So I’ve been trying a bunch of different solutions for my bed, attempting to eliminate the potential-ness of a warped bed. Especially if it ‘levels out’ while heated.
I originally printed with PETG, but to confirm that it was the bed, I switched to PLA Pro (Polymaker) and did a bunch of larger prints (10-15 or so, two rolls worth) with varying bed adhesion and ‘band-aid’ solutions. I took one of my larger prints that has been warped before, cut it down to 20mm (so 100 layers), to test these solutions futher. Every single one of them bowed exactly the same.
I then realized that every single print I’ve either run over night, or waited until the bed completely cooled. Today I took the same 100 layer object, and within seconds of the printer ‘finishing’ I carefully removed the magnetic bed, and ‘popped’ the print off, and immediately measured its flatness with the same straight edge ruler. It was ‘perfectly’ flat in every axis, and thus my realization earlier today came to fruition. If my bed is warped while cooled, and I let the printer cool down completely, (running over night generally, or preoccupied when it finishes) the bow is being added AFTER the fact its completed the print. In fact, I’ve never paid much attention to it, but while the printer was cooling tonight after this test, I heard the bed ‘creak’ almost as if it was ‘re-bowing’
While this isn’t conclusive, I plan on running a longer print tomorrow evening, and testing it again immediately after printer, and see if I get the same results, at which time I’ll update with my findings.
This also doesn’t make it any better that the bed is still warped.
That right there may be the ticket if you can access the screws without removing the OEM BL sticker on the aluminum. Shim between the aluminum and the base at the screws, or just remove the screws, add some thread sealant (or blue locktite) to the screw and then insert the screw and tighten until the deviation starts. IIRC there is another bed - Voron maybe that this is the procedure for getting the bed flat by adjusting the mounting screws.
I know there are pictures in this mile long thread of the build plate but I don’t have time in the mornings to research it.
Introducing FullMetal™ plate: which provides 8mm more metal than legacy Bambu PlasticSides plate . I made a 256x256mm 1mm thick aluminum plate with holes for mounting screws (not traming screws which are now embedded inside) and glued it with epoxy glue to original hot plate while pre-tensioned on steel frame. Trying to make super thin glue layer I screwed up a bit. I run the heater to make glue more liquid and it worked well. The problem was that the moment it hardened while hot was very rapid while I was still moving it back and forth to get excess glue out. So I misaligned it by about 1mm on one side and somewhat screwed up flatness on one corner too which I later fixed by bending the plate in glued state and scraping the surface. Got it down to about 0.05mm flatness when cold and hot, it changes shape a little bit when heating but stays within this figure.Then when I put on Trianglelab magnetic sticker it screwed up flatness which became about 0.15mm along right edge and 0.1mm overall. This garbage is somewhat thicker along the edges. Also it’s 254.3x254.5m instead of claimed 255x255mm. Should have bought larger sticker and cut it myself. Still much better flatness than originally but I’m thinking about getting a different magnetic sticker. This construction is just 0.4mm higher than original bed, so build plate alignment ribs still work. Also as it’s lightweight there are no sensor warnings.
BTW Bambu sent me a new bed which should arrive in a few days.
My replacement bed have arrived and it’s warped. It’s better than original one, but still far from perfect.
I’d like to measure it cold and heated, but I’d prefer to heat it up outside the printer, before even trying to replace it.
Have anyone tried to heat the bed by connecting mains power directly to the heating element? What is the power drawn from it? What is the duty cycle needed to keep it at let’s say 70-80 degC?
I measured it’s DC resistance, it’s around 51.6 Ohm. My DC PSU can give 30V, the bed takes about 0.58 A at that voltage, that’s too little power to warm it more than somewhere at human body temperature.
I ran it from autotransformer. Did not make exact experiments but I think it needed about 80-90V to keep it at such temperature after it reached it at higher voltage. This is equal to about 10% duty cycle if you run it from switched 230V. I don’t suggest running it from mains voltage for more than half a minute without interruption. It’s important to measure temperature to avoid overheating as there is 130°C thermal fuse. If you have ~120V version, divide voltages I wrote by 2.
Good point, thanks for warning me, I didn’t know that.
I figured out my PSU has two truly independent outputs, so I can double the voltage by putting them in series, which makes 60V DC, that should deliver about 70W of heating power, I’ll try that before doing experiments with mains voltage.
Today I received a new V3 bed from Bambu. Same Banana garbage, 0.4 mm cave in X direction, mostly flat in Y direction except area near to calibration board. However I expect it to be much worse when hot but did not try running its heater yet. So I was right to not wait for Bambu and modding original bed with FullMetal™ plate. Printed a few things yesterday evening and today and results were simply godlike compared with printing on banana garbage.
Wow, I just checked bed flatness again on my modified plate, and stupid 0.1mm thicker edges on magnetic sticker have miraculously disappeared. So now bed is perfectly flat. No wonder my prints were flawless in this regard.
Hi all,
I have bought 2 Babmulab X1 Carbon Combo printers form the EU store (first one on 14.January.2023, second one received today).
Both of them have super flat beds - I don’t see any issues even when I shine light under the ruler (when heated up thay are perfectly fine too !)
Im 3d printing for over 10 years now and I have seen a lot of printers that had much worse heated bed flatness.
Even in my RAPTOR XLS 360 printer which I’ve built a few years ago the bed is not that flat across the whole surface when heated up (Heated bed was made from special and expensive CNC milled Gleich aluminium plates that have superior flatness across the whole surface - the cost of the alu plate was around 200$ and it is 8 mm thick ! )
Good for you. BL must have put some leftover Kickstarter beds in yours. They then switched contractors and stiffed the rest of us with garbage banana beds, and now are doubling down.
Yes I have heated both of them to 100 deg. C (I have waited for 10 min for temp to stabilize) and they are still flat
EDIT:
Even the auto bed leveling is not correcting for any flatness issues - after calibration the Z axis is not moving at all when moving the X/Y axis across the bed
So I have disassembled my replacement bed and figured out that the warp is the matter of aluminium heating element only. Plastic wrapping is actually there only to cover electric parts, no other mechanical role, it’s soft and thin.
Aluminium plate is warped around 0.3mm in X and 0.1mm in Y.
I’m going to get new glass plate, this time it will be tempered glass 257x257x4mm with fillet edges and round corners (radius is 6mm). I’ll glue it directly on aluminium plate using neutral silicone in the corners and I’ll fill the gap between the glass and aluminium using thermal paste for better heat conductivity.
Then magnetic sticker and new PTFE coating, because I’m experiencing significant mechanical degradation when magnetic sticker is left unprotected.
True, but the reason is the edges are not aligned with the aluminium bed. I will have to use some washers as underlay for the bed, so the glass doesn’t sit on plastic edges at all.
In my case the plastic edges are 0.2mm higher than heatbed, so I’ll use something about 0.3mm to get 0.1mm clearance.
“It’s within Spec” - 1.7mm off. Bambu better hope my next printer isn’t this jacked up. That’s a USA Quarter. This part is 128x73mm in size, printed in the middle of the bed.
. I made a 256x256mm 1mm thick aluminum plate with holes for mounting screws (not traming screws which are now embedded inside) and glued it with epoxy glue to original hot plate while pre-tensioned on steel frame. Trying to make super thin glue layer I screwed up a bit. I run the heater to make glue more liquid and it worked well. The problem was that the moment it hardened while hot was very rapid while I was still moving it back and forth to get excess glue out. So I misaligned it by about 1mm on one side and somewhat screwed up flatness on one corner too which I later fixed by bending the plate in glued state and scraping the surface. Got it down to about 0.05mm flatness when cold and hot, it changes shape a little bit when heating but stays within this figure.Then when I put on Trianglelab magnetic sticker it screwed up flatness which became about 0.15mm along right edge and 0.1mm overall. This garbage is somewhat thicker along the edges. Also it’s 254.3x254.5m instead of claimed 255x255mm. Should have bought larger sticker and cut it myself. Still much better flatness than originally but I’m thinking about getting a different magnetic sticker. This construction is just 0.4mm higher than original bed, so build plate alignment ribs still work. Also as it’s lightweight there are no sensor warnings.
