[MacMarty]

I'm rebuilding my 3-axis DIY machine because of what seemed like mechanical binding issues. Now I suspect that my primary problem is NOOBishness.

PROBLEM: I can cause any moving axis to go into a resonance mode by grabbing the drive coupling with even a moderate amount of force. Right now the axis assemblies are running with NO LOAD. removed from table, just running the drive nuts back and forth. It's not mid-band resonance, as it happens while moving continuously at the highest speed I can (seem to) command - 2.1 inches per second.

I suspect I've made some fundamental oversight out of simple ignorance.

Steppers: KL23H2100-30-4B (382 oz 4-wire bipolar from Kehling/Automation Technologies)
Driver: Geckodrive G540
homebrew Power supply:
- 30vDC (measured while axis running)
- caps 2200uF 200v (qty 8 connected in parallel)
Current limiting resistors 62Kohms between DB9 pins 1 and 5

Axis shafts 2-start ACME with Lovejoy couplings and CNCrouterParts bearings (Per CNCrouterParts design)

"I am only an egg....."

[ger21]

Where did you come up with 62K ohms??? They should be 3K ohms. You're supplying too much current to your motors. Also, those motors have a rather high inductance, and would run about 50% faster on 48V.

It's pretty easy to stop a stepper motor by grabbing the coupling. And resonance is far more likely with the motors not mounted to the machine. In most cases, motors will run better once they are mounted.

How are you commanding 2.1ips?

[MacMarty]

Well, Gerry, it's like I said, some swarf-for-brains noob set up this thing, including the power supply and current limiting resistors. (I am that noob.) It seems likely that a 62K current limiting resistor is like having no resistor at all. The motors are NOT fried, (only slightly warm,) so I suspect that my existing power supply is under-built in regards to Volts -and- Amps -and- amount of capacitance. (I failed to mention that I do hear some hum when the system is on but stopped.)

The 2.1 ips figure is coming from LinuxCNC v2.4.3, currently running on a Dell Optiplex GX520, with no efforts to tune the machine for minimizing base jitter. Setup indicates that the current max step rate is 25,641 Hz. Motors are 200 steps per rev, microstep factor of 10.0, spinning a 5TPI 2-start ACME-threaded shaft. When I started down this path I was more concerned with getting a high amount of force, rather than potentially higher speeds.

Mechanical stall at end-of-travel I can understand. I don't understand why a seemingly small amount of physical resistance (MUCH smaller than hitting end-of-travel) can force an axis into a resonance state. Doesn't happen on other systems I have seen. (Tormach PCNC770, NextWave Automation [Rockler] Shark HD 2.10 .

[MacMarty]

OK, based on Gerry's comments and another read-through of the Geckodrive "Stepper Motor Basics Guide", I'm going to replace the existing PSU with a 48vdc 7.5A unit from Automation Technologies. And yes, I'm going to get a 3K resistor onto each motor's DB9 connector. While that's on its way, I'm also going to look into finding a computer with better base jitter characteristics. Thanks for the guidance, Gerry!