Loud knocking during printing. Suspected Z axis but dunno anymore

Bambu Lab H2 Series
1

Hello,

I am having a super wierd loud knocking during printing When I started noticing it it seems like it was happening during Z-Hops and coming from the rear screw location:

This sounded very metallic and somewhat originating in the rear. I started looking for the source. I tilted the printer so I can take a look and listen form the bottom of the printer, and it sounded the worst from there :

This does not happen when I move the bed up/down from the screen. Only during rapid bed direction changes, It is somewhat similar to the knock during bed levelling when the nozzle tap the bed. Actually it souns almost identical to that, not sure if this is a clue or not.

Also nothing when pulling the Z belt by hand.

I started suspecting screw/nuts/motor, as those are the things that are moving when this is happening.

I started taking things apart. Took the Z belt off to decouple everything so I can have a listen to each part individually.

Bed moves silently and smoothly when pulled by had with no z belt, so that kind of ruled out my theory about screws/nuts.

I tightened averything from the bootom around rear screw. No differnece.

I took out the Z-tensioner to inspect for anything out of order. Seems fine.

The Z motor turns silently when operated from the screen with no Z belt.

After mutliple z belt removals, manipulations, attepts at finding any and all slack anywhere, the knocking changed from sounding very metallic, to a bit more dampened, but is still there and I cannot find the source.

It resonates thorugh the chasis so much it sometimes sounds like it comes from the XY axis, sometimes like it comes from rear of the bed and sometimes like it comes from the bottom rear screw/motor/tensioner area.

Some Z hops dont do it also… its not intermittent, very repeateaple during a print, with certain z hops of that print do the knock and others dont. Every layer creates the exact same pattern.

I am loosing my mind. Bambu Support is beyond useless, incompetent and disinterested in helping me solve this. Started with “this is normal”, moved on to “we will ask Advanced Support Team for assitance” and never heard back from them.

Seems I am on my own.

Please help.

2

Can it be your rear nut starting to fail? Can you push up and release from below your bed close to the rear rod, do you have significant backlash from the nut?

3

Kind Sir, I was hoping one of you will pop in to help out (I follow the Stripped Nut thread closely).

Because of that thread, this was my very first suspision. This printer has been acting up Z axis wise from new. Whenever bed was moving, it sounded very wine-y, reasonate-y and rough. Immediatelly I thought - AHA, nuts are going. Especially since the grease was turning black withing hours of cleaning and regreasing …

Some more background info: Printer has 200hrs on the clock. I should add this is my 4th H2C, last 3 had to go back due to other faults that came up within 30 days. I was not willing to mess about with a brand new failing printer, especially since I both lack the competnecy and also because I had the option to return for a replacement, saving me a lot of time and hassle.

For this issue, I asked babmu for new screw nut set. But before it arrived, the knocking started.

I was convinced it was the screws/nuts but started doubdting the more I listened to it. I asked them for a new Z motor, which I also have on hand.

Since they are not responding to me, I bought a new Z belt and Z tensioner, which I also have.

I have not swapped anything yet. Just investigating and nothing seems to make sense. The biggest mystery to me is that with no Z belt, nothing seems to be causing any issue. I did focus on screws/nuts the most and cannot find a fault with them, without actually removing any that is.

As for the backlash - no, seems small and about the same as the two fronts.

Please help. Please tell me what would you like to see me do and video it for you to assist you with helping me finding the cause.

4

I am loosing my mind here.

When printing the same exact object with PA the knocking is worse than swapping the filament for PLA. The â– â– â– â–  ?

I just disconnected all 3 motors form the board in order to manipulate them vigorously. Neither Z, X nor Y are making any abnormal noise on rapid direction change, let alone knocking.

5

It very much sounds like a shaft connection that’s a bit loose - it spins freely very briefly, than makes a loud TICK when it hits the stop. Or possibly even a stepper that isn’t mounted rigidly and is torquing in it’s mount.

If it was me, I’d try to isolate it to specific movements - if you did the same print twice, does it happen at the same spots both times?

If you’re capable, making a custom G-code file to test various movements until you find the specific one may be a good approach, and then you can make a program to repeatedly do the same movement, allowing you time to better find where it’s coming from.

6

thank you for advise.

100% incapable. Very new to 3d Printers and not the brightest star in the sky generally speaking. Improvising and applying basic logic and common sense (within my own little limits).

Really need advise and guidenance form someone who knows what their doing.

As far as the movements go - yes appears to me that it knocks only during certain ones. There are plenty of Z bed moves that are prefectly normal. And then some that will make the knock at the same spot on every layer.

Each layer has the same exact “sound pattern” to it. With normal Z moves and knocking Z moves.

Having said that, as per my previous post, I disconnected motors from the board not to damage anything with induced voltage from manual motor manipulation, and went literally to town with jerking the Z belt back and forth, trying to replicate the knock manually. Cannot do it. Did long fast swipes, super rapid short ones, nothing.

When you say shaft - do you ment the bottom of the screw ? Or the motor shaft ? (sorry, I dont even know if I am using the correct terminology)

I do have a new Z motor, Is it worth throwing it in even if during manual manipulation it is not showing any symptoms? Equally it sounds perfectly normal when driven from the screen with no Z belt on it. And also normal when moving bed up and down from screen.

Its only when its doing those rapid movements during printing that the knocking happnes.

EDITED

7

umm, the plot thickens ?

I posted the same on reddit . I just had someone reply to me :

“I also have the same loud Z-Hops you are describing.
I don’t know exactly when it started but I noticed it a few days ago. It’s so loud, if it was like this from the beginning, I would have noticed for sure.

My H2C has around 600 hours. Like you, I checked the Z-Axis, the bed and everything else that would be mechanically responsible for any movement of Z. Without any change.

My guess is actually the firmware, because it startet around the latest update and Bambulab says they have „improved accuracy of build plate offset detection“ and „Improved motor noise related surface defects at low speeds“ which makes you recalibrate the motor noise calibration after updating. Which I did three times by now since updating.

I also suspected the Slicer, but old G-code files resulted in the same noise when a Z-Hop is performed. To be more precise, if I’m seeing and hearing correctly, the sound happens exactly when the Z-Hop is done. Like, when the nozzle gets in contact with the part again. I think it’s also for gap fill only. With normal hops the sounds are normal.

I might try rolling back on firmware and share the results.”

8

Try the files below. They are .3mf to be able to upload them here but they actually are .gcode. Download the 3 files, replace .3mf by .gcode (for example the first one will become “H2 X moves.gcode”).

Put the 3 gcode files in your USB key and select to print from the printer menu. What they do is for the X file perform several moves operating the X motor only, Y file only the Y motor, Z file only the Z motor. You can at least hopefully then pinpoint which has the issue.

H2 X moves.3mf (1.3 KB)

H2 Y moves.3mf (1.3 KB)

H2 Z moves.3mf (665 Bytes)

2 Likes
9

wowzers !

Thank you kindly Sir !

I will give that a go with wild curiosity. In the meantime however, as I reported, a user on Reddit haas reached out saying he is experiencing the same exact thing. He went on to downgrade the firmware and the kocking went away.

In the meantime I gave that a go as well and the knocking is indeed gone in mine as well. The bed levelling (which I also noticed a change in behaviour giving me reasons for concern, as its all Z related) also went back to how I know and remeber it.

The question then is - is the knocking intentional, are other users also experiencing it ? If not intentional, can this lead to damage/wear?

I am pushing the latest firmware back in, but I am not sure if I should.

May I please ask for your thoughts on this ?

10

Two things can be true at the same time - Bambu certainly appears to have tuned the stepper curves, but at the same time a more aggressive acceleration profile is going to highlight mechanical issues.

We aren’t seeing widespread reports of this, if it were purely down to the update we should be, so I’m still leaning toward mechanical issue.

11

Indulge me. https://www.youtube.com/watch?v=kCHlkCpwa40&t=47s at this very point it looks like your rear jack screw is moving up and down. Try to move that up and down with a bit of force. It should have no play up and down. Next, the next time it’s doing it again, place a finger on that jack screw and see if you can feel any up and down movement.

12

Thank you all for continued support.

X and Y axis sound perfectly normal.

This, however is my Z axis:

May I please torouble you to comapre that to yours? Perhaps even post a video of yours? if it is different? I would like to use that with Bambu Support, which is yet to get back to me..

As for play in each screw, please see below:

Front Left:

Front Right:

.And Rear:

13

I cannot feel any play in the screw itelf …

Generally speaking, if there is play in something, screw, nut, motor shaft, shouldnt this be causing print quality issues ? Cant say anyhing has changed for the worse yet.

I just ran the same print again. All speeds are lowered on it so its Normal !00% speed is slow.

I changed the speed form the screen during print. Going up does make it sound worse. But even slowing down to 50%, the knock is still there, even though the bed is moving really gently …

Should I record a Video, is it of any value ?

14

To be perfectly honest, the last video doesn’t sound abnormal at all.

I hear a heavy toolhead being accelerated and decelerated very hard, which is what needs to happen to achieve the print speeds this printer works at. Yes, when you turn up the speed the printer goes even faster, so it will have faster accels and decele, so and any movement changes will be louder.

The physics term for rate of acceleration change is “jerk” - it’s applicable for highly dynamic kinematic systems, like rollercoasters, and CNC machines, including our printers.

Some part geometries will result in more extreme jerk than others, if you want to post your original model for us to look at and run ourselves that may be the way to go.

15

The user from Reddit I mentioned, who is experiencing the same exact thing, has been in contact with me and has been doing some intensive testing on his end. He is very experienced with 3d printers.

He increased the hop distances just to try to pinpoint when does the knock happen. Below is his video:

the noise profile of the knock is identical to mine. He also says this is originating from the rear, motor/screw area but hard to pinpoint. FW revert takes it away, same as mine.

He has done some very interesting testing. He suspected the new motor noise cancellation to be a part of this, and going from older FW (with no knocking) he flashed the new one but didnt recalibrate the motors. Unfortunatelly the knocking came back immediatelly with new firmware even without recalibration, which in itself, has no effect on knocking after it is performed.

He says his screws and nuts are fine, no issue with those - same as mine.

So the last logical piece was the Z motor, which I had a replacement on hand.

I swapped that in. On a quick visual inspection, the replacement motor is evidently different from the one originally supplied with the printer. Turning its shaft by hand reveals a further difference between the two. Original motor has steps far more pronounced than the replacement one. The difference is very significant.

After installing it, I immediatelly noticed it sounds very different, it is far more quiet. I have been questioning my Z axis sounds from the start, sounding scratchy, resonate-y and rough. All that went away with the replacement motor.

However, the knocking is still very much there, ableit not as harsh or metallic sounding as on the video from the bottom of the printer I posted originally.

As for the model I am printing - it really does not matter as a specific type of Z hop, that is present in anything I print, is triggering the knock. But to answear the question, I was priniting sparkerjc’s replacement lead screw nut when I first noticed it, and i just kept using that as my test model. To save on time and filament I changed it so it prints hollow with a bit of infill, instead of solid as the solid internals dont do hops and there is no knocking, making me wait for a hop (and a knock) when testing. Basically each time the head jumps to another of the 3 screw openings, there is a knock. This is the exact model I been using (plate 5)

H2D+Lead+Screw+Extended TEST.3mf (447.8 KB)

Additionally, Bambu has finally got back to me. Well, I made the same exact post on Reddit, BambuSupport opened a chat there asking for my ticket number, and soon after I finally got a response to my Ticket. They asked for logs and nothing else at the moment. Awaiting reply,

Having said that, more often than not, the community will provide far better support than Bambu themselves. May I please ask for some more thoughts on the matter?

We have two exactly same printers, both bought in Europe, within the same time frame that are exhibitng the same exact issue. I am sure there are more, just peopple not paying attention.

16

Just a quick update.

I am happy to inform that Bambu Support is on the case and are working on a solution with R&D team.

They said they are commited to provide one, so its not an “if” but a “when”, at least as far as I understand it.

Looking forward to it; and if Bambu is reading this, I would like to express my gratitude for their help and a solution soon.

17

Any updates in this case, I found out that since few days my H2C also started having the same issues. 688h of printing time, so it’s not much, compare to my other printers which has 3k+ without any issues like this.

18

Yes and no.

Bambu is adamant the issue is not due to any specific part defect or malfunction.

They are also insisting this will be fixed via a firmware update and keep asking for patience…

Having said that just yesterday I received a response with a gcode asking me to run it and see if the noise is gone.

The gcode appears to run the calibration with some additional parameters via codes which are not documented. They also asked not to share the gcode as it’s been prepared for my printer only and can cause issues if run on other printer.

I am away from home until tomorrow and have not had a chance to run it yet.

Once I do I will post an update. I do not want to speculate on any outcomes yet. In the meantime I would suggest submitting a ticket and mentioning our cases to them to link all of this together.

Are you from Europe by the way ?

19

UPDATE:

After running the gcode there is zero change.

Not going to lie - I am really upset with Bambu at this point. Well over a month of investigating and RnD’ing and this is what they come up with ?

20

I’m also experiencing this knocking sound on my H2C with the new firmware. I hope they find the bug soon, I’m really annoyed!