got it - so with a ~1mm out of square I should be able to pop the dust covers off and check those rails for square from top to bottom to see if they are good - which I will do. That said, I’m also thinking that if that mounting plate behind the moving plate seems to be dead square (using machinist’s square not dial indicator) then the rails should be square right?
What matters is not whether each individual component is perfectly square or vertical in isolation, but whether the machine’s axes remain square to one another throughout their full range of travel. CNC accuracy is determined by the geometry of motion, not by the geometry of the individual structural members. As long as the linear guides, rails, and drive systems produce orthogonal, repeatable motion, the machine can cut accurately even if some supporting components are not perfectly aligned.
The spindle is the primary exception. It must be perpendicular to the machine’s plane of motion to achieve accurate machining and good surface finish quality. Spindle orientation can be adjusted independently through tramming or shimming, so it is not necessary for every supporting component to be perfectly vertical.
In your case measuring the rails could get you reasonably close, but the most accurate method is to establish a reference surface that is aligned to the machine’s XY plane, then jog the machine along the Z-axis while monitoring a dial indicator against a known vertical reference. This directly measures the squareness of the spindle’s motion relative to the machine’s coordinate system, which is ultimately what matters.
That mostly makes sense to me, and I really appreciate your detailed response.
Two things - knowing that this whole thing started with a terrible tramming test with 1.5ish mm ridges which led to the 1mm ish measurement of the forward leaning nod mostly on the moving plate.
Based on everything I’ve read, chasing down whatever mechanical problem exists that produces that much of an out of square situation is the first thing to do, then dial indicator test like you described should be within a few thousandths or something where could then tram with small shims.
Am I thinking about that process/progression correctly?
The one thing I’d clarify is that a Z-axis geometry issue and a spindle tramming issue are two different problems, but related. I found that once I had dialed in my Z-axis geometry, I no longer had to tram the nod in the spindle, only the tilt.
If the Z-axis itself is nodding forward ~1 mm through its travel, I’d investigate that first. Tramming shims on the spindle only change the spindle’s orientation relative to the Z-axis—they don’t correct a Z-axis that isn’t moving square to the machine’s XY plane.
My approach would be:
-
Verify the Z-axis motion is square to the XY axes
-
Fix any significant nod/tilt in the Z-axis travel via loosening the bolts I mentioned earlier, or identify if there is something else going on in your assembly
-
Then perform the final spindle tramming adjustment.
The dial indicator test I described measures #1. I do something similar to @corbin here for #3: https://www.youtube.com/watch?v=wP0vGhGectg
I probably take this stuff more seriously than most avid users because I do a ton of multi-sided operations that require a very very tight tram in the spindle and Z axis. For people only cutting sheet goods I think using a precision square off of a surfaced spoilboard and some feeler gauges could get you 90% of the way there, however a cheap dial indicator, precision square, and leveled surface plate would be my preferred way to measure Z axis tram.
Thanks again for the detailed response and direction. I’m honestly really hoping there’s something that I missed on that moving plate since it seems to be the only thing that’s off on the z using precision square+spoilboard. I did end up mounting a digital dial indicator on a magnet base today and ran the z up and down almost the whole length of the mounting plat that rides on the bearing blocks - there was .04mm difference between the top and bottom which is not perfect but I think a very small difference overall.
I was hoping to do inlays and some simple two sided carves (I think) for things like kitchen utensils - for those things I’m not too sure what kind of tolerances I need to be chasing here - obviously better than the 1mm nod but is .04mm that I should be able to shim good enough? I would think so, but I don’t know what I don’t know on this stuff.
Anyway - thanks again!