Printing a cylinder - there are gap

Talk about anything.
moreupper
Posts: 30
Joined: Tue Jan 05, 2016 10:08 am

Printing a cylinder - there are gap

Post by moreupper » Tue Aug 09, 2016 12:41 pm

This cylinder is 1,5mm in wall thickness but as showing on the photo it is printed as two separate circles why no filling at all?
sw:windows, v2.18

Picture of my up2 settings and the cylinder.
Attachments
picx1.jpg
picx1.jpg (44.76 KiB) Viewed 17471 times
picx.jpg
picx.jpg (19.29 KiB) Viewed 17471 times

Owen Sparks aka Marksman
Posts: 610
Joined: Tue Mar 15, 2016 9:56 am

Re: Printing a cylinder - there are gap

Post by Owen Sparks aka Marksman » Tue Aug 09, 2016 12:54 pm

Hi.

I've seen the same thing. I ended up thickening the wall to 3mm and all was good. It would be interesting to see where the sweet spot is. Say a trial print with a range of thicknesses from 1.5 to 3mm in .25mm increments.

Owen S

moreupper
Posts: 30
Joined: Tue Jan 05, 2016 10:08 am

Re: Printing a cylinder - there are gap

Post by moreupper » Tue Aug 09, 2016 1:03 pm

If it can not print 1,5mm it is for nothing!

User avatar
3DWP
Posts: 194
Joined: Wed Dec 30, 2015 4:00 pm
Location: Netherlands
Contact:

Re: Printing a cylinder - there are gap

Post by 3DWP » Tue Aug 09, 2016 2:33 pm

This is very common (sadly) - a preview of the slice file is the solution to prevent this. Also in the ideal world the Up software could detect this when slicing and adjust extruded line thickness and/or infill so any wall would be filled :cry: Sadly this is not the case.

But there have been a lot of these topics on this forum, if you search for it I'm sure you can find threads that covered this. There are also models on Thingiverse to check infill for different wall thickneses and accuracy.

example: viewtopic.php?f=13&t=1164&p=12907&hilit=wall+gap#p12907

moreupper
Posts: 30
Joined: Tue Jan 05, 2016 10:08 am

Re: Printing a cylinder - there are gap

Post by moreupper » Tue Aug 09, 2016 3:55 pm

Hi

If printing z= 0,20 or 0.15 will it do the wall better?
I have always believed that Z is the fineness of printed and not how it printed a thin wall...?

Is there any better software (i think not because the handling of data is custom made) or what printer is better than the up2 in same price scope?

User avatar
Tiertime-Jason
Posts: 286
Joined: Mon May 05, 2014 4:31 am

Re: Printing a cylinder - there are gap

Post by Tiertime-Jason » Wed Aug 10, 2016 12:55 am

Because the single wall thickness is a about 0.6mm, if the gap is filled you will end up with something thicker than 1.5mm and ugly surface.
The algorithm put dimensional accuracy at a higher priority so the gap is not filled. You can try print at 0.15mm layer thickness, see what happens.

This is the limitation of extrusion based printing, that thickness of wall is not a total freedom.

Elenc
Posts: 8
Joined: Sat Aug 06, 2016 8:46 am

Re: Printing a cylinder - there are gap

Post by Elenc » Sun Aug 14, 2016 10:28 am

Over the last 2 days I have been trying to print a cylinder with wall thickness of 6 and 4mm and all I get is a bunch of wafer thin walls with no joining fill.
I'm using Mac OS X El Capitan New Build and Up to date with UPStudio V1.0.26. The result, with no wall strength, is unusable. This is with 0.1mm Layer thickness, densest infill, fine quality and pre-heat (had to use the switches on the up Box to start pre-heat as it does not work through software).

As you can see from the images the part just fell apart with no effort including the base.



The first attempt (IMG_1991.jpg) using 0.15mm Layer thickness, densest infill, normal quality and pre-heat started to produce a solid object but it had errors on two sides where it clipped off the disk and put in to very thin walls. This model was check to have no errors in Rhinoceros V5.0, Simplify3d and UPStudio before printing.
Attachments
IMG_2001.jpg
IMG_2001.jpg (60.75 KiB) Viewed 17374 times
IMG_1999.jpg
IMG_1999.jpg (76.12 KiB) Viewed 17374 times
IMG_1998.jpg
IMG_1998.jpg (70.04 KiB) Viewed 17374 times
IMG_1997.jpg
IMG_1997.jpg (76.67 KiB) Viewed 17374 times
IMG_1996.jpg
IMG_1996.jpg (112.49 KiB) Viewed 17374 times
IMG_1995.jpg
IMG_1995.jpg (58.76 KiB) Viewed 17374 times
IMG_1991.jpg
IMG_1991.jpg (58.88 KiB) Viewed 17374 times

Owen Sparks aka Marksman
Posts: 610
Joined: Tue Mar 15, 2016 9:56 am

Re: Printing a cylinder - there are gap

Post by Owen Sparks aka Marksman » Mon Aug 15, 2016 5:12 am

Hi.
Have you tried V2.17 of the software?

I've used that to print approximately 50 cylinders of 3mm wall thickness without problems on an upbox. (.2mm and medium infill).

2.18 crashed if trying to print cylinders but 2.17 seems good.

If that fails I could try to print your stl if you like.

Cheers Owen S.

User avatar
wackojacko
Posts: 667
Joined: Fri Nov 12, 2010 5:46 am
Contact:

Re: Printing a cylinder - there are gap

Post by wackojacko » Mon Aug 15, 2016 7:10 am

Also put a small radius on the top, this will seal it off.
Bruce
http://www.3DPrintingSystems.com

Need help with your printer, check out our Youtube channel.
https://www.youtube.com/user/3dprinting ... /playlists

arhi
Posts: 256
Joined: Sun Mar 08, 2015 10:51 pm

Re: Printing a cylinder - there are gap

Post by arhi » Tue Aug 16, 2016 9:37 am

Tiertime-Jason wrote:This is the limitation of extrusion based printing, that thickness of wall is not a total freedom.
actually - NO!

this is the limitation of teartime slicer, not the extrusion based printing!

the TT slicer has fixed relation between layer height, nozzle size and wall thickness, but this relation DO NOT NEED TO BE FIXED. I'm able to print, on my own printer (my own gcode interpretter, and ooooold skeinforge as slicer) any ratio between the three, I can easily print 0.1mm wall with 0.35mm nozzle and 0.2mm layer just as well as 0.2mm wall on 0.2mm layer with same 0.35mm nozzle, or 0.3mm wall with 0.8mm nozzle on 0.4mm layer etc ..

the limits are that you can't properly print layer that's higher then 80% of your nozzle dia but that's about it, you can stretch ABS without any issues, you just need to get acceleration way up and top speed way down and add some pauses on corners if you want to have sharp corners without pulling ..

User avatar
Tiertime-Jason
Posts: 286
Joined: Mon May 05, 2014 4:31 am

Re: Printing a cylinder - there are gap

Post by Tiertime-Jason » Sat Aug 20, 2016 9:17 am

Elenc wrote:Over the last 2 days I have been trying to print a cylinder with wall thickness of 6 and 4mm and all I get is a bunch of wafer thin walls with no joining fill.
I'm using Mac OS X El Capitan New Build and Up to date with UPStudio V1.0.26. The result, with no wall strength, is unusable. This is with 0.1mm Layer thickness, densest infill, fine quality and pre-heat (had to use the switches on the up Box to start pre-heat as it does not work through software).


should not be like that, something is wrong, looks like under extrusion.

Elenc
Posts: 8
Joined: Sat Aug 06, 2016 8:46 am

Re: Printing a cylinder - there are gap

Post by Elenc » Sun Aug 21, 2016 9:26 am

I have been asked by support to try Windows 10 with Up 2.18 so I have purchased a windows 10 license ($299 Aus) and VMware Fusion pro ($180 Aus) as I no longer have a windows compatible PC and it's not much better. If I print at 0.15mm with normal speed and average infill I can print toys. But if I try to do dens infills, solids and fine resolution with or without 0.1mm Z I get very rough prints and 99% of the time it stops printing within 10 to 50 layers. I have used both UP filaments and third party filaments with the same results. If it's a under extrusion problem it's within the slicing software and not the up box.

User avatar
Tiertime-Jason
Posts: 286
Joined: Mon May 05, 2014 4:31 am

Re: Printing a cylinder - there are gap

Post by Tiertime-Jason » Sun Aug 21, 2016 9:34 am

Hi elenc,

Would you share your model, so that we can try it on our machines.
I personally made a similar cylinder model and did not have any problem printing it.

Elenc
Posts: 8
Joined: Sat Aug 06, 2016 8:46 am

Re: Printing a cylinder - there are gap

Post by Elenc » Sun Aug 21, 2016 11:04 pm

The STL file is 10.5Mb in size. I have your email can I email it to you.

User avatar
Tiertime-Jason
Posts: 286
Joined: Mon May 05, 2014 4:31 am

Re: Printing a cylinder - there are gap

Post by Tiertime-Jason » Mon Aug 22, 2016 1:52 am

Elenc wrote:The STL file is 10.5Mb in size. I have your email can I email it to you.
my email: jasonwu@tiertime.net

Magic
Posts: 6
Joined: Fri Aug 26, 2016 4:34 pm

Re: Printing a cylinder - there are gap

Post by Magic » Fri Aug 26, 2016 5:57 pm

Just 'thinking out loud'...

It had been suggested to me that the best print results are achieved from the highest resolution STL.

Putting that into context I use 'Solid Works' a professional industrial 3D CAD system. My STL files are generated from Solid Works native solids, hence they are converted to STL. To this end I'm able the set the conversion resolution (for want of a better word). Given 'a' part model with conversion set to 'fine' that will produce an STL of lets say a few 100k. If I set the conversion resolution to maximum it will produce an STL file in the order of Gb's for the same part.

My point is this, experience has proved that the highest conversion resolution typically does NOT produce the best print result, quite the opposite as the printer 'seemingly' tries to produce the STL exactly I've ended up with voids and 'extra' feature that should not be there!

Of course a 'printed copy' of the STL is what we might expect, hence I would guess it 'could be' the problem is the STL data itself, perhaps the void between the inner and outer of the cylinder does exists.

I'm on holiday for the next two weeks so I cannot test my theory just now ;)

arhi
Posts: 256
Joined: Sun Mar 08, 2015 10:51 pm

Re: Printing a cylinder - there are gap

Post by arhi » Sat Aug 27, 2016 2:42 am

DaveARM wrote: My point is this, experience has proved that the highest conversion resolution typically does NOT produce the best print result, quite the opposite as the printer 'seemingly' tries to produce the STL exactly I've ended up with voids and 'extra' feature that should not be there!
it depends a lot on the slicer. for e.g. the "graphical" slicers like skeinforge do not have issue with stl files with "too high resolution" as they convert every slice of the object into image and then convert that image into g-code. the image is create in resolution closed to resolution of the printer so the "overly high" resolution of the STL will only slow down the slicer but will not affect badly the output. this type of slicers are lot slower then the fully vector based ones but allow for easy manipulation ..
the "faster" - fully 3d / vector based slicers on the other hand can create "too complex toolpath" if they lack all the necesary optimizations (to optimize toolpath). most g-code interpretters will suffocate with this overly complex toolpaths and will produce ugly parts with overextrusion, filament jamming, part nocking .. on the other hand, some high end interpretters will optimize this nasty toolpaths and again produce great print ...

I stopped investigating and comparing slicers long time ago so have no clue where we are now in the low end market with slicers (do they know how to optimize path based on printer specification) but the initial view of the few modern ones show they did not advance as much as would be expected in last decade .. the intepretters are also crappy, the atmel based have no umpf to handle this anyhow, and the other ones are mostly based on simple milling algo's so again probably not that smart neither :( .. and since you really don't want to wait skeinforge to finish (100-1000x longer then a regular vector slicer) it's still better to create STL in resolution that makes sense for your machine then to expect slicers to handle them properly :(

Magic
Posts: 6
Joined: Fri Aug 26, 2016 4:34 pm

Re: Printing a cylinder - there are gap

Post by Magic » Sat Aug 27, 2016 11:48 am

@ ARHI

Interesting and educational read there Arhi ;)

Although I've been using an Up2plus for many years it still remains that my general experience of this printing method is limited to the Up2 and the GUI that comes with it. We've just (a few days ago) bought an UP Box but unfortunately I had to return it to the supplier as the extruder fails after its been printing for a few hours... but that's another story.

I've always had a good level of success with the UP2, 99% of what I use it for is a prelude to injection moulded parts, suggesting to me that the 'design rules' for plastic injection seem to suit 3D printing. I just need to compensate for the fact the the 3D print cannot produce the same strength and intimate fine detail of injection moulding.

I haven't as yet had call to print a thin wall cylinder as such, however I've not had any particular issues printing thin wall pillars (for screw fixing PCB's etc.) and thin webbing. Until recently I've only ever used STL files generated by Solid Works in its default setting. While increasing the resolution may make rounds more round and drafts smoother, winding it up too far produces unwanted results. Hence it made me wonder what other issue may be induced by the quality of the STL.

User avatar
Tiertime-Jason
Posts: 286
Joined: Mon May 05, 2014 4:31 am

Re: Printing a cylinder - there are gap

Post by Tiertime-Jason » Sun Aug 28, 2016 7:31 am

I printed the cylinder of Elenc, and it printed well with UP BOX.
So I think may be something is malfunctioning.

arhi
Posts: 256
Joined: Sun Mar 08, 2015 10:51 pm

Re: Printing a cylinder - there are gap

Post by arhi » Sun Aug 28, 2016 3:15 pm

@davearm, all 3 default stl settings for solidworks are ok .. the "fine" settings in solidworks is still not "too dense" for 3d printing .. but you can go custom and wind it up to be huge and unusable :D.
netfabb app (unfortunately only the expensive professional version) has option to easily "optimize" any stl to fit your needs
other apps that can manipulate STL's (art of illusion for e.g.) can also do it, not as easy and fast but more then acceptable. you basically optimize mesh to remove points/triangles while staying inside the resolution of your printer (for e.g. you say optimize to 0.05mm, this can for e.g. reduce a 2GB stl to under 1MB !! the stl will look "uglier" looking at it in 3d programs but the print will be exactly the same or better)

now the other part of the equation (slicer and printer) is way worse then the stl generation tools ...
the stuff I experimented with 10 years ago are still not readily available in common slicers and firmwares :( ... I never had time to polish all that stuff but I pushed bunch of examples and scripts showing how to get 10x more resolution while getting 10x more speed using exactly the same hardware

Post Reply