Steep Overhangs - advice and help?

This is for general discussions. Also FAQ can be found here.

Moderator: speedake95

Post Reply
stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Re: Steep Overhangs - advice and help?

Post by stratogavster »

Hello all....

Still having issue with overhangs....

Basic holes and 45 degree chamfers not liking to be built!! I have proven that a small layer height and 'skin' helps a lot, but i'd rather benefit from the faster build times on offer with thicker layers.....plus many others seem to be building thicker layers using default profiles....

Accepted my 'holes' scope for success is a function of diameter and layer height, but would be great to get these spot on without support material.

Chamfers are disappointing! :(

Any other ideas of suggestions?

Some images of latest findings....
1.jpg
2.jpg
3.jpg
4.jpg

Kjetilei
Posts: 334
Joined: Fri Mar 16, 2012 8:26 pm
Location: Stavanger, Norway

Post by Kjetilei »

Hi stratogavster.

Although it is not surprising that you're struggling with the big diameter holes/cylinders, I'm a bit puzzled about the problems with the 45-degree chamfers!

Sadly I don't have a direct tip that might help you solve the problems as such but:
- could you rotate the object so that the holes/cylinders are printed upwards instead of sideways
- could you redesign to get around the problems? (are the prints the final products or just prototypes of parts to be produces with other machines?)

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

It's puzzling me too....

I have designed parts to omit chamfers and tricky features, but that's not really the way I should drift too much. Sometimes my parts inevitably need these features in not 'ideal' places. I've seen other printers and machines cope with these features pretty generically...

After all 3D printers are meant to make the unmake-able! :P

I had this same issue with a gcode provided for the fan duct....my print failed 3 times but one guy had success first off. After a new file was uploaded with different backlash settings my print worked....

I can only think there is a overall calibration issue with my machine (steps/mm vs pulley diameters etc) which means X/Y positioning is not correct...thus layers are 'falling off' their respective edges...this may explain my X/Y dimension issues (though I seem to have sorted it with tweaking profiles...but I feel thats a fudge!).

My next target is to calibrate the machine properly and then repeat all tests again.....hardwork :( but needed :(

satoer
Posts: 436
Joined: Mon Oct 01, 2012 8:00 pm
Location: Netherlands

Post by satoer »

New idea… have you tried to change the perimeter – loop – infill option?
Fill Button => “What order to print” and then select: Loops – perimeter – infill. This way it first prints the inside loop. After that the outside (perimeter) loop. This way the outside loop can stick to the inside loop which is on top of the outside loop from the layer below.
My Felix 1.5 printed designs on Thingiverse
E3D V6 Hot end upgrade for your Felix
Tweezer bracket for your Felix
Led bar upgrade for your Felix

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

I did try this in an effort to prevent aesthetic problems (with some success) but will certainly be trying with overhangs.....thought i'd flag it up because Guillaume provided 2 differing 'duct' gcodes....one worked and one didn't....the only difference was backlash! :S hence the mystification of what is happening.

I'm sure i'll get to the bottom of this one, but it's at the expense of it being 'machine specific knowledge'....I truly think I need to calibrate as accurately as possible to create a definitive 'base' to start with, then any other tweaks there after should be repeatable and in theory they could be readily and successfully shared with the community

:)

User avatar
gfeliksdal
Site Admin
Posts: 405
Joined: Sat Feb 25, 2012 10:40 pm
Location: Netherlands
Contact:

Post by gfeliksdal »

I found that the backlash value is very important to obtain best prints (only for Felix 1.0), and the values indeed are machine specific. You can see the difference when printing the file attached to this message, with different slicing settings.

Also I've observed that with the newer hot-ends this value became more critical, for some reason. The only difference is that the new hot-ends have a sharper tip. So, the old hot-ends somehow gave better results by sweeping the just extruded plastic more flat and the lines next to it. To see if my hypothesis was right, I filed the tip of a new hot-end a little and the results became noticably better.
So, after doing some more testing to get to the bottom of it, I found out that varying the backlash value made the results as I would like them to be. On the different printers Felix 1.0 printers I had, they all showed acceptable results at backlash value 0.1mm.

If the sharper tip is the exact reason, I'm not 100% sure. The position inaccuracy was already there I guess, but the sharper tip seems less forgiving. Due to the lack of time, I was content with the backlash value optimization. The new linear guides don't have this issue anymore.
test_tower.zip
(435 Bytes) Downloaded 24 times

Kjetilei
Posts: 334
Joined: Fri Mar 16, 2012 8:26 pm
Location: Stavanger, Norway

Post by Kjetilei »

Guillaume: Thanks for sharing your experiences with us!

I would never have thought that minor differences to the shape of the tip would make a difference to this problem...

But the backlash tweaking and tip tweaking truly show how delicate successful 3D printing can be. The margins of error can be very narrow...

User avatar
gfeliksdal
Site Admin
Posts: 405
Joined: Sat Feb 25, 2012 10:40 pm
Location: Netherlands
Contact:

Post by gfeliksdal »

No problem, I've came up with another possible solution. It could be that the nozzle exit is slighty damaged or dirty. It might have hit the table accidently or just dust which or other contaminents are in the nozzle exit. See the difference between prints after cleaning the nozzle out with a 0.35mm drill.
uncleaned nozzle
uncleaned nozzle
Cleaned nozzle
Cleaned nozzle

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Good tip!! When my bad electronics where still being used I had the case where the machine went crazy and started to drag the hot-end across the print a number of times....so this i'm sure has done something to the tip profile :(

Will try and very carefully clean it....though have been very cautious of going anywhere near it as i'm sure it's been drilled to a very high tolerance and there is no guarantee the hole would remain round when touched :(

Is there any scope of a hot-end tip of a harder material? Some grade of stainless-steel perhaps?

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Forgot to ask....how did you print that gravity defying feature anyway? Has there been any tweaking of profiles at all? What profile was used? what speed? what temp?

I'm certain my machine could never print that as seen from the holes previously displayed, though in theory they are less challenging than this floating horizontal feature.

ps. didn't know there was a new hot-end?

Also could you post images of the Test Tower results :)

Post Reply