View unanswered posts | View active topics
|
Page 1 of 1
|
[ 14 posts ] |
|
| Author |
Message |
|
puzzle_weaver
|
Post subject: questions... Posted: Tue Jul 24, 2012 12:48 pm |
|
Joined: Sun Mar 25, 2012 7:12 pm
|
|
Hello,
Lately I have been playing around in freecad. I have made a puzzle and have many questions: Is freecad compatible with Shapeways? If so, how can I submit a design? How much tolerance should a puzzle have to be functional?
Thanks, AW
_________________ Boring monochrome signature
|
|
| Top |
|
 |
|
KelvinS
|
Post subject: Re: questions... Posted: Tue Jul 24, 2012 12:50 pm |
|
Joined: Mon Mar 30, 2009 5:13 pm
|
puzzle_weaver wrote: How much tolerance should a puzzle have to be functional? See this thread: viewtopic.php?f=9&t=20475Especially this post: viewtopic.php?p=248427#p248427
_________________ I'm going wherever they value my loyalty the most.
Last edited by KelvinS on Tue Jul 24, 2012 12:55 pm, edited 1 time in total.
|
|
| Top |
|
 |
|
bmenrigh
|
Post subject: Re: questions... (user search!) Posted: Tue Jul 24, 2012 12:55 pm |
|
Joined: Thu Dec 31, 2009 8:54 pm Location: San Jose, California
|
puzzle_weaver wrote: Hello,
Lately I have been playing around in freecad. I have made a puzzle and have many questions: Is freecad compatible with Shapeways? If so, how can I submit a design? How much tolerance should a puzzle have to be functional?
Thanks, AW This has been asked many times before. Please use the search. For example: http://twistypuzzles.com/forum/viewtopic.php?p=248609#p248609The short answer is that Shapeways accepts several file formats but STL is the preferred one.
|
|
| Top |
|
 |
|
Rubikscuber1
|
Post subject: Re: questions... Posted: Tue Jul 24, 2012 2:22 pm |
|
Joined: Sun Nov 16, 2008 5:57 pm
|
puzzle_weaver wrote: Hello,
Lately I have been playing around in freecad. I have made a puzzle and have many questions: Is freecad compatible with Shapeways? If so, how can I submit a design? How much tolerance should a puzzle have to be functional?
Thanks, AW I used SolidWorks Student Edition and I had to convert my files from .prt to .stl by using eDrawings. Shapeways accepts many formats and it can be found somewhere on their page. As far as tolerances, I didn't give any tolerances. I just wanted to see how much the pieces varied. That allowed me to sand the WSF material with 400 grit sandpaper to get it to my liking. For the next model, I use the sanded lengths for the pieces
_________________ Rubikscuber1
http://www.youtube.com/user/Rubikscuber1
|
|
| Top |
|
 |
|
bmenrigh
|
Post subject: Re: questions... Posted: Tue Jul 24, 2012 2:35 pm |
|
Joined: Thu Dec 31, 2009 8:54 pm Location: San Jose, California
|
Rubikscuber1 wrote: As far as tolerances, I didn't give any tolerances. I just wanted to see how much the pieces varied. That allowed me to sand the WSF material with 400 grit sandpaper to get it to my liking. For the next model, I use the sanded lengths for the pieces This is only a viable option for extremely simple mechanisms where all surfaces that interact with other surfaces are exposed and easily reached with sandpaper.
|
|
| Top |
|
 |
|
Rubikscuber1
|
Post subject: Re: questions... Posted: Tue Jul 24, 2012 10:21 pm |
|
Joined: Sun Nov 16, 2008 5:57 pm
|
bmenrigh wrote: Rubikscuber1 wrote: As far as tolerances, I didn't give any tolerances. I just wanted to see how much the pieces varied. That allowed me to sand the WSF material with 400 grit sandpaper to get it to my liking. For the next model, I use the sanded lengths for the pieces This is only a viable option for extremely simple mechanisms where all surfaces that interact with other surfaces are exposed and easily reached with sandpaper. yes, such as puzzles like the master quartet. But for puzzles that would be hard to sand, such as ones with many layers or complex curved internal surfaces, it's good to make tolerances, in my opinion, less than .2mm
_________________ Rubikscuber1
http://www.youtube.com/user/Rubikscuber1
|
|
| Top |
|
 |
|
Splinter
|
Post subject: Tolerances and accuracy Posted: Thu Jul 26, 2012 2:35 am |
|
Joined: Mon Dec 05, 2011 3:38 pm Location: Eindhoven, The Netherlands
|
Rubikscuber1 wrote: bmenrigh wrote: Rubikscuber1 wrote: As far as tolerances, I didn't give any tolerances. I just wanted to see how much the pieces varied. That allowed me to sand the WSF material with 400 grit sandpaper to get it to my liking. For the next model, I use the sanded lengths for the pieces This is only a viable option for extremely simple mechanisms where all surfaces that interact with other surfaces are exposed and easily reached with sandpaper. yes, such as puzzles like the master quartet. But for puzzles that would be hard to sand, such as ones with many layers or complex curved internal surfaces, it's good to make tolerances, in my opinion, less than .2mm I understand that it's advised to add tolerances to the nominal dimension modelled. (e.g. 54.4 mm + 0.3 mm -0.1 mm) But as far as I know, 3-D printing e.g. at Shapeways only looks at the model, in casu the nominal dimensions of it. Or isn't this true? Nowhere on the Shapeways website I see an advise that given tolerances to nominal values are taken into account. The best info I get from Shapeways, are some test results they've performed on the tolerance and position accuracy of slots and pins: link ShapewaysAnd these results are all quoted from nominal values. Anyone experience that tolerances provided in a model add to the actual accuracy when 3-D printing?? Or is it only the advise (which I do) to add -from nominal values- enough distance ('tolerance') between parts for assembly reasons?
_________________ Maybe I'm amazed... Splinter's amazing Shapeways shop, my E-book, maze overview page (Dutch), Mfave's puzzle shop
|
|
| Top |
|
 |
|
bmenrigh
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 2:52 am |
|
Joined: Thu Dec 31, 2009 8:54 pm Location: San Jose, California
|
|
The idea behind adding tolerances is not to improve accuracy. It is to account for inaccuracy.
If you designed and printed a part with a 3mm diameter circular hole in it and a 3mm cylinder to fill that hole, the cylinder WOULD NOT fit in the hole. You'd need to either make the hole slightly bigger on the cylinder slightly smaller.
The same goes with rails and grooves and other cases when parts would be sandwiched between two surfaces.
This is true for all forms of manufacturing.
Beware though, you can't just go and subtract off some tolerance from all measurements. You only adjust certain types of surfaces. Eric explains it well in his post. Understanding and getting the right tolerance for each particular application is a rather complicated endeavor and a lot of science has gone into picking the right way to tolerance parts in various different situations. There are mechanical engineering books that have huge tables of tolerance values for different materials in different interactions.
|
|
| Top |
|
 |
|
Splinter
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 4:16 am |
|
Joined: Mon Dec 05, 2011 3:38 pm Location: Eindhoven, The Netherlands
|
|
To stick with the pin & hole tolerances (I'm a mechanical engineer, so I know the books):
for me it's not the issue how to tolerate dimensions (e.g. can write H9-f7 fittings and such), but when I e.g. intend to have a pin of diameter 3mm I have no clue whether Shapeways:
- will always print it as 3mm plus tolerance. Say 3mm +0 +0.3 mm - will always print it as 3mm minus tolerance. Say 3mm -0 -0.3 mm - will print it always somewhere between 3mm +/- 0.2 mm
And similar for the hole.
So when an item has to fit within an assembly, and I would know (but I don't) that shapeways always has tolerances of +/-0.2 mm, then my hole would have to be dimensioned 3.21 mm and the pin 2.80mm. Or an identical combination.
If the hole and pin are then printed respectively 3.01 mm and 3.0 mm, it'll fit exactly. BUT, if the outer tolerances are taken the hole is 3.41 mm and the pin 2.60 mm. A difference of 0.81 mm!! THAT'S HUGE!!
Conclusion: to dimension to fit with a process at Shapeways with unknown tolerances is difficult / can give every new order suprising results (exact fit or far too loose/ strong fit)
Any suggestions to prevent this with this example are welcome. (In my new design I've made a snap-fit which works OK)
_________________ Maybe I'm amazed... Splinter's amazing Shapeways shop, my E-book, maze overview page (Dutch), Mfave's puzzle shop
|
|
| Top |
|
 |
|
Gus
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 4:38 am |
|
Joined: Sun Mar 15, 2009 12:00 am Location: Jarrow, England
|
|
From the Shapeways website, the basic accuracy of the laser sintering machine (used for WSF) is +/-0.15mm (and then +/-0.15% of longest model axis). For the "detail" materials, which use UV cured photo-polymers, it is +/-0.1mm. These are manufacturing tolerances, and have to be taken into account when designing your hole and peg.
_________________ Most of the things worth doing in the world had been declared impossible before they were done.
|
|
| Top |
|
 |
|
Splinter
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 6:30 am |
|
Joined: Mon Dec 05, 2011 3:38 pm Location: Eindhoven, The Netherlands
|
My first prototype has channels modelled of diameter 3.6mm, which turned out to be 3.2mm. So far worse than 0.15mm accuracy. Any idea whether it can be due to it being made from Black Strong and Flexible (and not WSF)? From the parts received, I got the impression that with BSF the base material is still WSF. But with some kind of top layer or coloring added. (Internal parts cleaned out from loose black processing powder turned out to be white) Anyhow, I've currently ordered 1 modified part in WSF. And I'll measure that against the dimensions of the model also. So I can provide the answer in some weeks myself 
_________________ Maybe I'm amazed... Splinter's amazing Shapeways shop, my E-book, maze overview page (Dutch), Mfave's puzzle shop
|
|
| Top |
|
 |
|
Gus
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 7:07 am |
|
Joined: Sun Mar 15, 2009 12:00 am Location: Jarrow, England
|
Splinter wrote: My first prototype has channels modelled of diameter 3.6mm, which turned out to be 3.2mm. So far worse than 0.15mm accuracy.
Any idea whether it can be due to it being made from Black Strong and Flexible (and not WSF)?
From the parts received, I got the impression that with BSF the base material is still WSF. But with some kind of top layer or coloring added. You are correct, BSF is just WSF which hase been dyed (you can do this yourself if you want to save some money) As to whether the dying process affects the WSF material, well it does absorb the water during dying, which may make it swell somewhat, but it would shrink back again when it dries. As to your channel dimension errors, you should take that up with Shapeways. But remember, Shapeways disclaimer is that all models are just "toys" for "entertainment" and not intended for anything else.
_________________ Most of the things worth doing in the world had been declared impossible before they were done.
|
|
| Top |
|
 |
|
Splinter
|
Post subject: Re: questions... Posted: Thu Jul 26, 2012 7:40 am |
|
Joined: Mon Dec 05, 2011 3:38 pm Location: Eindhoven, The Netherlands
|
|
I'll take it up with Shapeways when my improved model (now in production at SW) still has some unexpected deviation.
It is still largely the same as with old-school 'rapid prototyping': the material received is more to show you how it would look ('decorative purpose'), but you couldn't load it with masses (was very brittle in those days).
_________________ Maybe I'm amazed... Splinter's amazing Shapeways shop, my E-book, maze overview page (Dutch), Mfave's puzzle shop
|
|
| Top |
|
 |
|
Splinter
|
Post subject: Re: questions... Posted: Tue Jul 31, 2012 1:57 am |
|
Joined: Mon Dec 05, 2011 3:38 pm Location: Eindhoven, The Netherlands
|
Gus wrote: As to your channel dimension errors, you should take that up with Shapeways. I received my part in WSF yesterday (Shapeways was rather quick this time; but on the other hand: transport is only from the other end of town), and this was quite according to the dimensions given. I've the following theory when noticing my BSF parts and WSF part: I found several loose processing powder sticking in the BSF parts, and also some in the WSF part. I think that some of the loosely attached powder is fixed to the part when Shapeways has dyed it to black. Making some of my hole/channel dimensions smaller than expected. So, I'll stick to ordering in WSF, adapt the parts (cleaning out loose powder thoroughly) and when I want black I've to learn the dying process as indicated in this forum. We'll see how all works out with my new design. Still some steps to make.
_________________ Maybe I'm amazed... Splinter's amazing Shapeways shop, my E-book, maze overview page (Dutch), Mfave's puzzle shop
|
|
| Top |
|
 |
|
Page 1 of 1
|
[ 14 posts ] |
|
Who is online |
Users browsing this forum: Majestic-12 [Bot] and 6 guests |
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum
|
|
|