Take representative samples and look at them in cross-section. I'm sure there is no 'exact' thickness but a range of thicknesses. Take samples from different lots, test them and apply the usual statistical techniques.

Thank you for your reply, and it is really helpful to me.

I didn't know that M16x1 was a standard thread. Sound very fine to me.

M10x1.25 was one very common thread that I was involved with in electronics. (Potentiometer necks)

Also, it would depend on how your thread was formed. Rolled threads would have a different outcome than standard tapping where material is removed.

I also suspect the thickness would be somehwat random and to achieve an "exact" thickness would require appropriate heat treatment process sequence. somehting like anneal, heat, quench/cool or expose to controlled atmosphere.

Thank you for your reply, and it is really helpful to me.

A structure modification could only occur in case of thread rolling.

The starting diameter of the bar before rolling is around the mean thread diameter and has to be defined according to used material. You can assume that the flanks have a variable hardness since the highest deformation is at the root and the peak and the lowest about at the mean diameter. You could assume that the root is hardened about 1...2 root radius. This is ONLY an order of magnitude for your case you should make several samples ans measure hardness with a micro tester on a diametrical cut.

I am not home but when I shall be I will bring some pictures which could be used as guide lines since I am on a long leave it could be several weeks before I can do it.

Thanks, you are the best!

There is not going to be a uniform answer, as the hardening is a result of a separate process and is a side effect and not an intended condition arrived at by definite purpose....If you want an accurate answer you'll have to retreat the screws with an intended purpose....

http://www.mmsonline.com/articles/rolling-threads-has-advantages/

Thank you for your reply, and it is really helpful to me.