Come back with a little information about what you are concerned with.  

The problem is the premature failure due to fatigue of a pedal axle on a bicycle.

The threaded hole in the crank should be 90 degrees, but in this instance the hole is at 93 degrees. Consequently when the pedal spindle is tightened up to the shoulder, one side makes contact before the other and induces an unusually higher tensile stress on side of the spindle.
I need to know how significant the stress increase is and will it noticeably decrease the fatigue life of the spindle.   

The problem is little clearer now.
Yes, it will increase the likelihood of fatigue as the situation you describe will severely overload on side of the threads.  The seating surface, whether a shoulder or fastener head, should be perpendicular to the centerline of the threaded hole.  As your are probably dealing with fine threads this will make this situation worse.

This is somewhat like a standard test to check the head and fillet on a fasteners.

If you could come back with a little more information such as the thread description and material and where you are experiencing failures on the fasteners you could get a better answer.

Is this an alloy crank? I woudl guess that if its the standard 3-pc alloy crak common on most modern bicycles that the threads will pullout from the crank first. The spindle is likely 4140 or similar steel and much stronger than the crank.

nick

Nick
I love materials science!

The cranks are one piece steel and the pedal axle failed due to fatigue

well than unclesyd is as usual correct.

Nick
I love materials science!

Are the threads on the crank arm parallel to centerline of the crank bearings to the point you could spotface the crank arm?
All it would take just enough spotface to make a landing that is square to thread axis for the shoulder on the petal axle.

Another less elegant way would be to shorten the threaded hole if the centerlines are parallel. This would make the petal threads bottom out and take the wedge from under the head/shoulder of axle threads.

Unclesyd- usually on 1pc steel cranks the hole for the pedal is drilled and tapped through. It looks like they have a problem with the setup on the machine used to put the hole/tap into the steel. Not to mention how funny it would feel to ride a bicycle where the pedals are not perpendicular to the crank.

nick

Another reason to make sure the shoulder seats and is square with the crank is that the bending stress at the axle/crank interface will be reduced as opposed to pure cantilever without the axle being seated.

Jesus is THE life,
Leonard

Seems to me the main problem of fatigue is going to be in the high stress concentration area somewhere about half-way along the calf muscle, since the pedal axis is not parallel to the chainwheel axis.

good call yates!