threaded rod design and static load control

If you need thread geometry for FEA, it should be modelled.

I only use cosmetic threads for fasteners. If the threads are functional (like feed screws), I like to have them modelled.

If you model them, which function do you use? I've been working with SW for +1 month now and I'm not familiar with all of it.

Regards,

Geoff

That70sTick wrote:

Go to the "Help" Menu option and work through using the "Thread" items where you will learn to use the "Helix" and "Sweep" items.

Bo

Geoff wrote:

Thanks

That helped me a lot.

But when mating two parts together with cut-sweep and a boss-sweep I can't mate the thread.

I presume I have to mate the thread to do some FEA afterwards.

Best regards,

Geoff

Bo wrote:

I do NOT think there is any technique to allow mating of thread faces, one against the other, or at least I've not heard of any. Doing a mate that allows a FEA stress analysis would be tricky in that situation.

Collision detection will work on threads, but that doesn't establish a mate.

Maybe someone else knows more.

Bo

Geoff wrote:

Are you saying that you cut threads on the outside of the rod with a sweep and then you cut the same threads on the inside of the nut and then you had two swept surfaces to attempt to mate together? I'm not sure if two such surfaces would mate.

Something that you may try first is to apply the forces to the cut thread surface. This may not work. But, then you could try cutting a series of annular rings with the same pitch as the threads. It probably won't make much difference if the rings are not connected because approximately the same amount of material should be involved. What is nice about the annular rings is that you should be able to apply forces in COSMOS to only the number of rings that would be engaged in the length nut that you have and the approxomate distance from the end of the rod.

In the end, this may be one of those problems where the answer may be easier to estimate from a handbook. Probably along the lines of something like the mean thread diameter x pie x effective nut length divided by 2 x (some fudge factor like) .9 times the worst case expected tensile strength of the mateiral and since the failure is probably in sheer the resulting value should probably be divided in half....(which I'm not sure that COSMOS takes into consideration. But you should definately look this up and verify a calculated estimate agains COSMOS. Which is always the best practice.

And if this is a critical application, (like holding up a bridge or a beam in something like the new World Trade Center Building) you will probalby need to do a fair amount of physical testing.

The other alternative for this is to contact a supplier of such fasteners and get them to tell you what the specs are.

Just a few ideas... hope this helps

Ed Now if this works then

"Bo" wrote in news: snipped-for-privacy@i42g2000cwa.googlegroups.com:

Generally, threads and worm gears and such are not mated directly. At least one of the components will have to be mated using some kind of construction geometry. Something like a sketch point in one part mated to a helical face on the other.