As I said last time, I am going to make multiple knives in parallel, because I think that a non-trivial amount of time can be saved just by that. I had all my steel already straightened, so I could go right away to drawing, drilling and shaping. First thing to be done was to draw and drill one blade on one half of one steel bar. That I have subsequently used as a template to drill holes in the other half and in all the other steel bars.

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

For the drilling I clamped the template firmly with the drilled steel bars and before proceeding to the other side I inserted a steel pin in the drilled hole so the bars do not shift and slide. This has saved some time, despite it being a step with negligible time in the analysis, but my main hope here is that this work on multiple pieces in parallel brings better reproducibility regarding the hole positions and and that better reproducibility could save me some time later on when shaping the handle scales.

I have learned two lessons, one of them rather expensive. First lesson was that for this I have to take the 6 mm drill bits as de-facto consumable material, because I blunted and subsequently broke one towards the end despite using copious amounts of cutting oil. And I do not think it can be avoided.

Second lesson was that I really have to buy a new drill press, and preferably one with continuously regulated speed at that. My jury-rigged press with a very old hand drill has only two speeds and both of them are apparently too high for drilling 16 mm hole in 1,8 mm steel. Towards the end of the work the step drill bit overheated and it got irreversibly blunted on the 16 mm step. Ouch. That drill bit costs 40,-€. I think this investment is unavoidable, if I want to save time I cannot take the bit out and dunk it in water after each drilled hole, slower rotations would be better. And I cannot buy 40,-€ drill bit for each dozen of knives I make – that money be best saved for grinding belts. Hobby or not, at this rate I would spend the price of an acceptable small to middle sized drill press (300-400,-€) on the drill bits rather quickly.

Cutting the blanks in half and roughly cutting the outlines of the blades with an angle grinder did not take long at all. and working on multiple blades in parallel seemed to work comparatively well. But for the sake of precision, next time I will halve the blanks individually – two came out somehow too short and two too long. Not by much, but it is noticeable. The end result were three stacks of four roughly cut blades.

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

Before this step took me 15 minutes per blade, now it was 10 minutes per blade (teasing out broken 6mm drill bit and fruitlessly trying to drill a hole with blunted 16mm drill bit included). So I would call it a semi-success. I learned some do’s and dont’s and I achieved my two goals – the holes in the tangs align across all 12 blades nicely and I have saved some time without even really trying to.

First thing first – today I tested the very nearly finished knife when I was cooking shrimp for lunch.

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

It handles well and cuts OK, but I really suspect it won’t hold an edge as well as it should. But the cheapo wood looks way more posh than I expected it, ammonia fuming really, really improved its looks.

And now to the boring stuff.

What I am doing here is actually a small-sized DMAIC project – an abbreviation for Define-Measure-Analyze-Improve-Control. It is a process used in industry to bring some logic and use of scientific method into improving manufacturing processes. Although as everything in today’s corporate culture it is used wrongly and heavily abused and misunderstood all over the place, because american-trained managers …. wheef, do not get me started on american-trained managers.

Aaaaanyway, in the first phase, Define, you should either define your problem or your goal or both. In my case, I have a specific goal – to get my manufacturing time of this type of knife under five hours of manual labor.

In the second phase you should acquire all measurements that you need in order to do something about it – in my case I have measured the manufacturing time of each distinct step in the process.

And now, in this post, I am performing the Analysis of said data. Total manufacturing time: 10:46, or 646 minutes.

A picture is worth a thousand words, so here is a picture.

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

There are of course multiple approaches one might use, but in this specific case I think that this suffices ample enough – it is so-called Pareto Graph. The balks are actual times in minutes for each step, ordered from the highest to the lowest. The black line is a sum of the relative proportions of these times to the total as one progresses from left to right. As you can see, I have ten distinct manufacturing steps and from those five steps constitute 90% of manufacturing time. These are the steps where I have to concentrate on actually reducing said time, because here my efforts have the biggest payoff. That does not mean that improvements in the other five steps are not worthy pursuing at all, but they are not worth pursuing at this time.

I have done one thing that is not normally done, that is I sorted from get-go the steps into two categories – low hanging fruit, where I think I can do improvements without too much hassle and without obtaining expensive or complicated equipment, and high-hanging fruit, where I can save time only through acquiring new skill or new equipment or where I think saving time is not actually possible in a meaningful way.

Now comes the next phase, which is  to improve the process. I will try to implement some of the ideas for that I have expressed and make a batch of multiple (~10) knives with improved process. We will see how it turns out.