Hey all,
I finally got my custom sized 24x24 standard router going over the last coupe of weeks but I have been having some weird lost steps problem I can't seem to track down. The machine uses the standard linear motion system, but the Pro R&P drives, as well as the DiY nema 23 kit and an ESS. It is currently running off of my laptop, 1800mhz proc, 4gigs of ram, fully updated win 10, and I have as much background stuff shut down as I can. Getting no errors or anything.
When I first got the machine up and running as was playing with it to get used to how everything works as this is my first CNC machine it ran great. Then I had to take the gantry back off as I hadn't installed the sweepers. I ordered t-bolts instead of t-nuts and they interfered thought I could do without the sweepers, but that was wrong so I ground the down so they didn't interfere anymore. After I got it back together and squared up good I went to flatten the table top, but about halfway through, 20-25mins, I started to have a problem where everytime the Yaxis (this is the axis the gantry moves on for how I have it set up, slaved with the A) would start moving again when it came around the corner in the spiral the whole gantry would jerk like it had gotten out of square and was forcing itself back in.
After some playing I figured out that it was probably the Y motor losing steps as after it would jerk if I slowly went back towards 0 the gantry would hit the soft limit before it actually got to the end of the gantry, and the A side was always a bit ahead of the Y. At first I thought maybe it was the Gecko getting hot as it was pretty hot to the touch but not enough to burn you, I put a fan on it and let it sit to cool off but it didn't help. Then I decided to check the pulleys on the motors to make sure they were tight, they were and locktighted, no signs that they had been slipping. My motors have a flat spot on one side of the shaft and i made sure that one of the set screws was on that flat. The belts were also tight. When I had them disconnected I found out thtat the gantry was incredibly hard to slide, I think I over clamped the bearing slides when I reattached the gantry, so I loosened them up and readjusted everything and it rolls good, maybe 10lbs at most to move the gantry back and forth.
This seemed to fix the problem, I could jog it back and forth over and over again, and was able to restart the table surfacing run and finish it with no problems.
that was a week or so ago, I spent the last week making jigs and prototypes of the things I am making out of .070 phenolic sheet, had no problems, machine cut great. Today I went to make my first full run making 9 of one type at a time, they are 6.5x8 in rectangles with rounded corners and some holes drilled in them, nothing fancy. About halfway through the run I started losing steps again, the gecko was warm but not really screaming hot. I disconnected the motors the gantry still moves nice and free, and everything is nice and tight. So I started thing maybe there is a problem with the one of the Gecko drivers.
I zeroed my x-axis, and setup a dial indicator and zeroed it as well, then jogged the x back and forth its whole run about 30 times and then went back to zero and it was spot on. I swapped the x and the z motors to make sure the z was working as well and got the same results, so now I am pretty sure that the ports that the x and z are attached to are good, so I swap the x with the y, and the z with the a. Still got the binding on the y axis, and the x and z axis both work flawlessly.
At this point I was really stumped, I went and checked the motor tuning to make sure that the y and a were set to the same accel and steps per. The only thin I haven't really done is swap the motors themselves around. What is weird though is that as of when I left the shop I can zero the axes, and jog the y down to the other end and almost every time before I get back to zero it will bind, and if it doesn't happen the first trip it will within a handful every time. Thing is it wasn't doing this all week and I have probably jogged the thing back and forth a couple of hundred times during that week, checking fixtures and such.