It’s common for technical debates to get a bit heated, but it’s all in good fun. I say stick to your guns and see whether there’s a way to prove it one way or the other. That way we all may learn something, perhaps something unexpected.
1 Like
Well said sir. This is true - Sometimes we get heated but its all one big 3DP family
Indeed, your print results do not show anything wrong. 
The question is why? What is the Bambu Firmware doing that the wobble is not having an impact on print quality?
In my opinion there are two options which could explain why the X1C is remaining a decent print quality, while other printers wouldn’t:
- capping the max acceleration after the calibration way below the capability of the X1C
and/or - applying a very aggressive input shaper method after the calibration
I mean this is not wrong doing so, especially for a printer which is used by a wide spectrum different experienced users. It would be very interesting to get the technical insights.
Tom Sanladerer set out to test the effects of some different kinds of damping and printer orientations, and in some cases he found that results were actually worse than with no damping at all:
That guy is a 3D printing god (lower case LOL, just a minor god). I butcher his name every time I try to say it 
It’s the Vibration Dampening feature. They do a major test when you go on the printer screen Temperature/Axis > Utilities > Calibration it does a set of calibrations that includes Vibration tests.
You’ll see it shake the carriage (hotend) and it will run up through a wide range of frequencies , some will make your dog complain. 
They then map the most violent resonance points of the printer and store it.
If you ever move your printer or change the table it’s on in a major way, re-run the tests. Not a bad idea to do it monthly just as ‘good practices’ to catch anything that may have changed.
You’ll also notice at the start of each print it does a minor version of this - it centers the hotend and does a couple seconds of shaking. They’re looking for any changes, like filament reels changed from almost empty to full in the AMS sitting on top or on the table.
It’s amazingly accurate and one of the big guns in BL’s holster.
The way it works is it will avoid doing carriage motions at the speeds that would vibrate the machine in the mapped vibration regions. Pretty cool really, as it has to account for both the X and Y motion and avoid movements that are a sum of their resonances. 
1 Like
I’am all aware of that - this is called resonance compensation/input shaping.
Still, it doesn’t answer the question what exactly they are doing.
We cannot know, since the firmware is closed source, but there are only a couple of reasonable approaches.
The first and most likely is that they measure resonance frequencies and what motions cause them, and simply avoid the motions, aka using slightly different speeds and acceleration values.
The second and much more complicated and more unlikely option would be to use the drive train to actively counter vibrations by moving in the exact opposite directly, similar to active noise cancellation on modern headphones. It would require a constant monitoring of the acceleration values, a predictive algorithm to best catch the vibrations before they even mount into any motion and a extremy fast acting drive system. I doubt it would be possible for a printer in this price range, but who knows.
LOL! It’s called magic, my friend! I too would like to know, but closed source and all. It is one of their advantages over competitors so I understand their silence.
I’m relatively certain this is the approach. Other methods would be too costly, as you said.
It is not any magic and nothing which BambuLab has founded… known in Klipper since years.
I’m out. 
It would be interesting to know what kind of support surface Bambulab uses for the development and testing of it’s X1C, because if you had the same identical support surface underneath your X1C at home, then you could be pretty confident you’d be well positioned to get the same results.
2 Likes
Now there’s a vibration nightmare. Kinda puts this whole complex discussion to rest, doesn’t it. Those are sturdy shelves with lots of mass.
Off topic.
In the second video. What is that white machine? It looks like a hydraulic press. Or just a fancy but big scale. 
1 Like
Looks like a old hardness tester for Rockwell/Brinell/Vickers
1 Like
Kind of. I agree with parts of your concept (and disagree with other parts). You’re spot on about some flexibility under the base (in your case the table) allows for the printer to move as a whole unit during printing. On a rigid base, the forces created by a fast tool head would stay within the printer, which causes vibrations in the printer and the frame, which can transfer into print deformation.
It’s somewhat similar to your earthquake building example. If the energy from the movement of the building has no where to go, it’s going to stress the building. If it can float it’s going to dissipate. I’d assume it’s the same for the printer as it moves back and forth. But since your table isn’t doing it as efficiently as a paver and foam would, your CoreXY is having to work overtime to keep the print looking good.
The second video that @ThanksForAsking posted talks about this at about the 11:30 mark. He specifically notes that a printer with a tool head at the top of the printer is going to have more leverage to cause much more vibrations than the Prusa would withe movement at the base (much like the top of a tall building would if it were moving back and forth at the top).
I’m not actually sure if your table is dampening vibrations, it’s probably causing some. And the size of it probably isn’t helping either. A large table moving back and forth takes much longer to return to its original state after those vibrations resonate through it. That’s why the foam is ideal, the vibrations resonate through it and dissipate rather quickly, which keeps them at a minimum. So the concept of movement under the printer is solid theory, as CNC Kitchen has pointed out. I just don’t think a moving table is the best way to go about it.
The vibration compensation in the X1C is so good, it’s able to keep your prints looking good with the table moving so much.
It certainly does about a rigid base causing harm to the printer. 
From a physics standpoint there are two major things you can do to lessen vibrations.
a) Add more mass to the non moving parts
b) Have some soft material that absorbs vibrations
c) Have a vacuum between the enclosure and the moving parts with no connection whatsoever
So, clearly the best option would be to simply invent anti gravity and have the printhead float in a vacuum.
More mass to the non moving parts will resist the vibrations better, but combined with some vibration dampening material it should give some decent results. The pave stones you often see are usually seperated by some vibration dampening materials and are mainly to prevent the whole rack from jumping around, as the added mass with resist the impact of the vibrations better. Adding some pave stones on top of the printer should also help quite a bit, but appears to be neither a thing of beauty or practicality.
When I received mine, I did some tests with the X1C on the concrete floor and heavy load on top vs X1C on the cardboard box it was shipped in. Recalibrated each time printer placement changed.
Zero difference in print quality.
1 Like
Shows the great input shaping of the printers!
You have excellent results using foam pads and slabs of concrete, and that’s fine, but there’s people who also have excellent results without.
I don’t use foam, rubber or concrete. All my printers sit on the same 18mm birch plywood table which is bolted into 2 walls (corner) which means it can’t move in either XY direction. My walls are my “slab of concrete” if you will, only much better. Zero vibration even with all printers running at full tilt.
There are industrial printers which can print with quality while sitting in the trunk of your car while you drive over a bumpy road, but they use closed loop motors and high precision ball screws instead of belts to ensure consistent relative motion (the only thing that matters is the motion of the nozzle relative to the bed), and they usually have a bulky steel frame.
Whatever floats your boat is the solution.
This is my first printer so I probably overdid it worrying about wobble… I laminated some plywood to make a stiff stand and then bolted (3) 10"x10"x3/4" steel plates (~21lbs/ea) directly under the top to counter shaking during high speed prints. It hardly moves even on the occasional ludicrous print.
4 Likes