Something really bothers me with my printer: I get really inconsistent "slips".

Something really bothers me with my printer: I get really inconsistent “slips”. I’ve tried all the tricks I could think of: Current, acceleration, speed, smooth filament feed… I have no clue to why it jumps like this. What bothers me is that sometimes it works flawlessly!!! (And I do mean the exact same gcode from the SD card and talking about long prints). It does not happen on small prints. The drivers are properly cooled. Raps 1.4 setup. Oh, yeah, sometimes the slip is 1-2 steps, sometimes it’s waaay more: 400-500. Any ideas what so ever? Thanks guys!

Chances are that you are loosing steps on one axis. Try to lower the acceleration and/or jerk.

Oh, yes, only on the x-axis. I’ve gotten down to a=300 and j=12… Why it sometimes happens and other times not? With the same file. Not even removing the SD card…

Sounds like à mecanical problem. Remove the belt and move the axis by hand. Try stoppning and starting to find if it binds somewere. Linearbearings can do this when you run on unhardened rods.

Did that too, because I thought, initially, that it’s due to some rod defects. They are hardened. Ended up buying high end linear bearings (I thought the Chinese ones might be at fault… 12€ a pop :’( )

Is the axis belt tensioned and secured? What type of gears are driving it, are they the right ones to match your belt? Can the belt catch on anything during operation?

I guess it`s somewhat hardware related. Perhaps your belt sometimes rubs somewhere which increase friction and let your stepper drivers get hot over time. Or your motor becomes hot.

Yes and yes. It’s a t5 belt and the proper gear. The belt does not catch on anything. I designed the entire axis. It’s on thingiverse. https://www.thingiverse.com/thing:409932 
I’m losing my head around this…

I might see the problem. To avoid locking of the axis you need a certain distrance between the bearings on a rod. It seems you are using a triangular setup with only one bearing on aone of the rods. This could be the culprit, since it severely limits the amount of force you can use on the end of the lever. I only found this in german, but the images speak for themself http://www.igus.de/_wpck/pdf/global/2zu1Regel.pdf

Thanks! I’ll look at the PDF. I thought that, by placing the belt symmetrically in between the rods, that would be avoided…
I’ll read the file and get back with an update. Again, thanks!

I had a very similar issue.

I spent a lot of time rechecking all the mechanics and never resolved it.

It turned out to be an issue with my electronics board. The one stepper control channel (it wasn’t the stepper driver, I swapped that out) on the RUMBA caused random skips like this. If you have an extra channel, try swapping it around in the firmware.

I have exactly the same issue. I think maybe too much current to the stepper (always x axe) and too much heat therefore.

I had the same. Problem and it turned out to be a single loose crimp on the X motor connector. It took a few failed prints to work out what the the problem was.

@Bjorn_Marl ​​ I read the PDF, it’s not the same setup, I had that and I changed it to the current, because of the torque.
@lynnroth ​​ I’ll do that! I have the second extruder slot empty, hope it’ll work!
@Raul_Barral ​​ the motor is still child enough after increasing the current (I had a lot of tests on that side of the matter)
@Mike_Dalton ​​ i hope that swapping the driver position will do the trick; the wiring is good

A simple thing to check would be to see if the nozzle is getting caught on any part of the print while moving (infill or more likely perimeters.) Could be preventing the hot end carriage from moving the full distance and thus missing steps.

@Mark_Rehorst ​ I had a micro SD card with its SD adaptor… Changed it to a regular SD, from Samsung. Can you tell me what is the “default” down you usually print at, the acceleration and jerk? I’m at 40, 300, 12.
@Paul_Merry ​ yeah, that was my first thought, but it would do it for a (small) print when it was centered on the need, and not when it was close to either side… Not to mention that the same big print would work sometimes, other times it wouldn’t.
https://docs.google.com/file/d/0B2MWsxvLpodAelkxLV9DUVFfakk/edit?usp=docslist_api
https://docs.google.com/file/d/0B2MWsxvLpodARHFxVmd6NDFES2c/edit?usp=docslist_api

I also had a similar issue after a while when printing ABS. I had LEDs running on a separate PSU for a while, then I decided to wire them into my printer’s supply and noticed that when it skipped the LEDs would blink out while it was skipping but since I was printing from my computer the board would not loose power and the print would resume after the supply powered back up. So I replaced the supply and haven’t had an issue.
I my case it was the supply.
Also running RAMPs and a 2560.

I was also doing 50 mm/s, 600 mm/s² but the jerk was at about 20… So I sleekly took it down to 12. You think that’s still high?

@Danut_Lipsa ​ 50 is the print speed, 600 is the acceleration.

check belt tension might be jumping a tooth