I have a new A350T, the machine is brand new and my bed is already moving up and down a lot when changing the direction on the Y axis. Acceleration is already set down to 300 mm/s and Jerk to 3
Print speed on this video is 24 mm/s
That’s a huge issue in my opinion, i think i need to have a serious discussion with the Snapmaker Support but of course they haven’t responded to this ticket for a week now…
@Snapmaker-Support @Riskey @Jade This is an issue with MANY of the 2.0 machines, and is apparently not limited to certain versions of the linear rails. Seems to mostly affect 350-series models, but there are reports of the 250-series also experiencing this. Most users with the issue have resorted to user-modifications involving installing support rails, but it really does seem to be a pretty serious design flaw with the 2.0. Is there any chance of an official fix (more than just replacing rails, or tightening bearings), modification, or upgrade that Snapmaker as a company can get worked out for this? Certainly we will all love the quick attach heads and build plates, and the vibration compensation coming this year, but considering the volume of affected users and their discontent with the machine and company over this, I would think this should really be at the top of the priority list too.
So how do we gather the data on it?
Can we some how measure the play in the fresh machine vs “experienced”?
I am sure we could; though it requires precision measuring equipment and techniques for the data collected to be of any usable accuracy. Considering the machine is supposed to be accurate to within 0.04mm, I would say that any amount of bearing play or bed wobble in excess of 0.04mm is unacceptable. This would not include the flatness deviation of the bed when static; that is a combination of separate issues.
Several users have posted videos and photos in the forum and Facebook group of them holding a ruler up to the bed while it bounces, and it is often in excess of 1mm at the front and back edges.
With the power off, push the bed all the way to the back, then all the way to the front of the Snapmaker. See if the issue still exists. If it does, the Platform may not be properly aligned to the Linear Modules. To fix this:
Side Note: If you do completely remove the Platform, you can run a powered test of the Y-axis Linear Modules. Both Linear Modules should be able to resist any normal manual resistance that you apply with your hands. While using the manual controls to move the Y-axis, apply some manual resistance with your hands to see if either Linear Module stalls or skips. If either stalls or skips, it is bad, and needs to be replaced. DO NOT FORGET TO TURN THE POWER OFF BEFORE REASSEMBLY.
From what I have read, heard, and personally experienced, z-wobble in the bed has always been an effect of the y-axis linear module bearings being loose or worn. I can’t see how tramming the platform to the linear modules would fix that, but I certainly could be wrong. It’s worth a shot, even if only for the tramming aspect.
There aren’t any bearings in the Linear Modules that could possibly cause what your are describing, so what you’ve heard is false. The video clearly shows binding, which is what is causing the platform to move vertically. If the above instructions do not correct the issue, or if one of the Linear Modules is failing, the only option is to contact Support. Nobody in these forums aside from Snapmaker personnel, can help, and Snapmaker Support rarely visits these forums.
No Mxbrnr is exactly right (if not quite stated in prefect terminology) the issue pretty much effects all snapmaker A350’s and is a real fundamental problem with the way the linear modules are designed. The gantry carriages run along internal rounded shaft raceways on 4 U-form guide rollers which are tensioned outwards against the rail. The problem is that even if the guide wheels are pre-loaded correctly or even tighter to increase system rigidity, the small single bearing U-wheels are not designed to take that much axial loading. The wheels are rated to handle radial loads five times greater than when axially loaded. With the changing direction and acceleration of the big A350 bed it causes a larger moment load than the wheels can handle without adding premature wear to the system. This might be ok for something like the tool head but is greatly exaggerated because of the length of the beds lever arm. Ideally they should have used a recirculating rolling element for the guides or added another guide rod and two more wheels to the bottom of the carriage.
Both the Roll (Mr) and Yaw (My) of the machines bed motion is very well constrained but the Pitch (Mp) is lacking. You can tighten the preload on the bearing rollers to help reduce the issue but you risk decreasing the lifespan of the wheels, bearings, and rails. The only real fix for this inherent problem is to add external guide rails to the system so that the bed does not rock back and forth when changing directions. This most likely has nothing to do with with binding but its worth checking before you go down the rabbit hole of upgrades to fix a machine that should have been designed more thoughtfully in the first place.
@CNC-Maker, what @Elliot said. You may be right that in this case it is binding, and that sounds like a pretty easy and quick fix, so definitely the first place to start, but there is also more than one possible cause, and the bearings (“roller bearings”) inside the linear rails (yes, they do exist) are a very common culprit.
I can tell you that I personally had z-wobble in my bed, and actually had wobble in all of my linear modules, though the y-axis was the worst. The roller bearings in all 5 of them were worn; I replaced the bearings with new, and no more wobbles. There is even an official support article on how to tighten the linear module roller bearings, though it does not help if the bearings are actually worn, just loose.
I installed linear support rails for my Y-axis as well to prevent future issues, and will soon be installing additional mods to support the X and Z axis sliders too.
What you are describing, and what this thread is about, are 2 completely different things. I am solely referring to the topic of this thread and nothing else. As for your statement that this affects “all snapmaker A350’s”, that is a false statement. The issue only affects some owners, and depends upon a number of factors that are off topic for this thread.
FYI, I am one of the original Kickstarter owners and I’ve been contributing to these forums since. I know exactly what is inside the Linear Modules, which is why I am trying to stop the dissemination of false information. Opening a Linear Module should NEVER be done on a machine under warranty, unless someone at Snapmaker specifically tells you to. It is too easy to destroy some of the small parts inside of them. If Snapmaker tells you to open the Linear Module and you break something, they will replace it. They won’t do so, if you break something and they didn’t tell you to open it up.
Just kinda of off topics, will it be a cause of “nozzle hitting the infill?”
I have tried all my best on the z-hop / avoid printed parts, but it occasionally the nozzle bump into the infill and make some scratch sound and sometimes tear the figure out of the heat bed.
Yes, the z-wobble in the bed very well could be causing your nozzle to hit the print. You could try using a larger z-hop, but it would have to be more than the amount of bed wobble.
The thread is about bed wobble. If both binding and loose bearings can cause that, then they are both relevant to the thread, along with anything else which could even possibly relate to it. None of that is really anything to argue about in the public post though, so let’s move on and discuss the troubleshooting.
All that the OP video seems to show is a bed that wobbles when it changes direction. I would think that a binding would present with additional symptoms, such as a noise and/or a chattering movement along the whole length of the travel, not just at the start or stop. As you mentioned, removing the platform and re-attaching it with both linear modules in the same location is a great thing to do. In addition, to eliminate or verify other possibilities, when the platform is removed and the power is turned off, you can grab the individual slider/carriage on each linear module and see if it wobbles or twists in any directions; if it does, then it would indicate that the roller bearings are not providing the internal stability that they should be.
