Exocad Model coming out "muddy"

[vcapone]

I'm new to Exocad model builder and when trying to make solid models the analogs I have set in the scan become very muddy and unclear after I hit the Run button. Any reason as to why it does this? I'm not changing the setting because I was told not to, but if it helps prevent the muddiness I will.

 

[LarryRDC]

Sorry that didn't work. I understand, it's a PITA when you can't get things to work the way you want! Who is your Exocad reseller? Maybe they can try and change the parameters to help get what you want and not have any errors.

 

[CoolHandLuke]

If you want to print models with the analog socket holes you’ll need to make them from the project designs that used scan markers and the model analog libraries that are related to them.

the correct way? nah bro just make some crap and remake it a few times, then claim the "technology isnt there yet"

 

[vcapone]

I guess right now the solution is to make the models without the analogs, bring in the analogs in as a "Generic visualization mesh" and save out the scene as an STL, then use Netfab to repair it to one shell... Exocad still muddies the models when it generates which is annoying, why can't it just work like its supposed to?

 

[CoolHandLuke]

cause you didnt use scan bodies bro wtf

 

[Artin]

cause you didnt use scan bodies bro wtf

I thought I was the only one losing it out here

 

[Exelero]

I can understand that it's cheaper to print digital analogs than to actually buy and insert them into a correctly made model, but I would not trust a model made this way. Resin analogs don't provide the necessary stability and precision to do a decent job.

And yes, sometimes exocad does have a hard time making models when you try to align multiple meshes.

 

[sirmorty]

But...but... Something something. 3 point alignment.

No point trying to save a dollar or two.. you are better off ordering the analogs and ask the Doc to give them back to you after they are done with the model. Usually they just throw them out in the trash.

 

[CWilliams]

or maybe, JUST maybe, instead of 2876 steps, you make a simple order for the right parts necessary and charge the doctor accordingly, like maybe how it's supposed to work. Result? You won't lose a dollar, and wont have to try to manipulate software into cutting your corners, while wasting a bunch of time. But by all means Dr, torque those screws in WWE style bc your model cant' be verified.

 

[bigj1972]

Good thread

 

[vcapone]

We don't use scanbodies for the application we make at our lab. I understand that the Exocad would make the analog holes for physical analogs, but that's not the issue here, the issue is Exocad makes the model muddy and unusable and there has to be a solution to that. The question isn't "why am I not using scanbodies?" the question is "why doesn't Exocad model builder work properly, how do we get it to output the file 1:1?"

Here's a file ran through 3Shape model builder for a test and it looks just fine because it didn't alter or muddy anything. Before we get the "then just use 3Shape" comments, it still doesn't solve the issue of Exocad not outputting the file 1:1

If you can't wrap your head around the no scanbodies thing then don't bother because that's not going to happen, its not needed in our scenario.

 

[Car 54]

We don't use scanbodies for the application we make at our lab. I understand that the Exocad would make the analog holes for physical analogs, but that's not the issue here, the issue is Exocad makes the model muddy and unusable and there has to be a solution to that. The question isn't "why am I not using scanbodies?" the question is "why doesn't Exocad model builder work properly, how do we get it to output the file 1:1?"

Here's a file ran through 3Shape model builder for a test and it looks just fine because it didn't alter or muddy anything. Before we get the "then just use 3Shape" comments, it still doesn't solve the issue of Exocad not outputting the file 1:1

If you can't wrap your head around the no scanbodies thing then don't bother because that's not going to happen, its not needed in our scenario.

Gotta hand it to 3Shape, that scan, model, does look good.

 

[CWilliams]

You're comparing two different software's, expecting them to play the same. All while complaining you don't like ones set of operating parameters, that it needs to operate like the other and then getting mad that its not OpERatINg HoW iT SHouLD. If you can't hack it how you want, then call up your service rep. Pick up what people are putting down- there are helpful solutions offered to aide your problem that allow you to stay within the boundaries of Exocad- you just don't like them because it's not what you want to hear and that sounds like a you problem.

The 3shape model does look nice, but still the point stands- it's resin.

 

[CoolHandLuke]

you see the little black bits on your scan here

those are called non-Manifold triangles or Nodes. they are rendered on screen as complete surfaces, but if you were able to see the polygon's triangle normals or view as a wireframe, you'd find these are inside out, or dots in space joining edges, or just lines leading to points in space with no actual material. https://www.sculpteo.com/en/3d-learning-hub/create-3d-file/fix-non-manifold-geometry/

exocad's model program prevents you from trying to build a model with non-manifold edges. this is because it is actually possible to have an edge so large that it does not physically join the rest of the model except on screen.

you are comparing render quality with print quality. you are then comparing another company's sloppy render-of-a-print with exocad.

go ahead and try to print the 3shape model. it will absolutely NOT print as rendered.

exocad simply shows you true to life results by eliminating non printable geometry.

if you had used scan bodies you'd have real analogs in place. you arent complaining about tissue quality from any of the posts, just the implant quality.

now you may be asking yourself why does my scan have a non-manifold edge or node? implant geometry manufacturers aabsolutely do not want you to get quality prints because they want you to buy their parts to put in your model instead. with the implant companies parts in your model the quality is assured. if you just 3d print whatever you want, you can do like you have been doing - complain online about the parts not being good quality, and shame the wrong people through ignorance.

so go ahead, print that 3shape model. 3shape will not support you when you can't get good physical quality.

exocad simply renders your print most true to life.

 

[Car 54]

you see the little black bits on your scan here

those are called non-Manifold triangles or Nodes. they are rendered on screen as complete surfaces, but if you were able to see the polygon's triangle normals or view as a wireframe, you'd find these are inside out, or dots in space joining edges, or just lines leading to points in space with no actual material. https://www.sculpteo.com/en/3d-learning-hub/create-3d-file/fix-non-manifold-geometry/

exocad's model program prevents you from trying to build a model with non-manifold edges. this is because it is actually possible to have an edge so large that it does not physically join the rest of the model except on screen.

you are comparing render quality with print quality. you are then comparing another company's sloppy render-of-a-print with exocad.

go ahead and try to print the 3shape model. it will absolutely NOT print as rendered.

exocad simply shows you true to life results by eliminating non printable geometry.

if you had used scan bodies you'd have real analogs in place. you arent complaining about tissue quality from any of the posts, just the implant quality.

now you may be asking yourself why does my scan have a non-manifold edge or node? implant geometry manufacturers aabsolutely do not want you to get quality prints because they want you to buy their parts to put in your model instead. with the implant companies parts in your model the quality is assured. if you just 3d print whatever you want, you can do like you have been doing - complain online about the parts not being good quality, and shame the wrong people through ignorance.

so go ahead, print that 3shape model. 3shape will not support you when you can't get good physical quality.

exocad simply renders your print most true to life.

Nice informative post, CHL, so the real deciding factor can be the printed model in hand and why Exocad and 3Shape would be different in that regard. Some very sharp members here who replied with all of their posts in this thread.

 

[sirmorty]

Thanks for the explanation Luke.

It drives me a bit crazy that people think that whatever they seen on screen can be manufactured.
Or that what they see on screen hasn't been rendered and "fixed". Looking at you IO scans.

Sure your scanner is better or more accurate.
It's usually just the software fixing the mesh so you don't see that hole that it can't scan.
But sure tell yourself that it's better.

 

[vcapone]

I'm not expecting it to print perfectly? I'm well aware it won't, but the 3Shape file will print better than the botched file Exocad provides. I will continue to save the scene as an STL because that's been working fine for what we need the model for. It just baffles me that that's the quality Exocad puts out and there's no way to solution to prevent that...

 

[Toothman19]

 

[Karl60]

Have you checked the scan resolution settings in Exocad?

 

[tuyere]

I'm not expecting it to print perfectly? I'm well aware it won't, but the 3Shape file will print better than the botched file Exocad provides. I will continue to save the scene as an STL because that's been working fine for what we need the model for. It just baffles me that that's the quality Exocad puts out and there's no way to solution to prevent that...

If I understand this all right, this isn't my area of expertise: You are ultimately getting bad results because you are not following the intended and assumed workflows for the software. It doesn't matter if it 'works for you', the software isn't designed around your kludgy workflow specifically, but one that is widely-practiced, gets good results for a minimum of labour, is supported by manufacturers and complementary software, etc etc.
Consider an example from my own lab: I get bad results when I try to mill blocks, prisms, geometrically-defined forms etc using millbox, which I have to do to make parts for other departments occasionally, it's frustrating. It's also not surprising at all, because I'm nesting a block as an onlay and the toolpaths are intended for 3d contoured, organic forms, and not prisms with flat planes and 90-degree angles; the tools are ball-nose endmills intended for 3d contour milling and not the squared-off endmills you want for more conventional milling operations, and so on. This is not the fault of the software, I'm just using it in a weird and awkward way the developers did not anticipate or intend.


You are trying to drive nails with a wrench and complaining that the wrench isn't good at its job. Or, more accurately + glibly, you're trying to drive nails with a wrench and complaining that it doesn't work nearly as well as driving nails with a crowbar does. You're imposing a workflow and functionality requirements that neither tool is really intended for, and one of them may appear to work better than the other, but neither of them is going to do a good job because of how you're using them. To get good results you either need to use the tool how it's intended to be used, or pick a more appropriate tool.

 

[Exelero]

You could check the 3Shape model in Autodesk Meshmixer using the Analysis and Inspector option. It will highlight errors in the model that are hard to notice at first sight. I always use this program to find repair mesh defects before printing.