The simplest target is a cylinder. A small cylinder with a vertical axis gives higher sensitivity than a larger cylinder. It is also simpler than a cone and can be the rod that you have or are going to turn. The patented device with the laser on the outside of a chuck i.e. say 30mm from the axis of rotation produces very low sensitivity on targets of 1-20mm. My suggestion in the two year ago thread implied that the laser is much closer to the axis of rotation. Indeed this is what Chris made and exhibited at AP. He spun the device in a Myford 3 jaw witha beam just off axis to demonstrate. The beam hit the tailstock centre and gave a nice slanted line across the cone. Not quite the same animal as the patent, but the patent allows drills & mills to be held while being aligned.
The geometry of the device is the key, Think of the axis of rotation then a beam crossing the axis at 45 degrees and hitting a round rod also having the same axis as the beam rotation. We would see the circle drawn around the rod. If we move the rod 5mm at a right angle to the axis of rotation then the circle is tilted into resembling an elipse, with the high point 5mm up and the low 5mm down on the other side of the rod.
Now if we move the pointer so that it makes an angle of 5.7 degrees ( roughly 6 will do fine) we only have to move the rod 0.5 mm to get the same apparent movement of the beam . If we can set about 1.1 degrees then the 5mm displacement happens with 0.1mm ( 100 microns) of movement so this a sensitive arrangement but demanding on beam quality.
The vertical displacement for a given displacement is only related to the beam angle from the axis of rotation so a smaller target makes the sloping elipse proportionally steeper so easier to see.
Fuzzy spots do limit shallow magnification, as suggested earlier increasing the background illumination will help. There are some well known optician tricks such as stopping the source down or placing an occluder in the beam path to produce sharper spots, worth a try.
The beam colour will not make any great difference for a given beam quality, absolute sensitivity is not an issue.
Engineer's Buttons don't need mods, you can use a shallow angle on the sides – which should have a lovely smooth surface. But a sensitive arrangement might have a very round plug in the hole as a target , a turned step around the hole or use the hole( the top feature allows several buttons to be located without moving the buttons which could be bolted down.) .
Another suggestion that was made originally was to be able to change the angle by changing spindle speed to avoid the need to keep stopping and starting. This idea was shot down by a few people in the original thread without much ado they said that they thought that vibration and other issues would afflict the device. It actually works very well indeed, the implementation was a bit more complex than I described but it allows much easier actual use when working with shallow beams, in effect the speed control is a remote beam angle control .
So if you want to make one of these gizzmos I would suggest that you fix the pointer to a rod so that you can move the pointer from the axis of rotation outwards and that you can adjust the inclination to an almost vertical angle.
Billy.
PS Many thanks MG for the quadrant detector link, I was mighty impressed by their price tag, thinking of knocking out a few….