My very first 3D print result out of my Robox

[René @3dnerd]

Dear all,

after unboxing, and some calibrations (see thread: http://www.cel-robox.com/forums/topic/robox-ready-to-start-or-not/#post-8744) I started my first test print (see attached image).

I created the object using Inventor and exported it as STL file using the medium settings for tessellation. (See section A of attached image).
This file I used in Automaker, placed the object somewhere on the print bed, used ‘Draft’ settings and hit the Start button, the printer heated the bed, then the nozzle and started to print.
As you can see in the attached image the result is there. Not excellent quality but for 1st result I think it’s quite OK.
BTW, the Inventor object has a size of 15mm x 15mm x 10mm. The difference of the final result is 0.3 mm in each direction.

Now I have a few questions on my first result:

1.
See section D of the attached image. After the printer has heated the bed and the nozzle the print head moves to the front and cleaned the nozzle using the vertical rubber pad. It drops out a little amount of the filament. This filament fell down inside the bottom of the Robox.
Is this a problem - do I have to remove it? - How?
I think if I do hundreds of §D print jobs and every time a little filament fell down then a lot of filament will stay there.
I believe this is not so good - How do you deal with this?

2.
As you can see in the attached image section B and C, all walls bend inwards a little bit (compare to the red vertical line).
So the cube is bigger on the bottom compared to the top - how to correct this?

3. The bottom of the cube is not flat on the corners. All corners lift up a little bit. You can compare the red horizontal line of the attached image section B. Is there a way to improve this?

4.
See section C of the attached image. The printer created two thin inner walls with small gap. But the Inventor file shows only one ‘thick’ wall.
Why the printer created TWO walls instead of one filled wall? How get the result as the Inventor object with only one thick wall?

All in all I am very happy with my first result. But of course I have to learn a lot more, but my way with Robox has started…

 

Attachments:

You must be logged in to view attached files.

[Medusa @medusadelft]

Apparently, the purge before printing should end up in this hole. More people (including me) thinks it’s strange, but it should be easy to remove by removing the front plate (don’t know how) and turn the printer over so it will slide to the front. Please search the forums for the right topic(s).

The infill/two thin walls is a slicer problem if I remember correctly. Again, search the forums for a more detailed explanation. Some people have changed the current slicer version with 1.2 and get a bit better results. I’m not sure, but your problem might have been solved with slicer version 1.2.

About the first layers, maybe others can share their knowledge, but from what I have read in these forums, it might have to do with nozzle height for the first few layers.But other people can help you probably more (I don’t have my robox yet).

Hope this helps a bit.

I agree with you, for a first print it’s not bad!

• This reply was modified 1 month, 3 weeks ago by  Medusa. Reason: typos

[Aimdy @mistsoul]

For the purged parts which fall in the hole you can also collect them with the pliers that came with the robox without having to remove any of the mechanical parts. Hope this helps.

[René @3dnerd]

Hi,

thanks for reply, 1st I will try to remove the purged parts using the pliers.

Regarding the ‘two walls’ I found out a littlebit more. I installed ‘Repetier Host’ to see what the gcode is.
See RepetierHost_1.jpg, it shows how the outer wall should be printed.

- What I wonder is that the gcode shows 3 path for the outer wall, so using the 0.8 nozzle (from draft setting) it should completely fill the wall or not? (the wall in Inventor STL file has a thickness of 2mm).

- And what happened to the last layer? There is missing a part of the path (see RepetierHost_2.jpg).

- Regarding the version of slic3r.
If I go th the directory ‘C:\Program Files\Robox\Common\Slic3r’ and start the slic3r.exe it tells me version 1.1.5.
The current stable version of slic3r is 1.1.7. Should I update? Or does AM have problems with newer versions of slic3r?
(There is also an experimental 1.2 of slic3r).

 

Attachments:

You must be logged in to view attached files.

[Clicky @clicky]

Let me try to answer some of your questions. BTW I am not an expert - I started learning about 3d printing with robox:

1. So far it didn’t cause any issue. Just collect those pieces from time to time.

2. I think this is due to ABS shrinkage. Bottom is flat on heated bed as as such might not cool down as quickly as upper layers. So, upper layers shrink more than lower. Similar is with 3 - I think it is due to ABS properties and shrinking. I tried to print RPi case and all four corners ended up exactly as you explained. Similarly with my RPi camera case: http://www.cel-robox.com/forums/topic/abs-prints-and-what-i-learned/

PLA might be better in that respect.

4. is definitively Slic3r’s issue. BTW - you didn’t say did you check .gcode or post processed version - robox.gcode file.

Beside of that I used http://gcode.ws for gcode preview.

And for versions of Slic3r - I didn’t get chance to try it out but I think it is going to be quite seamless to switch from 1.1.3 to 1.2. Someone here did try to switch it from 1.1.3 to 1.1.7 and it worked. What I did is to invoke Slic3r directly in exactly the same way AutoMaker does and check the output - for 1.1.3, and 1.2. They didn’t differ at all!

Wall wise - my RPi case had T junction - walls 1.9mm thick and expected to see 6 passes of 0.3mm nozzle. L part of T shape was filled as expected and other branch had only 2 passes with gap in the middle! Version 1.2 of Slic3r handled it properly. And I always inspected robox.gcode file…

HTH

[René @3dnerd]

Hi Clicky,

thanks for reply. The gcode is from robox, as you can see in the title of the Repetier Host window

2:
If ABS shrinks on upper layers how to compensate this to get exact results? If this is a know problem there must be a solution also for other 3D printers - or not?
In the near future I want to create a housing and cover for a small electronic device. If the housing shrinks on upper layers the cover won’t fit anymore!
It would be OK if lower layers shrink same as upper layers, then I easily could increase the scaling to correct this.
But currently I don’t know how to fix this. May set the bed to a lower temperature that lower layers will also shrink more? But then we will get a problem with bed adhesion for 1st layer.
How other users deal with those things?

3:
For the bottom corner I thought that a ‘brim’ could solve the problem. But I cannot find any option for ‘brim’ in AutoMaker 1.00.15. And the manual (1.2) is for older versions of AM :-(.
Where has it gone?

4:
How do I update from Slic3r 1.1.5 to 1.1.7 (or 1.2)?
Do I have to completely replace the slic3r folder inside Robox->Common->Slic3r?

 

I tried to open http://gcode.ws, but I just got following (tried with Safari and Chrome):

403 Forbidden

• Code: AllAccessDisabled
• Message: All access to this object has been disabled
• RequestId: 6DE422D825823C7D
• HostId: lKHUXbKt0d8YYEXDBPertAmV8LudcmGgnNNMG3wqaVWXiLIk2FeLFTq3CELvSqZG

 

What I am missing?

Thanks for further help!

 

• This reply was modified 1 month, 3 weeks ago by  René.
• This reply was modified 1 month, 3 weeks ago by  René.

[Clicky @clicky]

2. I genuinely don’t know. I checked my box and it seem that it shrunk evenly at the bottom as well. But while printing it looked pretty much like line you drawn: wide at the base, slightly in in middle layers, and wide at the top few layers

3. It is there. In case of predefined settings you find it here:

and in case of custom settings here:

4. Yes. That would be the way. As I said - I haven’t got that far. I just had 1.1.7 and 1.2 downloaded separately, figured out how slic3r is called from AutoMaker and invoked 1.1.7’s and 1.2’s binary from command line and checked outputs. They didn’t differ so I concluded that you could easily swap it to the latest (providing that internal post-processing doesn’t get confused anyhow). If you give it a go - let us know the results! I would do it myself - but got distracted writing my own gcode preview…

I just tried the link you posted here and it works well for me. I normally use firefox, but I just tried it with Safari and Chrome - both work perfectly fine. Strange… Maybe it was down when you tried it.

 

• This reply was modified 1 month, 3 weeks ago by  Clicky.

Attachments:

You must be logged in to view attached files.

[René @3dnerd]

Hi Clicky,

thanks again for reply. Regarding the brim I understand now. In german it is not translated very well so I did not know what that was for.
Now I know - thanks!

I have a further question regarding the above mentioned ‘thick wall’ problem.
If we look at the gcode postprocessed by Automaker (see RepetierHost_1.jpg in one of my previous post in this thread) we can clearly see 3 path for the wall of the cube. Since I printed in draft mode it uses only the 0.8 mm nozzle. If Robox prints 3 paths at 0.8 mm for a wall thickness of 2 mm then all 3 paths should overlap and built a solid wall. This is what I expected.
But the print result has only two paths for the 2 mm wall and in between a gap of 0.4 mm.
So Robox did just two paths and not 3 as the gcode says. I thought Robox exactly follows the gcode -but it seems not???

[Clicky @clicky]

Could you attach that gcode here so I can see it myself, too?

I, kind of, doubt that Robox decided not to follow gcode properly. Maybe it is just our interpretation of gcode that is wrong…

 

[René @3dnerd]

Hi Clicky,

finally I could also put my gcode onto gcode.ws. (seems it was really down yesterday).

As you requested attached you will find the gcode files for the ‘cube’.

Thanks for further help!

• This reply was modified 1 month, 3 weeks ago by  René.

[René @3dnerd]

It seems I cannot upload gcode files due to security reasons?
How to upload gcode file to this forum?

I put them here:

581da3f9d228466c_robox.gcode

581da3f9d228466c.gcode

 

[pelgrim @pelgrim]

@3dnerd : zip them.

java (EE), JavaFX, HTML, GIS) programmer, database wizard, framework inventor, looking for a job ! http://roboxing.com/wizards

[Author]

Posts