It’s an incentive thing. So far they seem to be open to trying to fix things and make things right, but only in private. I suspect if people keep blasting them (for instance, by posting about warped beds in threads about other topics made by Bambu Lab employees), it could have the opposite effect and they will participate in public forums/social media less. I don’t think they really care about the public complaints.
I get it. But you don’t actually know if the bed is warped once installed and heated, correct? Did you install the first replacement bed they sent you that wasn’t as bad? Seems like having a flat bed when it’s installed and in use is the most important thing.
And I would guess if you installed it and it was still warped, they’d probably give you the refund you wanted given how you’ve already received multiple replacement beds.
Well I am not spending 4 hours running on speculation that a visibly bowed bed will magically change its structure when installed. Funnily enough, that’s the same line of reasoning Ciprian used as well (you sure you don’t work for them?)
Bambu is welcome to swap out my printer otherwise I will be refunded one way or another.
Not going to argue this silly line of speculative reasoning.
I feel like this is at least partially about me in the P1P announcement post but that’s okay I stand by it. I refuse to believe that keeping quiet and dealing with them in private is the way this gets resolved, especially since the customer support experience is different for everyone
Why were we not even offered to wait for one of the “flat” beds
Why were they okay with offering full refunds on both of our x1cc, when one of them and the $800 worth of spares returned were outside of the return window, but other customers are refused refunds?
Can you say with confidence that you think Bambulab would have even addressed the issue to begin with if people weren’t going to public forums to express their dissatisfaction?
Well, plastic by nature is somewhat flexible, so it’s not far fetched that it might straighten out when firmly attached to a more rigid base. Given that materials also “magically” change their structure when heated, I don’t think it’s all that speculative.
I am sure, but I’ve worked in plastics manufacturing for almost 20 years. Ciprian sounds like a smart person who understands material science.
I want you to have a flat bed and I’m disappointed that some people (you included) still haven’t had this issue solved. But again, I’m simply asking if getting a flat bed before it’s installed is all that important. Seems to me that once a bed is installed, it’s going to flex, especially at min/max temp, so those are the only measurements that really matter.
Of course not. This thread already has 13k views and hundreds of posts. It’s gotten plenty of attention and will continue to until the problem is solved.
I didn’t say that you should “keep quiet.” I said that they appear open to trying to fix things, but in private. It’s been 3 months since this thread has started, a they’ve made 1 blog post about it (without much info in it), that’s it. Employees haven’t responded to this thread. They didn’t respond/acknowledge to warped bed posts in their P1P announcement thread.
If you feel like posting about warped beds in more places is going to bring about a faster resolution, then have at it. I just don’t think it’s going to bring about a resolution any faster, and it will make them apprehensive about participating in public more than they already are.
No idea. But these things usually differ from case to case, which is why I feel like they are better handled in private. But that’s just me. Others should do what they feel is best for them.
Is there a way to design a printed bed such that when it is heated, the thermal expansion is controlled in dimension? I can believe that something bowed becomes more bowed or less bowed after heating. It would be ideal to have a flat plate expand in one plane as opposed to having unpredictable behavior. Maybe a good reason to have a multipart or multi material bed where thermal contraction and expansion is directed into non-printable regions
Idk, it is starting to feel like beating a dead horse so I’ll likely chime in less and less (until we order another one😈)
I have mostly been waiting for an update on the new flat beds so I could order another machine + AMS because I stand by it likely being the most reliable multicolor system available right now, but it seems like I’ll be waiting for one/some of you intelligent, resourceful, gentle ladies/men to come up with an aftermarket solution for the beds before ordering another for us, that or a “Pro” line, as again I’d gladly pay the premium
Fun bonus fact: Ciprian was also our support agent for the warped bed issue. He was great, and if you catch him at the right times he’s very responsive.
Prusa is doing something similar to what you’re thinking of with the multi-segmented heatbed on the XL, in theory it’ll work to keep warping down but it’s still too new to say for sure
Make it all one material. Then it’s pretty predictable. Or if multiple materials are needed, then the one with the lower CTE needs to be the dominant material and make up the majority of the part.
These materials will always expand laterally. They are warping because one material is expanding and contracting more than the other ones they are attached to. The unpredictable nature of how they are warping probably has more to do with slight variations in the manufacturing process.
I feel you. It sucks after all this time that some are still having issues (although it does appear that some had their problem solved with new beds).
I’m working on it. I should have one warped bed in my hands in the next week or two. Ideally I find another 2-3, but I should have a prototype pretty soon.
This was kind of my point about getting things sorted privately if you need help (I could have worded it better). As @Thrawn pointed out here, it very well could be cultural differences that are causing a lack of public comment. Everything I’ve seen from support seems like they are very willing to make things right when you contact them through the proper channels. If they weren’t, I’d have a much different opinion on calling them out on social media/forums.
I wish it was so simple. It’s not. Even if the bed was made from one single material, it would still warp as long as it is heated from one side, just because of temperature gradient. And the effect will be stronger with higher thermal expansion coefficient.
As I understand the structure of the heated bed, it is basically PCB heating element attached to aluminium plate. It is heated from the bottom side. It means, bottom side it hotter then top side. Bottom side expands more, thus the plate warps its edges up.
Maybe it would be better to place the heating element between two aluminium plates, so the expansion (warping) forces work against each other?
The problem is the aluminium. We need a glass/mirror or another material with a much lower expansion under heat. Like Pyrex-, Borosilicate-, Quarzglass or a metal like Invar, or Meissner porcelain.
α in 10−6 K−1
Alu: 23,1 !!!
Quarzglass: 0,54
Invar: 0,55 bis 1,2
Pyrex: 3,6
Borosilicate: 3,3
Meissner porcelain: 3 to 5
I accept what you’re saying. Although heated from one side, wouldn’t the aluminium plate stabilise over a period of time? So the pcb is heating from below to 100c, the plate expands as you describe. Once the aluminium plate has heat soaked up to the same temperature then would we expect the plate to return close to flat? The bottom may still be slightly hotter as the heat comes in that way.
In short, those with bed issues. Do they lessen if you heat soak the bed for a period?
In my opinion, not really, unless the surrounding air has temperature very close to bed itself.
Let’s have typical example. Let’s print something with bed temperature set to 100deg. The chamber stabilize at around 50deg, the heat is radiated through the printer casing into the room. It means the bed looses heat constantly. Let’s assume it means the temperature of the bed on its upper surface is 10deg less - 90deg.
If the bed was 250mm wide and expansion coefficient was 26e-6, you get 250.46mm on the bottom side and 250.42mm on the upper side, it means curvature somewhere over 12m radius, it’s pretty simple math and nobody is able to do anything with this, the physics is unbeatable.
The solution is
a) use material with lower coefficient (at it was said thousand times - glass is way more suitable)
b) use different design of heating
c) make it work with much warmer surrounding air to minimalise heat loss (not feasible for filaments with low glass transition temperature), but it has also consequences in shortening lifetime of some components of the printer - timing belts, linear bearings, steppers etc.
There is one interesting thing - some people experience that the warp is actually smaller when the bed is heated, which is a bit confusing to me, I believe it has something to do with desing of surrounding plastic part of the bed. Or maybe they simply measured it wrong, we don’t know. Anyway, my bed is actually way worse when heated (0.3mm gap in the center when cold, 0.8mm when heated).
With a hell a lot of effort I bent bare heater plate to <0.05mm flatness. Really tough job without having some sort of press to apply a lot of force in controlled manner. Plate is very stiff, certainly not some soft aluminum alloy. Now when hot plate stays dead flat in X direction where it previously became concave and becomes slightly concave (~0.05mm) in Y direction where it previously was somewhat bulged.
Also with thermal camera I noticed that sides/corners are noticeable colder than center of the plate. Which of course contributes to warping. It’s not surprising all due to how copper traces are laid out without making center more sparse and edges more dense. It’s exactly opposite, same density everywhere and no traces on corners where screw holes are located. This is without plastic base, so maybe some thermal insulation it provides makes the difference smaller. Did not check temperature uniformity before taking it apart. Disregard temperature reading in red color since it’s measures black screw temperature while readings are corrected for lower emissivity of aluminum plate.
Its even worse then just that. Since the plate is usually screwed, the added tension of the screws will warp the metal even more to relieve that stress.
There are quite many factors, and most of them add up to make it worse instead of playing nice and cancel each other out. It is really not easy to construct a good, lightweight and flat heatbed.
Wish my X1C was delivered so I could find out what boat (banana?) I am in
If anything this discussion is showing, a community solution or a Bambu solution to having a print bed with a flatness tolerance relative to the motion of the print head to print to is very desirable and will differentiate printers just like how high quality print speed differentiates today.
For my CNC router I know how important it is to align the gantry to the spoilboard and we do this by surfacing the spoilboard so that it is uniformly flat w.r.t the spindle.
For 3d printing, where the material AND the platform is heated, it’s a more challenging problem worth solving to the degree of setting expectations and hopefully software can help with the contraction of the model but definitely can’t help with a warped print bed.
I got my X1C in Feb should be a mass produced one, but still comes with a V2 bed, warped, support asked for the pic of the underside, sent me another V2 bed, still warped. If the V3 beds tend to be more flat or better, can I just ask for a V3? or it simply won’t fit my machine?
V3 is not better at all and also requires replacing power board as it has different connector. Or you can crimp different connector but it’s not Bambu approved. V2 and V3 are basically the same except power connectors and different power connector cover. But in regards what affects performance and flatness, it’s the same thing. There are plenty of banana v3 beds reported, including mine. And apparently replacement v3 beds they send are mostly bananas too.
