Modeled thread diameter problem

[Nick]

I am having an issue where when I model a thread on a part that is almost to diameter, the entire piece changes dimensions even though i am only doing a partial length.  An example is using 5/16 tubing modeled with a 5/16-18 thread on one end.  I set the base model to 5/16" nominal as the material would be purchased.  Then I modeled the 5/16-18 thread on one end.  When I made the print, the entire model has been reduced to the nominal of .307"

I am currently trying to made a threaded rod out of .156" material with an 8-32 thread on each end.  I will be doing a straight thread and do not care that the major diameter will be a few thousandths below spec.  However, as soon as I put the thread on, the entire rod becomes sizes up to the nominal .160".

I can model a short step up/down section on the ends to apply the thread to but it seems if the thread is not modeled full length, it should not affect the rest of the part diameter.  It's important on the manufacturing side that these changes do not occur resulting in unwanted changes of the part.  Am I missing a step here?

[Phil.E]

Hi,

We are aware of this problem. Thanks for asking about it.

Currently the best way to control this is to offset a workplane and use it to Split Face where you want the thread to end and the rod diameter to remain as modeled. Then apply threads to the thread portion of the split face.

Thanks,

[autodeskHCGNN]

This is awkward, illogical, and unexpected behavior.

If I am using a partial thread to thread only the ends, I DO NOT expect the rest of the rod to have it's dimension reduced. Why not fix the function rather than offering a kludgy workaround????

[TheCADWhisperer]

@autodeskHCGNN 

Can you File>Export your *.f3d file to your local drive and then Attach it here to a Reply?

[autodeskHCGNN]

OK - I created the simplest of models to show 2 examples of threading a 10mm x 50 mm halfway (i.e. 25mm of threads).

The one on the left uses "Auto" sizing which chooses a 10 mm ISO metric profile. Note that even though I thread only 25mm  out of 50mm, the entire rod is reduced to a diameter of 9.85mm which I assume corresponds to the major diameter of the 10 mm 6g thread auto-selected.

The one on the right is set manually to a 5 mm ISO metric profile threa.. Note that again even though I thread only 25mm  out of 50mm, the entire rod is now reduced to a diameter of 4.901 mm (!!!) which I assume corresponds to the major diameter of the 5 mm 6g ISO metric thread.

The simple (default) user expectation would be for the reduction in diameter to *only* effect the length of the rod that is actually threaded. It is unexpected to the point of being absurd that for example in the second case the unthreaded portion is reduced by more than half in diameter (from 10mm to 4.9mm) -- thus completly destroying the original dimensionality of the rod.

Now I know there is a kludgy workaround where you can split the rod in 2 pieces, then thread, then join but this seems to defeat the purpose of the nice option to not thread the full length (i.e., by unchecking the "Full Length" checkbox and setting an "Offset" and "Length" for the threaded part)

[selcuk_bolukbasi]

Dear Phil Eichmiller, You're a software engineer as I read. This fail isn't related with design method or technique. This is a programming fail. You can write lots of causes for this problem's solutions. But we used lots of software before FUSION 360 anysoftware else doesn't make this fail when you make thread any shaft. we hope that FUSION SOFTWARE TEAM solve this problem as soon as possible. Thanks for eveything your made.

[TheCADWhisperer]

---

@autodeskHCGNN wrote:

Now I know there is a kludgy workaround where you can split the rod in 2 pieces, then thread, then join 

---

I was able to do it without splitting into 2 pieces and then join after threading.

[patl1]

When I used the suggested method of splitting the face then modeling the thread. I get a "2" at the start of the thread callout on the drawing. I have deleted everything and tried several times and the drawing callout apparently still thinks the thread is called out in 2 places. This happened on both of my modeled portions of my threaded rod component. Can't figure out what is causing this.