Hi, the  formula that came with the machine is .695 x 200 x 8 = 1112 steps per  inch. I found a video for calibration but I am not sure I under stand it. It measures the distance of a known part and compares it to the DRO readout and then adjusts the steps per inch setting accordingly. I will have to watch the video a few more times.  If you have any ideas please let me know.

200 is the steps per rev of the stepper motor, the 8 is the microstepping and the .695 must be the distance traveled in one rev of the motor so assuming that is correct then the Steps per should be as stated 1112.
  If it had ballscrews then the calculation would definitely be the way to set the steps per but as its a timing belt there may be some slight errors in the pitch which is not working in your favour so if you have an accurate means of measuring you could use the "Set Steps per Unit" utility on the settings page but that will only ever be as accurate as your measurements.
Hood

Hi, I found the axis calibration tool on the setting page of Mach 3. After calibration I was able to use  G code to move the axis a set distance and the dial indicator show the correct distance was moved. The DRO showed the movement to be short by .0003 and when I returned it zero the dial indicator returned to zero, but the DRO would be 0.0009. When I run a cut with a lot of x changes the error adds up and  creates a real noticeably  affect making the part useless because it is a rotor and it is out of round.  If you know of a way to correct this problem please let me know. Thank you for all your efforts. Best Regards

Your problem will be the low resolution of your system, 1112 steps per inch equates to a resolution of  0.00089928...... which just so happens is what your DRO is showing
  In reality your resolution wont even be that good because microsteps are not uniform or exact, if you want better resolution you unfortunately need a better machine.
 Having said that  it should not increase in error as Mach keeps a track of it, it is likely to be that your steps per unit are not quite right if you are getting an increase.
Hood

You also say "out of round" that could be causeg by backlash if the out of round is at the cardinal  points.
Hood