I start with guidelines for half and whole pitch, and the dimensions D and d3 (divided by two for radii) for the outside diameter and bottom of the thread groove, and draw lines for the slopes of the thread at 30° from the vertical at the pitch distance 0.4mm…
They should intersect the outer diameter to leave a flat of P/8 or 0.05mm, but the measured dimension is 0.117mm, more than twice too big.
And the distance from top to bottom of the sloping lines, H, should be P * 1.08253, or 0.433mm. It isn’t. It’s 0.346.
When I model screw threads I use the major diameter, D and the minor diameter D1 and ignore d3. I also draw the crest of the thread and use the 60° included angle but allow the bottom to be whatever width it needs to be. So far that has worked for all the screws I’ve modeled and 3D printed.
Thanks, Dave, for such a prompt reply, and your suggested simplification. The printed threads have to fit into a manufactured metal M2 nut, and your suggestion will probably work - I expect there’ll be enough “give” in the plastic to allow the nut to be threaded onto it.
But I’m also interested to work out WHY my drawing using the official diagrams and dimensions isn’t making sense.
I have been successful modeling threads as described and having them fit commercially manufactured threads, both printed external threads and nuts and printed internal threads and commercially made screws.
I think the thing with the dimensions you get in those tables is that they aren’t as precise as they really need to be. If you use the minor and major diameters and P/8 and P/4 for crest and root widths, you won’t get 60° for the angle. Something’s got to give.
For M2x0.4 external threads I would use 1.9335 for the major dia. and 1.4905 for the minor dia. Use 0.05 for the crest width and let the root be what it will.
Stack as many as needed of the latter, then explode, remake as a single component, close the open top, then run Solid Inspector 2 to clean up internal faces.
Not sure if I’ve used enough segments in the helix - only 12 - but not too difficult to redraw with more if this doesn’t print smoothly enough. M2 thread - solid.skp (454.5 KB)
I’ll have to look when I get home in a couple of days. I’m pretty sure I managed to draw an accurate thread from the standard proportions. Note, however, that there are clearance allowances that slightly tweak the ideal dimensions to avoid binding.
I am with @slbaumgartner … when you do a model based solely on the proportions (see the first figure of the original post by @john_mcclenahan) then everything works out (more or less).
Here is a screen shot of a model of the profile developed using the proportions.
The only thing dimensioned was the 2mm Pitch (and the angles) … the rest of the model was based on the geometric ratios.
But even then … as @john_mcclenahan points out, when the model is measured, d3 (and d2) do NOT yield dimensions that agree with the tables.
And then there is also the question that jumps off the page at me: How can h3 be greater than H1 … as shown in the table? (Look at the original figure.)
Any ME’s want to weigh in on this? … and give us a short course on thread nomenclature?
Meanwhile, I have a rather pedestrian FDM printer … with a “Fine” mode resolution of only .12mm … so printing M2 machine screws is probably not in the cards.
Thus, to check out the usefulness of this model, I scaled it up and used the profile to model and then print a useable M3 machine screw. 241103C_12mm_ M3_Machine Screw.skp (1.0 MB)
Happily the printed 12mm M3 thread seems to fit a standard M3 nut without any problem.
