Review of Tormach and new PCNC 1100


Page 1 of 2 12 LastLast
Results 1 to 20 of 23

Thread: Review of Tormach and new PCNC 1100

  1. #1
    Registered
    Join Date
    Aug 2011
    Location
    USA
    Posts
    6
    Downloads
    0
    Uploads
    0

    Default Review of Tormach and new PCNC 1100

    Hello,

    I am writing to share my experiences with ordering, setting up and getting started with a Tormach PCNC 1100. It might be helpful to others who are thinking about getting one. Forgive the long story. You may want to skip to the conclusion at the end.

    My background: degreed mechanical engineer (CAD and FEA with 15 yrs experience). Close to zero machining experience. Good with tools.

    Goal: To be able to machine complex 3D shapes in soft materials, like wood and polyurethane modeling boards, machinable wax, maybe aluminum. Before buying I spent a couple of weeks using the SprutCAM demo software proving that I could mill a complex surface STL file that is typical of what I want to do.

    Milling machine: chose PCNC 1100 over 770 for the larger work envelope. Gave up higher speeds. Chose the standard package, except no Kress, 4 jaw chuck, touch tool and edge finder.

    Odering: Tormach customer support was excellent: they answered my questions promptly and helped me select the components- even telling me I didn't need some of the stuff I thought I needed. The UPS and freight shipments arrived 7-10 days after they recieved my check.

    Other stuff: I bought a selection of carbide end mills from Mari and American Carbide. I got a sample kit of polyurethane foam materials from Freeman Supply.

    Setting up: Ran a new 220V circuit in the garage. Rented an engine lift. Followed Tormach directions and watched u-tube videos of assembly. With only a little help from my 120lb girlfriend, I had the PCNC on the stand and fully assembled in one day. Spent another day or so installing the 4th axis, organizing the tooling, becoming familiar with all the parts.

    Getting started: The following weekend I set up the computer, became familiar with the Mach3 controller, jogged all 4 axes and did the "First Part" exercise. Cool! Everything was working perfectly.

    The first snag: I wanted to try a 3 axis program with a STL file. With some learning, trial and error, the SprutCAM program looked good. But when the cutting began, oops, not what I expected, not the right shape. My first phone call to Tormach Tech Support. Eric was very helpful and easy to work with. I sent him the SprutCAM file and the G-code. He quickly diagnosed the problem: I was using an old version of the PCNC 1100 post-processor that has a bug for 3 axis programs. He sent the new one and it worked perfectly. Thank you Eric and Tormach! Back in business.

    The second snag: I wanted to try a small demonstration of a 4 axis program with a STL file. With some learning, trial and error the SprutCAM program looked good: 4 indexed roughing operations and a rotary machining finishing operation. The A-axis indexing operations worked fine. But 5 min into the finishing operation, the continuous A-axis motion got stuck. Stop the machine! Looked in the manual for the 4th axis. Checked rotary table motor coupling. Found that a small allen wrench set screw had worked loose and prevented the motor from turning. Followed directions, disassembled the motor coupling, tightened both set screws, reassembled. Back in business.

    The third snag: New workpiece, same 4 axis G-code for STL file. Indexed roughing operations went fine. A-axis finishing operation did not cut the right shape. Contacted Tormach Tech Support. Eric was very responsive again. He suggested that I change the SprutCAM setting from Circular trajectory to Linear. This is where I am now. I will experiment with his suggestion and indexed finishing operations over the next few days.

    My conclusions: Tormach is a great company to work with. Customer support is excellent. They will help all along the way, from ordering through hardware and software issues. The PCNC 1100 is very well designed and a good value. SprutCAM is powerful and great for my needs. As should be expected with such a complex undertaking there have been some challenges. But no show stoppers. The PCNC 1100 has given me the capability to do some amazing things! Thank you Tormach!

    Wayne

    Similar Threads:


  2. #2
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Post some pics of the STL files. I gave up on SprutCAM and STL files. That was my only disappointment is I bought SC for it's STL import but it crashed miserably every time I tried one. I gave up on the project but if you have STL's working I may give it another shot...



  3. #3
    Registered
    Join Date
    Aug 2011
    Location
    USA
    Posts
    6
    Downloads
    0
    Uploads
    0

    Default STL files

    Hello Magnum164,

    I have tried several STL files. The first one caused SprutCAM to crash occasionally due to high memory usage (over 2GB). Since SprutCAM is a 32 bit application, even in 64 bit Windows 7 it cannot allocate more than about 2GB of RAM.

    Since then I've been more careful about the number of polygons in the STL file. I start with an OBJ file from my modeling software because I am doing organic shapes. The OBJ file is converted to a STL file using MeshLab (good free open source software). My most recent OBJ file has about 90,000 polygons. That leads to a SprutCAM memory usage of about 250MB, which is fine. I expect to slowly increase the number of polygons in my models as I refine the milling process and figure out how much detail is necessary.

    If high polygon count is not what's causing your problem, give me some more details and I'll try to help.

    Wayne

    Quote Originally Posted by Magnum164 View Post
    Post some pics of the STL files. I gave up on SprutCAM and STL files. That was my only disappointment is I bought SC for it's STL import but it crashed miserably every time I tried one. I gave up on the project but if you have STL's working I may give it another shot...




  4. #4
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by DrDucati View Post
    Hello Magnum164,

    I have tried several STL files. The first one caused SprutCAM to crash occasionally due to high memory usage (over 2GB). Since SprutCAM is a 32 bit application, even in 64 bit Windows 7 it cannot allocate more than about 2GB of RAM.

    Since then I've been more careful about the number of polygons in the STL file. I start with an OBJ file from my modeling software because I am doing organic shapes. The OBJ file is converted to a STL file using MeshLab (good free open source software). My most recent OBJ file has about 90,000 polygons. That leads to a SprutCAM memory usage of about 250MB, which is fine. I expect to slowly increase the number of polygons in my models as I refine the milling process and figure out how much detail is necessary.

    If high polygon count is not what's causing your problem, give me some more details and I'll try to help.

    Wayne
    A program should not crash due to high memory usage, with system paging it should give you a notice long before it crashes stating you are running low on memory that you need to free up more memory. Been down that road already. A crash is generally due to an error in the program.

    I did have fairy large files and using small end mills which would produce a LOT of cutting paths. SC's memory management was just not what I thought it should be.

    For now I no longer use STL files, instead converting to true organic models and creating IGES files for the CAM. More time, but less trouble in the end.



  5. #5
    Registered
    Join Date
    Aug 2011
    Location
    USA
    Posts
    6
    Downloads
    0
    Uploads
    0

    Default A couple of questions about STL files

    Yes, SputCAM shouldn't crash so abruptly with large STL files. But I've found that it works OK with my models so far.

    I'm trying to find the best strategy for tool paths: rotary, waterline, plane, etc. Do you have any advice on this? Or a reference?

    Could you tell me how you "convert to true organic models"? And how do you get IGES files? What is your starting CAD file format?

    I thought that OBJ was the standard format for interchange of organic surface geometry. And STL was the standard format for fabrication- stereolithography, 3D printing, etc.



  6. #6
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by DrDucati View Post
    Yes, SputCAM shouldn't crash so abruptly with large STL files. But I've found that it works OK with my models so far.

    I'm trying to find the best strategy for tool paths: rotary, waterline, plane, etc. Do you have any advice on this? Or a reference?
    No, I never got that far But I was planning to use the waterline as that is what I liked about SC and figured it would be best.

    Quote Originally Posted by DrDucati View Post
    Could you tell me how you "convert to true organic models"? And how do you get IGES files? What is your starting CAD file format?
    Actually I said that wrong, its converting from true organic. Starting format does not matter too much. I can import STL files into Rhino3D then edit and correct the mesh several ways, even reduce poly count if needed. Sometimes it's complicated and is basically retracing the STL (or OBJ) and placing meshes over the existing model (retopology). I can then import the Rhino file into Alibre or export as an IGES for CAM. I used to use 3d-coat a lot for retopolgy when I was doing game design.

    Quote Originally Posted by DrDucati View Post
    I thought that OBJ was the standard format for interchange of organic surface geometry. And STL was the standard format for fabrication- stereolithography, 3D printing, etc.
    OBJ files were originally Wavefront files, not really standard however since it was easily edited with a text editor, a lot of programs would read/write the format. The actually standard being developed and becoming more/more accepted is Collada. You are correct, STL was developed with 3D printing. only problem is a STL file is generally not great for subtractive type machines, works create for additive though. At least that is true for me. I prefer to convert to IGES.



  7. #7
    Member Don Clement's Avatar
    Join Date
    Jan 2007
    Location
    Running Springs, California USA
    Posts
    1332
    Downloads
    0
    Uploads
    0

    Default

    I use IGES exported from Solidworks to SprutCAM. In fact SprutCam is a Solidworks partner and a plug-in to Solidworks (32bit or 64bit version)automatically exports IGES files and calls up SC.

    from the SC website:
    The full associativity of SprutCAM machining to the SolidWorks design model reduces errors when the model changes and supports the process where updates are received for models already machined. When the geometry used to define a machining operation is changed in the SolidWorks design, SprutCAM enables the user to automatically synchronize all machining operations with the updated geometry. The system will make it easy to program all machines in one system and work towards lean manufacturing by optimizing both milling and turning methods, while minimizing the levels of software investment.



    Don



  8. #8
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    VisualMill is also a Solidworks partner, and will open Solidwork files directly as well.



  9. #9
    Gold Member MichaelHenry's Avatar
    Join Date
    Jun 2006
    Location
    Chicago suburbs
    Posts
    3063
    Downloads
    0
    Uploads
    0

    Default

    Thought I'd read somewhere that STL files model the object surface as a bunch of triangular meshes and that for that reason are better suited to lithography or 3D printing than they are to subtractive CNC milling. Is that the case or did I misunderstand?

    Mike



  10. #10
    Registered
    Join Date
    Aug 2011
    Location
    USA
    Posts
    6
    Downloads
    0
    Uploads
    0

    Default Two worlds of 3D

    Yes, STL files are composed of triangular meshes. But that is currently the only way to describe complex organic shapes for the purpose of fabrication, whether stereolithography, 3D printing or CNC milling.

    There are two worlds of 3D: mechanical design (CAD) and organic modeling. CAD software (SolidWorks, Pro/Engineer, Alibre, etc.) is geared toward simple parametric shapes (flat surfaces, cylindrical holes, simple rounds, etc). The most commonly used file format for machining (IGES) is suited for these types of features. IGES files can be converted to STL files for stereolithography or 3D printing but individual feature information is lost.

    Organic modeling, on the other hand, is geared toward complex non-parametric shapes (people, animals, trees, etc.). Modeling software (maya, modo, zbrush, etc.) is primarily used for visualization (movies, advertising, video games). The underlying geometry and the most commonly used file format for geometry exchange (OBJ) is composed of polygons (including triangles). OBJ files can be converted to STL files because the geometry is similar. OBJ files are not usually converted to IGES files because you can't take polygon data and extract simple features like flat surfaces, cylinders, rounds, etc.

    So, although STL files may not be ideal for CNC milling, they are the only game in town for organic surface models. Thankfully, CNC programs like SprutCAM, Mastercam, etc. can create tool paths from them. So we can fabricate organic shapes on our Tormach milling machines. Cool!



    Quote Originally Posted by MichaelHenry View Post
    Thought I'd read somewhere that STL files model the object surface as a bunch of triangular meshes and that for that reason are better suited to lithography or 3D printing than they are to subtractive CNC milling. Is that the case or did I misunderstand?

    Mike




  11. #11
    Gold Member MichaelHenry's Avatar
    Join Date
    Jun 2006
    Location
    Chicago suburbs
    Posts
    3063
    Downloads
    0
    Uploads
    0

    Default

    DrDucati,

    Thanks for the education. What happens if an IGES model of a part with organic features is used to mill the part? Does it fail to work, produce an inaccurate part, make zillions of short move segments, or something else?

    Mike



  12. #12
    Registered
    Join Date
    Aug 2011
    Location
    USA
    Posts
    6
    Downloads
    0
    Uploads
    0

    Default

    Mike,

    Not sure because I haven't been working with IGES, just STL. But my guess is that the CNC program will produce an accurate part, just as it does with STL files.

    Magnum164 says that he uses Rhino modeling software to generate IGES files for organic shapes. But I believe that this is unique to Rhino.

    I'm afraid I have exhausted (and maybe gone beyond) my level of expertise in these things.

    Wayne



  13. #13
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by DrDucati View Post
    Yes, STL files are composed of triangular meshes. But that is currently the only way to describe complex organic shapes for the purpose of fabrication, whether stereolithography, 3D printing or CNC milling.

    There are two worlds of 3D: mechanical design (CAD) and organic modeling. CAD software (SolidWorks, Pro/Engineer, Alibre, etc.) is geared toward simple parametric shapes (flat surfaces, cylindrical holes, simple rounds, etc). The most commonly used file format for machining (IGES) is suited for these types of features. IGES files can be converted to STL files for stereolithography or 3D printing but individual feature information is lost.

    Organic modeling, on the other hand, is geared toward complex non-parametric shapes (people, animals, trees, etc.). Modeling software (maya, modo, zbrush, etc.) is primarily used for visualization (movies, advertising, video games). The underlying geometry and the most commonly used file format for geometry exchange (OBJ) is composed of polygons (including triangles). OBJ files can be converted to STL files because the geometry is similar. OBJ files are not usually converted to IGES files because you can't take polygon data and extract simple features like flat surfaces, cylinders, rounds, etc.

    So, although STL files may not be ideal for CNC milling, they are the only game in town for organic surface models. Thankfully, CNC programs like SprutCAM, Mastercam, etc. can create tool paths from them. So we can fabricate organic shapes on our Tormach milling machines. Cool!
    Here are some examples....
    http://www.cnczone.com/forums/786135-post17.html

    And a good video on Organic modeling.

    The Secrets of Organic Modeling



  14. #14
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by MichaelHenry View Post
    DrDucati,

    Thanks for the education. What happens if an IGES model of a part with organic features is used to mill the part? Does it fail to work, produce an inaccurate part, make zillions of short move segments, or something else?

    Mike
    It will work fine. However the resolutoin may not be as detailed as in the original organic model. Organic models can have MILLIONS of triangles to describe the surface. To go to IGES you will have to reconstruct the part with fewer faces and you will loose detail. However, that is probably not a bad thing.

    Wanna see some very high poly Organic models?

    Pixologic :: ZBrush :: Gallery

    Now think of how you would make a mold for one of these



  15. #15
    Gold Member MichaelHenry's Avatar
    Join Date
    Jun 2006
    Location
    Chicago suburbs
    Posts
    3063
    Downloads
    0
    Uploads
    0

    Default

    Magnum164,

    Interesting models. I'd sure hate to have to design one of those in Alibre, let alone machine a mold.

    Seems like detail of the machined part would be limited by cutter size with a CNC mill and by resolution of the 3D printer. From what little reading I've done on 3D printers it looks like the hobby class of machines is limited to 0.003" or more of resolution. That might be good enough for the models on the the page you linked to, but I'd want something better for certain model engine parts.

    Different tools for different projects, of course, but it would be a lot of fun to play with a 3D printer.

    Mike



  16. #16
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    I do want a 3D printer for that very reason and a Color one at that.


    One thing that should be noted, is STL's for 3D printing have to have a thickness associated to the sides of course, while STL's for machining can just be the exterior surfaces.

    So for 3D printing, you will in effect have double the resolution by default.


    Quote Originally Posted by MichaelHenry View Post
    Magnum164,

    Interesting models. I'd sure hate to have to design one of those in Alibre, let alone machine a mold.

    Seems like detail of the machined part would be limited by cutter size with a CNC mill and by resolution of the 3D printer. From what little reading I've done on 3D printers it looks like the hobby class of machines is limited to 0.003" or more of resolution. That might be good enough for the models on the the page you linked to, but I'd want something better for certain model engine parts.

    Different tools for different projects, of course, but it would be a lot of fun to play with a 3D printer.

    Mike




  17. #17
    Member
    Join Date
    Jul 2004
    Posts
    595
    Downloads
    0
    Uploads
    0

    Default

    A 3D printer that resolves near .001 is about 100K. Our work just purchased one... really nice machine.

    David



  18. #18
    Gold Member
    Join Date
    Jun 2006
    Location
    Suffolk, UK
    Posts
    2512
    Downloads
    0
    Uploads
    0

    Default

    I guess the Chinese Knock-off's going to be about $159.95 then.

    Phil

    Quote Originally Posted by David Bord View Post
    A 3D printer that resolves near .001 is about 100K. Our work just purchased one... really nice machine.

    David




  19. #19
    Registered
    Join Date
    Mar 2010
    Location
    USA
    Posts
    816
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by David Bord View Post
    A 3D printer that resolves near .001 is about 100K. Our work just purchased one... really nice machine.

    David
    Would be interesting in knowing the brand/model...



  20. #20
    Member
    Join Date
    Jul 2004
    Posts
    595
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by Magnum164 View Post
    Would be interesting in knowing the brand/model...
    I was just told its a "Projet HD3000 plus"

    We make prototype fans, and parts and they are strong enough to be used for testing.

    David



Page 1 of 2 12 LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Review of Tormach and new PCNC 1100

Review of Tormach and new PCNC 1100