What is your definition of fallout?  Do you mean product that does not conform to the specification?  Do you mean literally parts that are lost during the heat treating operation, and are no longer recoverable?  What is the specific quality problem?

Parts that are lost during the heat treating operation and are no longer recoverable.

Thank you for the clarification.  What is the heat treat process being used (quench and temper? annealing?) and what type of furnace is being used?  Your range is not unheard of for small parts in a continuous, mesh-belt process, but would be extraordinary for parts processed in vacuum furnaces with high pressure gas quenching.  Are you required to be certified according to QS-9000 or ISO/TS 16949?  If so, continuous improvement is a requirement, and therefore you should be working on methods for reducing the fallout rate.

the wire is already annealed when we get it after forming parts we send out to get heat treated. Heat treating is to take it up to 25-35 Rockwell. Their furnaces are gas batch furnaces. The parts are placed into basket sent through then
quenched. We are not ISO or QS but the heat treaters are.

The problem is usually in the quench tank, especially when the parts are small or thin.  If your heat treater would drain their quench tank, you would find almost all of the lost parts at the bottom.

The parts are not lost they are distorted where they are out of tolerance.

That small percentage is not un-common.  The amount of distortion can depend on the part geometry, material and how it is processed.  

You could take the customer to the heat treater and test a batch before and after processing for distortion; they could watch the fallout for themselves.  Or, do it yourself and show them the data and stand your ground.

If they are not willing to accept that then show them estimates for fixture quenching and they ahould not bother you much after that.

_______________________________________
Feeling frisky.........
www.tailofthedragon.com