Sadly this happened to me yesterday, just gonna copy my post from Reddit:
Brand new P2S - wanted to try out the Biqu Glacier plate and something really weird happened. During the preparation step before the print, the toolhead went to the cleaning / purge area in the back, suddenly (it hadn’t even purged any filament yet) the bed lifted all the way up and the nozzle scratched right over it. I cut power immediately and was able to get the nozzle removed and then later lower the bed.
I used the “Cool Plate” build plate type as recommended by Biqu, turned on auto bed leveling. Did I mess up here? Please be gentle, I’m new to Bambu and came from a very cheap printer.
Shouldn’t there have been anything in place to prevent this? Should I be worried that the printer is damaged after this “crash”?
So far I’ve learned that this issue has happened to some other P2S users as well, for example I found a post from Germany: Redirecting..., where the bed crashed into the nozzle in the same exact way (even the same scratch pattern) when he used the Cool Plate setting (he used the official Bambu plate).
Must be a firmware issue with the cool plate input, I have always used smooth plate with the glacier which is what the sticker code is on my H2D glacier plate.
That must have been a disappointing thing to happen to a new plate.
A couple things, if you simply swap out the plate and the printer doesn’t know about it the Z height will be from the last plate. Did it do the slow probe in the middle first?
From the video, it looks like the plate is lifted (as it wasn’t seated properly along the edge near you) but that might be just a perspective effect.
Thanks! Yeah it’s quite „meh“, luckily a plate can be replaced or simply flipped. But I did not dare to use any other plate than the stock PEI after this happened.
I did select the Cool Plate option (recommended by BIQU), removed the textured PEI plate (just saying - someone suggested I may have put one plate on top of the other lol) and then put the Glacier plate on the raw bed and it snapped on perfectly. I agree from that perspective and with the camera lens it does look like it is lifted, but that was not the issue.
To be honest I can’t recall exactly what the printer did, but I don’t think it did the proper probing at this stage. The toolhead was rested in the back for a while and then all of a sudden this happened.
You can also see that it goes right back to the cleaning area after destroying the plate - strange?
After this, it continued to run into the plate with force (ended up in the center of the plate), which is when I acted and quickly cut power.
Given that there are other reports about it, I no longer believe it was my fault at all. Something about the combination of failed plate detection (that’s on BIQU for not providing a proper compatible AR code for a plate they advertise as compatible with the P2S) + using the cool plate option on the P2S.
Issues like were a problem with ALL Bambu Lab printers …
I had this with P1S where it murdered my new Engineering Plate while moving the head.
Then I had a similar thing with H2D couple days ago, but I caught that before it actually moved into the bed.
Those are extremely rare issues, but they do happen with BL and I don’t know why. But all of them have exactly the same problem - printer “forgets” the homing position, and sometimes FW doesn’t catch that which results in bed either going “all in” into the nozzle, or “all in” down into the floor (which happened so many times with me).
With H2D 3 days ago it was even funnier - I just did a homing procedure, then proceeded to move the bed all the way down just to realize that the printer already forgot its own position and rammed into the floor on full speed.
Submit a ticket, bbl won’t be able to replace your BIQU plate but when something like this happens and they recognize their fault they would provide you with a bbl plate, good thing they have multiple plate options to choose from on the P series
If you suspect it has to do with the gcode from plate selection, you can inspect that from machine → machine start gcode
I doubt it is that simple though. Theres a convention that textured plate gets mashed down in z by 0.04 to account for nooks and crannies, but that’s about all I’ve ever seen in there.
Update: Took a couple days, but props to Bambu Lab! They fully acknowledged the issue, confirmed that it’s a known issue. Checking on the eddy sensor is recommended in the meantime:
After reviewing your description and the related reports, we can confirm that this issue is related to abnormal data from the eddy-current coil sensor, which can occasionally cause incorrect Z-axis movements during the pre-purge step.
To prevent this issue from recurring, we recommend manually adjusting the position of the eddy-current sensor. This adjustment has proven effective in resolving similar cases. Please refer to the following guide for step-by-step instructions: Adjust P2S Nozzle Eddy Sensor | Bambu Lab Wiki
They also mentioned that there will be a firmware fix available (though no ETA).
Not sure how that relates to the manual adjustment of the Eddy sensor, maybe it’s just improved detection and you will be prompted to check the sensor or something like that…
And finally they offered to send me a textured PEI replacement plate.
Good to know that a solution is documented, but what’s Come in mind is … Is it needed to be checked directly out of the box on a brand new machine, or as said in the documentation, if all calibrations goes well then it’s ok to go ?
This had just happened to my p2s. I still have warranty, how would I create a support ticket. The scratch is so deep on the plate I am surprised it dint go through the print plate….
