Weird thing happened to me with LinuxCNC and a 6040z: A few pauses happened on X-axis while it was routing. For a couple of seconds the axis will stop moving half-way of a long G1 move or when I was manually jogging the axis. In both cases the axis location number kept updating on the screen while it was not really unexpectedly paused. Of course the part being milled did not end very well. As I am just setting up LinuxCNC may be I have forgot something. Or maybe I have a loose connection on the parallel port or … (you name it).
was the problem seen in the video? I couldn’t tell.
are you sure you aren’t loosing steps?
Your feed looks really slow, possible rubbing causing built up edge on bit, will lead to higher force and missed steps if motors are barely adequate.
It sounds like a stall to me. Might be bad step signal, or just trying to run too fast. Could be an overrun too due to system latency. Though I think LinuxCNC would detect that if that was going on.
@Paul_Frederick Stepper will stop for around two seconds to resume motion later for reasons unknown. When I have seen missed steps on other systems it was more like a bump and never a full stop for a while. I think it might be a bad contact but asked just in case some misconfiguration may cause a similar effect. @Alan_M Sorry, video is misleading as it does not show the effect, it only shows the beginning of the work where, later, I found the error.
@Bruce_Hoover and @Cayenne I was being super cautious about feedrates and depth of cut (or so I think: 200mm/min and 0.2mm DOC for a 3mm single-flute HSS bit cutting 6061-T6 Al). It did not feel like the machine was suffering with the job. Plus it was not a stutter but a two-second pause midway a long movement (the side a of 150mm square).
run a air cut, is the pause repeatable?
if so try g64 p0.02
I add this to the header of all my gcodes.
It will maintain set feedrate while sacrificing accuracy up to 0.02mm
200 mm/min @ what rpm?
if you are at 24k rpms, you need around 560 mm/min to maintain 0.022mm chipload. If you are using a chinese end mill, it’s probably not that sharp, and would need higher chipload to avoid rubbbing.
I would try 24k rpms, 1200 mm/min, and adjust DOC to match machine rigidity. I would think you could do that with atleast a 0.5mm DOC
What is length of your bit sticks out of the collet? use the shortest bit possible. even with carbide your 3mm bit is like a noodle, It should stick out of the collet no more than 13mm If it is sticking out 25mm, then 0.2mm DOC sounds about right. you need to keep the feedrate up though.
HSS huh? I’d go around 18k rpms, .12mm DOC and 500 mm /min feed. with a 25mm stickout of collet. If you have 13mm stickout you can run 1mm DOC with the same 500mm/min and 18k rpms.
@Alan_M thanks a lot. It did not repeat with a air cut with the spindle off but it did when doing manual move with the spindle on. Not easy to repeat though. I am puzzled.
Do you have any toolchange routines or any custom mdi buttons? I have had problems with laggy response to manual commands immediately after a probe routine through a mdi button, but never during a program running.
how is your machine grounding? Make sure there are no ground loops, and vfd wires are not running alongside motor wires or parallel cable.
you may want to do many tests with the spindle off, if you never have the problem, start the spindle and retest. it may be interference.
That might be it as wires go through the same chain plus metal frame seems not to be properly grounded. I will check that tomorrow. Thanks a lot for the tip.
I have forgotten to check the grounding but after securing the parallel port connectors on both ends I was not able to reproduce the problem with the spindle on. Thanks to all for your answers.
Hi, i think you’re steppers might be stalling. You should try lowering your feed rate some on on the axis that stalked and see if it helps. Also, there are many known steeper driver boards that produce this failure, do some research on your driver, it might be faulty.
Hi, just realized there was a video attacked so i guess high feed rate might not be the issue. I would check the axis and make sure they are not getting jammed by aluminum chips.