Nozzle Knocking Print Loose

Hi All,

Hoping someone can help be get to the bottom of the issues I’m having. I’m looking at a 95% failure rate at this moment in time. Had my A1 since Christmas 2024, got a hold of the FDG miniature and terrain profiles. Adjusted the temperatures to suit Sunlu PLA meta filament and was having great results with minimal stringing for about a week. Since then I’ve been having a lot of trouble with the nozzle knocking my prints off the bed and causing spaghetti.

I can pre-answer some common suggestions for this.

I have recently lubed up the rails.

I have Zhop when retract enabled, setting the threshold distance to 0 for that makes no difference, the prints still get knocked.

Adhesion seems perfectly fine, I wash my bed every 3-4 prints and checking the first layer looks smooth and well laid. Bed temp is 65. I print first layer at 200 and the rest at 185. (I should note that I have also tried to print at 200 and 220, which made no difference)

I haven’t ever used Grid infill, I only ever use Gyroid or Lightning and it’s always set to around 5%.

I’ve checked the screws behind the nozzle, all tight, couldn’t even move one 1/8th of a turn and doesn’t seem to be any play in the toolhead.

For smaller, less bed contact models I’ve been using a 10-15mm brim, and while that sometimes helps, it’s not a guarentee.

I have tried lowering the flow ratio from 0.98 to 0.95 just incase I was over extruding.

I have tried Zhop at all levels from 0 - 1.4 (overkill i know, but it was worth a shot)

The knocking happens regardless of nozzle size, I have used both a 0.4mm stainless and a 0.2mm stainless.

The smaller the layer heights, the sooner the knocking occurs, which makes sense to me due to less tolerance for error at small heights, which suggests the problem is with the Z axis.

I have reduce infill retraction disabled and avoid crossing wall enabled.

I don’t know if this is important but if I try to print multiple models at the same time my failure rate is about 100%.

I can’t be 100% on this but from what I’ve witnessed the knocking seems to occur during travel moves, rather than layer or infill.

At this point I feel like I’ve tried just about everything I can think of but my printer just refuses to behave. I’ve also seen posts about Bambu Studio potentially not generating Zhop movements properly, which I’m hoping is not the case since that would mean this issue is over a year old with no software fix in sight.

Any help you could offer would be greatly appreciated. I’m relatively new to 3D printing but I’ve already learned a lot so don’t feel like you would need to dumb anything down for my benefit. At this point I will settle for a print completing, I can always fine tune for quality later. Thanks in advance.

1 Like

There are only two ways a print can get knocked over:

  1. The filament is stringing, so a blob builds up on the nozzle which is eventually deposited on the top of the print and when the nozzle comes around to that spot on the next layer it snags on the lump and knocks the print over.

  2. The print is warping so one or more corners are “higher” than the actual layer height.

In the case of #1, most often stringing is due to filament needing to be dried. You covered a lot of details in your list of troubleshooting but did not include “I dried the filament” so you might consider that.

If the print was warping, I’m guessing you’d have noticed. But if that is the problem, your only fix is to print in a warmer, less draft space or put the printer in an enclosure.

I’m betting it’s #1 because you specifically mentioned “stringing”.

6 Likes

Hey, thanks for the quick reply. I realised after I wrote my post that I hadn’t mentioned drying. I have the creality space pi and typically dry a spool for about 8 hours. Humidity levels in the chamber are around 14% during this time.

Do you think I’m drying for long enough? How often would you recommend I should dry a spool? I’ll admit it’s been maybe 2 days since I dried it before posting today. Humidity in my home is around 50%, I live in a damp country unfortunately.

I have noticed a thin coating of PLA around the nozzle during swaps and fault finding a few times. When I say thin though, I’d bet it’s still thicker than my layer heights. I’m hoping you’ve hit the nail on the head. Drying is still something I haven’t really found a lot of information on, other than it varies vastly due to environment, manufacturer and material/colour etc. Thanks again.

Humidity and Dryers are often misunderstood.

When we talk about Humidity, we generally mean “Relative Humidity”, where the “relative” part is “relative to temperature”. If you increase the temperature of the air, without making any changes to the moisture content of the air, the RH will still go down proportionally with temperature increase.

And what the dryer is reporting is the humidity of the air in the dryer, that’s not the moisture level of the filament.

And filament “diffuses out” moisture in a dryer very slowly.

Because of this, it’s not the absolute RH that’s reported by the Dryer that you really care about. What matters is if the RH is still decreasing. Initially the RH will drop quickly because the temperature is increasing. You have to give it time and watch it carefully to see when it’s really bottomed out.

Leave the filament in the dryer until the humidity bottoms out and then, go another 8-12 hours.

Or, you could just throw the filament in the dryer a day in advance of when you plan to print and 24 hours will generally be enough for any filament moisture level.

How often do you need to dry? It depends on all the factors you outlined above. I take a roll I think is dry and start a print and if it looks too stringy, I stop the print and dry it.

3 Likes

Thanks for taking the time to explain that, makes sense to me. I’ll give the filament a longer dry time and maybe invest in an airtight container or two with some silica sachets. I’ll report back if things go well.

1 Like

Hi Rocket, so I already had my filament drying when you messaged back last. I had taken the spool out after 18 hours of drying and did a test print with it. The nozzle is still dislodging my prints. I’m at a loss here, none of my settings were changed from day 1. I did have 1 “successful” print using the stock 0.12 bambu profile whereby the model completed but the supports on the model were pretty much welded on and it was a complete mess, I had enabled “on build plate only” it’s almost like my Z axis gets worse as the print goes on, which is why there were almost no top interfaces on the supports.

I’ve manully trammed the bed, and tried changing the Z offset in the machine code, but neither seems to have made a difference. That seems to only leave warping, but the environment is pretty warm, no drafts, print bed is at 65, the weird part is that when the print fails I can’t even see any telltale print defects on the topmost surfaces. Think I’m going to have to make a video to really see what’s going on.

1 Like

This suggests you are over-extruding. If the filament is dry, then the flow is too high. Properly tuned, supports should almost fall off the print leaving no material behind. Every time I’ve had a problem with supports, reducing the filament flow rate a few 100ths has fixed it.

Note that wet filament will over-extrude, which is why I went with “drying” first. But even dry filament can be sub-optimally tuned and over-extrude all on its own…

2 Likes

So I just had a succesful print at 0.1 layer height, I disabled “independent support layer heights”. All I know at this stage is it stopped the collisions, but I have yet to figure out why. I’m assuming, that it’s because the support layers were printing marginally taller than the object itself which would eventually lead to a collision due to the uneven height. This would also explain the fact that I was seeing no surface defects, because the nozzle hasn’t been contacting the surface but rather the side of the supports. If I wan’t to reenable independent support layer height, which setting would I need to adjust to avoid this happening in future for smaller layers?

2 Likes

what you guys are talking about of the TH knock the print over isnt it

I had that happen in the past. Independent support layer height means it’ll max flow on those layers. In particular in combination with a low wall count (supports), this promotes curling. Curling in turn places material in the path of the fast nozzle, knocking over the weak supports.

If you want to have avoid that, you can try to increase the support wall count, but that would not eliminate collisions. So it’ll just move failure higher. Note that with increasing height, the moment arm increases, so you do not gain much with that.
To tackle curling itself, you’d need to lower flow, but that would also apply to the model itself.

=> In my use cases, it was always better to abandon the small print time gain from independent support layer height.

1 Like

Well, the independent layer height for supports did not safe me much and gets you to support printing issues. So I’d recommend abanoning that hope.

What could make a big difference is, well, did you thoroughly dry the filament? It has been around the 7 seas and probably moist out of the bag. Moist filament tends to curl, warp, ooze and string much more than dried filament. So that could help. Another indicator is worsening print performance (which you indicated may be a factor).
Hopefully that can already improve things.

Removal is usually more difficult the smaller the top z-distance. But .2 should usually be enough unless you have insufficient cooling and/or excessive sagging (moisture).

Are you already using adaptive layer height? I have found that to help both quality and likelyhood of success.

Finally, if you have an AMS Lifht, you may be interested in using (dried) PETG as a support interface (top z-distance 0mm, top separation 0mm). It may be overkill and extend print time significantly, but it does improve overhangs significantly.

Filament should be dryer than the Sahara, fresh out of the Space Pi after 24 Hours. I just had a successful print (yay), minimal next to no stringing, but I put another similar print on afterwards with the exact same settings, which failed after layer 40 or so just now, which is kinda weird, maybe it’s a problem with the models? I don’t have an AMS unfortunately, don’t have the space for it nor did I think I’d need anything other than PLA at the time of purchase. I’ll give the adaptive layers a shot, done that on some of my bigger models with a good degree of success.

1 Like

Yeah, 24h should have done it :sweat_smile:

I am wracking my brain as to what other things come to mind. But other than trying a slow print at 50%, nothing really sticks out. Better one good print than two bad ones per print time and all that.

But that stringing is a bit strange for dry filament.

Oh, the stringing was minimal like 1 or 2 wisps where the support interface was, i think it’s a side effect of meta being a low temp PLA.

1 Like

So I now had a second success on a small model. Changed the layer height to 0.12 instead of 0.1 to see if was maybe the stepper motors increments that were causing the issue, 0.04mm increments I read. But, slicing in the 3rd model, knocked loose again, exact same settings. I’m super confused now.

1 Like

Do you have another brand of PLA available? Just to rule out that it is simply a dodgy roll/particularly tricky or even bad batch.

You may also want to consider pasting the 3mf here.
I myself will not be able to have a detailed look before the weekend, but it is quite likely somebody else will have a detailed look and be able to advise based on what can be spotted in the file.

“Sorry, new user cannot upload attachments”. I’ll try once my account has gone up a level. Thanks for your suggestions so far though, been helpful.

1 Like

With this account, you have enough reading time to copy-paste. You may just need open a few more threads and scroll through them. Shouldn’t take more than a few minutes.

The you should be able to just copy from your drive and paste straight here.

So I increased wall loops to 2 from 0 on supports, seem to be having more success. I also tried to reduce the wall loops to 1, the print failed, I’m thinking it’s likely to be curling as you said, but I’ll see if it starts failing on the 2 wall loops next :sweat_smile:

Estou tento o mesmo problema com a A1 aqui. Começou quase no mesmo período de tempo, cerca de 2 meses depois de chegar a impressora.
Sou novo na impressão então vou acompanhar esse tópico.
Agradeço a todos