Jump to content

Welcome to eMastercam

Register now to participate in the forums, access the download area, buy Mastercam training materials, post processors and more. This message will be removed once you have signed in.

Use your display name or email address to sign in:

Glad I'm Paid By The Hour


pro grammer
 Share

Recommended Posts

7 minutes ago, pro grammer said:

MC2020 stock model op in process for the last 30 minutes. 21 ops. 6 of them being simple flowline surfaces. 1 being a small optirough. Part is about 4" x 3" x 2". 2nd attempt. Tasked out of the 1st thinking it locked up.

Run it through Verify

Save As STL file

Use the STL file to define the next stock model

You lose the associativity but it's much quicker and much more stable

  • Like 4
Link to comment
Share on other sites

Tolerance and overall filesize of the NCI for the ops? 

Stock model generation set to "5 axis tip only" in Source Operations page?

If you need the holder in the Stock Model generation and cannot turn the above option on, is it an imported holder with a million segments to represent radii/etc you don't need? Can it be simplified? Same question, but for the tool- is it geometry linked to a level? These are probably the most common causes of massive generation times that I see.

 

  • Like 1
Link to comment
Share on other sites
13 minutes ago, Chally72 said:

but for the tool- is it geometry linked to a level?

so tool defining geometry linked to a level, or an external file (stp, dxf, mcam) will have a negative influence stock model processing time?

Link to comment
Share on other sites
26 minutes ago, gcode said:

so tool defining geometry linked to a level, or an external file (stp, dxf, mcam) will have a negative influence stock model processing time?

The answer is usually "yes" because of how most tooling vendors supply files for tools and holders. A Ø0.250 ball mill created from scratch in the Tool Manager might consist of 4 entities that make up the cross section to sweep, but download a Sandvik or Iscar 2D DXF and that same straight shank ball mill might have 40-60 entities, with the shank itself made up of a few dozen small linear segments where one line could have done it. These models are often exported en-masse from their native software for these generic online libraries, and leave a lot to be desired.

Every added facet in that tool or holder is an added surface that has to be drawn and checked against at every single point in each toolpath, so especially in the case of a complicated holder that models collet nut back angles and radii, you're adding potentially hundreds to thousands of surfaces to the calculation overhead, vs what is normally a 10-20 surface calculation for an extremely clean and basic holder and tool.

Here's an example of a holder from a recent customer file where we were doing some complex paths and recalculating rest models quite a bit:

Holder.jpg.4aa9e2b79ce72f96f877ee7c3bc3a317.jpg

This is made up of over 150 segments. Realistically, we don't need to calculate against the lower collet nut undercut because we'll never actually use that for clearance on the part. Similarly, the upper gradual radius that consists of 20+ segments could perhaps be simplified to a single taper if we weren't cutting clearances that close. 


So, if we were doing a lot of regeneration and testing and HAD to use the holder as part of the stock model calculation and not flip that "5 axis tip only" switch, (which calculates the stock based on the flutes of the tool only and thus exponentially speeds up calc time) I would create a simplified version of this holder to swap in and use for the operations that the stock model references- which is exactly what I did on this one.

Hope this helps explain a bit more about streamlining for stock models.

 

  • Thanks 2
  • Like 6
Link to comment
Share on other sites
2 hours ago, pro grammer said:

MC2020 stock model op in process for the last 30 minutes. 21 ops. 6 of them being simple flowline surfaces. 1 being a small optirough. Part is about 4" x 3" x 2". 2nd attempt. Tasked out of the 1st thinking it locked up.

Have you set "Mastercam.exe" Processor Priority in the Windows Task Manager? (Setting to 'High' can boost performance.)

Same question for the "View > Managers > Multi-Threading > Multi-Threading Manager"? Inside that dialog box, you can set the "maximum number of threads" (Set '1 per physical core on your machine'), and also set the 'Default priority' for each Thread. (Set it to 'High'!!!)

  • Thanks 1
  • Like 2
Link to comment
Share on other sites
14 hours ago, Colin Gilchrist said:

Have you set "Mastercam.exe" Processor Priority in the Windows Task Manager? (Setting to 'High' can boost performance.)

Same question for the "View > Managers > Multi-Threading > Multi-Threading Manager"? Inside that dialog box, you can set the "maximum number of threads" (Set '1 per physical core on your machine'), and also set the 'Default priority' for each Thread. (Set it to 'High'!!!)

Yes. I already set the Priority to high. I am also using all cores. This usually only happens on a few parts. I just tasked out then imported an stl from verify to get it done.

16 hours ago, gcode said:

Run it through Verify

Save As STL file

Use the STL file to define the next stock model

You lose the associativity but it's much quicker and much more stable

Yup. Ended up doing that. I try to avoid it when I can.

Link to comment
Share on other sites

"Stock Tolerance" is a double-edged sword with Stock Models.

The reason is that if you 'open up the tolerance' for creating the initial stock model shape, then you end up with a model which geometrically 'not geometrically accurate'.

Ideally, what we want is to be able to control the "tessellation tolerance". Tessellation is the process of taking the 3D model, and "meshing it", or creating a "skin of the part, made up as a collection of triangles" with a "normal direction" on each triangle, to indicate the "outside vs. inside" of the STL model.

The issue is that when you change the Stock Tolerance, the "outside shape" can change size, relative to the model that you are choosing. (Solid or surface model)

I have found that the "STL Export" function (independent of the Stock Model), can actually output a "geometrically accurate STL", while the STL Tolerance essentially just changes the "size of the triangles" that are created. However, when you export the STL, the vertex points of all of those triangles will "physically lie on the model surfaces/edges". This isn't necessarily true with the Stock Model creation function (for initial stock only).

Basically, if you take the step of independently exporting your model (surface or solid) as an STL File, and then "Importing the STL as a Mesh Entity", you are able to control the tessellation tolerance for the triangle size, while having the skin of the model still be 100% physically coincident with the model that was used to create the mesh.

As an example, to show you what I'm talking about, try the following procedure:

  1. Take any "moderately complex" model, and perform 4 different STL Exports: one at 0.040", one at 0.010", one at 0.004", and one at 0.001".
  2. Export the STL File Names (add the tolerance to the file name to keep the files separate), and make sure you use the "options" button to change the export tolerance.
  3. Use File > Merge to Import the STL models. Use a new Level for each Import. Import each model 2 different times. One as a STL Mesh, and one as 'Lines'. This will give you 8 different levels. 4 with mesh entities, and 4 with "lines".
  4. By importing the STL as 'lines', you can physically measure distances, and also create arcs (3 points method).
  5. Now you can examine the results and see how the models will change the accuracy of how the curves and holes are represented, but the linear surfaces and flat faces (edges), are still matching between all 4 different export tolerance values.

You'll see that the models will still measure "accurately" for linear distances (flats and surfaces will be coincident with the original model that you used to create the mesh.)

By Exporting/Importing manually, you will remove the 'Stock Model Mesher' from the equation, and will have full control over the accuracy and tessellation tolerances used to create the initial stock shape.

I will create an 'initial stock shape' that uses as few triangles as possible, while still giving me the accurate rendering of curved surfaces (and holes!).

  • Like 2
Link to comment
Share on other sites
14 hours ago, jlw™ said:

you can open the tolerance to .002 and make it much faster.

If you're copying stock models, always set it = last model then only have to apply new toolpaths.

image.thumb.png.5af961e6a30a390ba37a9e0c26e1a922.png

I had already changed the tolerance to .005.

11 hours ago, Colin Gilchrist said:

"Stock Tolerance" is a double-edged sword with Stock Models.

The reason is that if you 'open up the tolerance' for creating the initial stock model shape, then you end up with a model which geometrically 'not geometrically accurate'.

Ideally, what we want is to be able to control the "tessellation tolerance". Tessellation is the process of taking the 3D model, and "meshing it", or creating a "skin of the part, made up as a collection of triangles" with a "normal direction" on each triangle, to indicate the "outside vs. inside" of the STL model.

The issue is that when you change the Stock Tolerance, the "outside shape" can change size, relative to the model that you are choosing. (Solid or surface model)

I have found that the "STL Export" function (independent of the Stock Model), can actually output a "geometrically accurate STL", while the STL Tolerance essentially just changes the "size of the triangles" that are created. However, when you export the STL, the vertex points of all of those triangles will "physically lie on the model surfaces/edges". This isn't necessarily true with the Stock Model creation function (for initial stock only).

Basically, if you take the step of independently exporting your model (surface or solid) as an STL File, and then "Importing the STL as a Mesh Entity", you are able to control the tessellation tolerance for the triangle size, while having the skin of the model still be 100% physically coincident with the model that was used to create the mesh.

As an example, to show you what I'm talking about, try the following procedure:

  1. Take any "moderately complex" model, and perform 4 different STL Exports: one at 0.040", one at 0.010", one at 0.004", and one at 0.001".
  2. Export the STL File Names (add the tolerance to the file name to keep the files separate), and make sure you use the "options" button to change the export tolerance.
  3. Use File > Merge to Import the STL models. Use a new Level for each Import. Import each model 2 different times. One as a STL Mesh, and one as 'Lines'. This will give you 8 different levels. 4 with mesh entities, and 4 with "lines".
  4. By importing the STL as 'lines', you can physically measure distances, and also create arcs (3 points method).
  5. Now you can examine the results and see how the models will change the accuracy of how the curves and holes are represented, but the linear surfaces and flat faces (edges), are still matching between all 4 different export tolerance values.

You'll see that the models will still measure "accurately" for linear distances (flats and surfaces will be coincident with the original model that you used to create the mesh.)

By Exporting/Importing manually, you will remove the 'Stock Model Mesher' from the equation, and will have full control over the accuracy and tessellation tolerances used to create the initial stock shape.

I will create an 'initial stock shape' that uses as few triangles as possible, while still giving me the accurate rendering of curved surfaces (and holes!).

Accuracy was not important to me. I am merely creating pictures for setup sheets.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.

Join us!

eMastercam - your online source for all things Mastercam.

Together, we are the strongest Mastercam community on the web with over 56,000 members, and our online store offers a wide selection of training materials for all applications and skill levels.

Follow us

×
×
  • Create New...