Printer performing 'jerk' moves randomly mid print

[Roxy]

This program is very helpful on this kind of thing. I think you can "Try it Today" and see what you think. If you like what it does, there are similar programs that do the same thing for free:

http://www.prestosoft.com/edp_examdiffpro.asp?flm=1

[printbus]

This could have nothing to do with configuration settings.

If USB is being used to print, make sure the host computer isn't doing much of anything else during a print, especially if it is an older/slower machine. The command buffer in Marlin isn't very deep, and the printer could simply be running out of things to do while it waits for new gcode data to be received and interpreted.

I'd also scrub through all the mechanics, looking closely for anything that could be momentarily catching or binding. Move everything manually through the entire travel range and make sure things move smoothly. Remove belts etc from motors and repeat the testing if nothing is felt. Perhaps something like a bearing is bad and occasionally catches, torque builds up and then the movement breaks free...

[DBFIU]

This could have nothing to do with configuration settings.

If USB is being used to print, make sure the host computer isn't doing much of anything else during a print, especially if it is an older/slower machine. The command buffer in Marlin isn't very deep, and the printer could simply be running out of things to do while it waits for new gcode data to be received and interpreted.

I'd also scrub through all the mechanics, looking closely for anything that could be momentarily catching or binding. Move everything manually through the entire travel range and make sure things move smoothly. Remove belts etc from motors and repeat the testing if nothing is felt. Perhaps something like a bearing is bad and occasionally catches, torque builds up and then the movement breaks free...

Prints from SD card.

The motion occurs at different height intervals whenever it chooses. If it was a mechanical snag, it would repeat in certain positions and be somewhat predictable.

I checked all the points that I thought caused a snag, everything moves smooth. The snag looks to be motor controlled. The steppers spin and then momentarily and abruptly just stop for milliseconds, and then continue to the very last segment of motion, which is ends up being a few millimeters of travel. It always happens at that last leg of travel.

I will try to look for more mechanical snags, but I couldnt find any on either machine.

[DBFIU]

I uploaded the same firmware to my 3rd machine, and it does the exact same thing. It performs the jerk movement right before the G28 on the way to travelling to its center home position. Then it will do it seldomly after that.

It has to be software. I will try to think more about it. I did not use the comparison program yet to look at the .zip firmware vs my firmware after I touched it.

[Roxy]

How about you take a quick video of the jerk motion and put it up on YouTube for us to look at. It might be something very simple, but we kind of need to see what it is doing.

[DBFIU]

ok, i will do that, thank you Roxy.

[DBFIU]

I just took the video. Let me know what you think.

[Roxy]

I can hear it. I can just barely see it. I'll see what I can find in the G28 code....