Received carbon viola bow from p&h.  It looks like they use an inexpensive hydraulic wire crimper with a special purpose set of dies to shape the hair ends.  A short length of copper tubing is the likely source of stock material.  See photo below for detail.

@Fiddlerman and others.  Currently playing the viola bow on a cello, and it seems perfectly suited to that use.  No clue what a brute their cello bow might be.  For the interest of @bocaholly, the cloth to resin ratio is likely 50:50.  It is the most responsive (admittedly cheap) carbon fiber bow I have yet tried.  The unbleached white hair is not pretty, but there is a lot of it and it take rosin well.   The ebony frog is well done and the combination imitation whale bone and metal wrap is attractive.  Grip is imitation leather (similar to Glasser).  Logo on stick looks to be simple paint transfer.  The stick geometry is designed to quickly tension the hair and I think that it takes more turning torque on the button than normal (no noticed issues with button screw thread or frog nut position).  

I also purchased replacement hair sets at $7 each.  The manufacturer paints the tip end of hair red, but I have yet seen a difference in the ends.  They must use a hot wire to cut the hair bundle ends because they are perfectly even.  That must stink.

My first reverse engineering goal is to make a set of crimper dies so I can use my Harbor Freight hydraulic wire crimper to make hair sets (I plan to use corian sink stock).  Then I will see how easy it is to replace the hair.  It looks like a five minute job.  

@bocaholly and others.  According to my wife’s food scale (do not tell her), the bow weighs 65 grams.

@starise and others.  I think that you have misread the thread.  The P&H bow is intended for home rehair.  The only tool needed is the frog button.  The company sells new hanks of hair with installed clip ends for about $7.  

I only brought up the design of the hair end ferrules because I want to use the technology for rehairing other bows (the traditional method of wrapping with string is tedious).  

I am aware of the incredibow.  I would like to try one but have not found one cheap enough to warrant experimentation and do not feel like spending $150 on something with nylon hair.  

I am going to wing two birds with one post.  Here goes.

@starise, I sent you an pm regarding where I got mine.  Good luck.

@peter .  It looks to me that the hair is cut about 3 mm beyond the crimp (likely with a hot wire) and the horse hair hank version is fused with what looks to be ca glue.

The patent refers to the use of pvc containing monofilament that they found to have similar rosin holding characteristics to horse hair.  I procured three specimens of this product for a violin.  It looks like the ends are heated in a cylindrical die and compressed into a fused button.  The process is likely similar to that used to butt weld polypropylene water main pipe.  The patent has a diagram but no description for this.

I took dimensions of the crimp and it looks like a #4 AWG is a close approximation for the lower portion.  I could likely form the upper indent section by silver soldering a small steel rectangle on the other half of the die.  My approximation of the copper tube wall thickness is 0.6 mm, so I do not think that complex metallurgy is required to create an acceptable die for this.  I have been wrong before.

I have no doubt regarding your seamanship but the crimp is likely to have a more svelte mortise profile due to the amount of compression obtained by the press.  Hank length accuracy is also likely better if a jig was created to provide a consistent stretch while the hair is clinched.  

I approximated the diameter of the crimp at 1/4” diameter (which makes sense as the originators were English engineers).  Standard refrigeration tubing of that diameter has a thickness of 0.030” .  K&S item #8131 is 1/4” diameter brass tubing with a wall thickness of 0.014”.  I am going to try that first.  

@peter et al.  Per provided photos in the thread above, it appears that there is no extant “flash” from the die set.  The measured width of the crimp is 2.5 mm.  The measured width of my die set is 10 mm.  

My faulty color vision skewed my previous comments (which I will edit).  The instructions clearly state that the red dot (side of crimp opposite of indent) faces forward on tip and faces backward on frog.  My wife pointed out the dots.