Lines on side of print (but worse with S3D)

[mprenzler]

Hi All,

I recently bought a Malyan M180 (rebranded as Duinotech here in Australia) second-hand, but only a few weeks old. So far I have got reasonable prints out of it with S3D, but of course would like to improve them if possible

I have been using a simple 20x20x10mm block to make comparisons between the compiled G-Code provided by Malyan and S3D. So far I have not been able to reproduce the quality of the Malyan design using S3D and I am puzzled why. As far as I can see, I am printing under the same conditions (e.g. preheat, printing temperatures etc.), but I get a significantly worse side surface finish (i.e. "lines") with S3D. As you can see from the attached snapshots (using a USB microscope; S3D print on the left, Malyan print on the right), printing at 0.3mm layer height in S3D is significantly less even than with the Malyan file:

S3D 0.3mm (left) vs Malyan (right)

If I reduce the layer height to 0.1 in S3D, the print quality is better, but still produces more pronounced "lines" than in the Malyan file:

S3D 0.1mm (left) vs Malyan (right)

(It is especially noticeable when zoomed in):

S3D 0.1mm (left) vs Malyan (right) - zoomed

Following the troubleshooting guide I suspect I might have larger than acceptable fluctuations in extruder temperature, so intend to carry out a PID tune. What doesn't make sense to me, however, is why the difference between the S3D and Malyan prints? If there were a problem with the printer, then surely it would be noticeable in both groups of prints? It is especially puzzling that the Malyan print is finished in half the time of the S3D and still exhibits better quality.

Any suggestions on what I am doing wrong, or further tests I can make to troubleshoot would be appreciated.

Cheers,

Michael

[CompoundCarl]

Well there's a lot of things that can affect this. First of all, can you post a zoomed-out picture of the cubes as a whole? It's hard to tell some things when you get so close. Also, it may be useful to post 2 gcode files (one from each program), since that is also an easy way to compare things. Finally, I would also recommend posting your S3D factory file since that will allow people to recreate and test this easily

[mprenzler]

Hi CompoundCarl,

Thanks for taking an interest (and congratulations on the 1000th post!).

As suggested, here is the photo of the parts (from L to R: Malyan, 0.3mm S3D, 0.1mm S3D):

There's not a great deal between them so I'm probably just wasting people's time . Still, I'm the kind of guy who likes to understand things, and this has got me puzzled.

Measuring the prints with calipers, the overall dimensions are:
Malyan: 20.00 x 19.96 x 9.96mm.
0.3: 20.06 x 20.20 x 9.84mm.
0.1: 19.94 x 20.10 x 9.86mm.
So again, the dimensions of the S3D prints are a bit out, but not by much.

Unfortunately the Malyan part is not available as G-Code (I asked, but it's only available in compiled format), but I have attached the x3g file provided with the printer (zipped, because I wasn't allowed to upload the x3g). I have also attached the S3D project file and G-Code (also zipped).

Cheers,

Michael

P.S. One other thing that might be significant is that the Malyan part has been created with honeycomb infill, while the S3D parts have been done with cross-hatching (I just kept whatever default was used for a given setting).

[CompoundCarl]

Wow, those 3 cubes actually look really similar! The biggest difference I saw was from the start point locations. Try going to the Layer tab and changing the start points to use "choose start point closest to specific location". That way you will align all of the start points which will make it look better. You also probably need to adjust your retraction and coasting settings to help reduce the defects at the start/stop locations on your perimeters. They have a great guide talking about that here: https://www.simplify3d.com/support/prin ... s-and-zits

[mprenzler]

Thanks CompoundCarl, I'll have a look at those settings and print out another test piece(s) to see if things improve.
Yes, the pieces do look similar. If you think I'm being too picky feel free to let me know (I'm new to 3D printing, and what can be achieved with the current printer models)

However, it doesn't seem unreasonable to me to want to replicate in S3D the quality/performance I get from the manufacturer-supplied files. I would like to understand exactly what is causing the differences, rather than try and make improvements by trial and error.

Anyway, I'll have a look at your suggestions and let you know how I go.

Thanks again,

Michael

[Choumarin]

Hi,

What is your layer height in Maylan?
I would suggest that the issue is linked to the layer height not being a integer multiple of z steps.
Your pics looks a lot like in there: https://www.evernote.com/shard/s211/sh/ ... dcdd2d687e

[mprenzler]

Hi,

What is your layer height in Maylan?
I would suggest that the issue is linked to the layer height not being a integer multiple of z steps.
Your pics looks a lot like in there: https://www.evernote.com/shard/s211/sh/ ... dcdd2d687e

Thanks, that link you sent does look a lot like my "problem". I don't have G-Code from Maylan so I can't say exactly what settings they are using, but I will try measuring the print and the print bed to get an idea. Thanks for the lead!

M.