I think the idea is ABL takes care of figuring that out. And then the 2nd layer will be completely flat.
Certainly a possibility – even when using tape with air gaps in between. The maker of that video followed up with this test to show the honeycomb doesn’t introduce heat transfer issues.
FWIW, I haven’t attempted this yet, but commenters on that video are saying it works well for them.
It wouldn’t - the Bambus do no fade-out ABL at all. The bed mesh is applied until the very last printed layer.
So if the bed is warped, even the very top of a 256mm printed cube is „warped“. Worse, if you have a warped bed, you can not even use the whole printsurface to the max height of 256mm.
Mine is low in the middle by 3.2mm left to right.
Acceptable front to rear so no shimming is going to help.
Are you sure you didn’t put the ruler on the lidar calibrate plastic party?
This I call a DoA and I wouldn’t accept any asking for log data or anything else.
Almost flat !! better than mine
Quick update: received my replacement bed. The second bed is cratered in the middle, and that’s visible to the naked eye. Support wants me to spend 3+ hours installing the new bed and send them data. Clearly I am not going to spend 3+ hours replacing a bed which is cratered to the naked eye with the already defective one in my printer.
What a mess. As of right now I only have 150x150 square in the middle which prints reliably. Anything bigger than that and hello errant layering issues
I ordered this tape reading comments between you and Rovster. Professional Grade Aluminum Foil Tape - 2 Inch by 210 Feet (70 Yards) 3.6 Mil - High Temperature - Perfect for HVAC, Sealing & Patching Hot & Cold Air Ducts, Metal Repair, More! https://a.co/d/0iXmznr
Seems like this is the way I will go as well.
Thank you for the inspiration! I am doing this as well.
I do want to know why you did the entire bed in aluminum tape as per your picture? Reading your comment above it seems like you only targeted low spots, starting with the bigger ones to smaller ones. But how did that end up with tape all over the bed?
Sorry if it seems like a stupid question 
Relatively simple… everything you see taped was a lower area 
A couple of things I’ve noticed regarding the warped beds
#1 it is primarily made of plastic
#2 three lead screws rather than 4
With those two things I’ve pointed out which we all already know it is of no surprise to me that many people are having issues with a warped bed
That being said when the bed screws underneath are tightened it will be extremely easy to either tighten too much, not enough, or unevenly
I would suggest using a straight edge and adjusting the thumb screws underneath that hold the plate to the Z frame
Being that the bed is primarily made of plastic it will easily go out of true if the fasteners are uneven which in turn should be able to be resolved by adjusting the tension on the thumb screws
I’m not a fan of the instructions of the Wiki page showing pictures of the bolt head flush with the thumb screw underneath and calling it good
As always this is just a suggestion which hopefully will alleviate the extreme dish in the center of the bed
If my bed was warped that’s what I would try first 
For what it’s worth. I measured my 4 - X1C beds. All were good. I’m including the bed serial numbers in case those having issues want to look for correlations of those having issues. Maybe there is a particular batch of beds that having the issue?
FYI. Printers were shipped from December through February. I’m in the US.
00500A291109660
00500A291107758
00500A291106901
00500A2B0808061
Well, as a person, that has tons of problems with Bambu printer, I can tell you my personal impression: their current stance is “we don’t care”
To stay on the subject, I have a concave bed on my X1Cc with ~0,5mm dip in the middle. Submitted logs and videos to support. Their response was:
"We have checked the bed leveling data and your heatbed measurement is in our accepted heatbed deviation.
The heatbed doesn’t need to be replaced, so if you have any issues with printing, please let me know and I will provide support to solve those problems."
Another good example of how they treat their customers:
Just scroll down until you see the response from BL team.
If they don’t care, you shouldn’t care either. File for chargeback with your credit card. The product is clearly defective. If you have proof then you have proof.
Force companies to provide better customer service. If they push back, this is what I will do.
Actually read the page instead of just looking at the pictures.
From a “mechanical” point of view it is impossible to bend the bed by unevenly tightening the three (!) bed screws. This would be different if the X1C had four levelling screws.
Some notes about bed leveling, attachment points, and why a flat bed matters. Many of you will know all this, but I thought it important for those haven’t really thought about it, or just flat out don’t know.
In basic geometry, three points define a plane. You can find any number of other points on a plane, but exactly three (x,y,z) points define a plane. You can move any of them in any direction and the defined plane will change but still be a plane. This isn’t the case with more than three points. If you move one, you likely no longer have a flat plane. So, for those wondering, three support points is absolutely the simplest method for supporting a build plate and ensures that leveling the plate cannot deform the plate. Of course this means it can’t correct for deformation either.
So why should I bother to level my bed? The printer has automatic bed leveling, correct? Yes, yes it does and but the evidence in this thread, it does an incredible job a correcting for deformed and unlevel beds. But its important to have a level bed anyway. To be more precise, it is important to have the bed exactly parallel to the plane the XY gantry moves in. If the bed and gantry movement planes are not exactly parallel, all angles involving the Z axis will be off. It’s a bit more complex to draw than I care to undertake this morning, but as an example, if your print bed is 1mm lower in the front than back, but otherwise flat, and you print a cube all the angle should be 90 degrees, however they will actually be 89.776 degrees in the YZ plane. For figures, single plate organic shapes, etc, this really won’t matter. However if you are printing anything that moves, fits together, or needs accurate measurements, this will cause significant problems.
A bed that is level, but not flat will have a similar problem except the amount of deviation from 90 degrees will vary across the bed and be quite complex, based on the shape of the bed surface. As noted above, without ABL fading, the top surface of the print will have the same curvature imposed on it. That is to say that a print, with a top surface parallel to the gantry plane, will have a curvature that exactly follows the curvature of the bed. Again, only an issues for functional prints, though as others have noted, a very significant curvature can cause adhesion issues at the outer edge of the bed where the the compensation mesh is the least accurate.
A side effect of a non flat bed is that you may not be able to properly level it either, at least not without significant effort. Take my bed, for example. It has a height variation of 0.75mm, left to right and 0.43mm front to back. The actual shape of my bed will be a very complex curve that would be almost impossible to measure manually. For sake of illustration, lets assume the outer edges are flat in the Y direction and the front any rear edges are bowed down. The lowest point is in the center of the bed and is 0.75mm lower as reference to the outer left/right edge. Because the Y axis varies by 0.43mm, that means the center back of the bed is 0.32mm lower that the front right and left corners. Those three points happen to be the mounting points for the bed. If I tram the bed, using those points as reference, the back of the bed is actually 0.32mm higher than it should be. This means that the left and right edges, which we said are straight, are no longer parallel to the gantry plane of motion. So now we not only have a warped bed, but we have bed that isn’t and can’t easily be leveled.
Again, none of this matters if you are printing things were dimensional accuracy isn’t a big deal. It’s not going to matter one bit for that cool crystal dragon print. However, that helmet you want to print, that you have to break up into half a dozen pieces because it’s just a hair to big to fit on the print bed, it will matter there.
I submitted my support ticket on Sunday, Feb 19th. In my initial submission, I included the log from a print, a video showing the bed moving smoothly up and down as the nozzle moved across the bed, and and the picture I posted above, showing a comparison of the warp between the X1C and my Prusa Mk3. Today, I received a response, asking me for my shipping information to send me a replacement bed.
Great explanation!
My bed is now ±.2ish mm off over the whole surface. Not too bad in my opinion.
I re-did the bed tramming yesterday and double checked the nose distance to the bed in all 4 corners of the bed. And unfortunately, I found out that the gantry is racked. Nozzle distance is in 3 corners the same, but I have +1mm in the back left corner. Means Y is in the front is either -1mm or in the back +1mm.
Don’t know if I can anyhow fix this - would be an easy task on a Voron or any similar built printer.
In the meantime I am certain that Bambu Lab is and was fully aware of these issues and this is the reason that they do not provide any bed mesh information to us users anywhere in the fancy UI‘s and I am certain this is why they fully encrypt all the log files. I am very disappointed.
Bambu Lab! Start a honest communication about QC issues and stop sending people through a fucking hassle support process when things are clear and obvious off! It starts to feel like a scam!!!
I was of the opinion that BL listens to the problems and needs of its customers and offers appropriate solutions. From what I have read in this thread - however, that seems to be more of an illusion.
The trouble seems to be related to the new release of the P1P. The problem has been more noticeable for 1-2 months. Perhaps a change of manufacturer?
The question is whether the success is too great or whether the production costs are higher than expected or wanted and now they are desperately trying to save money at the sacrifice of quality.
Yesterday I wanted to buy the BL Carbon X1. But since I don’t want to make figures, but rather mechanical parts, the deviations described are too big and not acceptable for me.
If BL comes around the corner with a solution in the coming days, I’ll reconsider
