NEED HELP, Mazak 18MS C axis instability, Mitsubishi drive problem?

Hello, My name is Jayson, I purchased a used Mazak 18MS, with a T32-3 control and have been trying to get it operating correctly. In a turning application there is no issues the spindle is smooth and quite, in a C axis mill contouring process lies the problem. I had Mazak come out and work on the machine for a day and that was a complete waist. The initial and repeating problem is in a C axis move, the servo shakes back and forth a few degrees in any called positions, but the machine position when its advanced to the point shows constant on screen. Mazak came out and tried to adjust the BKO-NC6572-H02 preamp, and the BKO NC6572-H21 sensor. No improvement was made. I was told I need to replace both. @ 10K. I bought an Oscilloscope, and the procedures from Mitsubishi for adjustment of the MBE90K set, I just got a new H21 sensor, installed it, wave forms are the same as before, C and D phases are average 10.5V P to P, spot on to spec, Y pulse is only coming in at a negative pulse, (constant 5.1 V, pulse drops to almost 0V then back to 5V level, when I try to adjust to a positive pulse, 2 pulses appear, ring is clean, the single slot on the ring is clear, the width of the pulse is spot on to the duration and coming in and out on the negative amplitude of the C phase just like in the Mits. spec, but just going the opposite direction, and my C axis consistently orientates to the same position. If I adjust to the positive pulse C axis will not repeat to same position, so I have it set to the drop pulse, It shakes occasionally here but repeats, and If I call out a line cut to lets just say 10 degrees to 15 degrees the servo shakes a little during the move and at the stop point. I can move the chuck by hand in rotation a couple degrees with minimal fight back from the servo. I just changed the Preamp BKO NC6572 H02 today with no improvement. I have went through the FR-SF 18.5 drive parameters and the parameters on the NC, comparing the to the factory paperwork that came with the machine. Some of the values were moved, I incrementally moved them back to the specced values with no improvement. It looks like this drive has been replaced or rebuilt in the past, but there is no paperwork to support this, and I have read about bad drive rebuilds. When the spindle is shaking in a stopped state in a C axis move, the Load meter on the NC on C axis is bouncing up to 30 to 40 percent with no work in the chuck. I can grab the chuck jaw and stop it and the load meter drops to 0. Please help. C axis is fairly quiet in movement too.