If less is more... then this could go horribly wrong.

Before we talk specifically about the bearings, let's first cover why we're using bearings.

The reason is fairly simple really: bearings seem to be the best solution in terms of function and reliability

I'm going to confess something to you...

I've not done much research on knife bearings (and by "research" I mean: searching on Google and seeing what random people on the internet are saying about knife bearings).

I'm approaching the bearings part of the knife cold and doing what makes sense to me.

The way I see bearings working on a knife is pretty much like most mechanical things in life where there are two opposite requirements at the same time.

Two examples:

-- You want a bicycle to be as light as possible ...while being as strong as possible (or at least as strong is it needs to be)

-- You want the battery on your phone to last as long as possible ...while being as small as possible

As far as I can see, bearings are the same:

You want the blade to be as rigid as possible ...while being as smooth and easy to deploy as possible.

So we have two opposing requests we're asking of the bearing at the same time. It seems we want to tighten the pivot as much as possible so the blade is held as firmly as possible ...but we still want the blade to be able to move freely at the same time.

If I picture have just three balls in the bearings -- this is the minimum you could have and still have the knife function -- this seems like there will be a lot of pressure on each one. And this pressure will in turn press into both the blade steel and the scales.

My feeling here is that, when you tighten the pivot, there is not going to be a lot of free movement. Not only that, but there would be a lot of space from one ball to the next - which would likely result in less rigidity in the blade.

Let's say we double the number of balls from three to six. Now we have half as much distance from one ball to the next ...and we've also just halved the pressure on each ball (if we maintain the same tightness on the pivot).

Does this make sense?

By doubling the number of balls from 3 to 6 we can still maintain the same force when we tighten the pivot ...but the pressure on each ball is half of what it was. And, with the force on the balls half of what it was, the balls should move more freely.

So let's say we double the balls from six to twelve. We should be able to maintain the pivot tightness and also increase how freely the balls move. Or another way of looking at it would be that you can maintain how freely the balls move ...while at the same time increase the tightness of the pivot.

The more balls the better.

I could be totally wrong about this - but it makes sense in my head and I'm going to run with it. And this is why I've squeezed as many balls into the bearing as is physically possible.

Here's what it looks like:

Why 57 balls? Because 58 is too many.

Now you're probably wondering how I know 58 is too many, right?

Well, if you remember from a few sections ago, I said I've failed at making a knife multiple times in the past. During one of those "failures" I designed and made a bearing. In theory (and by this I mean the drawing on the CAD model on the computer) it actually looked like the most I could squeeze in was 60 ...but, in practice, this did not work.

Now, before we go any further...

We really need to discuss ball material, ball size and bearing cage material first.

We're going to use ceramic balls in the bearing because this seems to be the best solution. The hardness is really the determining factor here.

We will use Silicon Nitride coated ceramic balls. Steel balls can't really shatter under force - but ceramic balls can. However, this really is almost impossible in the way we are using them. If you took them out of the knife and hit them with a hammer on a hard surface ...yeah, they'll probably shatter.

Steel balls just seem like a less good solution.

The size of the balls is one of those things where you're trying to strike a balance between two opposing things.

You want the balls as large as possible because it will spread the load of each individual ball over a larger surface area (we're talking on a microscopic level here) ...but, if they're too large, you can't fit many into the bearings cage ...and it would also result in the knife being very thick (trying to squeeze those large balls in).

The upside to using balls that are as small as possible is you can fit a LOT into the bearing cage ...which, if my theory holds up, means you are spreading the load over more balls. The problem with small balls is that they are likely to roll less smoothly over surface of the metal (particularly if any debris found its way into that area of the knife). This, of course, is just theory and conjecture on my part - but I feel it's valid enough to start with.

The "Goldilocks" size is around 1.5mm (or close to it as they may end up being whatever is the closest to this size in imperial measurements). This is probably around as large as we can go without having the knife end up being thicker to accommodate them.

At 1.5mm diameter, this is where we can fit 57 balls into the cage.

Like I said earlier, I've not done a huge amount of research into bearings - so my knowledge is probably somewhat limited in this area.

There are a few options available here. I'm going to go through them quickly to explain why I think none of them are the best solution.

First up is having no cage to hold the balls. This is a great idea ...in theory. But I've owned a knife like this. The balls go everywhere (even if you're careful) when you disassemble the knife.

Remember, if we're making the world's best knife, we must allow of easy maintenance of it. To have balls flying everywhere is an absolute deal-breaker ...no matter how smooth the opening of the blade is!

This is the cheap-and-cheerful way to do it. Not something I would consider putting on this knife.

Using a type of plastic such as acetal (a version of this I've seen used is Delrin) is, to my mind, a good solution. Low friction and somewhat easy to manufacture. However, I don't like it mainly because it seems a "cheap" type of material to use on a knife such as this.

High lubricity is what this material is. It also looks pretty damn cool. This seems like a good material to use - but I feel there is an even better material...

Just in case you're not too familiar with Zirconium...

It's similar to Titanium in that it's non-magetic and won't rust or corrode. It is much softer than Titanium (certainly Grade 5 that's the most common type).

But here's the crazy thing:

It has this freaky thing it does when you heat it to a high temperature. Zirconium will turn black and acquire a VERY HARD (but thin) surface coating on it.

I don't know the physics of it - it's just incredible though. In fact, it's so incredible, I've been using this feature of Zirconium for the last couple of years in the product of fidget sliders (which, as I mentioned previously, are the most popular fidget sliders in the world).

The friction of two heat-treated Zirconium parts sliding on each other is very, VERY low.

Thus, making the bearing cage from Zirconium is a no-brainer (but, again, this is in theory!). The ultra-hard Silicon Nitride coated ceramic balls moving against the ultra-hard surface of the heat-treated Zirconium should result in a near-undetectable amount of friction.

If it works how I hope it will, these will literally be the world's best knife bearings installed in the world's best knife. How is that for absolute confidence and arrogance!

Back to the number of balls we touched on earlier...

Like I said, 60 was what the CAD model showed was possible. But, once it came to the CNC machining and the reality of working with metal, 60 was not possible. This was because the "walls" between the holes we were machining for the balls would break through to each other. The metal was just too thin.

We then reduced the number of balls from 60 down to 58 ...the same problem was happening (but it was close to not happening). So we reduced it by one more to 57 and it worked. So 57 is the magic number for now.

As we know, bearings are the "middle man" between the blade and the scales. It's what allows the blade to rotate freely - but at the same time remain rigid.

There is a decent amount of force (or pressure you could say) exerted on the bearings from the tightening of the pivot. This force on each bearing is equally matched with the same force by the blade on one side and the scale on the other side. This is, as you already know, Newton's Third Law: "For every action, there is an equal and opposite reaction."

You're probably wondering why I'm now throwing Newtonian Physics quotes at you, right?

Well, there's a reason. The force on the very hard ceramic bearing balls -- 57 of them -- is matched on one side by a fairly hard blade - the blade will likely be in the region of 60 to 61 HRC (this is not fully decided yet, but it's an adequate placeholder number for what we're talking about right now).

On the other side of the bearing is the Grade 5 Titanium scale. Typically Grade 5 Titanium is around 36 HRC -- which, relative to the blade, is a lot "softer". What I'm thinking is that the bearings are not going to have much effect on the blade (no matter how hard we tighten the pivot) because it's very hard balls pressing onto very hard metal.

However, with the titanium scale, it's very hard balls pressing onto relatively not-hard metal. From what I can tell this is what most knifemakers do. They allow the balls of the bearing to create a "track" into the titanium scale. From my understanding of metals (which isn't huge if I'm honest) the titanium will essentially compress initially a small amount where the balls press into it ...but not much after this.

This is actually not a bad solution. Using some sort of hard shim (which is really just a thin washer) between the bearings and the titanium scale is something I like the idea of - but I'm not convinced it's required.

I have a theory... (yes, another one!)

If there were just 10 balls in the bearing (like most knifemakers seem to use), then I can see the balls creating a "track" into the titanium scale. However, because we're using 57 balls, the pressure from the tightened pivot results in a significant increase in the distribution of the pressure. And, in turn, I reckon there will not be much of a "track" made into the titanium scale from the balls ...even though the force from the tightened pivot will be significant.

As will all my theories, I could be flat-out wrong. I really need to test it though. If we put shims into the scales of the world's best knife when they really don't need to be there ...that's definitely a reason to hang my head in shame and sneak quietly out the back door.

So, yeah, we're going to try the bearings directly onto the titanium scales and see if the theory holds-up.

To get the balls in there we just press them in. Have you ever put a jumper on? Of course your have. Well, it's exactly the same as that. Your head goes through the neck of the jumper - with the neck stretching temporarily for you to be able to do this. It is exactly the same with the balls going into the bearing cage. The cage will "stretch" (on a microscopic level) as the ball is pressed in. The ball is then held inside the cage because the "neck" is smaller than the diameter of the ball.