slight layer shifts on manual filament change (with pause) #1411
Comments
|
How long did it take for you to change the filament? The motors get powered off after 15 minutes of idle time, that's when the shift happens. It should not happen if you make it in this time and handle the toolhead delicately. |
|
Not caused by slicer and gcode. |
i printed those little cubes for testing and i stood right next to the printer, as soon as the printer paused |
|
@suit It looks there is problem with your printer, could you please create a ticket with related LOG? |
|
@sodachen22 since i've printed dozens of this cubes, i can't find it in the log right now - but i'll reprint them later this week and create a ticket, there |
|
I can confirm that if I disconnect AMS and try to manually change the filament, I get a layers shift as well. I also noticed that if I ran out of filament and there is not any suitable filament in the AMS and it takes me an hour to get back to the print, the layers shift happens as well. I'm afraid it's due to powering off motors after 15 minutes. |
|
I can confirm this too. On one day the manual filament change (with pause) works perfekt and on the next day after the update i cant print anything with it. Additionally i can see at which point the layer shift happens. Its right after the extrusion of the new filament and the beginning of the new layer. The head jumps a little bit, perhaps 1mm and then the new layer starts. |
|
Based on the g-code, it appears this is for the P1P, and based on @DeKra90's comment, this seems like it might be related to the latest firmware (it worked before, but after the update, it did not work). Can @suit confirm which firmware you've used and noticed this issue? For the manual filament change, are you using the unload and/or load button on the P1P? |
|
this happens on 01.02.01.00 - and yes, i'm using the manual load/unload feature introduced in this version i cannot confirm if it worked as expected before the firmware update though, sind i've nevertried it |
|
A similar layer shift occurs if the "step recovery on power loss" is turned on and the power is cycled. Perhaps related to the same thing - the motors powering down, |
in this case the issue also occurs when you change the filament right away - to my informationen, the steppers power down after 15 minutes, i did not exceed this time period, but the shift still occured |
|
Same happens to me on my P1P with no AMS. When it runs out and pauses, the print head moves near the chute and the whole printer starts vibrating violently and making an awful noise. This persists, until i feed new filament into the head and resume the print, at which point the layers shift. Maybe the head "hits" something on its way to the chute or the motors gets jammed ? Untitled.video.mp4 |
|
I am having the same issue. The time interval of the filament change does not change the outcome. With the latest firmware update, I get a layer shift every time a filament runs out. P1P no AMS |
|
I have this issue each time I insert a pause into the print. The first layer after resuming always shifts. P1P |
|
Having this issue as well. Very annoying.
|
|
Further Investigation done
both 3a and 3b yield the same result - so the shift happens before that during the unload is there any news on that? |
Someone "solved" it by adding customer G code found here (https://github.com/EukaDesigns/Bambu_CustomGCode/blob/bfdd8d4e9dce94def22d21b0e4ac8b5e49ebadeb/change_filament_noAMS.gcode_ What happens is in slicer sw you add filament change at your desired level. When you run part and it gets that layer, head positions over the poop chute and unloads, the pauses. You can then manually load new filament in (making sure nozzle is hot enough and pressing extrude button a few times); doing this, the layer shift hasnt happened to me. I THINK I also tried doing the LOAD option on the P1P control as opposed to manually loading, and the shift didnt happen doing it that way eiter. |
i'm aware of that - i've just now added a comment on the reddit post of the same guy problem that i have with this: it is a workaround that is based on an older version of the gcode provided by bambu studio - so there is quite a lot missing other thing is, this code turns off the fans, heats up the nozzle to the higest temp, cuts and flushes and then pauses the printer - so if you don't change the filament right away, the nozzle stays hot and the fans stay paused i'm aware that the code is "on your own risk" but, this is not a suggestion that should be made without crosschecking the code since this is appearantly a printer bug that occurs during homing on the filament unload (not the load) it should be solved on the printer - but the slicer can workaround this if you don't want to break the AMS in this process, there needs to be condition to check if there is an AMS present or not - and then add a manual pause and a filament cut those issues are closely related to #1409 and #1408 which can be probably solved with the same solution a) add a pause on manual filament change |
|
My P1P had layer shifts after manually changing filament since I got it, which was at the beginning of February. So the problem exists at least since firmware 01.01.01.00. |
|
i found another workaround and can confirm now, that this issue is indeed a result of the UNLOAD function Steps to avoid the issue
i printed about 20 testpieces so far, all of them have a perfect layer transition without a shift if i do the same, and use the UNLOAD function: layershift after layer shift @SaltWei @sodachen22 is there any news from the printer team about this issue? |
|
Just happened to me. Pause for layer change Slight shifting of the layer, maybe only half a mm or so but still. The best resolution I've seen floated is, instead of using a normal pause, add custom g-code - So it homes after the pause. Am going to test it now to see if it actually works. I see there's a long 'custom gcode' solution in this thread but I'm not sure how it's used, and it seems pretty in-depth and convoluted. Will leave it for now. Edit: Seemed to work fine, so it's a workaround for the time being. |
|
@SaltWei any news on this from the printer team? It‘s been more than two month since it was reported. Pretty annoying bug. |
|
Yeh this was fixed ages ago. Like 3 firmware updates back or something.
…On Thu, 11 May 2023, 19:57 Spybyte, ***@***.***> wrote:
@SaltWei <https://github.com/SaltWei> any news on this from the printer
team? It‘s been more than two month since it was reported. Pretty annoying
bug.
—
Reply to this email directly, view it on GitHub
<#1411 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA4T22NG3K5UGYDCE3ODXSDXFUZDDANCNFSM6AAAAAAVQGLCLM>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
It was not. They just added feature to load/unload but it happens as well (at least from other bugs). |
|
The load/unload feature came before the bug.
I thought the bug was fixed by an update which added a home after layer
pauses, but maybe I just did that myself... I can't remember now, I haven't
had any layer shifts in ages and just forgot about it.
Scratch my last comment then, maybe it's not fixed officially, but you can
fix it for now by adding a home command in your pause / filament swap
gcode. Sorry my bad, poor memory.
…On Thu, 11 May 2023, 20:04 Jiří Málek, ***@***.***> wrote:
Yeh this was fixed ages ago. Like 3 firmware updates back or something.
… <#m_7352415902838322240_>
It was not. They just added feature to load/unload but it happens as well
(at least from other bugs).
—
Reply to this email directly, view it on GitHub
<#1411 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA4T22JGKCONGL6TXYRQQRDXFUZ27ANCNFSM6AAAAAAVQGLCLM>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
No, it was not fixed. I have the latest firmware on my P1P and I still have layer shifts. The last one two days ago. Yes sure, I can do that fix. But it should have been fixed by now. |
|
Change your pause gcode in your slicer. Seems to work as a fix until bambu
do it properly
…On Thu, 11 May 2023, 20:21 Spybyte, ***@***.***> wrote:
No, it was not fixed. I have the latest firmware on my P1P and I still
have layer shifts. The last one two days ago.
—
Reply to this email directly, view it on GitHub
<#1411 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA4T22K6DJENJ6XN3ZBJWXLXFU32JANCNFSM6AAAAAAVQGLCLM>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
I found this thread, read it and thanks to you guys I know how to do this work around. But my question still remains, when will it be fixed? I would say a Bambu printer is a premium consumer device for that price. Because of this I would expect that such bugs would not take more then let’s say two to three weeks to fix. But maybe my expectations are way to high. |
|
This is the github for the slicer though, not actually sure it's even on
the main firmware teams radar. Might be worth sending an email to the
actual support team for bambulab and linking this thread there.
…On Thu, 11 May 2023, 20:40 Spybyte, ***@***.***> wrote:
I found this thread, read it and thanks to you guys I know how to do this
work around. But my question still remains, when will it be fixed? I would
say a Bambu printer is a premium consumer device for that price. Because of
this I would expect that such bugs would not take more then let’s say two
to three weeks to fix. But maybe my expectations are way to high.
—
Reply to this email directly, view it on GitHub
<#1411 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AA4T22MOIA2GUCVHGB2WWTTXFU6BRANCNFSM6AAAAAAVQGLCLM>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
You are of course right. This is not the proper way. But @SaltWei assigned it to the printer team (as he stated somewhere above). That’s the reason I asked him. |
|
Hi @suit
Tips:
|
My dude, you just posted a huge workaround post for the X1 touchscreen and ams interface. This entire thread is for a bug with the P1P. What can be done for P1P ??? |
|
Hello @HobinBambu as posted earlier, there is already a much faster and easier workaround directly on the printer - this huge wall of screenshots and steps is most confusing, since the issue is pretty clear: simply don't use the load/unload features so you don't screw up the homing and you are good to go i've posted this 6 weeks ago: #1411 (comment) |
|
@suit @Sandermand |
|
I just tried the workaround on my p1p, as in:
And I still got the layer shift of ~2mm to the left. Did I do something incorrectly above? |
|
Severely disappointed in BambuLabs and close to simply sell these printers and go back to Prusa. The development is far too slow. Bug fixes take forever. |
I agree, this is frustrating to say the least. I myself am waiting for fixes for 2 bugs I've reported almost 3 months ago. 🤦♂️ |
|
The issue was listed as fixed in the latest firmware. You still
experiencing this?
…On Mon, Jun 26, 2023, 09:40 MadKiefer ***@***.***> wrote:
Severely disappointed in BambuLabs and close to simply sell these printers
and go back to Prusa.
It takes FAR too long to get a suitable fix for the layer shift. I've not
had one such issues on any of my printers in the past.
Same thing, no matter how a stop is produced (nozzle stuck to a support or
Spool running out), the resume always ends up with a shift, ruining long
prints in the process. When I have prints that are detailed (we work with
art stuff), this is a waste of material, horrible for the environment and
financially not agreeable.
The development is far too slow. Bug fixes take forever.
—
Reply to this email directly, view it on GitHub
<#1411 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AGUKWEG5IP5M662NKCIDFMLXNE4GVANCNFSM6AAAAAAVQGLCLM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Yes. It is still an ongoing issue. Both of our machines have this issue whenever the spool (NOT AMS - yet) is empty. |
|
@thedodus @MadKiefer sorry for the issue,we fixed the layer shift issue caused when pause the printer and use 'load/unload' commonds in P1P OTA version01.03.00.00 ,and this issue is caused by the 'load/unload' not the 'pasue/resume',If you only use 'pasue/resume' in printing but it still cause layer shift, It is recommended to check whether the PTFE tube or the TYPEC cable will affect the movement of the tool head, or re-tension the belt, if the issue still exists ,we encourage you to reach out to our customer service team Click here to open a new ticket in our Support Page,thanks! |
|
Testing was conducted using version 01.03.00.00 on a P1P, and it appears that the problem of layer shift I previously encountered no longer persists. Here are the steps I took that led to this issue in the past:
No layer shift anymore. Hope this helps. |
|
I am having this exact issue even with the most recent firmware |
Still having this issue... It is not printer mechanics related because if I do a manual pause with the printers physical button it does not cause a layer shift. It only happens with the add pause layer in bambu studio. What is the fix? |
Describe the bug
when the filament is changed manually during print, there is a slight "random" layer shift (x/y), that is not a result of the gcode itself
this offset appears do not have something to do with the gcode itself - see attached code, the pause and filament change is on L25 and L26, the X/Y values are exactly the same for all the layers - so the resulting gcode is in order
i assume that the filament change, nozzle wiping and moving around the toolhead screws up the homing information by a hair and thus causing layer shifting
i've printed multiple parts so far, the layer shift seems to always lean to the rear/left side of the printer - it ranges from about 0,1 to 0,5 mm
i tried this with a few of those test 10 mm³ cubes and i always get a shift - i've also been very carful to note move, touch, or shake the printer in any way - if you use the same gcode and don't change the filament and resume right away, you don't get a shift
3mf File for This Bug
pause+filament change_next_layer.gcode.gcode.txt
To Reproduce
Steps to reproduce the behavior:
Expected behavior
X/Y position stay exactly the same on resume
Screenshots
left side is also warped due to another issue already reported, but the layer shift is also pretty substantial here
The text was updated successfully, but these errors were encountered: