Can't get decent prints

Snapmaker 2.0
1

Hello,

First of all, let me introduce myself. I’m not new to 3D Printing but this is my first Direct Drive printer and let me tell you this is giving me headaches to calibrate properly. For those tests, I used SM A250 with the official enclosure. The enclosure fan is turned on at 100% and I’m using SM fillament. I used cura and my test profile is provided at the end of the post.

All images here since a new member one could put one image: Teste - Album on Imgur

So, at first glance, I leveled the bed and checked it. Per the following image, it seems I have a nice first layer:

Couldn’t do PID Autotune cause it simply “crashes”.

Then, I decided to create a baseline cube for further comparison (yes, its not good):

At next step, I decided to check the extruder e-steps. I had a value E235.79 initially, and after extruding 100mm from 120mm, it extruded 11mm more, so I decided to change my values to E216.32.

Next step was checking the flow rate. Made a vase cube and this is the outcome:

The flow is almost perfect, the walls are smooth as silk. The average ratio was 0.4175mm per wall, so I changed my flow to 95.81. So far so good!

Next step was to check some retraction settings. After print this test, wich is made from segments of 0.20 from 0.40 to 1.40 as retration, I almost saw no change, so I left it with the default value of 1 and was too far from being so smoth as vase cube .

Finally, made a temp tower. I noticed the best temp is 200º, but it seems I’m under extruding and not even smooth as the vase cube.

As the last test, I’ve made a cali cat. As you can see, it’s bad. Very bad and not smooth.

Cura Profile: https://drive.google.com/file/d/1CWEXi8t0NrKCYSix20a4WdJfGrONbLKl/view?usp=sharing

So, I’m out of ideas in what can I do more. Would love your help to try to solve those issues.
Hope that I made enough tests to get your help.

Thanks.

2

Hi, and welcome!

You can get more detail on my experience here: Snapmaker 2.0 Initial Calibration Settings / Experiences

Others have their own posts.

Don’t forget about linear advance, that can explain some of your issues.

Unless there’s a new extruder gear design E216 sounds way too low, in my opinion. I don’t disagree with your measurement of 11mm extra, just saying that sounds weird to me.

If you changed from 100->95.8 flow because of the vase wall, and that causes underextrusion elsewhere, then change it back. Wall thickness is less important than overall quality. 4% change is huge.

Double check your actual filament thickness. Make sure there’s no clogs, even partial, in the nozzle.

Your stringing tower seems reasonable. I wouldn’t worry about that until you get other things dialed in a bit closer. Retraction is less important on that than Z hop, which definitely should be enabled.

You have quite a lot of elephant foot, your Z is calibrated too close to the bed I’d guess, bump your Z offset up 0.05mm and try the first few layers again.

Your temps sound reasonable. I had the best luck with 205, but 200 is probably close enough if you’re happy with it. I think 205 was a bit shinier of a finish for me, and had a bit better layer adhesion. Make sure and pull apart/break a few tests to ensure the bonds are strong enough for what you need.

3

Hello,

Thanks for your feedback.

Will double (or triple) check my extruder calibration. Was measured twice but dont know. Let’s try again.

My calibration was based in the following process:
Auto bed level 5x5. In the last step, when I feel resistance in the card, I just set Z up 0.05mm, right as you said. Should I do another 0.05mm to 1mm? It seems too high from the point I get resistance.

Thanks.

4

0.05->0.1 would be reasonable. It’s possible your calibration is bad (or at least not optimal). @Tone has a check level macro for moving around the bed and verifying the calibration is good. Easy to do manually too with the G42 command. If you go to Z=0 at any given coordinate it should be the perfect tension. It’s likely not right now. Bumping Z up or down when you save the calibration is just a trick, what really matters is that at the center of the bed if you go to Z=0 it still has the perfect tension.

5

Well,

I think I still need your help.
Per my calculations, I set E-Steps to E231.37 after new tests.

I can’t do the manual bed leveling since I don’t have a laptop for it.
With Octoprint I’m not able to adjust the last “point” manual per excel instructions.

After that, I check acceleration test and almost all seemed equal for me. Stucked with 750 that seems to have a little less ghosting than 800/900/1000.

Finally, set Linear Advance to 0.06.

Despite I think I have better prints now, it seems yet too far from what I saw multiple people printing.

[Album] imgur.com

Any other suggestions?

6

The best tip i can give you is 3d printing is a skill. practice and patience. experimentation

there are plenty of general best practices, but every machines gona act a little different, every spool is gonna act a little different etc.

i think it looks pretty good to me to be honest

if you really want to see a perfect one, do a slicer flow calibration and then run it SLOW.

smaller nozzle helps too.

7

Compared with other people with same settings (slicer) this is like crap. I must have hardware problems somewhere (I already had one dead rail on arrivel).

I already did everything I could to do better prints :s

8

slicer settings are not exactly universal.

if you feel you have defects, contact support@snapmaker.com

they will probably ask you to run their 13 hour calibration cube and send pictures of the results for what they think

can you share a link to this model? id love to try it. currently doing an 8 hour print then ill do it.

also for funsies, what material are you using? id be interested to see what happens if i try to slice you some gcode from my simplify3d settings.

9

I already did everything I can remember. It seems impossible I can’t even get smooth walls. I even start doubting myself :pensive:
I know it’s not universal, but the differences are too big for being only slicer profile.

The file is Cali Cat

I’m using PLA. Settings with 0.20 layer, it takes about 35min with 15% infill. My cura profile is attached in first post if you want to check it.

If you think that print its good, please see the attached photo in that album that have flash enabled. You can see (and feel) every layer…

10

It is not atypical to be able to see and feel the layers.

Try to think of it as fun, not something that you are failing at.

It takes a long time to perfect things, and theres always a million variables. Overtime you will begin to learn what minor tweaks can be used in different circumstances.

Try this out for fun sake:

calicat.gcode (549.6 KB)

thats actually a pretty neat little calibration checker, lots of little things going on there to make things difficult.

11

Yes, it’s a nice calibration tool :slight_smile:

I’m trying right now the default luban profile. It will take much longer, but will check this one first.
After that, will check your gcode :slight_smile:

12

Just would like to make some questions since I couldnt find yet the problem:

1 - I notice that I need to make excessive force if I want to move any axes manualy. I need to secure the printer with other hand so it doesnt move due the force I need to apply to move it.

2 - I noticed my aluminium covers of Z axis sometimes “lift” while the printer moves. Its supposed to happen or I need to check them with the tool provided? If so, should I tight them “forcing” them or just tight them?

13

is this while machine is powered up?

it should take a little effort to move in any case, but that might be part of your problem. i dont think its a good idea to force it with the power on.

the dust covers may be loose a bit, which isnt the end of the world, but can cause extra noise and possibly the vibrations are affecting the quality i suppose.

did you try the gcode i gave?

also, despite whatever age or brand of filament you may have, it may be partially responsible. could be wet or just not good. even brand new spools.

it took me a long time to dial things in to satisfaction.

14

The axis move was done with everything turned off. I can’t move it with one hand without properly secure the printer with other hand so it don’t move.

I tested the gcode file with the same results. Forgot to take a picture.
Already tested with 2 filament from different brands (without counting SM one). So I guess it’s impossible it could be the filament. I even see “flaws/gaps” in the same spots, but that could be related to any hardware problem.

I only can remember something like this, any problem in the rails or something.
Just to remember when I purchased it, one rail was dead. Why should I think others might not have problems? How can I make sure rails have the correct tension between them? I don’t feel safe to open those rails and make something wrong.

If I just could see/feel how force is needed to move them and how to make move it properly, maybe this could be the cause of my problems.

15

id say your required force to move sounds about right.

16

did you try the teaching tech slicer flow test?

is your filament a decent brand that you are trying with?

17

Tried with SM, noulei which I had good results in other printers and Devil Design.

The flow I did the test and it gave me 4.175 in the walls. Per teaching tech calibration page Ive change the flow to 95.81. Got worst results so revert back.

The only thing I need to do next is disassemble the printer and assemble it again and check the rails. Maybe change them from Y to Z to check if there is any difference. Or maybe tight better those screws with the tool provided (if I manage to know how its done).

19

Thank you your comments. Im surely I have some problems somewhere. If I just could make anything to test them out would be great.

At this moment Im at work so I cant tell you my settings but my retraction is at 1mm. I made a stringing test from 0.4 to 1.4 (in 0.2 increments) and didnt saw any visible difference from 0.4 to 1 so kept this setting. Did everything with PLA (without enclosure, with enclosure, with enclosure and enclosure fan 100%).

Made my cura profile public to check.

About PID autotune, it crashes my SM, but didnt made bed autotune. Didnt noticed any variations that could affect or maden the prints unstick from bed.