Home » Community » General Discussion » Wall Thickness (plain wall vs. supported wall)
Search Search  
Show: Today's Messages    Show Polls    Message Navigator
Wall Thickness (plain wall vs. supported wall) [message #46531] Thu, 05 April 2012 16:08 UTC Go to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
So far I haven't seen anything in the forums dealing with this issue, so here goes... The basic formula for dealing with Big Parts in White Strong & Flexible is well defined and certainly makes sense for bare stand-alone walls. But what about when your walls have support structures behind them to keep them rigid? I'm thinking in terms of oversized parts that use an internal framework for support instead of a heavy wall thickness.

Is such a design something that would automatically get rejected as not printable? Is the thickness formula a hard-and-fast requirement regardless of a design's inherent strength?

Thanks for your input.

--Josh T.


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46549 is a reply to message #46531 ] Thu, 05 April 2012 19:10 UTC Go to previous messageGo to next message
avatar stonysmith  is currently offline stonysmith
Messages: 2094
Registered: August 2008
Go to my shop
Shapie Expert
moderator
It may NOT be a hard-and-fast rule, but one thing you must remember.. the prints happen only one slice at a time, and therefore thin walls have to be able to "stick" to the layer below them.

There is also the issue that models are oriented in (somewhat random) directions, and what you think of as "up" is not always so.


Patience, Persistance, Politeness - the 3Ps will help us get us to Perfect Printed Products
Re: Wall Thickness (plain wall vs. supported wall) [message #46562 is a reply to message #46549 ] Thu, 05 April 2012 20:32 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
Thanks, but I don't think that material sticking to itself is a problem for the type of scenario I have in mind; I'm considering models that would be thicker than the .7mm minimum, but thinner than the thickness calculated using the rules for big (>117mm) parts. Specifically, I'm wondering if the calculated minimum thickness for large models can be avoided by using a good self-supporting internal structure.

OR... will Shapeways automatically flag such a model as unprintable? No point in spending time making a complex structure if it will be rejected.

Thanks again,

--Josh T.

[Updated on: Thu, 05 April 2012 20:36 UTC]


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46563 is a reply to message #46562 ] Thu, 05 April 2012 21:19 UTC Go to previous messageGo to next message
avatar stonysmith  is currently offline stonysmith
Messages: 2094
Registered: August 2008
Go to my shop
Shapie Expert
moderator
One method that has worked well for me is to setup a sample model with the questionable peices and then send that to customer service requesting a "Thin Walls Check".

They can pass it to production and take a look without acutally printing the model.

I realize that this will cause you some work, but perhaps you can work out just one section of the model for them to take a look at.


Patience, Persistance, Politeness - the 3Ps will help us get us to Perfect Printed Products
Re: Wall Thickness (plain wall vs. supported wall) [message #46564 is a reply to message #46563 ] Thu, 05 April 2012 21:21 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
Thanks. I appreciate the suggestion. Smile


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46631 is a reply to message #46564 ] Sat, 07 April 2012 13:32 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
Just as a quick experiment, I uploaded a part with a thinner wall than per calculations (just under 1.3 mm instead of the calculated 1.5mm) and the automatic part checks did not flag it.

Are thickness problems (for oversized models) typically flagged during upload or is it something that is only caught after an order is in process?


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46634 is a reply to message #46631 ] Sat, 07 April 2012 15:21 UTC Go to previous messageGo to next message
avatar stonysmith  is currently offline stonysmith
Messages: 2094
Registered: August 2008
Go to my shop
Shapie Expert
moderator
It's done with manual (human) checks just as the item is sent to the printer.

Most of the automated checks only determine whether something can be printed in a specific printer or not. Steel and Glass are a larger printers than the plastics. I one goofed on the measurements of a model and it ended up about 18 inches long.. Wouldn't fit in any one of the plastic printers, but it fit nicely in the Steel printer... for $2500. !!

The item was supposed to be about 1.5 INCHES, not 1.5 feet (oops)


Patience, Persistance, Politeness - the 3Ps will help us get us to Perfect Printed Products
Re: Wall Thickness (plain wall vs. supported wall) [message #46740 is a reply to message #46634 ] Tue, 10 April 2012 17:01 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
Update...

After some correspondence with customer service, the wall thickness issue is non-negotiable. Apparently, it does not matter what sort of internal structures may be in place to make a model sturdy; minimum wall thickness (as calculated for parts longer than 117mm) is a requirement.

I'm guessing that this is a business/pragmatic issue and that Shapeways just doesn't have the resources to judge thickness on a case-by-case basis, so it's a lot easier and simpler for them to have a hard-and-fast rule to go by.


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46755 is a reply to message #46740 ] Tue, 10 April 2012 19:39 UTC Go to previous messageGo to next message
avatar NormL  is currently offline NormL
Messages: 33
Registered: March 2012
Go to my shop
Member
I also fought this issue. Their linear solution to a 3D problem does not make sense, but, they need a rule that someone without an engineering background can assess. I had sent them stress tests done from within my software which did prove the print ability of what I was proposing and did comparisons to other thicknesses and structuring. Point loading and fixing the same geometry in every case. To no avail, I ended up cutting my model into 6" x 6" pieces and printing it to glue together. Very stupid, but, they need a rule they can enforce. I even thought about posting on here to see if there were others fighting this like me and if we could create a review group for them. I didn't do it because in a quick perusal I didn't see any postings.

This is not an unknown problem to Shapeways and I feel, maybe naively, that we will see a modified rule at some point in the future. The fact is the current rule completely ignores structure. This is a hard pill for me to swallow as I can just print a stress analysis that shows how it will do. That is going to be a hard thing for them to develop a review policy for though.

I did speak with an Eosint representative and printing with structure and not material width is the correct answer. We just need to help them develop a rule so that our large models are not blowing up during printing and disturbing other prints going on at the same time. Heat is the issue and my understanding is that they are pausing for an hour every vertical inch, +/- regardless for dissipation.

Another thing I would suggest, is rotating the grid on your model before you create a copy for uploading so the faces line up the way you want. I wasn't doing this and my models, for instance, were saying they were 7.4" x 6.4" x 1.2" when they were in fact 6" x 4.3" x 1.2". They take a bounding box from your grid to create the size that is listed on your model properties ... Not a big issue and some human still has to review it and will see the bounding box skews the model, but, I don't want to depend on how much coffee they have had.

Love your screen name!

[Updated on: Tue, 10 April 2012 22:33 UTC]

Re: Wall Thickness (plain wall vs. supported wall) [message #46758 is a reply to message #46755 ] Tue, 10 April 2012 19:51 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
CPConsult wrote on Tue, 10 April 2012 19:39

Love your screen name!


Thanks. And thanks for such a thorough description of your experience. For one of my larger models I plan on separating the bottom cylinder and make it a two-part model within the same file (hopefully they won't have a problem with that). But I agree that it seems silly to have such an arbitrary rule on thickness when the result is higher cost and weight. Too bad there's not an easier and practical workaround.

Now I just need to hear back from customer service on why I'm getting a "unknown_size" error after uploading when Netfabb shows a perfect model.


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46765 is a reply to message #46755 ] Tue, 10 April 2012 20:50 UTC Go to previous messageGo to next message
avatar stannum  is currently offline stannum
Messages: 1052
Registered: May 2009
Go to my shop
Shapie Expert
Add one more case: sprues.

You have a model that has been printed multiple times, and next order you get two emails. One shows a picture measuring two distances, one small and another big, and the text in the email says you have to increase the small zone to a given size or bigger, but no hint about what is the big measurement about. The other email has a picture showing a similar small measurement, and the text says it has to be thicker but not how much, and that it would be better if some parts were single part instead of two.

How do you really fix all that? Ignore the emails and rearrange the sprue layout.
Re: Wall Thickness (plain wall vs. supported wall) [message #46767 is a reply to message #46758 ] Tue, 10 April 2012 21:02 UTC Go to previous messageGo to next message
avatar stonysmith  is currently offline stonysmith
Messages: 2094
Registered: August 2008
Go to my shop
Shapie Expert
moderator
Meatball Rocketry wrote on Tue, 10 April 2012 19:51

Now I just need to hear back from customer service on why I'm getting a "unknown_size" error after uploading when Netfabb shows a perfect model.


A question for you: did you turn all of your polygons into triangles?
I recently have gotten a couple messages of "file not found or empty mesh" and it turned out to be that the (WRL) file was not fully triangulated. I don't know if this is a older problem, or something new, but I triangulated the shells, and poof.. the upload works fine.


Patience, Persistance, Politeness - the 3Ps will help us get us to Perfect Printed Products
Re: Wall Thickness (plain wall vs. supported wall) [message #46769 is a reply to message #46767 ] Tue, 10 April 2012 21:32 UTC Go to previous messageGo to next message
avatar MeatballRocketry  is currently offline MeatballRocketry
Messages: 40
Registered: June 2011
Go to my shop
Member
stonysmith wrote on Tue, 10 April 2012 21:02


A question for you: did you turn all of your polygons into triangles?


I'm uploading STL files after repairing them in netfabb, so they're already triangulated. Oddly enough, this "unknown_size" error is something brand new for me as of this morning.

In the middle of writing this reply, I tried to upload two files that were yielding the error several times this morning, and they just went through. So I have no clue what the problem is/was. Hopefully the rest of the files will work out.


Josh T.
Re: Wall Thickness (plain wall vs. supported wall) [message #46863 is a reply to message #46769 ] Thu, 12 April 2012 14:55 UTC Go to previous messageGo to next message
avatar NormL  is currently offline NormL
Messages: 33
Registered: March 2012
Go to my shop
Member
I don't know if Shapeways has somebody on staff qualified, but, if they did, maybe a design review fee would be in order if you wanted to go below their thickness rule. It saves us a lot to go below the rule, so, a review fee would be better than what we have. In essence, if you want to print outside of their rule, the burden of proof is on you and you need to pay this fee and provide these documents. If they want to proceed into rapid prototyping of larger articles they need to do something.

As far as the current rule, I do feel that a rule that compares material mass and surface area might get a little closer than the current one, but, that certainly has issues all its own. Any rule like this would also have to factor in model height and then would get very complicated.

I had done a fender design that by their rule had to be 3.72mm thick (928.8 cm3) and I also did a 1mm thick (384.17 cm3) design that was just a s strong, but, obviously heavily gusseted. 41% of the material and a better product Rolling Eyes .

There would be issues, such as models that have mesh issues and can't be tested and where exactly are you going to load and fix my model? Right now they have the ultimate say and should continue to do so, but, as it stands right now, you can give them the most robust piece that will fit in a bounding box of their minimum dimension for your material width and it can get kicked for a long diagonal face. That is amazingly frustrating. Basically a set of rules applying a force at one side and fixing another on any suspect geometry.

It may be that the issue is, that large objects comprised of more air than material don't suit the business model. I haven't heard this is an issue, but, it would not surprise me.

Just looking for thoughts as I would like to use Shapeways for prototyping and I am finding it just does not suit the environment for larger parts.
Re: Wall Thickness (plain wall vs. supported wall) [message #46867 is a reply to message #46863 ] Thu, 12 April 2012 17:27 UTC Go to previous messageGo to next message
avatar stop4stuff  is currently offline stop4stuff
Messages: 3154
Registered: June 2010
Go to my shop
Shapie Expert
Shapeways services are more aimed at 'the masses' than the serious RP types, so Shapeways rules need to be fairly strict in order to keep costs down.

If you weren't aware Shapeways is more of a middle-man. The rules ensure a very decent price compared to dealing direct. For example, ProMetal, who produce Shapeways stainless steel models, have a minimum order of $150 (USD). Shapeways is $6 flat fee + $8 per cc + shipping. Don't forget the shipping fee is per order, no matter how many models or weight Wink

There's a discount on WSF for large (cc) models with a more than whatever specified % density volume, the details are in the material page linked to at the bottom of each and every Shapeways website page. Depending on the geometry of a model, increasing the volume can actually reduced the cost quite substantially (even just chucking in a solid cube of the right size can invoke the discount costing less for more material Confused )

Shapeways has its place in the 3D printing market and they enable people like me to bring designd to life and maybe earn a few $ along the way with their most excellent business model, however using Shapeways as a rapid-prototyping service is going to be a bit tougher although very much cheaper.

Paul

[edit] This doesn't really help the OP's ossue though - sorry Rolling Eyes

[Updated on: Thu, 12 April 2012 17:28 UTC]

Re: Wall Thickness (plain wall vs. supported wall) [message #46868 is a reply to message #46867 ] Thu, 12 April 2012 17:34 UTC Go to previous messageGo to next message
avatar NormL  is currently offline NormL
Messages: 33
Registered: March 2012
Go to my shop
Member
I was PM'ed that they were a middle man of sorts for a lot of the printing. My bad, as I did not know this part.

Please don't get me wrong, I do appreciate their business model and can see their position rapidly growing. Still to some extent printing is printing and if structure concerns can be met, then the OP, others and I can print larger models

[Updated on: Thu, 12 April 2012 17:35 UTC]

Re: Wall Thickness (plain wall vs. supported wall) [message #46871 is a reply to message #46868 ] Thu, 12 April 2012 18:13 UTC Go to previous messageGo to next message
avatar stonysmith  is currently offline stonysmith
Messages: 2094
Registered: August 2008
Go to my shop
Shapie Expert
moderator
Per their advice, I have on more than one occasion sent files to customer service and requested a "Thin walls check", which they've been happy to do for free. It might not be the "final" check done by the production team, but I've gotten pretty good results from their advice.

I have suggested, and I still maintain that they should charge a SMALL fee for doing this service (like $1 or $2). Doing this check consumes some manhours of their time, and I'd like to see them maintain profitability.. I'd like to see everyone at Shapeways get a raise at the end of the year. Very Happy


Patience, Persistance, Politeness - the 3Ps will help us get us to Perfect Printed Products
Re: Wall Thickness (plain wall vs. supported wall) [message #46931 is a reply to message #46871 ] Fri, 13 April 2012 13:10 UTC Go to previous messageGo to next message
avatar lensman  is currently offline lensman
Messages: 1629
Registered: December 2009
Go to my shop
Shapie Expert
stonysmith wrote on Thu, 12 April 2012 18:13

I have suggested, and I still maintain that they should charge a SMALL fee for doing this service (like $1 or $2). Doing this check consumes some manhours of their time, and I'd like to see them maintain profitability..


Agreed. I suggested this one-on-one model checking a long, long time ago and said they should charge a nominal fee for it. It would save a lot of embarassment when a customer orders one of your models only to get it cancelled as unprintable.

Glenn


Glenn ------ My Website Third Dimension Jewellery
Re: Wall Thickness (plain wall vs. supported wall) [message #46944 is a reply to message #46931 ] Fri, 13 April 2012 14:33 UTC Go to previous message
avatar stop4stuff  is currently offline stop4stuff
Messages: 3154
Registered: June 2010
Go to my shop
Shapie Expert
The thing is though, anyone can do their own checks and know a model will pass by using NetFabb Basic's measuring tools - ok this may be a bit of a work-up for colour models especially if the model needs repairing after being coloured.

I suggest that a premium thin wall check service should cost no less than $10 per check. Twisted Evil Rolling Eyes


 
   
Previous Topic:wow! what a detail...
Next Topic:What Has Changed?

Logo

Hello.

We're sorry to inform you that we no longer support this browser and can't confirm that everything will work as expected. For the best Shapeways experience, please use one of the following browsers:

Click anywhere outside this window to continue.