Just wanted to show off my T-REX skull. This is not my first rodeo, I have several other 3D printers but the SnapMaker A350 is my largest of this writing (I’m upgrading my Ender 3 Pro to 400x400). I’m pretty impressed once I got this thing printing (the original nozzle was clogged and the second hotend didn’t work, but luckily I had other nozzles and I could go on how poorly designed the hotend system is but that’s for a different post.) I sliced using Simplfy3D with 25% infill and supports from the platform I’m pretty sure I could have gotten away without these but better to be safe…) I used Atomic Filament’s PETG Bright White (Atomic is the ONLY filament I use in any of my printers (Quality, ie made in the USA and it is awesome, over being cheap))… I don’t know what I missed, ~135 hrs total print time (95 for the main part and 45 for the jaw). oh yeah, I lost power 2 times during the print which you can see in two prominent layer shifts. They are my own fault due to using a smart outlet to monitor power and me not purging smart schedules from prior uses of the outlet. But I will say the SnapMaker did a really good job of picking right up… Oh I almost forgot, it was printed at 200% from the original https://www.thingiverse.com/thing:308335
You need to play around with your flow rates. The print looks nice, but could look a lot nicer. I am still waiting on my 350 so I don’t know the full reason why your print has a waviness to it. The most common reason is over extrusion. I have been building custom delta printers for about 8 years now and decided to venture over to the dark side and get into cartesian printers. This printer has a lot to offer for the price so why not! Start out by printing test cubes and adjust the flow rate for each cube. .4mm nozzle, .2mm layer height, 100mm/s print speed. 40% infill
His waviness may also be due to bed wobble in the Y-axis as it appears the issue is more prominent the higher up you go in the print of the top half of the skull.
I would print something else out and watch the bed closesly during moves where it is reversing direction frequently (i.e. printing top/bottom layers) to see if your bed is moving wobbling in Z. I currently have this problem as well. The fix might be to open the rail up and tighten the cam nuts on the carriage, or it could be due to bearing wear on the carriage as ive seen in other posts here on the forum but If you do see bed wobble I would suggest filming it and sending a ticket to support.
In the mean time, slowing print speeds will reduce print flaws related to bed wobble.
thank you @C.Harris that is exactly what I would contribute to the banding in my print, I was going to reply but I haven’t had the time to type it out. This isn’t my first rodeo in printing, I have 6 other printers including 2 that are really high quality at work (one can print 1mx1mx1m and costs $250k). 100 mm/s can work on delta printers but since this is a lead screw printer you really have to slow it down. I think I was around 45 mm/s and that is actually fast for this style of printer. I know you can run a delta really fast but you run into the fact a delta is usually not as big as this printer, at least not in the XY. you also don’t have the other two functions of the printer which I have done some cool stuff with the laser. I haven’t had the time nor the enclosure to run the CNC but I guess it will run pretty good withn reason. I did try and put my drag knfe (DNOEKl) and run it with Vectric Vcarve Pro but had some issue with the gcode and need to figgure out some things. Just like I have had issue’s running lightburn (they call for M8 or M9 cooling which sometimes SM just stops or it ignores) but the results with lightburn are heads and tails over SnapmakerJS. I’m not trying to say this print couldnt be better because I think it can (the nozzle is a huge problem which now I’m running a E3D nozzleX and think it will help a lot). Just be ready coming from a high qualty delta things are differnt on the cartesian side of the house. I’ve always been a skeptic of the cartesion and I see the downfalls, but the fact this machine can do much more it’s acceptable
With testing I found having the acceleration at 500 fixed the Y ringing, but 750 was acceptable in Y for me and had no X ringing.
The default of 1000 I determined was insufficient for what I print on this machine.
50mm/s nominal speed with the reduced acceleration has been an acceptable trade-off.
And yes, ha, my $250 delta can easily move at 250mm/s. That lightweight tool is a bit scary at the speed. I can’t imagine what a $250K printer is like.
Oh, and that T Rex is out of this world. Incredible.
It’s something we have been crying to S3D to add for a while writing scripts for each of my printers gets old. I will try something with reduced acceleration and see what happens… These are things that should already be in the firmware IMO…
Yea, but you know someone would complain about prints taking too long if it was always acceleration limited. I think 1000 is an acceptable trade off for many people. ¯\(ツ)/¯
I don’t blame all of my bandings on acceleration, maybe 750 might be appropriate. If people are worried about prints taking too long this might not be the printer for them 
…
But I will say it is more than acceptable for someone who is unexperienced to do some cool things…
These acceleration settings I assume are corrected from the command line? Or is there a slicer setting for these?
If command line what is the Gcode I should use to play around with these?
If you’re going to change this, be warned that your machine may start malfunctioning by skipping steps. When a “step” command arrives on the wire before the previous step motion has completed, the second step might not happen. The linear modules have no distance measuring device in them; they rely solely on counting steps correctly to know their position. After a missed step, everything after it will be shifted in the direction of the missed step.
And the situation is worse for the Y-axis because of it’s doubled module. When one of the sides misses a step and the other doesn’t, it rotates the table slightly and increase the friction on the lead nuts, meaning a future missed step is more likely.
I’m not advising you not to try it. I’m advising realism in expectation about the capabilities of the machine. When you push the limits, you’re more likely to get print defects.
These machines are jerk limited and run an S curve acceleration profile to minimize the chance of this, but yes, at some point it’s inevitable.
Technically it’s not “jerk”, but junction deviation, but it still serves to limit rapid changes in acceleration.
Both. It’s loaded from EEPROM at boot and can be changed by the slicer for advanced applications.
Check out here https://teachingtechyt.github.io/calibration.html#accel
As noted in there, high acceleration is hard on components, and with the mass of this machine if anything I think it should go down, not up.
Thanks for the info @brent113 and @eh9. I was planning on reducing acceleration rather than increasing so should be alright I would assume
Amazing. Would you mind sharing your Simplify3D settings? Not getting the quality that I expect.
Regards,
D…
will do, I’ve been traveling for work and will get it uploaded here as soon as I can.