I received my A350 in late January and fought for months, and spent hundreds of dollars, to get a level bed with virtually no help offered from support or personal correspondence with @Edwin. I just saw the attached photos on FB and to my amazement the bed mounting points are now machined flat and not just the rough casting I received. How do I get a “new and improved” bed frame replacement and a new print sheet from the issues with the original shipped firmware?
I’m in the same situation as you. I also have a crooked non machined carriage.
It’s impossible to print big things on it (which is the purpose of the A350).
I also tried many different things PEI, glass.
Best result I got was putting rubber rings on the screws under the heatbed and then calibrate, tighten or loosen the screws and recalibrate and so on. It took a whole day to get everything flat to get a decent print.
And if I wanted to do laserwork or cnc, I had to start it all over again.
Most of the time I’m trying to get the machine to work as it should instead of creating things.
I had a faulty end switch and a broken proximity sensor from the start (which are both ok now).
In short this is not the machine that was promised it to be.
Snapmaker should indeed get there act together and offer some real support.
I’m still waiting for a new printsheet that was promised months ago because the original was damaged
during the first calibration and the faulty proximity sensor.
Funny, I did silicone o-rings back in March I think but abandoned them for the same reasons. I tried many things to include, two different thicknesses of mirror glass (custom cut of course) and finally ended up with a custom cut Mic6 cast tooling plate with an EZ Stick sheet on it. It’s still not perfect because the frame is bent from the initial bed gouging but it’s closer.
I did get an email from Edwin and I have responded, so we shall see what happens next.
Did you ever find a solution for the frame leveling issue - the aluminum frame that SnapMaker refers to as “the platform”?
Yesterday I finally got my X-axis horizontal, that is, the distance between each Y-axis and the X-axis was the same - I have photos - where the measurement is identical on a 1/100mm dial
If only that was enough.
I started wondering if I should run an end mill across all attachment surfaces, until they’re all “identical”, but was going to hold of on such a move until I’d asked here.
I’ve been thinking the same, and I’ve on Facebook seen videos of people doing exactly that. I guess - aside from using shims - it is the only long-term solution. I yet did not get around to sourcing the necessary bits, but it’s on my list
Interesting thought, but I’m not sure how that would go with the 3D printing bed heating up.
My biggest - self imposed - obstacle is that there’s no “UNDO” if I get this wrong. I’ve not seen any way to purchase the platform as a spare part and I’m sure I’ll be voiding the warranty if I take an end mill to it
The other challenge is that I don’t know yet just how repeatable my leveling experience is.
In the end, I pushed both Z-axis to the top end-stop whilst powered off, then loosened the bolts holding one side and pushed that side “up” whilst tightening them. That was enough to get it level, since it really only needed 25/100mm to get both sides at the same height.
Took a bit of time to figure out, but got there in the end.
The platform can be sourced through Snapmaker support - I seem to remember that it was not even too expensive. Still, if you still have warranty, I’d pester support with “My bed is not level - what shall I do?” until they send you an new platform. Some people did this and succeeded. I first thought that my platform was OKish and only after end of warranty realized that I was wrong… Otherwise, I’d have tried to get a new one.
It’s a long shot, but perhaps shimming between the linear rail carriages and the platform will “distort” it just so, to make it “flat” on the top. Might need some finite element analysis
Meanwhile, I’ve asked SM support for price and availability of a replacement, since my unit is (just) out of warranty, even though it’s not actually been used - yet - but I’ve been having fun getting to this point!
It’s a bit overkill, and not perfect, but mostly works.
The added baseplate is milled flat so all inaccuracies related to the build plate are gone. And the other ones are easy to replace and are flat by design.
You lose some height, and it adds some weight, but you gain a lot of comfort. With the print bed it is not perfect as the mdf does start to deform after some time. Especially for higher Temps. It’s pretty stable if you don’t go over 50degrees.
Humidity might have an impact as well. I have been on the lookout for another material for the printbed.
It’s been a massive inspiration and I’ve been working on work holding solutions as a result.
As my first fixture, I’ve managed to come up with a combination of a vise and a chuck to center round stock in the same location and I’m now thinking in terms of work holding platforms, or fixture plates, that are attached to the platform in a repeatable way.
I have been contemplating your use of the spoil board as a base, but as you point out, MDF warps over time. I figured that if I have the platform actually flat, then my job of creating fixture plates becomes much simpler, regardless of which material I use.
I am considering using an aluminum plate between the platform and the fixture, so I can setup a grid with standard bolt holes, rather than requiring M4, but that’s based on not (yet) knowing how flexible the platform is, as in, how much does it flex. That’s all in the future, but discussion in that thread and with several machinists have lead me here.
I do owe you a debt of gratitude for voicing the issue and formulating a plan. It’s been marvelous, so thank you!
I don’t know if you saw this thread about it. I especially like the idea of @sdj544 of just screwing on a piece and then mill the fixture out of it. Easy to reuse afterwards as well and definitely aligned with your machine.
The bottom plate is basically still as flat as on day one. It’s just the one the heated bed is attached to that warps at the corners at high Temps. (and starts to stay that way now). For the other platforms it is still good. The machine is in my basement with a max temp deviation of 4 degrees summer to winter and ventilation that makes manages the relative humidity as well.
So I am on the lookout for material ideas for the printbed plate.
(and I did do the linear guide mod that helps with the weight and especially cnc jobs)
I am just glad it’s of some help and I am looking forward to see what you come up with
Thanks for the tip. based on the cheap blank pcb sheets I’ve been getting I’m a bit concerned about the flatness, but those are only 1.5mm thick. A bigger piece would probably be significantly better. Not really a big fan of milling it as it is essentially fiberglass. But definitely an option. I’ll see if I can get a reliable local source for it. (garolite is one of the latest hypes as an alternative printsurface btw)
An alternative I was thinking about was polypropylene.
So far MDF has worked good enough for me to not make it an urgent need. And quite frankly, it’s quite cheap compared to the other alternatives
Glad you said that. I was trying to figure out if that’s what brought it to mind but couldn’t find a link I’d thought I’d saved about that or if I’d mis-remembered.
Obviously McMaster Carr is not the cheapest supplier. Probably alternative sources.
Garolite is a brand of epoxy-fiberglass composite. Phenolic is a different chemical class of resin than epoxy, and as a rule higher quality. It may be that the vendor of Garolite is dyeing the product brown to make it look like phenolic. Tufnol sells a phenolic composite, for example. The Tufnol site, though, recommends polyester-glass composite for operation at elevated temperature.
Composites aren’t known for extreme flatness unless they’re cured in a mold. A single-faced mold (cured on a metal plate) might have one face flatter than the other, and sufficiently flat to make a good print surface. It’s certainly flat enough for woodworking tolerances; maybe not for FDM.
After receiving the price details from SnapMaker support, the platform for the A350 cost me USD $31 + USD $15 shipping when I ordered it today. I’m not sure which “version” I’ll get, presumably the latest, but I’ll know when I’m looking at it.