Wayne,
Manual peening is definitely a peening operation, and it does uses nozzles. However I think the foundation of the problem is that the AMS-S-13165 has been too widely used, it is currently used for both critical and non critical shot peening applications.

The use of automated methods for shot peening obviously has a huge influence in getting the process right, now that we understand the importance of the right amount of coverage, and uniform coverage on the shot peened surfaces. I understand that some suppliers may have experienced operators being able to perform really uniform and controlled shot peening, but automated equipments will always be more repeatable and controlled. Specifications have to evolve as technologies and knowledge evolve, and what's happening wiht AMS-S-13165 is a great example of this process.

From my point of view, it is unacceptable that parts like flight critical component be subjected to manual shot peening. However I understand that some other parts, let's say cut ends of automotive valve springs may be peening manually, but even then it is not a proper way of doing things because of poor repeatability and productivity. I think suppliers out of the aerospace inductry may want to start working on a spec for shot peening on non critical components, if they really waht to use that technique. Again I don't believe it is a good choice.

It is also important to note that currently, the effect of shot peening on improvement of fatigue life is still rarely taken into account in the design of parts. More often, the part is designed around the known operating conditions, and shot peening is only used as an additional margin of safety which doesn't add weight. I think shot peening has to be pushed to the point where the technology is there to provide a stable, controlled and repeatable process so that designers can take it into account in part design, so that it can help lower the weight of the components and the overall efficiency of mecanical systems. From my point of view, this can only be obtained by implementations of new technologies, starting with automated and robotic applications.

I agree with Walter that involvement in SAE AMEC is really important to get the specifications to the point they need to be, and at the moment a lot of work still needs to be done...