rigid tapping fail advice and "discriminant" err

What is the acceleration of you spindle? I am trying to get this to trip in sim and am not having any luck so far

skunkworks wrote: What is the acceleration of you spindle? I am trying to get this to trip in sim and am not having any luck so far

Sam,

In the INI file (also attached) is this:

MAX_VELOCITY = 2.5
MAX_ACCELERATION = 50.0
# The values below should be 25% larger than MAX_VELOCITY and MAX_ACCELERATION
# If using BACKLASH compensation STEPGEN_MAXACCEL should be 100% larger.
STEPGEN_MAXVEL = 3.125
STEPGEN_MAXACCEL = 62.50

Or were you talking about VFD accel parameters? In the VFD I am accelerating from 0-3400 rpm in 1.0 second (linear ramp).

-Tom

I should also add that sometimes, AFTER completion of the cycle, I get an error that says (paraphrasing) 'Linear move on line 35 would exceed joint 2's physical limit'. I get this sometimes along with the other error, and sometimes by itself.

I just noticed this error today after running this quite a few times.
-Tom

Also, also... I noticed that even though my deceleration in the VFD is set to 0.8 second from 3400 to 0 (again linear ramp), at 275 RPM (stops very quick) it looks from this attached video that the spindle doesn't stop when motion in Z stops. You will see slow motion just at the point of transition. You can see, from black sharpee lines i drew on the workpiece at the collet that the piece is screwed up the tap by the spindle - and hence pulled out of the collet a bit. When the cycle finishes and retracts the tap, it isn't fully out of the workpiece. I previously had a G53 Z0 X0 which retracts and [with tap partially inserted] SNAP...That is what happened the first time I am sure. I have since removed the G53 after finish so it just sits there...



-Tom

Could we have your whole config? Rob it thinking the discriminant error is a red hearing. Either you are missing steps or there is some sort of configuration error.

skunkworks wrote: Could we have your whole config? Rob it thinking the discriminant error is a red hearing. Either you are missing steps or there is some sort of configuration error.

Sam,
I will post the whole config later today if II get to the shop. I could be missing steps. I upped the speed recently, running at 150 ipm. I spent several hours running a long series of movements the other day and didn't have any lost steps (measured a return position with a dial indicator), and I have run some programs that have run fine so I was fairly convinced that wasn't a problem. But, I have seen some other odd issues and am now wondering if that could be the problem. If I get a chance I will also slow the machine down and see if that makes any difference.
-Tom

Could you also plot motion.spindle-revs vs z axis.2.motor-pos-cmd?

sam

skunkworks wrote: Could you also plot motion.spindle-revs vs z axis.2.motor-pos-cmd?

sam

I assume you mean with halscope, motion.spindle.res on one channel and z axis.2.motor-pos-cmd on the other?
-Tom

yes

Here is the plot...