I milled slots into 2x1 aluminum with a 5/8" bit and ended up with the slots off center by about 1/20". I used a manual edge finder with a .2" diameter to calibrate the DRO and set the y value at the edge to 1.1". I then moved the mill to y=0, where I milled the slots. I was very thorough when calibrating it and when milling, but when I was measuring the distances from the edges of the slots to the ends of the 2x1 yesterday with calipers, I noticed that one side was showing a distance of ~0.72" and the other was showing ~0.62". I can’t figure out for the life of me what could have possibly gone wrong. Anybody got any ideas? Is this amount of error normal for a manual mill?
Perhaps your 2X1 is not actually 2". If you want to ensure you are directly on center no matter what the actual width of your material is:
- Use the edge finder on one side and zero the DRO
- Then use the edge finder on the other side, divide the number on the DRO by two and enter it.
- Move you axis to zero and you should be on center.
As a sanity check, get a pointer and move the machine to XY-0. If the pointer is directly on the corner than you should be fine.
Perhaps you are not using the edge finder correctly or maybe it’s off?
Is it due to backlash?
Are the glass scales for the DRO clean/not broken?
Are your vises indicated? Also what kind of mill are you using
I very much back the above 3 commented suggestions. You can always give me a call if you want, sent you my number.
I measured the aluminum with calipers before milling to ensure it was actually 2".
It sounds to me like your vice is not square to the X axis.
If you insert a dial indicator into the spindle and jog the X left and right, the dial indicator should not change when pressed against the inner face of the back vice jaw.
If you subtract the widths from the tube width… 2.00-0.62-0.72= 0.66. Either your cutter is on the large side or your table is walking?
What type of work holding are you using? A vise?
I’m pretty sure this was the issue. I was running some tests with the milling machine today and I found that when I used the edge finder on a piece of 2x1 aluminum, the y-value I found for the edge of the aluminum varied based on where I measured it. Specifically, the y-value changed by about .08" for a 12" horizontal change, which is about 0.067" per inch along the x-axis, or an offset of about 0.38 degrees.
I guess the moral of the story is two-fold: 1. Make sure you edge find as close as possible to where you are going to mill. 2. Make sure your vice is square to the x-axis.
Thanks for all your help everyone. I feel much better now that I know what went wrong.
Awesome that you see what went wrong. Doubly awesome that you stopped to check your assumptions, bounce ideas off people, and consider feedback. Keep it up. If 1 scrapped part = 1 lesson learned, then you’re getting a great exchange rate.
- Repeat 2 every so often during build–PARTICULARLY if, as we do, you happen to have a pivotable vice. (That also makes it much easier to re-straighten…) I usually target .000 over 6", but will settle for <.002" over 6" if I’m in a hurry and it’s taking forever.
I got bit that way last year.
Also: Make sure to check your table as being square to the head at the same time.
For my team, I check the table once or twice a year, typically right before build and at the start of the fall; the vice gets right before build and if we’re working a long part or previously adjusted the vice angle for other parts.
Do this with your drill press too. We had a whole bunch of problems with things not lining up in the 2019 build season because the table was not perpendicular to the axis of the drill bit.
Often it’s not perfectly square. It could be 2" in one place, but +/- a little in different places; try checking your manufacture’s specified tolerances. Also, was the .05" error consistent across the slot?