Square circles

[Sorbue 2]

Hello everybody,

I'am making my first prints with my new UM original + and i am quite happy with the printing quality. However the circles i want to print are a bit square, as it can be seen on the next picture :

So i had a look at several tips and i :

 

• Checked the belts tension
  
• Checked the XY axis alignment
  

 

I also tried to print a 100m x 100mm square. Measuring the printed part gives 98.5mm x 99.3mm. Is it normal ? Can it be the origin of my square circle ?

Thanks for your answers !

[Sorbue 2]

I found this and i'll try it as soon as possible :

http://blog.mcmelectronics.com/post/3D-printer-calibration#.VT9ziWOgalg

[IRobertI 514]

Did you check the short belts as well? Those are usually the culprits.

http://support.3dverkstan.se/article/23-a-visual-ultimaker-troubleshooting-guide#circles

(the pictures above show an UM2 but it's the same thing on the UMO)

[Sorbue 2]

Yes i checked the short belts, the motors are the lowest possible and i imagine the belts tension is enough ...

[SandervG 1,518]

I would also suggest the short belts are the issue.

Are the pulleys tightened sufficiently?

Do you see any other artifacts in your print?

[peggyb 169]

are the thin rods exactly square? compared to the thicker rods?

[Sorbue 2]

I will wheck again the pulleys.

Can a too high tension in the belts be a problem ?

What do you mean @peggyb ?

[Sorbue 2]

I will wheck again the pulleys.

Can a too high tension in the belts be a problem ?

I would say that, from time to time, lines are not touching properly, i can see small gaps sometimes.

What do you mean @peggyb ?

[gr5 1,999]

It's hard to tell from your photo but I think that might be normal. It's not a problem exactly with the printer but the PLA. It shrinks very quickly (milliseconds) as it comes out and while it is still liquid but shrinking it sticks to itself like a liquid rubber band. It's difficult to make vertical holes in this situation and they tend to be about .5mm smaller than where the head is positioned. However the holes although they tend to be smaller are usually mostly round but I don't always zoom in on them quite this much so I'm not sure if this is normal or not.

Yes, too tight belts can cause a problem - it can greatly increase the friction causing play/backlash. It's weird but having the belts too loose or too tight will cause backlash but "too tight" is much more rare and is easily detected by pushing the head around with power to servos off.

[Sorbue 2]

I am pretty sure that my circles are not good. I have seen many pictures of them (UM + PLA) and they look much better than mine.

[gr5 1,999]

I'd like to see more pictures. Is this the bottom layer or top? I'd like to see outer edges also on top layers. Some of the issues causing square circles only affect the bottom layer (leveling off slightly). Some affect only upper levels. Loose belts and backlash affect how well the solid infill touches the edges. Typically with loose belts one is worse than the other and you get a pattern of alternating diagonal infill that is very distinct and clearly a belt issue - if both axes are EXACTLY equally loose then it's possible to get a nice pattern like in your picture but then the other problem is the infill doesn't touch the outer walls. Also with loose belts if you print certain rectangular/blocky objects and Cura switches directions (clockwise versus counter) you get very distinct patterns that instantly (to my eye) say "loose belt".

Instead if this is a bottom layer picture it could be simply "move the nozzle closer to the bed by 1/4 turn on all 3 leveling screws".

[Sorbue 2]

Thank you for your help.

Here are two more pictures. The first one is the down side. The square shape is the same for each Z level.

[swordriff 119]

Table which the printer sits on vibrates, or printer has "DHL Elephant Sits on UM2-Syndrome", so is not stable on surface, if its not the short belts.

[SandervG 1,518]

Alternatively, you can also see some density change in the infill at the top layer.

What could also be is that 1 or 2 pulleys are not centrally drilled, like the hole is not completely in the middle.

It is quite a hassle to check, but if we are totally sure it is not the short belts, it is something to look into.

[gr5 1,999]

Ah!!! Yes! I see it now! Backlash. See how you have pairs of infill touching and then a gap? touching/gap/touching/gap - the pattern repeats. The is EXACTLY what I was looking for. I didn't notice this in the first picture.

This means you have backlash/play. It could be due to friction (consider loosening the end caps slightly on the 4 long rods or belts too tight) but more likely it's the belts and most likely the short belts (those going to one or both stepper motors).

[peggyb 169]

the calibration of the rods is on page 55 of the manual, using the calibration sticks to get everything aligned and square.

[Sorbue 2]

@peggyb Yes the calibration of te rods seems good to me, thanks to the sticks.

@gr5 I made a new test with more loose short belts : the circles are more and more square. So it means that your idea of a lack of tension in the short belts could be good. But the motors were placed at the lowest possible position : I can't give more tension to the short belts. So i'll try to minimize the friction with the end caps.

I've also noticed that moving the extruder head manually is not totally fluid ...

Different solutions to be explored

[SandervG 1,518]

We'll get there! Plenty of expertise in the forums

[IRobertI 514]

I know on my old UMO (three years old) I had to file one of the slots for the motors as it hadn't been cut through fully. That's probably not the issue here but you could still file the slot a bit so that you can push the motor down further if need be.

[Sorbue 2]

OK that's it ! To much friction because of the end caps !

Now my circle are almost perfect (and no more gaps)!

Thank you everybody