Autodesk and Kickstarter 3D Printer Torture Test - Calibration

Nifty

Printer VIP
Administrator
Joined
Nov 3, 2004
Messages
3,046
Reaction score
1,405
Points
337
Location
Bay Area CA
Printer Model
CR-10, i560 ,MFC-7440N
This is very interesting!

Autodesk and Kickstarter teamed up to create this torture test for 3D printers:

Info: https://www.kickstarter.com/blog/toward-better-3d-printers-a-new-test-from-autodesk-and-kickstart

Github info & files: https://github.com/kickstarter/kickstarter-autodesk-3d/blob/master/FDM-protocol/README.md

I think this thing is brilliant! Seems to cover almost everything needed. I love that they have a somewhat objective scale from 1-5 on the various tests.

I got a bunch of cheap filament that was on sale ($8 a 1kg roll + free shipping), so I think I'm going to run this test on the filament I'm currently using and then this filament... using the exact same settings... and see what we get!
 

The Hat

Printer VIP
Platinum Printer Member
Joined
Jan 18, 2010
Messages
15,617
Reaction score
8,691
Points
453
Location
Residing in Wicklow Ireland
Printer Model
Canon/3D, CR-10, CR-10S, KP-3
Autodesk and Kickstarter teamed up to create this torture test for 3D printers:
My environment is my biggest test..
I swore I would never print the Benchy test, but I broke that rule and surprising got great results from it, a few days later I tried it and again and I ended up with a Feckin mess the second time, did I learn anything from it.. ? Don't do it again...

I found all of these torture tests meaningless for my own 3D printing situation, because I can get great results from them one day but the next day they’ve gone to hell, which took a while for me to figure out.

What I reckoned was, it all depended on ambient temperature in and around the printer itself that day, and more so to have constant humidity, the filament can bubble in the hotend nozzle causing an erratic filament flow.. Less than expect output..

Because my filament comes out of the dry box (Humidity 50% + or - 5%) it travels anywhere between 50 cm to 1.5 metres before it gets to the extruder motor and the protection of the Bowden tube.

I have tried enclosing the filament in long Bowden tubes, but that is not very practical because all the tubes would have to be different length to reach from each box to each printer. (From 8 different rolls) !

Now all of these factors can and will produce different results, mostly fairly good, but sometime pretty awful, and the last factor is the filament quality itself, the filament rolls can be perfect most times, but can give poor results towards the end of the roll.

Now giving all of the above factors, any test that I might carry out will vary wildly, when some of these factors change, the daytime temperature here doesn’t vary much, but the humidity moves like a Feckin Yoyo by 50% in either direction in a matter of an hour.

When it comes down to quality output I get pretty reasonable results most times but there are days when I just can’t print anything and I have to suspend all printing till the weather improves, varying moisture/humidity levels are my biggest enemy because I don’t heat the shed…
 

Redbrickman

Printer Master
Joined
Dec 27, 2010
Messages
1,101
Reaction score
1,183
Points
293
Location
UK
Printer Model
MB5150
I ran that test on the CR10 a long time ago and was surprised how good it came out, especially the overhang.

If you are interested in doing some calibration tests and playing with a nice Slicer then check out SuperSlicer a variation of Prusa Slicer and Slic3r.

It has built in tests amongst other things and is reckeoned to be better than Cura and S3D. I have installed it a few days ago and it is nice to work with but I'm still learning about it.

https://github.com/supermerill/SuperSlicer
 

Nifty

Printer VIP
Administrator
Joined
Nov 3, 2004
Messages
3,046
Reaction score
1,405
Points
337
Location
Bay Area CA
Printer Model
CR-10, i560 ,MFC-7440N
Mine came out pretty good! Some stuff was even better with the cheaper filament, which is interesting!

1603069088813.png
 

Nifty

Printer VIP
Administrator
Joined
Nov 3, 2004
Messages
3,046
Reaction score
1,405
Points
337
Location
Bay Area CA
Printer Model
CR-10, i560 ,MFC-7440N
One thing that is clearly something I need to address is the z axis irregularities. My towers / z-axis sides aren't very smooth and even. The issue is that there isn't any pattern or consistency... which makes me think it's a filament feeding issue.
 

The Hat

Printer VIP
Platinum Printer Member
Joined
Jan 18, 2010
Messages
15,617
Reaction score
8,691
Points
453
Location
Residing in Wicklow Ireland
Printer Model
Canon/3D, CR-10, CR-10S, KP-3
I Feckin hate doing these torture tests and mostly try to avoid them, but seeing as my two latest printers never done any I decided to give them a go, warts and all.

First the little KP3, I had a bit of a problem with tracking or a code shift, I reckon cause by the poor micro card quality, it probably needs a reformat, but I continued never the less.

Here you can see the shift, the way it moved out and then back in.
Untitled-1.jpg click to enlarge...

This is the underside, and how it fairs out without a bridging..
Untitled-2.jpg

And here the bridging on the sides was less than satisfactory.
Untitled-3.jpg

This is the finished where it scattered all of the vertical little pins on top and only one cylinder was able to come out.
Untitled-4.jpg

Moving onto the CR-10s, here is the underside where it handle the bridging a lot better on the unsupported angle and the inside of the base.
Untitled-8.jpg
Here is the finish and all of the vertical pins are all perfect and again I only got one cylinder to pop out.
Untitled-9.jpg Untitled-10.jpg

On the side the bridging was pretty good on all 5 pieces.
Untitled-7.jpg

But none of this reflects on how each machine prints overall, when printing any project the quality is more than adequate and the pieces usually do the job that they were intend for, if something needs a bit of tweaking, I just reprint the piece again..
 

Tigerman

Printer Master
Platinum Printer Member
Joined
Jan 6, 2011
Messages
232
Reaction score
130
Points
207
Location
Jordan
Printer Model
Canon MG5640, iX6840
What good 3D printer for CAD models ?
 
Top