So I got home today check to make sure that the screws do not in fact hit the carriage, they don’t. I reran the same job again, could not reproduce the same results.
So I started to rerun the bit holder job noticed it was doing the stair step cut again, so I turned up the pot on the x and now it seems to be cutting good.
Hopefully it is good to go now. Thanks for all the help and the input from everyone.
I Think is A EMI Issue If is that go to You tube and look for (Eliminate EMI Issues From The Beginning ) hope it helps.
Kind regards:
Victor.
ups Just saw the msg after 9 month later 000 (o)(o) 000_
Possibly a broken wire in the drag chain?
Residue on the rail? Damage to the rail?
Absolutely, and the Y-axis is the main axis for driver issues (as the two motors share a single driver stage)
What are your current GRBL parameters for:
$111
$121
What is the voltage reference (Vref) for your drivers?
Also what steppers do you have, are they 1,7" or 2,3" wide?
For the time being derate $121 to equal 100 and retest.
A good test for overheating of drivers are:
Draw a shape of some size and do an “air carve” at max feed rate and very shallow depth per cut - you want the Y axis to move large distance back’n’forth. If the driver is triggering overheat protection is will start to halt, restart, halt etc. This will be a firm clue.
Base parameters:
$111=2500 and $121=100
Test, if no fault increase $111 in increments of 500 untill overtemp is triggered. Then derate that value to 80%.
Repeat process for $121 with increments of 100 untill you start stalling (it will make a grinding noise) and derate to 80% here aswell.
Ruling out possibilities usually end up with finding the solution (often as last item on the list…)
The gshield is inferior to the Nema23, let alone two sharing one driver. And torque drop at speed, just steppers 101.
We do know that your current configuration do not work, never have. Hence we need to break this down and start slow, pun intended 
What we do know:
Nema 23 140oz
gshield
8000mm rapid and 500 acceleration
Y axis fail, not in the same place (you posted elsewhere you tested a different area with similar result)
Y-axis have two motors, sharing a single driver, so each Y have 1/2 the power each compared to X/Z stepper
It isn’t conclusive though but all suggest Y-driver overheating, causing it to loose Y step sync.
Ruling out the possibilities though, ensure that;
But my bet is driver overheating 
Good to hear, glad you finally decided to measure Beats guessing all day long
Happy carving! 
