So for the life of me I cannot figure out why this is happening.

So for the life of me I cannot figure out why this is happening. I can print small parts just fine say < 2hr prints, but i am having an issue with anything longer than this or with bigger prints and by bigger prints i mean x-idlers and what not. It seems that at the last few layers or 20 mins given the x-idler the y-axis starts shifting really bad about every 5-10 layers rendering a 4hr print useless “at my current settings”

Now before you go asking why an idler is taking 4 hrs to print, its because my y-carriage is to heavy for my 48oz/in stepper to handle at full speed / acceleration. I have the jerk settings way down, the speed and acceleration are at 1000 and 1500 respectively. Everything rides smoothly the belts are not to tight as far as I can tell. Not to mention its hard to diag since it only happens on “Large Prints”

I have fans on all of my steppers and they are cool to the touch when it skips like this, there are no teeth missing in the belts or pulleys, and I have filed a flat in the y-axis stepper. I still need to pull that off and see if it has skipped somehow.

any ideas or suggestions, what settings or fix’s have you employed when facing this issue.

The LM8UU side of this print is facing the back of my printer.

Sall good everyone starts somewhere and all suggestions are good as I only have 2 eyes, so bouncing stuff off someone else usually helps me in the long run.

Cant be the spool, I designed the spool holder myself and it rides very smooth and the feed is positioned right down into the extruder as to get rid of the pulling tensions that other spool holders cause.

It could be the belts, but my y-axis stepper has been acting a little funny lately “least i think its the stepper” the y-carriage makes this horrid clunking noise when it does in-fill “fast infill not long infill” the smooth rods are straight without me taking a laser or gauge to them.

could also be that my LM8UU bearings are just toast and need to be changed… I think this printer is due for a complete tear down and rebuild “frame included” I just don’t have time for that right now… Thus I want to switch to a Delta style printer.

I’d say it’s either the motor or the belt skipping. Check your belt tension and the stepper driver’s temperature. If they scorch your skin, they’re probably a tad too hot and will go into thermal shutdown for a moment to cool down and then resume like nothing happened.

see the stepper driver slipped my mind, I just cant seem to find that sweet spot for the stepper driver on the y-axis…

I have beefy motors, so the way i do it for each axis is this:

  • Turn up the drivers until it goes into thermal shutdown (you’ll hear the motor’s noise pause for half a second, in which you can easily move the axis by hand)
  • Turn down the current about 10-20% to make sure there’s still a safety margin
  • Increase maximum speed, acceleration and jerk (in that order), each until the motor is skipping steps - test these by jogging the axis at full speed
  • Decrease each of those by about 20% for a safety margin


Thanks for the tip If i have time I will try that tomorrow and let you know how it goes. I plan on upgrading to I think there 60oz/in or 80oz/in steppers here in the fall. They should be able to handle my heavy ass y-platform.

I use two 0.48Nm motors, two drivers and two belts for my double-wide platform, consisting of a 6mm FR4 base, 3mm window glass and of course two heated beds. Works like a charm.

Mm so I see no reason that re tuning my stepper voltage and acell, speed and jerk shouldnt fix the problem then. What jerk value do you have for your setup. These slow print times are killing me. I’m running a I2 with a birch frog plate, aluminium platform, heated bed and 3mm mirror glass with LM8UU bearings.

The Prusa i2 is not a stiff printer, so high acceleration and jerk values will seriously decrease your quality (y is still a fair bit better than x in that regard).
But even on my Mendel90 with oversized rods and belts, i decreased my jerk value and print speeds just to get that extra bit of quality and strength to my parts - i’ll let you know what values i used once i get home.

Sounds good, Ill look for them if I’m not already in bed. What do you think of upgrading the frame to aluminium extrusions? kind of like the OB1.4? I’m really waiting for the plans of the 3DR Delta to be released.

The OB1.4 looks like a fine design at first glance - it seems to avoid the issue the MM has inherited from the Mendel where the extruder would crash into the triangular frame at higher z-heights. The MM gains a lot of its strength from direct screw connections between the t-slot profiles and i’m not sure the OB1,4 uses them at all. I’d try to get some feedback from folks that have built one before pulling the trigger.

sounds good, have you seen the 3DR delta at all? I don’t know why I am so intrigued by a delta system but I want one.

I’ve not seen the #3DR in person, but i do trust @Richard_Horne to come up with nothing but a great design. The one delta i have seen is @Florian_Horsch_flouS 's #DeltaTower (at MakeMunich) and i was pretty impressed - both by the sheer size of that thing but also by the quality it achieves.
So while i do find the concept interesting, i have very little experience with what they’re good at and what you need to look out for. I’m sure the delta veterans will be able to help you out with that.

Sweet, Thanks for all your help, and letting me pick your brain. If your on ill catch you tomorrow I need sleep.

My two cents: if you print the same part twice and fails exactly the same spots, it’s software/firmware related. If it fails about the same height but a little different, maybe your x is binding at certain heights. Or you have some cables/wires that are getting caught somewhere. Could also be power supply.

I concur with @Thomas_Sanladerer on the stepper pot adjustment. I just went threw this, chunky motor running hot,layers skipping. I just tore down my Yaxis and put it back together. Then tweaked the pot, and now it’s MUCH better.

Yep, I think that’s the going plan for this week/weekend. I really need to get some longer all thread and extend the frame of the printer a little, as the frog plate just barely squeezes past the frame on both ends.

@Thomas_Sanladerer so my friend and I came to the conclusion that the frog plate is completely not needed for my setup “which gives me back my 25mm z height :D” no springs needed as the aluminum thick plate is flat, just a simple y stage. I have to attach the belt and what not them I will give it a spin. My system should like this setup a little better add its not as heavy.

I’ll post pics of it when I’m done. And the y stage rides a lot smoother now.

@Jeremy_G_WeisTek_Eng i never understood why the Prusa was designed with such a complicated y-stage - all i used on mine was one simple plate with the bearings, belt and heated bed mounted to it. Bed springs are usually used for bed tramming as well as for correcting a warped PCB - how are you planning on accomplishing that?

My speed/acceleration/jerk values are 300/3000/15 for both X and Y as they are both pulling about the same weight per axis - for something like a Prusa, i’d suggest values in the magnitude of 200/1500/5 if your drive system can handle it.

@Thomas_Sanladerer Those settings are working pretty good so far, I think I can push it a lot faster now that the y stage is lighter.

I ended up going back to the solid aluminium plate bearings, and spring loaded heat bed, No frog plate and no 3mm mirrored glass. Added a belt tensioner to my y axis belt and moved my x axis belt tensioner closer to the x carriage as it was causing the x axis to skip.

I’m in the process of moving so allot of my stuff is packed or un-accessible right now so doing a proper modification of the y stage is kind of out of the question right now, although i did try. The aluminium plate heated up to much, and the thermister was way to unstable.

Oh and I straightened out my z axis smooth rods as well, as it had a nice slant to the build platform…

When I get back from Australia and when I get time I think its time to bust out the tools and start work on my own 3d printer derivative. I’m thinking H-Belt design only slightly modified.