full size.

Just like with my belt grinder, I started with a particle board leftover from kitchen renovations, which was the base for my old drill press. It is a nice >2 cm thick board, covered with waterproof plastic on the upper side. Stable, strong, simply ideal as a base for a machine. Because the motor lacked any flanges or wings or whatevers to fix it and only had 4 M5 threaded holes, I took two pieces of steel that had 90° angle, straightened them to about 120°C angle and screwed them onto the motor. This provided me with two ears, that could be screwed on wooden blocks connected to the base plate. The switch was attached to the plate by its standard holes, plus two wooden pieces to better secure the cable.

©Charly, all rights reserved. Click for full size.

As I mentioned, the motor lacked cam wheel. So I stood in front of a choice – to buy V-belt cam wheels and V-belts, or try something else. I tried something else, because I deemed it easier and cheaper. I bought two PP furniture wheels of different sizes. One got attached onto the motor in the standard way – with a steel wedge ground from a piece of spring steel and a lot of cursing. The diameter of the axis was slightly smaller than the inner diameter of the wheel, but I was able to fill the space with a piece of steel pipe. For the bigger wheel I had more luck – it had inner diameter 15 mm. So I could just buy a 15 mm steel pipe and further I could put to use two old ball bearings that also had 15 mm inner diameter. With a bit of banging and a lot more cursing I was able to fix the two ball bearings and the big PP wheel onto the pipe (the PP wheel is further fixed with a nail, so it dos not hold on only by friction). Next step was to fix the ball bearings on two wooden blocks onto a separate particle board plate.

At this stage I also took a strip of leftover flooring PVC and glued it onto the PP wheels for better traction. It is actually nearly impossible to glue anything on PP with reasonable strength, but there are adhesives on the market that manage this task strong enough for this kind of application (I think). If it goes pear-shaped, I can always screw it on later.

For the tumbling drum I took a 100 mm diameter PP pipe and again I glued on it a few layers of PVC for better traction. The PVC lays directly on the steel pipe in order to reduce the fast rotations of the motor as much as I can.

Two small furniture wheels aid in keeping the tumbling drum in place whilst allowing it to freely spin. The tumbler thus lies between the axis and the two wheels and holds in place purely by its own weight.

©Charly, all rights reserved. Click for full size.

Last week I cut a piece of thick leather 1 m long, 3 cm wide, and I cut the ends at an angle so they overlap whilst the overall thickness remains the same, and I glued it together with epoxy. Hide glue would probably be better, but the weather was way too cold for messing about with hide glue. Today I took last few hours in adjusting the positions of the two wooden plates against each other to have adequate tension on the belt without it wandering in one direction or the other. As of now, it has been running for an hour without problems.

It has about 120 rotations  per minute, which might be a bit too fast. I put in shredded walnut shells, a soft coarse polishing compound and a broken blade from my failed machete build.

We will see what comes of it.