Threading changed the shaft size.

[Bertho_Boman]

I have a 3.00mm diameter shaft, 80mm long.  I created a M3 thread, 6mm long on one end. 

The threading worked as expected except Fusion changed the whole shaft diameter to 2.93mm.  

[jhackney1972]

I have no explanation for the reduction in diameter but if you put a thread relief in place, you need it anyway, it will not reduce the diameter of the shaft.  Model is attached.

[Bertho_Boman]

Thanks John,

I forgot to mentioned that I as a bypass fixed the problem by first reducing the threaded section to 2.93 mm and then thread it.

My question is why does this happen.  Fusion should not change shaft diameters.  Do I have some configuration setting wrong?

Bertho

[jhackney1972]

Looking at the Fusion 360 Thread Table, ISOMetricprofile.xml, you will see the table specification calls for the diameter size you get.  Why it applies it to the complete shaft is beyond me.

@jeff_strater, do you have any ideas or do we need a repair ticket?

[TheCADWhisperer]

Split the face of the cylinder before adding thread.

@Bertho_Boman 

Attach your file here if you can’t figure it out.

What standard are you using? Can you link to reference page of the thread nomenclature for the standard?  What class of fit is your Design Intent?

Is it physically possible to assemble a 3mm cylinder into a 3mm hole? (Without using a press.)

Can you show Section View of assembled mating parts (threaded cylinder and hole)?

What manufacturing processes will be used to create the threads?

All of these are relevant questions when considering helical threads.

[g-andresen]

Hi,

If you measure the shank diameters of various standard screws, you will find that they are always smaller than the "nominal diameter".
For example, an M6 screw has a shank diameter of 5.89 mm.

günther

[Bertho_Boman]

Note,  I am not questioning the thread dimension.  The threaded section is correct.

My point and big surprise is that Fusion changed the whole length of the shaft, not just the threaded section.

That is definitely an unexpected behavioral to me.

I specified the length of the thread so the changes should only apply to that section.

Bertho

[Bertho_Boman]

As I mentioned, the generated threads are correct.

The changed dimensions of the shaft is not correct.

I understand how to work around the problem but I should not have to do that.

I finally caught the problem when I was preparing the documentations and drawings for the shaft and I was surprised that it had changed from the 3.00mm.

How can this thread be turned into an error report?

[TheCADWhisperer]

No work around required - only logic.

Attach your file here if you can’t figure it out.

The Fusion team is aware of this functional behavior.

It has been discussed here many times over the years.

[Bertho_Boman · Accepted solution]

That is good that the Fusion team knows about it.

As far as I am concern, having to separately create a reduced section in advance to threading, is a workaround.

I realize that threaded sessions normally  have separate areas or dimensions.

This also opens up a question of specifying the threaded length.

If my reduced section is 6mm long and I happen to specify the thread length to 6.5mm, presumably it will surprise me and change the whole shaft.

In the next project I will have to be careful and hopefully the issue will be corrected.

Bertho

[TheCADWhisperer]

---

@Bertho_Boman wrote:

As far as I am concern, having to separately create a reduced section in advance to threading, is a workaround.

---

@Bertho_Boman 
Not necessary to separately create a reduced section - that would be a workaround.

The best technique in current release of Fusion does not require that and is fully parametric.

[Fabbunny69]

I have just experienced this, any update?

After specifying a thread of a certain length, fusion changes the shaft diameter beyond the length of thread i have asked for.

How can it be logical for fusion to change something you have not asked for? 

Workarounds should not be needed to stop fusion doing something that must inherently be an error in internal coding.