thread modeling

[mrPelagic]

I am modeling some threads for 3d printing and eventually polymer injection molding. The threads do not look right to me, but the dimensions are nominal from the machinery handbook #27 (&28 as a sanity check.

Please see the attached picture, the purple threads (witch are internal threads). (the root of gray threads are another issue stemming from a known problem. please ignore!)

method of creation: Autodesk inventor (don't tease me, I hate inventor, which is probably why I am asking such a basic question)
5/8-32 un3A internal thread
1. create bore at nominal minor Diam (~.594")
2. draw 60deg triangle with .002" rad. the tangent to the rad is set to nominal major diam :.625"
3. use thread coil tool to cut threads at 32tpi

the tips of the threads look way out of proportion to me, and to make them look "correct" I modify the major diam to about 0.645". I dont' like what I dont understand!

any thoughts? It's been years since I had to actually model/design threads. HAve I forgotten something obvious?

thanks!!

 

[pylfrm]

mrPelagic,

5/8-32 UN-3A is an external thread, so I will assume you mean 5/8-32 UN-3B.

Your step 1 is correct; minor diameter is 0.59395 at tolerance median.

Your step 2 looks a bit questionable, although I'm not sure exactly what you mean. Instead, you should probably be using the pitch diameter (0.60645 at tolerance median) to define the thread flanks. This is the diameter where adjacent (opposed) flanks have an axial separation of 0.5*pitch, or 0.015625 in your case.

The root truncation (flat, radius, or otherwise) must fall within the zone between the minimum major diameter of 0.625, and the maximum diameter defined by a flat width of 0.041667*pitch with the flanks at least material condition. This works out to 0.63266 by my calculation.

Refer to ASME B1.1-2003, particularly Fig. 3 and Table 2, for the official definition.


Out of curiosity, are any 3D printers actually capable of producing this thread correctly?


pylfrm

 

[drawoh]

mrPelagic,

Your 3D CAD should not matter here. Your geometry is pretty basic.

I have done something similar in 3D CAD. I used my Machinery's Handbook. The important tolerances were applied to the pitch diameter. I used the geometry to model around this. The larger threads worked fairly well. The small threads were not within the tolerances of FDM.

--
JHG

 

[tbuelna]

Might be a problem with how the profile is positioned/oriented as it gets swept along the helical curve. Also could just be a graphic display resolution limit rather than a true geometry issue.

 

[Occupant]

I agree. If you don't align the profile, let's say with the X-axis, and use the "Sweep" command, it will always stay perpendicular to the generating path.

 

[mrPelagic]

pylfrm& drawoh, thanks. Ahhh. Funny , i often felt that I'd been hacking my threads incorrectly all these years. I'll try the drawing the PD instead.

Yes, totaly within resolution of some 3d printers (but true, i dont think Fdm could do it- at least not well) we use a polyjet system to print. It squirts out a liquid then UV cures it. It's pretty amazing- layers thickness of .0005", xyresolution around .001. Not very strong, nor hard. And terrible friction properties though.

I wish i could show you some of the tiny, functional mechanisms I've been printing. It is a different design/iteration cycle with printing because it's virtually a daily iteration.! Funfunfun.