Designed this dial arm to be used with a dial gauge and a 1/4-in bit to help measure and adjust the perpendicularity of the spindle and figured I’d share the files in case anyone is interested.
With this I was able to adjust my spindle to approximately ± 0.02 in (±0.5 mm) over a 30 cm span, which equates to approximately 0.09 deg.
I consider this pretty good for what I use the MPCNC - hope it helps others…
(the file is large because it includes a photo of the dial arm in action) dial gauge holder.zip (1.8 MB)
Just did some more calcs and ultimately, on a 1/4-in bit, my results are roughly equivalent to a 10E-5 mm runout - i.e., negligible… pretty happy, hope this eliminates some of the marks I was getting on some profiling cuts.
But, doesn’t this rely on the work surface being perfectly flat? I’m wondering if a way to isolate work surface irregularities might be to use a smaller flat surface like an aluminum plate sitting on the bed. Then doing Z-probing at three points on the plate and shimming the plate to adjust all the points to a horizontal plane. Then using this tool against the shimmed plate???
There have been a few threads diving into this subject recently. Another thing to watch out for is that while this tester does help you adjust your bit to perpendicular, it doesn’t really help with getting the z axis travel perpendicular… Just something to be aware of.
For that I suppose you could use the same tester but with the dial indicator arm in the horizontal plane, then indicate that by riding it up and down against a trusted square.
I think it would be possible to borrow an idea from here Automated measurement of Z axis perpendicularity and jog the x/y in a circle around a single point so a single point on the work surface is used. Then you’re not measuring spindle relative to plane of surface (which may be poor) but you’re measuring spindle relative to plane of x/y movement.
In the end it’s probably equivalent to surfacing the spoilboard or shimming a true plane parallel to the x/y movement plane, but could skip a step and maybe more direct – fewer opportunities for errors to creep in.
That is an interesting approach… If there was a way to ‘capture’ the indicator position so that the spindle rotated while the router was traversing I think you would get a good visual as the indicator needle swept while the router was rotating… although I guess the whole indicator would be rotating also, so not so good for visualizing the needle sweep…
Probably just easier to put a spot on the table and manual spin the indicator back over the spot.
Good point. I would think to test for that though you would want to have already made sure your spindle is perpendicular to the bed. So that you’re chasing one ‘error’ at a time…
Another interesting approach. If you setup a flat plate for the initial testing (instead of a single spot) you could then rely on that for resting the square…
But, doesn’t this rely on the work surface being perfectly flat?
You are correct - my work flow was to surface the spoilboard using a 1-inch surfacing bit - this made sure the spoilboard was parallel to the X-Y travel and gave indication of how off the Z-gantry was. I then used the gauge to adjust the perpendicularity.
Another thing to watch out for is that while this tester does help you adjust your bit to perpendicular, it doesn’t really help with getting the z axis travel perpendicular…
Yes, I’ve been trying to figure out how to adjust this but have not been able to come up with an “easy” way. As you can see in the photo, the shim is not that thick and is placed between the spindle holder and the Z-gantry tube.
Adjusting the z rails and z motion relative to the plane of x/y motion would not be affected by leg height. I think it can be fine tuned to some degree by the tension bolts ABC but I have not paid attention to the details of which bolts move which way.
When you level the spoilboard, XY will be parallel; however, if the Z is not perpendicular, you will see lines along the tool path caused by one side of the bit being slightly deeper than the other.
Alright, this will take some modeling (and trial and error) but I think I know what to do to adjust the Z-rails. I’m going to modify the top and/or bottom parts of the gantry (in my case, I only need to modify the bottom part) to add set screws that will allow the bearings on the top and bottom parts to be moved front and back (by no more than 1-2 mm) - specifically, the set screw will allow the bearing to be moved to the front. The Z-rail can be kept snug by loosening/tightening the screws on the front bearings.
I’d like to hear Ryan’s take on this, if he has any issues before I spend the time.
Don’t know if this will make more sense as to what I’m thinking, but the photo shows the top (XYZ_J_burlly - modified) partway printed to show where the two set screws will go plus the widened (by 2mm) screw hole.
This shouldn’t be needed. Tightening that screw should pull the other two bearings closer to pull that part of the Z back and tighten the grip. Are you trying to adjust the Z separate from the tension?
Yes. I originally adjusted the spindle perpendicularity using the dial arm from my first post and used shims to adjust such that the bit/spindle is perpendicular but the rail movement is still not perpendicular to the XY plane. The modification will (should) allow me to adjust the z movement while maintaining proper tension on the bearings.
While I’m confident the modification will work (the mechanics are straight forward and I did a test fit with loose parts), I’ve decided against installing it on my machine. I did a test cut a 80 mm (L) x 80 mm (W) x 40 mm (H) cube with facing operation on the top.
I could not measure a lack of perpendicularity between the top and the sides (XZ or XY planes);
same goes for the XY plane, all sides were perpendicular.
The facing operation on the top of the cube was beautiful and any marks could be removed with 220 grit sand paper.
This tells me that, for what I need, my MPCNC is dialed in to tolerances well beyond what I could use it for and I should just “let sleeping dogs lie.”
If anyone is interested in the part, let me know and I’ll post the obj. file - I only modified the top part, which allows the Z rails to be moved from the original position towards the negative X and Y (i.e., spindle movement in the positive X and Y direction).
Do you have the dual endstops setup on your machine? I ask because I didn’t have them on my machine at first and my original test of XY perpendicularity seemed really well dialed in. Later when I did some real parts (this was in metal, so less forgiving than just wood carving) I noticed some very real out-of-square issues (not horrible, but measurable and noticeable). The worst part, for me, was that the skewed-ness wasn’t consistent from run to run. I’ve recently bitten the bullet and installed the dual endstops and auto-squaring. I’ve not yet run any parts with that configuration, but I’m hoping it will hold tolerances better than I was seeing before.
