But wait - what's going on with this diamond knurling?

Something doesn't compute here, for me at least. It may well be that I'm not getting the memo but either way, I'm struggling with something here.

This tool has 2 wheels, each with "straight" knurling wheels. So the direction of the groove that is cut is a direct function of the angle at which the wheel "holder" presents the wheel to the work. In my case, that is 30 degrees for each wheel. And the lower wheel is positioned on the underside of the work, directly opposite the upper wheel. This results in the grooves being at 30 degrees in the opposite direction. The result is 2 sets of grooves, each 30 degrees from the (horizontal?) axis. Here's a video from the manufacturer that almost shows how it works:

All that seems pretty clear at first sight - and almost intuitive. The edges of the sharp teeth on the wheels are pushed along the surface of the workpiece, cutting grooves as they do so. However, the lower wheel seems to me to be getting dragged backwards through the work, rather than pushed forwards. These wheels are both supposed to be actually cutting the workpiece, although the tool is described as being capable of both cutting and forming when the right wheels are fitted.

And indeed, when the wheel has been in operation, you can see that the upper wheel has swarf on its "front" face (at the bottom of this photo), whereas the lower wheel has swarf piling up against its "rear" face (at the top of this photo). And in the process, surely the lower wheel has been pulled strongly away from the needle bearing, not against the (hardened) face of the bearing, as was so carefully described in the handbook. I smell German rats.

This is another view of the tool in its operating orientation, showing swarf buildup on the back of the lower wheel. 

And this is what you would see when viewed from the back of the lathe. The tool is fed from left to right in the view.

This doesn't seem right to me. It appears that the lower wheel is actually being required to form the material, not cut it. The only way that wheel could cut in its current orientation would be if the tool was moving from right to left in the photo. WTF is going on? 

The "QuickTM" cut knurling tool:

To get another insight into how cut knurling tools are normally operated, here's another (German) cut knurling tool system from Hommel and Keller:

They offer only 3 styles of wheel for all of their cut knurling operation:

• AA - "straight" teeth
• BL - "left hand" teeth at either 15 or 30 degrees
• BR - "right hand" teeth at either 15 or 30 degrees (corresponding to the BL)

And both wheels are effectively placed at the same position ie in front of the workpiece

Here's a still from a video showing how they cut a diamond knurl This uses a BL and a BR wheel to produce the cross hatching.

And although the German site doesn't seem to want you to see any user docs, there's a Korean site that hosts some almost helpful instructions which clarify matters somewhat.

And most insightfully of all is this video showing how the wheels are adjusted to suit the work diameter. Turning the knob rotates the spindles that the knurling wheels are mounted on, to compensate for the different angular positions at which they contact workpieces of differing diameters. But bear in mind that these tools present the knurling wheels at the same side of the workpiece, not diametrically opposite:

Very interesting - but this is a different concept to the Boeni tool. 

How does this thing actually work?

I started out with just the upper wheel to see what it does. By turning the spindle manually with a feed of 0.25mm per turn, I can take a reasonable cut that generates visibly sized chips.

About 10mm of knurl here. It seems to have made a pretty good job, with nicely cut grooves. The swarf coming off seemed pretty convincing to my eyes.

I then fitted the lower wheel. The end result is a bit scruffy but I can't pretend the setup is in any way optimised.

Then I removed the top wheel, leaving just the lower wheel. It definitely felt and sounded different. And those grooves are looking formed, rather than cut.

View from behind.

In order to make the "lower wheel" more visible, I reoriented the tool so that it's actually above the workpiece but with the same angle it would have been if presented from below. To do this, the wheel is flipped so that is at 30 degrees to the x axis as before - but this time on the other side.

This gives a better view of the action.

I'd like someone (possibly a German one) to tell me that this wheel is cutting the workpiece. Anyone can see that in fact it's being dragged backwards in the groove. And even if the front edge of the tooth was somehow making a radial cut to the cutting face, there's nowhere for the swarf to go.

One key difference between cut and formed knurls is that forming results in an increased final diameter, whereas the cut knurls end up at nominal diameter (at best) or under size.

Here's the starting diameter (3/4" or 19mm):

The lower wheel results in an oversized final result:

Whereas the cross knurled piece is undersized.

So it's very clear that the lower wheel is indeed forming its grooves. This perhaps partly explains why the quality of the cross knurling is rather underwhelming. I'm not sure the manufacturer has been 100% straight here in terms of describing the product as a cut knurler.

Here's another example of a cut / knurl tool that is clearly behaving in the same fashion. You can see the lower tool dragging:

Well that's rather disappointing but at least I understand what's going on now. The question is - how to make the best of what I've got?

What we can do with this:

• 30 degree "right hand" knurls, properly cut.
• 30 degree "left hand" knurls, formed.
• er, a crosshatch formed by the above.

What else could be done if I modify the tool, buy additional wheels or make up some additional parts?

• Straight knurls - if 30 degree wheels fitted.
• Proper "left hand" knurls - if we run the spindle in reverse - but ideally with the bearing facing the right direction to take thrust.

Here's a fairly workable scheme - straight knurling wheel, canted over at 30 or 45 degrees, cut one direction, then flipped over to cut the other, with spindle in reverse: