knocking sound while printing

[ohms 1]

Hi, I just got my ultimaker 3 extended and have done a few test prints that came out great, but there's a slight knocking sound that seems to happen at intervals coming from inside the back wall that I can't quite pinpoint. Is this a normal sound or could something be loose back there that I should worry about?

Thanks

[neotko 1,417]

Hi and welcome!

Maybe is the second spool holder that is shaking? Without knowing from where it comes I don't know...

Can you reproduce the sound while is off or only sounds while on?

[IRobertI 521]

It's super hard to diagnose without actually hearing it as people have very different ideas about how things sound. Are you able to shoot a short video to capture the sound? That would make it much much easier to figure out what it is and if you should be worried.

[ohms 1]

here is a video, I think you can hear it. I've gone around the printer many times, but I can't see where it's coming from and it is not the spool holder. there are 2 cores installed, but only using core 1.

 

[neotko 1,417]

IF is that small tickticktickticktocktocktocktick tick tock tock is the autolevel system compensating the z of the first layers.

To completely remove that you need to do a manual calibration only (doing a manual calibration & setting auto calibration yo Never). That's what I use since I don't like automagicall stuff that i can't control.

Edited November 13, 2016 by Guest

[ohms 1]

ahhh well that's good to know. Didn't know my bed was that uneven haha. I'll play around with that and see if it does go away

thanks.

[IRobertI 521]

It's not necessarily very uneven it's just that the printer is making lots of tiny adjustments fairly quickly as the head travels back and forth over the bed. But of course, the closer to "perfect" you can get it, the better

[Zwakie 21]

I was experiencing the same type of sound the last couple of print jobs and also could not pinpoint the source of that ticking sound. Good to know the auto-levelling mechanism could be the source, I will check once the 31 hr job that I just kicked of has finished.

I did notice that the pully belt on the front X-axis is a bit loose (not nearly as well tensioned as the rear one), and suspected this to be the cause. Needless to say that I have to correct that, anyone can tell me the proper procedure for that?

[neotko 1,417]

I did notice that the pully belt on the front X-axis is a bit loose (not nearly as well tensioned as the rear one), and suspected this to be the cause. Needless to say that I have to correct that, anyone can tell me the proper procedure for that?

 

If you mean the short belt:

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

Basically like the guide shows, unscrew a bit the x4 motor screws to leave it a bit loose, then push down and while keeping the pressure screw the x4 ones.

Edited November 13, 2016 by Guest

[Zwakie 21]

If you mean the short belt:

 

Thanks for replying @neotko, but that is not the belt I was referring to. I don't have a picture available, but it's the one running parallel to and just aft of the front panel. Any ideas how to increase the tension for that one?

Edited November 13, 2016 by Guest

[Labern 775]

In the white block the attached the shaft to the belt, there is a spring tensioner. If you loosen one of the pulleys it should allow it to pull the belt tight again. Then retighten the pulley.

[avluis 2]

 

If you mean the short belt:

 

Thanks for replying @neotko, but that is not the belt I was referring to. I don't have a picture available, but it's the one running parallel to and just aft of the front panel. Any ideas how to increase the tension for that one?

 

I think I know which one you are referring to since my printer had the same issue as you described -- all I did was to hold on to one of the side rods (to keep it from moving) while pulling on the belt a bit -- basically, you are getting it to skip a track in order to equal out the tension -- check that you don't need to do the other side as well~

[Zwakie 21]

I think I know which one you are referring to since my printer had the same issue as you described -- all I did was to hold on to one of the side rods (to keep it from moving) while pulling on the belt a bit -- basically, you are getting it to skip a track in order to equal out the tension -- check that you don't need to do the other side as well~

 

Thanks @Labern and @avluis for your replies.

Question for @avluis: basically what you are saying is that the belt is one tooth off, causing the bottom part of the belt (which is 'drooping' a little) currently having 2 teeth more than the top part of the belt?

Edited November 14, 2016 by Guest

[neotko 1,417]

Is better to just release the pulley screws and let them get tensioned by the spring inside the slideblock. Forcing the belt position will only cause problems.

After adjusting the screws you might want to check x/y alignment just in case something moved from the readjustment.

[Zwakie 21]

Is better to just release the pulley screws and let them get tensioned by the spring inside the slideblock. Forcing the belt position will only cause problems.

I agree to what you are saying @neotko, it seems to me this is the best option in case the number of teeth top and bottom part of the belt are the same.

I don't think we have another option than what @avluis is suggesting if the number of teeth are not the same, or is there?

(I guess I'll be counting teeth shortly and yes, x/y calibration is part of the plan on matter how this is fixed)

[Labern 775]

Loosening the pulley should make the teeth the same if one side is loose and the other tight as when it rotates it creates more teeth on one side and reduces them on the other.

It would be wise to check the axis are square after this so you don't get parallelograms

[avluis 2]

Loosening the pulley should make the teeth the same if one side is loose and the other tight as when it rotates it creates more teeth on one side and reduces them on the other.

It would be wise to check the axis are square after this so you don't get parallelograms

 

@Zwakie I would do this first just to make sure nothing gets broken~

[Zwakie 21]

Is better to just release the pulley screws and let them get tensioned by the spring inside the slideblock.

 

With getting the NFC issue out of the way I felt I was on the roll and have a look at the pully as well to quickly find that I thought of this task a little lightly. Maybe I am completely overlooking something, but where exactly are those pulley screws located @neotko ? From your wording i understand it's not in the white block, right?

[neotko 1,417]

The metal pulleys that rotate with the belts, they have 1 small screw each. After realigning them remember to tight the screw wuite hard to avoid any future slipping. Ofc remember to align x/y while doing all.

This is a cideo for UMO+ but the screw/align is the same basically

 

[Zwakie 21]

The metal pulleys that rotate with the belts, they have 1 small screw each. After realigning them remember to tight the screw wuite hard to avoid any future slipping. Ofc remember to align x/y while doing all.

 

Thanks @neotko for the quick response, unfortunately this confuses me a bit. Your answer seems to be the way to align the rods for X/Y. Where does the "removing droop in the lower part of the belt" fit in? Can you please elaborate a bit more on how to just do that then?

[neotko 1,417]

As soon as the screws of the pulleys are free, the tension of the belt will tighten. Inside the slideblock of um2/um3 there's a spring that holds the belt, so when the pulleys screws are loose the spring will make an even tension on both sides.

Edited November 17, 2016 by Guest

[Zwakie 21]

As soon as the screws of the pulleys are free, the tension of the belt will tighten. Inside the slideblock of um2/um3 there's a spring that holds the belt, so when the pulleys screws are loose the spring will make an even tension on both sides.

 

I see, I'll give it a try tomorrow and let you how that works out. Thanks again for educating this 'dummy'

[neotko 1,417]

A pleasure men, I learned all by reading pdfs and breaking stuff

Check page 12 of the um2 github assembly manual, the slideblocks of um3 are exactly the same but white.

 

https://github.com/Ultimaker/Ultimaker2/blob/master/um2%20assembly%20manual%20V1.1%20_english.pdf

 

Edited November 17, 2016 by Guest

[Zwakie 21]

Just loosening the screwon on the pully did the trick. I heard one shallow click (what I think to be one tooth flipping from the bottom to the top side of the pully) and tension top and bottom are now the same. Next performed the XY-calibration, resulted in 1 notch up for the X. This removed the last little sub-optimal thing on my UM, although after a firmware upgrade I noticed another -again non-critical- thing (but I will that write down in another forum topic).

Once again, thank you @neotko for your answers and directions!

[Zwakie 21]

IF is that small tickticktickticktocktocktocktick tick tock tock is the autolevel system compensating the z of the first layers.

To completely remove that you need to do a manual calibration only (doing a manual calibration & setting auto calibration yo Never). That's what I use since I don't like automagicall stuff that i can't control.

 

I think there is more to it than just the autolevel system correcting for sub-optimal manual leveling. I have two reasons for this assumption:

1. It keeps doing it, even after paying real attention to manual leveling and making sure that the layer-thickness of the very first layer printed is very consistent height-wise

2. The ticking sound does not stop after the 10 layers that the autobed-level system is supposed to do it's task; in my case it keeps ticking throughout the entire print, regardless of how many layers are being printed (done some prints > 700 layers and enjoyed that sound until the prints finished).

So I guess we have to start a quest to find out WHAT ELSE might be causing this ticking sound. As the original poster of this thread I cannot locate the source of the sound and have no clue how to pinpoint what is causing it. If anyone has any idea, please share it with us so we can start printing quietly

Edited November 23, 2016 by Guest