[GeoffDuniam]

Afternoon, all,

I've got an issue that I suspect may be noise related. My setup is

• Mach 3
• Gecko G540
• 48v 7.5a power supply
• Huangyang VFD / 1.5 kw water cooled spindle
• 1800 x 1200 table, loosely based on Joes Hybrid CNC plans (aluminium table and gantry, MDF Z Axis support structure)
• 4 axis drive (2 motors driving Y axis, axis A slaved to axis Y)
• All the cabling is shielded (Spindle cable, motor cables, EStop and limit switch cables)
• Estop and limit shielding is earthed at the controller end - earth on the power supply
• Spindle is earthed as per the the "Water cooled chinese sppindle disassembly" discussion on this site
• Spindle shielding is earthed to the VFD Earth
• VFD, G540 power supply and PC all running off a common heavy duty power board (which should mean a common earth)
• Motor cables are all fitted with the Homann Design GC-02 DB-9 Stepper Breakout Board and cable shielding attached as per the instructions
• VFD is controlled from Mach 3 via the famous RS-485 based plug in developed by royaumedeole and is working fine (cable between the RS-485 and VFD is 2 core shielded, but the shielding is not earthed at this point)

The problem is that the motors are hesitating when the spindle is powered up and running, most noticeably during fast travel - although the problem also manifests during relatively slow cutting operations as well. With the spindle start G-Code bypassed and the spindle not powered up and running, the travel appears to be smooth and hesitation free. There's a short video of the behavior (sorry for the poor quality!)


here

Just wondering is anyone has experienced similar issues, and if so, how did you resolve them?

Thanks in advance for any assistance.

Geoff

[kennected]

I see this is an older post, but I had the same issue. With my limited experience and understanding I looked through several posts and decided to change my "debounce" setting (in settings tab of Mach3) to 1000. It cured the hesitations that were very similar to yours, immediately. I think your theory of noise issues might be accurate and the remedy might be to simply tell Mach3 to ignore small values of the noise (which is my interpretation of increasing the debounce setting) or take the approach of curing the noise that is occurring by increased shielding to keep it from being an issue in the first place.