Banding/ringing on prints?
I didn't find anything on this on the knowledge base and since I was helped here so thoroughly, expertly and quickly the last time.. its your very own fault that I'm here again 😉
My MK3S+ has started to show some real bad ringing and/or banding on my prints. I didn't change any settings since my last successful prints (about 1-2 weeks ago), so I suspect its some loosening on the hardware that is occuring after a bit of use. I've attached a picture of the "offending" lines. They're regular and appear consistently around the print. Z is absolutely fine, the finish of the print is smooth as expected, so I assume some minor X or Y shift is happening. At one point it actually appears like under- or overextrusion...?
Best Answer by DirkB:
I seem to have figured it out!
It is embarassing and weird at the same time.What I found is that the heat break was not screwed in tight/fully into the heat sink. That means the heater block, nozzle etc. had a wobble in it. Which is really strange considering that it is a preassembled kit straight from Prusa... and considering that I had to do more than one full turn to get it screwed in tightly means that it was that way from the day I received it (which I also remember, but not knowing better... in hindsight that was REAL dumb). It was impossible to tighten it without disassembling the extruder due to the power and thermostat cables preventing that. How it managed to function for months I have NO clue.
I am currently in the process of printing another benchy - hopefully this one completes. The other 2 attempts aborted mid print due to a detach from the build plate and produced some spagetti (I think I put the PINDA sensor too low and it caught on something).
So that is something to watch out for and maybe add to the knowledge base, stupid as is 😊
RE: Banding/ringing on prints?
Another example on a similar part.
RE: Banding/ringing on prints?
And, lastly, the project that printed this file.
RE: Banding/ringing on prints?
Check the grub (Allen) screws on the X and Y stepper motors for proper tightness. One screw needs to be properly aligned with the flat on the motor shaft and tightened first. Then tighten the second screw.
Make sure the drive and idler gears are properly aligned and the belts are tracking centered, not dragging on one side.
Check belt tension on both the X and Y axis.
Your problem is likely one of these issues.
Cheers
RE: Banding/ringing on prints?
Yup, that was it!
I tightened the X-Belt after pausing my print (dont judge :P) and ran a belt test after. Original value was 204, went up to 280. So it was that belt that was eay too lax, current print is looking good again.
RE: Banding/ringing on prints?
Glad you got it sorted.
As a point of information, the numbers produced by the built-in belt tests are only indirectly related to actual belt tension values but 204 does indicate some attention is needed.
There is a printable tool to measure belt tension here:
https://www.printables.com/model/46639-tnsion-meter-for-the-gt2-belts-of-i3-mk3s-or-pruse
Please be aware that the designer states this must be printed in Prusament PETG to insure accuracy. FWIW, I printed one in Overture PETG and it seems accurate, as far as I can determine. YMMV…
When I get some Prusament PETG I’ll print another one and compare the two.
Cheers
RE: Banding/ringing on prints?
Thats a nice tool. While I have some PETG, I dont have Prusament...
IIRC the "ok"-values for belt tension run between 240 and 275. Dunno where I read that, but I somehow seem to have retained that info.
RE:
Well, not exactly. 275 is the target or ‘ideal’ value. Anything between 250 and 300 would be OK. but better stay within 260-290. A figure of 240 or below is already a bit too tight (yes, lower value means higher tension). Likewise, above 310 is too loose.
Keep in mind that the printer doesn’t measures the belt tension itself but the resistance felt by the X and Y motors to move the bed or extruder along the axis. So, if there’s anything hindering the advance ( stuck idler, interfering cables, X bearings squeezed by the X carriage,..) the test gives a low value even if the belt is way too loose.
The measurement tool gives you a better idea of the real belt tension.
RE: Banding/ringing on prints?
There we go. Dangerous half-knowledge! I'll read up on belt tension just to be sure.
It is printing fine right now, but its always good to learn. At least with this machine, its an easy fix. I dread remembering the Anycubic Vyper I had before... that machine was a nightmare.
RE: Banding/ringing on prints?
@artur5 - Thanks for articulating that so clearly and providing an explanation to the reference I made. Well said.
Well, not exactly. 275 is the target or ‘ideal’ value. Anything between 250 and 300 would be OK. but better stay within 260-290. A figure of 240 or below is already a bit too tight (yes, lower value means higher tension). Likewise, above 310 is too loose.
Keep in mind that the printer doesn’t measures the belt tension itself but the resistance felt by the X and Y motors to move the bed or extruder along the axis. So, if there’s anything hindering the advance ( stuck idler, interfering cables, X bearings squeezed by the X carriage,..) the test gives a low value even if the belt is way too loose.
The measurement tool gives you a better idea of the real belt tension.
RE: Banding/ringing on prints?
Looks like I didnt get it sorted after all 🙁
After a bit of a break, my last print came up terrible.
So I tried the belt tension - X 250, y 260 - so within working parameters.
A benchy turned out.. underwhelming.
So I went and checked the steppers, the idlers and the belts as recommended and followed the troubleshooting guide for layer shift. All seems fine to me. So while waiting on Prusa Support to give me a hand, I had a thought - I removed the foam I use for decoupling the printer from the desk and did another benchy. It was better, but still somewhat unsmooth. I also noticed that with the exact same GCode the same issues appear at the same height.
RE: Banding/ringing on prints?
The two benchys - up/left was with decoupling foam, right was without.
RE: Banding/ringing on prints?
Oh and to make things even more confusing (for me at least):
This is the print that I cancelled after an overnight print session.
It started out ok which is why I let it run, then went absolutely apeshit and returned to pretty good layers...
Have you checked the grub screws on your driver gears?
Cheerio,
RE: Banding/ringing on prints?
I didn't, but rectified that.
The drive gear turns great and the grub screw was quite tight.
I'm printing a calibration cube rn to see if that does tell me anything.
RE: Banding/ringing on prints?
When you save drive gear and grub screw, don't you mean gears and screws - plural ? The pulleys on both the x axis and y axis motors. Each pully has 2 small grub/set screws and those loosening are the prime cause of back and forth layer shirts that you have pictured.
Diem isn't talking about the one on the extruder motor, although that one does periodically need checking if you have extrusion issues mainly.
RE: Banding/ringing on prints?
Oh I'm sorry! My English is quite good usually, but I'm not that familiar with the more technical terms. I googled "driver gear" and thought those things look like the extruder gears 🤭
The grub screws on the pulleys are tight, I checked that when I went through the steps of the layer shift trouble shooting guide (but opened and retightened them now just to be sure).
Interestingly, the calibration cubes I printed look mostly fine concerning the walls. Except one thing - there's a line in the cube at the, or very close to, same height than a more prominent line on the benchys.
While the issue seems to have improved with my recent tinkering, there still seem to be minimal layer shifts to be happening - and one larger error at certain heights.
RE:
I seem to have figured it out!
It is embarassing and weird at the same time.What I found is that the heat break was not screwed in tight/fully into the heat sink. That means the heater block, nozzle etc. had a wobble in it. Which is really strange considering that it is a preassembled kit straight from Prusa... and considering that I had to do more than one full turn to get it screwed in tightly means that it was that way from the day I received it (which I also remember, but not knowing better... in hindsight that was REAL dumb). It was impossible to tighten it without disassembling the extruder due to the power and thermostat cables preventing that. How it managed to function for months I have NO clue.
I am currently in the process of printing another benchy - hopefully this one completes. The other 2 attempts aborted mid print due to a detach from the build plate and produced some spagetti (I think I put the PINDA sensor too low and it caught on something).
So that is something to watch out for and maybe add to the knowledge base, stupid as is 😊
RE: Banding/ringing on prints?
Make sure your build plate is clean or prints detaching will continue…