SoliForum - 3D Printing Community — Printing threads.

Re: Printing threads.

null@example.com (mconsidine) — Wed, 19 Jun 2013 16:43:25 +0000

At the risk of dragging this thread on longer than it needs to be ... Has anyone been experimenting with the "spiral vase" mode in the latest release of Slic3r? (I gather Cura has a similar mode.) I wonder if this is set if it will interpret helical patterns any more "nicely."

At the moment I'm just thinking out loud, but will give it a shot in a day or so.

Matt

Re: Printing threads.

null@example.com (mconsidine) — Mon, 17 Jun 2013 17:46:03 +0000

I was trying to just get *something* at .3mm for a proof of concept, recognizing that no matter what, I'm going to need to chase the threads. But even that didn't seem to work, which led me to think that the bigger issue was the strategy used to slice the model.

I did not have a fan on the bed to cool the plastic quickly, so maybe I should revisit this. How fine a thread have you been trying to print? Maybe I should re-read what the user who was trying to get down to 0.05mm did ...

Matt

Re: Printing threads.

null@example.com (nlancaster) — Mon, 17 Jun 2013 17:01:23 +0000

what layer thickness are you printing at?

maybe a fan to cool the print right away. I have been experimenting with thread printing and find that the threads are not cooling fast enough and warping. I am going to get my fan up and running soon to combat that.

Re: Printing threads.

null@example.com (mconsidine) — Mon, 17 Jun 2013 13:40:31 +0000

Thanks for the link. Those were nicely done. For the M42, the issue wasn't the diameter (in my couple of attempts) but instead the slicing. I tried printing it both horizontally and vertically and it just didn't seem to want to hold together (stringy). Maybe it's my settings, but by now I think I'm pretty well dialed in on backlash, motor steppings (x,y,z and e), extrusion multiplier and filament diameter. A single wall calibration cube comes out dead-nuts-on. So I'm suspecting that translating a fine helical pattern to an x-y-z layout process is the culprit in existing software.

At the same time, I wonder if knowing ahead of time that the pattern is helical would allow for a smarter slicing strategy. I assume that current slicing software is necessarily "dumb" about what it is trying to reduce to gcode. (An exception to this would be Slic3r's theoretical ability to use codes for arcs when it thinks it's trying to do a circle. But I don't think even that works ...)

Matt

Re: Printing threads.

null@example.com (elmoret) — Mon, 17 Jun 2013 01:30:29 +0000

M48x6 printed just fine on a Printrbot Simple:

http://www.matter-replicator.com/forum/ … &t=119

I'm sure you can do M42, but 0.75mm thread pitch is really fine.

Re: Printing threads.

null@example.com (mconsidine) — Mon, 17 Jun 2013 01:02:29 +0000

Well, I'll admit that I'm no pro with respect to clever designs or configurations for these printers. But I've come to the conclusion that an M42x0.75 thread can't be printed. None of the slicers I've tried seem to be able to deal nicely with it (in terms of travel paths). Or maybe its my configurations. In any case, attached is the SCAD and STL files if anyone finds a need to explore this rabbit hole.

Matt

Re: Printing threads.

null@example.com (mconsidine) — Thu, 13 Jun 2013 12:50:50 +0000

adrian wrote:

From back in the annals of time, this article from hack-a-day: http://hackaday.com/2011/11/15/a-techni … 3d-prints/
Still holds true....

Thanks for that. It would be nice if Slic3r (or one of the other, similar, apps) could add this sort of technique automatically. But I suspect anything beyond a simple interface layer to the bed would need special design attention.

Also, I can see the need to have to tailor where infill can be light and where it needs to be heavy. All this suggests the need for a slicing app that allows for more "art" to be added to the slicing process.

Thanks too for the links to the sample threads and settings commentary - they are helpful. I've got a feeling that getting the T-Mount adapter printed in a way that is usable will be a challenge. Because of the weak point noted above, I wonder if it might make sense to print the ring standing up? Certainly would be a good test of overhang and support capability!

Matt

Re: Printing threads.

null@example.com (saymo606) — Thu, 13 Jun 2013 11:28:46 +0000

The curling of large flat items is a real issue with this technology and I think we might be conflating the ability to "stick" with the tendency for solid plastic to contract. So I wonder if the solution isn't more in the fill design than it is in the type of hairspray used

less infill the better
design in some relief holes at the corners of your print if the design permits
5mm brim
cover the enclosure

all the above is good for combating warp. I pretty much print everything with a brim now... not many warps now at all..if they do, i reduce the infill.

Re: Printing threads.

null@example.com (adrian) — Thu, 13 Jun 2013 11:24:42 +0000

mconsidine wrote:

So I wonder if the solution isn't more in the fill design than it is in the type of hairspray used
Matt

From back in the annals of time, this article from hack-a-day: http://hackaday.com/2011/11/15/a-techni … 3d-prints/

Still holds true....

Re: Printing threads.

null@example.com (mconsidine) — Thu, 13 Jun 2013 11:19:44 +0000

Thanks all. Last night I spent some time coming up to speed (sorta) on OpenSCAD and ended up with an M42x0.75 design that netfabb's cloud service was happy with. So I'm going to try printing that and will post results.

I agree with the comment about the weakness being in the layers and am concerned about that.

'course, I also need to figure out how to successfully print an iPhone case to use with this adapter. The curling of large flat items is a real issue with this technology and I think we might be conflating the ability to "stick" with the tendency for solid plastic to contract. So I wonder if the solution isn't more in the fill design than it is in the type of hairspray used

Matt

Re: Printing threads.

null@example.com (2n2r5) — Thu, 13 Jun 2013 04:19:21 +0000

http://www.thingiverse.com/make:37155

Here are some very small threads that I printed.

Re: Printing threads.

null@example.com (adrian) — Thu, 13 Jun 2013 03:52:52 +0000

I have managed to succesfully print working threads on a large scale; I've made a few variations of http://www.thingiverse.com/thing:10189 and they all sealed and functioned...

But its not exactly an M3 thread

Re: Printing threads.

null@example.com (nlancaster) — Wed, 12 Jun 2013 21:14:18 +0000

http://www.thingiverse.com/thing:13923

I printed up one of those as a test. Works pretty good, and It is a neat puzzle.

That being said, I wouldnt want to use printed nuts/bolts for anything that needed to be rugged. the primary weakness of our printers is the layering, and you will bee applying a load to the screw that will be trying to break that layer connection.

Re: Printing threads.

null@example.com (neoblood3d) — Wed, 12 Jun 2013 20:28:06 +0000

Blender3d has a lot of features. Graphicall.org has a ton of modified builds that can help you create virtually anything through a series of conversational type variables.

Re: Printing threads.

null@example.com (mconsidine) — Mon, 10 Jun 2013 23:13:00 +0000

Thanks for the link and the advice.

Using the M42 nut in that collection, scaling it so that the thread width is 0.75mm and then placing it in a 2" O.D. ring, I get the attached SketchUp model. It is then exported as an STL.

It looks right but that's about where things go off the rails. When I try running it through either Netfabb Studio or the Netfabb cloud service, I essentially end up with a ring and no interior threads, at best. (Actually, I end up with a model that's all red in Netfabb Studio.)

So I'm thinking I've somehow modeled this incorrectly or otherwise missed some essential point to try to model and then slice a ring with fine detail.

If anyone with more experience than me in modelling has any suggestions, pointers or a fix, I'd be appreciative.
Regards,
Matt

EDIT : Removed large model and substitute one that netfabb is happy with shortly...