Results 1 to 9 of 9

Thread: How do I fix an Aspire model with some unintended negative z values?

  1. #1
    Join Date
    Oct 2009
    Location
    Elgin Illinois
    Posts
    706

    Default How do I fix an Aspire model with some unintended negative z values?

    Hello Shopbot friends:

    I am designing a 3D model, and I noticed that the model went .266" below the "Z plane" when it should not go below zero at all. I found a hand full of parts in this complicated model (it has 118 model components in 11 different levels) which have slightly negative minimum z values). When each of these parts are "turned on", the over-all Z height of the model changes, as the base gets pulled under the zero modeling plane.

    When I had a similar problem 2 years ago, I was told that I had messed up by using negative base heights and negative tilt angles (as shown in "component properties"). My model was so loused up (for this and resolution problems), I had to throw it away, costing me to lose over 40 hours of work. Ever since, I have NEVER used a negative base height, or a negative "tilt" angle. And now I am fearing another messed up, unrepairable model.

    Trying to isolate the source of my little negative offsets (which add up), I turned off different levels, and found that all of the negative offset came from one level. But within that one level, it seems as though 4 parts (not repeatably the same) contributed the negative offset. I spent over an hour, just trying to narrow down which parts were giving me the negative offset, and couldn't really identify which parts, as it was cumulative in small increments between several parts and the changes in negative offsets associated with each suspect component were not consistent. But it seems as though each suspect part had used a "tilt". (Again, NONE were negative, and I have avoided negative values like the plague!)

    I can't have my model change heights depending upon which parts I turn on or off. But I can't figure out how to identify this, and I don't know why this has happened since I have used no negative tilts or base heights in making this model. Looking at each part's component properties shows no negative values, but I can find small negative values by using the "Scale Z Height of Model" tool to inspect each part. This is murderously tedious.

    What do I do? Why do I get these tiny negative offsets on individual components, when I haven't used subtractions, negative tilts or negative base heights??????? Can I block the minimum Z value of my parts from going negative?

    Thank you for your suggestions! Chuck
    Attached Images Attached Images
    Chuck Keysor (circa 1956)
    PRT Alpha 60" x 144" (circa 2004)
    Columbo 5HP spindle
    Aspire 9.0, Rhino 5

  2. #2
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,986

    Default

    Chuck,
    It could be something as simple as those components need to be moved up or down the tree/stack in order to get the right combination. If you want to email the Aspire file I'll take a look when I get a minute today.

    Check your PM.

    -B
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

  3. #3
    Join Date
    Dec 2000
    Location
    Thorp, WI
    Posts
    2,845

    Default

    Chuck, I've had similar results when tilting a component from the center of the component. When you do this, it tilts the component so that one side raises above the plane and the other goes below the plane.
    Scott




  4. #4
    Join Date
    Oct 2009
    Location
    Elgin Illinois
    Posts
    706

    Default

    Thanks Brady and Scott!

    I've sent Brady the file.... And Scott, what you said makes sense and I hadn't thought of that. I have to go now due to an obligation, but later tonight I'll go and play with the model. I know that some of the problem parts had been "baked" so I couldn't tell if they had been tilted or not.

    Is there some way to tell a part to use only its absolute value, and therefor avoid this pesky negative dipping below the Z-zero plane?

    Thanks again, Chuck
    Chuck Keysor (circa 1956)
    PRT Alpha 60" x 144" (circa 2004)
    Columbo 5HP spindle
    Aspire 9.0, Rhino 5

  5. #5
    Join Date
    May 2009
    Location
    Philadelphia PA
    Posts
    93

    Default

    I would suggest that you use the "Create Component from Visible Model" tool and then use the "Replace Below" gadget to remove any unwanedt "material" that is below the Z-Zero plane.
    Michael

  6. #6
    Join Date
    Oct 2009
    Location
    Elgin Illinois
    Posts
    706

    Default

    Thanks Michael! I am off to a late start, but I will soon get back to my model to apply what you and Brady have recommended! Thanks so much! Chuck
    Chuck Keysor (circa 1956)
    PRT Alpha 60" x 144" (circa 2004)
    Columbo 5HP spindle
    Aspire 9.0, Rhino 5

  7. #7
    Join Date
    Oct 2009
    Location
    Elgin Illinois
    Posts
    706

    Default

    Hello Shopbot friends:

    a) I installed the "Replace Below" Gadget, and watched the related video.
    b) I noticed that there is a new version of Aspire, so I down-loaded that.
    c) Then I went to open my file so that I could then go to experiment with my new "Replace Below" gadget.
    But first, I moved around my cursor, just checking out the Z heights as shown at the bottom of the page, and to my utter bewilderment, I saw huge negative Z values, like -1.5inches! I selected the top component in my design, and used the "Scale Model Height" tool, and it showed the minimum Z height was -.0106, and certainly not -1.5 inches..... Then I turned the design on its side, and looked at it from the positive x direction, and very clearly, the vast majority of the part is entirely positive, making the -1.5 inch Z height value an absolute impossibility.
    Any idea as to what could have gone wrong???? Was this just a coincidence that I installed the newest Aspire update/patch before my z height measurements went out of whack????
    Thanks,,,,,,,,,,,,, again, Chuck


    Attached Images Attached Images
    Chuck Keysor (circa 1956)
    PRT Alpha 60" x 144" (circa 2004)
    Columbo 5HP spindle
    Aspire 9.0, Rhino 5

  8. #8
    Join Date
    Jan 2004
    Location
    iBILD Solutions - Southern NJ
    Posts
    7,986

    Default

    Chuck,
    It's not all that out of the ordinary for the Z value to be off or unreliable (at the bottom 'DRO') at times. Most of the time it's good, but there have been times it was off for me...and I've come back and it was OK. Go figure. I would not rely on it as the final word. Instead, select the component you want to flush below the zero plane, get its specs from the relief height tool and use the Gadget and so on.

    There comes a point with a model like you show where it gets entirely too complex to keep individual components all broken out. At that point, I save my original file (with everything broken out) and save a new one with everything merged or components that don't need further transformation merged/baked. This makes things a lot easier and WAY faster. You can tilt the whole model in one go or use a 2RS to kick it up on either a planar angle or bezier shape to get the perspective you need. You can always go back to the saved previous version should you reach a dead end on the workflow or mess up.

    There also comes a point where you reach diminishing return with modelling - often times something small you were worried sick about on the front end doesn't even show up on the CNC or if it does, it is so faint that only you know it is there. You have to 'zoom out' on the project scope once in a while to have perspective...This is a very hard lesson to learn with this stuff- where to be a perfectionist & where to let it ride. All of this of course is within the context of making money...not sure if this is for profit or charity work.

    -B
    High Definition 3D Laser Scanning Services - Advanced ShopBot CNC Training and Consultation - Vectric Custom Video Training IBILD.com

  9. #9
    Join Date
    Oct 2009
    Location
    Elgin Illinois
    Posts
    706

    Default

    Thanks Brady! Sorry for my slow reply,,,,, I keep getting tied up in other things, especially on the weekends......... But I appreciate your reply......

    a) My recent post was about the big negative Z height values for the cursor position....... I find that I get those readings to be as I would have historically expected, IF I check on Job Set-up, Z-Zero Position, Machine Bed! So either I am crazy (a real possibility) or the software patch I installed minutes before I started getting these unexpected readings, changed the way the cursor position operates. Well, I'll move on, which-ever the case may be.

    b) This is effectively a non-paying job. I saw the opportunity to design and make this great carving. I wanted to see if I could do it, to see if I liked doing it, and to see how long it would take (so I could estimate the cost of other similar work), and use this job as a selling tool IF I wanted to make more of these.

    I did make a kind of similar carving for an 1880's house 3 years ago, but it was a historical copy of a very abstract design, based on a really crummy photo. So there was no real right or wrong, as long as it looked like it could be the thing in the fuzzy, torn, blurry photo............ So that was simple compared to this........ However, my interest isn't in making blobby non-descript things, but to make fancy detailed carvings, which in my neck of the woods, is what most of the old fancy houses had. With so little experience designing and making such a non-trivial carving, I am still at a bad place on the learning curve. And for that reason, your experientially based advice is greatly appreciated.

    Thanks again, Chuck
    Chuck Keysor (circa 1956)
    PRT Alpha 60" x 144" (circa 2004)
    Columbo 5HP spindle
    Aspire 9.0, Rhino 5

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
  •