Last modified: 2024-07-13 10:51:26
< 2024-07-09 2024-07-16 >The plan was to get the machine finally mounted in the tub and start checking the axes for squareness.
However the M8 bolts I bought are just slightly too short. I should have been able to predict this. With rubber washers on, they end up flush with the underside of the table, which means I can't get a nut on the bottom. Rats.
It was very difficult to get them in as well, because I have to position the rubber washer underneath the very heavy machine. I'll just have to get some longer bolts and do it again.
So for now let's ignore bolting it to the tub, fit the stepper motors, and then try to check squareness.
Steppers are back on. I think the Z axis one wants the cable to be hung from higher up, and the X axis one wants the cable rerouting. Maybe Y axis wants rerouting as well. But, onwards, for now.
I think the plan is to somehow get an indicator on the spindle motor housing, and a square on the table, and then sweep the indicator along the square to square it up in one axis, and then see how far out a perpendicular axis is.
So first I'll square it up on Y, and then I'll measure X.
Hmm, can't stick an indicator to the spindle housing. There's not enough steel on this machine! For squaring X and Y I can stick it to a Z rail.
And now how do I clamp the square to the table? I think for now we'll measure off the edge of the table. Then I should be able to tell if it's at least substantially as square as the router was.
Y axis steps/mm is half what it was with the old machine. I think 800 steps/mm on each axis?
How do you reverse an axis in LinuxCNC? Set a negative scale? Looks like it.
The table is out of parallel with the Y axis by 0.25mm over 140mm. The front of the table is skewed left. So we'd expect the right of the table to be skewed forwards if both the machine and the table are perfectly square.
So what does X say? Almost exactly the same! About 0.24mm over 140mm. I'll take that.
And now the Z axis? Firstly the Z axis needs to be perpendicular to both X and Y, and secondly the spindle axis needs to be parallel to the Z axis.
I can use the cylindrical square that I made from the lathework book! This is relying on the table being broadly flat, but I think the table is probably flatter than the Z axis is square.
The top of the Z axis leans away from the table normal by 0.15mm over 77mm.
Let's try that again just to make sure it's consistent. Same direction but 0.09mm over 77mm. Could the "cylindrical square" be tapered?
22.99mm at the top and exactly the same at the bottom, so no, this isn't explained by the square being tapered.
I should measure that the table is flat. The table rises at the left by 0.58mm over 230mm. So I think that means the Z axis leans much further away from the X axis than I previously thought. I should get some plate on the table shimmed up so that it is level, then put the cylindrical square on that.
OK, I put a plate down with no shimming, and now getting 0.26mm over 220mm, higher at the left. I'll shim it with a piece of paper at the right.
Z axis now showing 0.35mm over 90mm. Top of Z is leaned right relative to the X axis perpendicular by 0.22 degrees.
The table is almost totally parallel to the Y axis (as you'd expect, with hardly anything in between). The Z axis leans forward at the top by 0.10mm over 90mm, or 0.06 degrees. And basically the same measurement when repeated.
I haven't measured the spindle axis at all, but we have:
And then I need: