Axis creep problem

[robert_cheal]

My X axis was out by about .05" after I cut a 30" dia. 3" wide (circle radius) molding by a manually created method similar to how the Extruder works.

I created the tool path with about 60 concentric circles offset .05" apart and then set to elvations so that the cove cutter would shape the profile that I needed. I connected these circles with a short vertical step and a X move to the next circle in the path. At 2" cut speed it jerks just slighly at each change.

I have a 4G PRT that started acting up and lost its position by several inches a few times that same day.

I am also waiting for the arrival of new replacemnent pinions.

I can't think of any other factors as to cause of my problem. Perhaps grounding for a 4G any, special needs?

Before I try to create another piece of $300 worthless MDF I thought I better seek out any advice tha might be offerd.

Thanks, Robert

[richards]

Robert,
I didn't notice when I visited your shop whether you had a common ground on all pieces of equipment. On my Alpha, I have all axes linked with 8-gauge grounding cable which attaches to a common point on the control box. The frame of the computer is also connected to that same point, as is the ground wire from inside the control box brought out and connected to that common point. In addition, I have a flat braided copper wire winding through my dust collector that is also grounded at that same point; however, the other end of the wire in the dust collector is left unattached. In short (no pun intended), a resistance reading from any point on any axis to the common ground point reads zero-ohms. In fact touching bare metal on any piece of equipment with a probe shows 0-ohms.

That takes care of part of the problem. To fix other problems, I replaced all signal cables in the system with a shielded cable whose conductors are twisted pairs. Since doing that, electrical noise has not been an issue. Also, some very reliable people suggest connecting circuit ground to chassis ground. I still haven't made that step. (Old ideas and habits die a hard death.) In theory, it should work perfectly, but I still prefer to keep circuit ground for the control computer totally separate from the chassis that holds that circuitry. If I had tried everything else and still had a problem, I would try connecting circuit ground to chassis ground.

It may be a coincidence that your part is off by 0.05" when each pass is also 0.05". Is it possible that the count needs to be incremented by one to make everything work properly?

[robert_cheal]

Mike,

I did not actually measure the .05" but when I drilled a new reference hole adjacent to a known reference hole I approximated the distance. The part being cut was first roughed out with a 1/2" straight cutter and then finished with the cove cutter using about 60 rotations around the part. The only reason I noticed that the part was off was that one of the straight roughing passes was also a meant to leave a 1/16" X 1/16" notch in the finished profile.

I was thinking that my PRT would only have to lose or gain .001" in each of the 60 rotations around the part be off at the end of the file.
I guess that I could break the finishing passes of the cut file into 4 files and recalibrate each time. The straight cuts of the part could go last and I could limp through this project. That kind of leaves me disappointed that at present that my 4G PRT can't perform the task without jumping through hoops to get there.

I was wonder about worn pinions causing part of the slop but form what you’re saying it sounds more like electronic instead of a mechanical problem.

And now to the grounding. I would say mine is poor, a simple 12ga. wire from the control box to the grounding slot of an outlet in, the same duplex that I am plugged into. Also a simple wire from one of the table legs to the same outlet. And of course no grounding on the dust collector. I had 8 ga. wires from each axis linking up to this same connection prior to my 4G upgrade but I was getting little shocks from my machine so I unhooked it so time ago. Am I the front runner candidate for “electrical noise idiot” of the forum?

I need spend the time to implement your recommendations. As far as replacing all signal cables in the system with a shielded cable is there any special place to get it and are the same connectors used?

I hope that fixing grounding solves this and when my new pinions arrive that would just make me feel better. If nothing works after that I hate to think this project exposed the certain limits of my machine.

Thanks for now,

Robert