It’s not about precision, or even power.
It’s about the geometry of all galvo laser systems.
The beam is always taking a path from the mirrors (the focal length of the lens plus about 40mm or so) and is 10mm wide at the mirrors. It will be about 15 deg from vertical at the edge of the workspace for any lens.
The true focal point has a 1/e2 dia of somewhere around 0.025mm-0.040mm down the centerline on an F160 lens with a 10mm galvo aperture. The dia is about 50% higher at the edge of the workspace, so best energy density is only 44% of workspace center. 1/e2 is a figure for where the laser energy drops below 13.5% of the intensity in the center and 86.5% of the beam’s energy falls inside the 1/e2 circle. It’s gaussian distribution, a “Bell Curve”, of energy density vs radial distance from the center.
That’s the overall energy distribution at the focal point. The actual spot size of the mark varies. It may be that only a 0.01mm dia circle has enough energy density to affect the metal surface so the mark is 0.01mm dia. It’s also possible that a really powerful pulse could penetrate deeply and create subsurface expansion that explodes off a crater of metal larger than the focal spot.
For a 110mmx110mm workspace lens, the beam dia increases by 0.0625mm for every mm past the ideal focal point along the lens axis. The neat magic of the F-theta galvo lens is that the focal point is not a constant distance from the lens, that would have made a focal plane into a hemisphere of about 200mm radius. But F-theta design means an F160 lens has a focal point 160mm from the lens’ virtual focal plane only when going straight down. It is longer focal length at nonzero angles (theta) that matches the hypotenuse so it creates a flat focal plane to work on.
At the center of the workspace, the focal energy density drops to 50% when you move 0.66mm below the focal point. The edge of the workspace, 55mm from the center, the beam is slanted about 15 deg from vertical, and the energy density drops by half at about 1.5mm from the focal plane.
The thing to see is that the beam energy is a converging/diverging cone, 10mm dia at the mirrors and near zero at the focus, but it also slants as high as 15 deg to reach the edge.
It can cut at the surface easily enough, about 0.05mm deep, perfectly. But then the energy cone of the beam needs to have a clear path carved out for it to reach the bottom of the cut. The geometry of what you’d have to carve out to make a path to the bottom of 1mm thick material is totally different at the edge of the workspace vs the center.
Additionally. the beam needs to be focused on the bottom of cut as that bottom moves. That can be done manually by just readjusting the Z focal height, but note that’s not the same as clearing the path for the beam to reach the cut bottom. In fact, when you lower the Z by 1mm, due to the slant of the beam, the point where the beam intersects the surface of the material will shift inward by 0.275mm at the edge of the workspace but none in the center.
Or, if you’re vectoring a line on the edge of the workspace, it will engrave a line about 0.035mm wide. If you do it 100x over, it will get slightly deeper and cut at a 15 deg slant but gets out of focus. Lowering the Z by even 0.064mm will shift the beam’s center point 0.035mm/2 towards the center. Now half the beam’s energy cone hits the surface outside the cut, and it’s somewhat out of focus there.
The galvo laser is fundamentally different than the CO2 or fiber laser cutting bed. To cut metal, the metal must be cold and then get hit with an extremely fast, dense energy pulse that explodes off a tiny particle of metal. If you keep drilling the same spot, the material will heat up and the surface starts to melt or at least become viscous and the same beam energy can’t explode a chip off liquid metal anymore. It will just deliver more heat and melt it but unlike a fiber laser cutter, there’s no air jet to blast a clear cut through this molten metal path. So all it does is melt, flow into the cut you’re trying to make, and actually close it off instead of deepening it.
So where the beam’s energy cone intersects metal outside the focal point, it doesn’t self-clear that path to reach the bottom. It just heats up the metal and it may even close the channel off.
Lightburn offers a “wobble” cut to try to get that depth cut by making a wider path. It is not a really comprehensive solution though- it’s able to get some depth out of a cut in the center, but much less so off-center, and it can make a problematic amount of heat. I’m not sure Lightburn really understands the problem, there actually is an algorithmic solution but it’s pretty weird.
