Interesting.
Are you thinking the time frame for deformation is relatively quick? If float glass that starts out acceptably flat but transitions at some point to becoming unacceptably flat, are you thinking that’s more elastic deformation or ‘flow’ plastic deformation. I always assumed it was due to the elastic properties, but I haven’t looked into much.
Would flipping the glass like a mattress counteract any plastic deformation if that is indeed the case?
The state of glass, like the article you reference eludes to, is a debated topic. Whilst glass displays some liquid like attributes, such as their molecules having the ability to shift due to their lack of a stable crystalline structure, at ambient temperatures the movement of the molecules is so slow that there is no practical effect when you consider the timescale and application.
The hubble telescope’s lens has an optical function, therefore variations in flatness/shape on the order of millionths of an inch will have a practical effect on its function. By comparison, in 3D printing variations on the order of thousandths of an inch are where you might see a practical effect on the printing process.
Clearly from your replies to brent you are capable of reasoned discussion so I really dont know why your reply to me was filled with such animosity… its a forum, people will present opposing points of view…
Id be interested in what else was wrong with my original statement, im always happy to learn!
I’ll throw in that plastic is also an amorphous solid, so if glass is reclassified to liquid then plastics would be as well, however according to the linked article their argument is that it is neither a solid nor a liquid, but something else. I’d argue that the term “amorphous solid” encapsulates these properties and that this is the reason that we have such a classification but I digress.
Regardless, significant short-term deformation should not realistically occur unless the temperature is above the glass transition temperature, which is several hundred degrees Celsius for soda-lime glass (a very common type). An example of this occurring is leaving a PLA print in a hot car and gradually becoming a blob. It doesn’t ever truly melt, but softens in the heat and slowly flows. If flow does occur at a lower temperature, it is extremely subtle. As noted by @C.Harris the misconception of extremely old windows flowing is false, and instead is due to the flawed manufacturing processes of the time. Realistically, deviations visible to the naked eye have not occurred even on glass that was produced hundreds of years ago.
I’ll also add that the small amount of flow deformation that may occur also happens in metals, which are (typically) crystalline solids. Again, this is not significant to everyday purposes unless the material is exposed to elevated temperatures, often very high temperatures if no external stress is applied other than gravity.
If anyone would like to continue this discussion further I would be happy to make a post in the off-topic section to debate and talk whilst also avoiding cluttering this thread further.
I’m a total noob and I’ve had no issues. I love it! Prints beautiful 3d stuff. Adhesion is perfect!!!
I am glad that you have good experiences with Snapmaker 2.0.
We have been collecting users’ feedback from different channels and been devoted to make this machine more stable.
Cheers
Edwin
Yes sir!! I’ve had no issues at all. I jumped right in with Snappy (A350), knowing absolutely nothing about 3d printers or printing. The packaging was excellent, the instructions were perfect and easy to follow, and the craftsmanship of this machine is impeccable. It prints beautifully and my friends think I’m an expert (this machine is so wonderful it makes me look good). Great job to the entire Snapmaker team. You all created an amazing product. I could not be happier with my Snappy!!
(I realize you didn’t ask for all that, I just can’t say enough good things about it!)
Thanks and have a great night!
Tracey Capehart
My SM has been gathering dust for a couple weeks. Part of that is just Spring keeping me busy outside. I am also a bit disappointed with the performance of the SM, or possibly with 3D printing in general - this is my first (and likely last) 3D printer.
Prints have visible layer lines and occasional blobbing, some prints split along layers (though that may be the specific filament), and all prints are out-of-spec in regards to hole diameters and such.
I know now that I have to spend a good bit of time calibrating the SM to get results that I consider satisfactory, and that is one of the reasons why it is gathering dust. I just have too many other things to do that are of higher priority than tinkering with a plastics-extruder. Originally, I bought the SM to have a 3D printer that didn’t require much tinkering, and because of the CAN bus. What I ended up with is basically the CAN bus, and even that uses nonstandard cable connectors.
@Edwin its good to hear that you take feedback serious and for me the 3D printing is not too bad but not out of the box good! You have to improve the bad frame its too unstable and and seems like the production of it is it too bc i heard so much differences between the level issues of people
I bought it for the sole purpose of its multi functionality. I am more practiced with CNC and had already bought Vectric Aspire as i have 2 other CNC machines, a Shapeoko and an X-Carve. However I wanted to have one in my office for little projects and after fiddling around with the post processor and some tweaking of it I’ve got Aspire tuned to the 2.0 A350 really really well. I haven’t tried laser yet, but am currently learning to do my own 3d models and am printing the second half of my first design of my own from scratch.
WITH ALL THAT SAID – I find that the 3D printing function is actually quite user friendly as long as you read, learn, and fully understand the basics of 3D printing. I prefer the screw driven design over belt driven, I’m more of an accuracy over speed type of person. Screw designs are by their very nature slower than belt designs but deliver far better prints. Belt designs have a lot of play, kind of like how your belt on the front of your vehicle’s engine obtains a little play and wear over time.
Be patient, and keep learning. I have an Ultimaker S5 Pro which is supposed to be one of the best on the market and I’ve had the Snapmaker print just as high quality prints as my ultimaker with very thorough tuning.
Learn the machine diligently, each machine is different and has its quirks but become familiar with YOUR machine and learn its quirks and you will be able to get it tuned to they way you want. Also what filament are you using? Has it absorbed moisture and become wet? Print quality most of the time isn’t necessarily the machine, especially if you have properly leveled the bed and done the E-Step calibration. Both of those are crucial for any 3D printer. Even my Ultimaker, which is quite high priced, had to do e-step calibration. I’ve yet to hear of a printer that you just set up straight out of the box and get the best results, it takes tuning and patience.
Lastly, with it being a 3-IN-1 machine, it’s not designed to be the best at everything and does have an annoying inconvenience of bed warping when heated but thats physics, a much thicker and more expensive build plate would be needed to remedy this which would drive the cost up more, but if you tune it right it will be fine. “Skilled at many, master of none” comes to mind… and for those other machines that cost 500-700 like you say, sure they COULD print better (the opposite is also true, it goes both ways) but will also require tuning and calibration and are only 3D printers. Other 3-IN-1 machines (there aren’t many) that I’ve seen aren’t as well suited as the snapmaker. 3-IN-1 is kind of a niche at the moment and not very many companies make them. The Snapmaker is an all metal frame which is what caught my attention and others are not. I’ve also seen some rip off designs of the Snapmaker at lower prices but I don’t know what their quality is like and am not willing to find out.
I’m of the same mind. I’ve never actually successfully printed anything on it. I’ve made suggested heating or adherence adjustments with no luck. 7 item sample size but now it just sits there. I really wanted the CNC and laser pieces but given the changeover work, i probably should have done a printer and a separate device.
I think it prints fantastically and this is my 3rd printer.
At first I was annoyed that Luban didn’t have a “horizontal expansion” setting, but now I’m pretty sure that the software does it automatically.
Most printers start the center of the outside layer CENTERED on the line of the outside of the model. 0.4mm nozzle gives ~0.2mm expansion. I have been designing and printing bolts and nuts and things that snuggly fit together with no issues. Honestly my only issue is when I forget one part was printed on the SM, and another was on my X-MAX, and the X-MAX part won’t fit because I haven’t perfectly dialed it in yet.
Great printer. Laser is cool but weak, and I have yet to try the CNC, but I’m also not really interested until I get a rotary module!
(I hope I can get an answer soon about my damaged heated bed wire. wink wink)
I keep remembering this thread when fighting with the many-times-disassembled-and-calibrated print bed and print head.
There’s a bit of the luck-of-the-draw happening here. Some people get well-performing machines, some get poorly-performing ones. Could be design, could be manufacturing tolerances aren’t met, could be improper or lack of QC (the most likely explanation, as these are shipped unassembled and it is unlikely the components were ever assembled together and tested before shipping).
There’s also a bit of expectation management. Consumer-grade 3D printing in general, and the Snapmaker in particular, should still be limited to enthusiasts. That is, if it doesn’t bother you to spend hours calibration and testing and re-calibrating and optimizing, then you’ll really get into this. If you see it as a means to an end, and just want the thing to crank out prints, then you’re likely going to be dissatisfied. I don’t think the presentation of the Snapmaker does it any favors in this regard: while the design is good and it looks nice, the design gets in the way of the constant calibration and maintenance, to the point where an ugly print head like on the Lulzbot starts to become attractive.
I feel this is true of 3D printing in general. Sure some companies out there have started to be able to ship fully calibrated and tested machines, but something, sometime, eventually, WILL go wrong, and you’ll be in that exact boat of trying to figure out what went wrong, how to fix it, how to prevent it, and how to recalibrate your fixed machine.
Most people who show interest in 3D printing, I tell not to bother, because it’s not as simple as it seems. If you don’t have hours of time to invest in reading and learning, there’s no point in buying a giant paperweight.
It is very frusterating.
Why I am having problems with overhangs and stringing at this point on basic PLA is beyond me. I’ve done so many things to try. I run test towers, frequently they look identical from top to bottom. Retraction, speed, zhop, temperature, cooling… everything. It is just so difficult to get things to run consistently.
You wouldn’t think that every single spool would act so differently. There is no one size fits all on anything, even a different color from the same manufacturer and same type prints differently. I spend hours and hours just adjusting things trying to get it perfect, and it just doesn’t happen most of the time.
Occasionally, I get things tuned in so perfect that I get so happy with the results, but its so rare. There’s always SOMETHING to fight with.
I think that this has more to do with 3d printing as a whole than the snapmaker itself though.
With that said, I’m starting to get curious about SLA printing now. Thats a whole new adventure, which I think would be nice for me to have both worlds.
@Orbnotacus i have one of them machines that are fully automated. Ultimaker S5 Pro. Sure… it’s convenient, but it does NOT justify its price. Especially with the small build volume it has. You’re paying out the wazoo for parts and it’s filament is proprietary.
I think I repoerted this in another thread, but I am reliably printing now, using the supposedly-dodgy SM filament even. Been printing calibration routines for two days now. The fix? Tearing the thing down and rebuilding it. Well, not completely: I re-seated the Y axis modules, I popped off the end caps of the Y-axis modules and re-seated them, and I did a teardown + rebuild of the print head. The machine just needed some hand-on luvin’ I guess, because I found no problems and now it works great.
So, in regards to the filament, I’ve been making a note of what settings are filament-specific, and creating either Gcode or STL tests to be run on every new filament. The e-steps one is paramount; linear-advance k-factor (minimal-fish test- that Marlin test suuuucks), a temperature tower, a retraction tower, and probably a single first-level square just to test the squish factor. The two towers probably should be considered optional, i.e. for debugging; the goal would be to get sane settings for a filament roll in like an hour or so.
With ya there on the SLA. Trying to get someone to buy me one 
What I think I might do is take a single layer from the retraction tower and the temperature tower, and use those as test prints. The idea here would be a quick sanity check: is the temperature suitable? do retraction settings need tuning? I think that combined with the e-steps + linear advance would go a long way towards verifying a new filament.
I am going to make macros in Luban for applying the settings that were determined by this initial, drawn-out calibration of the machine. I have had success in the past changing the e-steps and k-factor by small amounts on-the-fly, so the ability to quickly reset them would make this easier.
I made a little torture test for this purpose to expedite things.
26 minute run time the way i slice it.
shes got some bridging, overhang, dimensional accuracy, holes, pegs and retraction checks.
measurement checks are in freedom units. 1/8" .25" .5" etc.
3d test v4.zip (17.7 KB)
Gives me a quick gauge to where I’m at.
That’s a great little test! The angled bridges/overhangs in particular are showing something I need to address, which didn’t appear in the (filament cube or the Make 2017 test files. Looks related to the curling I was getting on the Lattice Cube Torture Test.
I should be done with basic test prints today, and then will swap filaments to nail down a process for that. If things continue to go well I may be able to print actual parts tomorrow. If I can remember what they are, months and months later… 
OK, I finally changed to a new filament and dialed it in last night. This is the process I used:
Purge old filament.
E-Steps calibration. The only thing to note here is one of technique: I have a bowden tube nozzle on the print head, and I use the nozzle as a register for the notch on the caliper’s depth-gauge. I fix the calipers at 120mm and trim the filament so it is flush with the edge of the caliper (i.e. the exact length of the depth-gauge stem).
Create Cura profile for the new filament.
Print center bed-level square. This is a Cura project which has a 25mm cube (via Add Calibration Part extension) resized to 1mm high in the center of the build plate. This is a very fast print and the purpose is to ensure that first-layer adhesion does not screw up the subsequent prints.
Live e-step and k-factor tuning. I use a resized (25mm) Calibration Minimal Fish for this. The printing is closely observed, with the values of e-steps and k-factor changed by small amounts (0.10 for e-steps, 0.01 for k-factor). For example, “M92 E236.40” to set e-steps to 236.40, “M900 K0.05” to change K-factor to 0.05. Repeat until satisfied.
Print the MooseJuice test, again closely observed and making any tweaks to speed, cooling, and retraction that are ncessary. Repeat until satisfied.
Print the Filament Test Cube as a final sanity check.
My notes also in included a one-level tmperature tower to test bridging, but the above process proved fairly quick (well, like an hour, but that can be cut down with practice) and seems to be pretty complete.