General dimensional accuracy and roundness of printed holes

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

General dimensional accuracy and roundness of printed holes

Post by stratogavster »

Hello All Felix users out there....

I'm currently experimenting with the 'high quality' profile in SFACT in response to the great print I did of the Octopus (which was sliced with this said profile).

I'm experiencing dimensional accuracy issues in hole features and circles.

Now this is obviously mostly due to the design of the machine - high friction in the linear slides, play in the slides, backlash and just general repeatability...but what makes it odd is that i've printed a couple of parts with incredibly accurate external linear dimensions but their subsequent hole features are very 'off'.

The example part I printed was 15mm x 40mm (15mm Z), with 2 x 8.1mm holes. After printing and measuring the 15mm and 40mm dims where amazingly spot on!! quite impressive. But sadly the holes came in at 7.8mm.

Mechanics aside is there anything within the slicing software that should be consider?

To reiterate the profile used was 'High Quality', with one amend. Printed line width was increased from the default 0.3mm to 0.38mm to match that of the other profiles used.

Layer height was default 0.1mm

Print quality was ok, though had bad 'blobbing' at the perimeter start area.

Any ideas or suggestions would be great :D

Trnquill
Posts: 66
Joined: Fri Oct 05, 2012 10:01 am

Post by Trnquill »

There's a setting for shrinking holes in SFACT called Stretch. You might want to tinker with that one. :) I usually draw holes a tad larger to compensate the "shrinkage" and/or just drill them out to exactly correct measurement after printing.

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

Post by stratogavster »

Din't know that option was in there.....will give it a go.....sadly i'm not able to re-drill some of the holes, so i'm aiming to get the machine printing holes of requested diameter.

Firstly i'm upgrading my machine to v1.5 to get rid of the inherently sticky IGUS slides and get this things printing circles properly, then will see what that produces. Next thing I guess is SFACT....though i'd prefer not to 'fudge' things in the design of my parts

:)

dognotdog
Posts: 181
Joined: Tue Sep 04, 2012 9:22 pm

Post by dognotdog »

Upgrading to v1.5 definitely improved hole quality, due to reduced backlash. However, I still need to make holes 0.1 to 0.3mm larger than what I want to stick through it. I mostly have to do two prints of a part that needs to be accurate in that respect, the first being a prototype for dialing in the hole sizes.

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

Post by stratogavster »

Based on feedback from Trnquill, the 'Stretch' plugin is something i'm going to seriously look at. Obviously shrinkage is a factor here....though I thought the software / firmware had all these factors accounted for?

I'm used to designing parts for injection moulding, and quite happily from my perspective the tooling designers take care of shrinkage when my parts are released to them, so when the parts come off the tools they are exactly as per the design intent....no need for me to fudge anything. I just assumed that the 3D printer world had accounted for these factors already....

Hoping to get my head around the slicing software pretty soon! wish there was a more proven and repeatable solution out there to cope with the super techie task of slicing and dealing with all the realities of 3D printing :)

Trnquill
Posts: 66
Joined: Fri Oct 05, 2012 10:01 am

Post by Trnquill »

stratogavster wrote:... the tooling designers take care of shrinkage when my parts are released to them, so when the parts come off the tools they are exactly as per the design intent....no need for me to fudge anything.
Yep, that's the way it goes. YOU are the tooling designer here. If software was able to do the stuff this specialized group of people do, what would we need them for in the first place? ;) In other words: if we need special people to transform a design to printable/moldable/whatever format in industrial world, I don't think there's a automated software for that in the open source world for some time.

I think you have the parameter in the slicer because not all the machines nor filaments are the same. You need to have this value adjustable to get right results. Just adopt it to your machine, filament, temperatures etc. There's not one single correct value. Otherwise it would be in the firmware as you said.

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

Post by stratogavster »

I generally agree but what I fail to see sense in is that these machines seem to not be able to cope with dimensional accuracy's in some areas but not others.....

After doing another couple of prints last night I always get X, Y dims almost spot on....maybe only 0.1mm too small up to 100mm, and the Z height is alway perfect! (needed a lot of tuning though!)....(but still have other quality issues)

...and as for hole diameters that seems to be another thing altogether....requesting the machine to create a hole with a diameter of X but then being given a diameter of X -0.5(+)mm is not the best...when confronted with a linear dimension of Y and being given Y-0.1mm (which is pretty good!)

It's well understood that interfering parts certainly need to be designed with process tolerances in mind, which is inherently linked to the geometry and function of the parts....so on that basis knowing that my particular machine needs to have holes enlarged or pins reduced by 0.5/0.6mm is no problem, but that doesn't help with any mass appeal to said designs if other machines are capable of rendering holes at a high accuracy, thus the apparent process change and increase in accuracy would likely cause the design to perform differently when produced by another machine....i.e looser fitting parts than is required

I'm going to start with low level stuff first....upgrade the linear slides and then do an incredibly detailed recalibration (if I can figure out how to do it)...then repeat the part prints to see if there is a dimensional improvement.

Trnquill
Posts: 66
Joined: Fri Oct 05, 2012 10:01 am

Post by Trnquill »

So, you tried the "Strecth" but it didn't help?

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

Post by stratogavster »

No haven't tried it yet.....just reading through the manual trying to understand it fully :)

Have tried clip and jitter and retraction....printing now seeing what they do on the same part...

'Stretch' will be next....

Post Reply