I recently had a Blaser R93 in my hands, and spent quite some time trying to figure out what could possibly happen that would release the bolt backwards upon firing. The rifle has now gone back with his owner, but I gained a much better understanding of the system.

Here are some thoughts on a possible failure mode.

If gasses exert sufficient pressure on the "tip" of the locking fingers, in A, these fingers can buckle within the space between the fingers and the enclosure, in C. This would release them from the barrel locking ring. Not all fingers would need to buckle to initiate a catastrophic failure.

Also, if gasses were to penetrate between the bolt and the fingers in B and the second annular space further rearward, this would precipitate the buckling/failure.

I do not know how gasses could penetrate there, but it is worth investigating.

In the pictures of failed bolts (my apologies for the poor definition, that's all I could find), the extractor is missing, and two fingers are missing just behind it. Coincidence? Also, on one of the bolts it is clear that the fingers have buckled, for whatever reason.

Such a failure would not be a collet ring failure, or a catastrophic material failure in a high-stress area: it would rather be akin to removing the locking pin on a mechanism already under stress. If you pull it out, a lot of unexpected things happen very fast...

This is just a theory. If you have access to a R93 or good technical drawings of it, please check and see if it leads somewhere.







Philip

That assumes that the area that the fingers engage into on the barrel doesn't yield. I'd like to add to you theory by saying also if the pressures reach a high enough point and also expand the rear section of the barrel this type of failure would be easier to achieve then just overcoming the strength of the collet (fingers) grip alone.

I'd like to see a picture of the barrel

Something doesn't look right. Isn't the outside of the diameter of the bolt head indicated by the red? In the drawing the expanded locking rings appear to be the same diameter as the bolt head. If this was so they would either not be locking into the barrel or the non-expanding bolt head would be in the same ring and therefore not able to by cycled.

In other words the bolt locking rings have to expand to greater diameter then the bolt head. You can see that in the pictures but in the drawing it appears impossible.

Howard, from what I understand the drawing shows the bolt not yet fully closed: the head is in closed position, while the fingers are yet to be pushed in their expanded position by the forward rotating bolt handle.

Ok that would fit with the drawing of the bolt. But the barrel appears to lack "room" for the locking fingers to expand further outwards. Perhaps I am not reading the drawing correctly.

Yep look at the position of the part that is rotated by the bolt handle it is not in the locked position.

As the bolt handle is pushed forward and rotated into it's locked position. the fingers will be cammed outward at what ever angle the bolt head is machined to.

The fingers are pushed forward by the cam, and outwards by the cone of the bolt. Once "open", they protrude very slightly around the bolt head, maybe 1.0mm.

In those cases the barrel has an radial expantion, with 2 or 3 longitual cracks locatet towards the two cutouts for the scopemount, and a third starting towards the rear barrelatatchment screw.
If one does a static presure test of this locking systen, and test it with an ungreased locking surface, it lets go about 15 ton of static presure. If one lubricates the surfaces it lets go at about 12.5 to 13.5 ton of static presure.. The isue is the apx 45 deg of locking angle transforming the axial force to a radial force. The harder and greacier surface, the more radial stress.

The scematic draving posted by Phillip A is not completly identical to the barrels used after 2000. On the barrels i have seen, there is only a groove for the fingers to expand into. This meening that the reed area around the bolthead, is almost non existing as a gap. It is pretty tight fitting with only a clearance of apx 1/100".
This fit "backfieres" if you have a casehead failure, leaking gass out around the bolthead. . Because of the werry limited gasrelieve hole, to vent this gass out. you suddently increase the gaspresure area from internal case area to the entire boltfront area. This means an increase in presurearea from 0.75cm2 up to 3.14cm2.
Calculating from a preaure of apx 4000 bar, the boltthrust goes up from apx 3ton of thrust to 12 ton of thrust. This thrust it pretty close to where the locking gives way.


PS Phillip
On your pictures they have remoowed the flesh, from the shooters face, that was shown on the original pictures