Re: [CAD_CAM_EDM_DRO] Resonance in steppers

Jon Elson wrote:

>rosnekcaj wrote:
>
>
>>I am losing and gaining position error as my system goes through the
>>gcode I feed it. I suspect it may be due to stepper reonance. Is
>>there any gcode that would pinpoint that this is the problem. In
>>other words how can I find the sensitve velocity that triggers the
>>resonance.
>>
>>
>You don't mention your software, but assuming it has a feedrate
>override, run the same program with different feddrate
>percentages. This should scale all speeds in the program.
>If the position shift is the same for multiple runs at one
>speed, but varies at different speeds, that would indicate a
>resonance problem. If the error seems to be really random, or
>is not affected much by the feedrate override, then it may be a
>step pulse to direction timing problem.
>
>With properly tuned Gecko drives, there is so little stepper hum
>that you aren't sure it is even a stepper system. With some
>other drives, especially of the non micro-stepping variety, the
>stepper whine is quite audible at almost any speed, and becomes
>REALLY loud right near the resonance. If you hear a rising
>whine followed by a loud GRRRONK noise (best I can do spelling
>it out) that is resonance. If there is little whine and no
>grinding noises, you probably do NOT have a resonance problem,
>but either a problem with noise on the step line or a timing
>problem between step and direction.
>
>Jon
>
>
>

Software is EMC2, motor driver is quadrature and not capable of
micro-stepping, using Axis.ngc. Motors are synchro-syn 840 in-oz being
driven at 5.5 amps and stepping down 2.5:1 in the timing belt pulleys to
a .2 in/rev ball screw. Setup is on a 70's vintage BP that had been
retrofitted with balll screws and a Bandit CNC sometime in its history.

I am using a DRO to determine what tie final position error is after
running axis. Homing is always accurate to within .0015 in. Error after
running axis is any where from -.09 to + .04 in.

Max vel and acceleration settings vel = 90 ipm (1.5 ips), acc= 90 ipm^2
(1.5 ips^2) with about 10% faster settings on stepgen and equal settings
in Traj section.

Using these settings and reducing to as much as 10% of original values
show no improvement in the random tracking errors.
Trying to see what it would do I could jog up to about 150 ipm on x and
130 ipm on y before the corresponding axis would freeze.

Homing is always accurate to within .0015 in.

Here is a portion my ini file setup:

# Trajectory planner section
###############################################################################
[TRAJ]
#+ machine specific settings
AXES = 3
COORDINATES = X Y Z
HOME = 0 0 0
LINEAR_UNITS = inch
ANGULAR_UNITS = degree
CYCLE_TIME = 0.010
DEFAULT_VELOCITY = 0.0167
MAX_VELOCITY = 1.5
DEFAULT_ACCELERATION = .1
MAX_ACCELERATION = 1.5
###############################################################################
# Axes sections
###############################################################################

#+ First axis X
[AXIS_0]

TYPE = LINEAR
HOME = 0.000
MAX_VELOCITY = 1.5
MAX_ACCELERATION = 1.5
STEPGEN_MAXVEL = 1.7
STEPGEN_MAXACCEL = 1.7
BACKLASH = 0.0024
INPUT_SCALE = 2500
OUTPUT_SCALE = 1.000
MIN_LIMIT = -13.2
MAX_LIMIT = 13.2
FERROR = 0.050
MIN_FERROR = 0.010
HOME_OFFSET = 13.2
HOME_SEARCH_VEL = 1.0
HOME_LATCH_VEL = 0.05
HOME_USE_INDEX = NO
HOME_IGNORE_LIMITS = YES

I'm pulling my hair out at this point trying to figure out what is going on.
Any ideas?

Jack Ensor