Mike,
I have a Chelton system that records data once every second. Having looked at the data, as expected there is some fluctuation in the manifold pressure, however it doesn't exceed 0.1 PSI. Fuel pressure fluctuations are from 7 PSI to 8 PSI., these fluctuations appear normal to me. Since the problem goes away with low boost off, that certainly has to be an indication that it's related to the boost pump. If it's sucking air when running wouldn't it also be leaking fuel when sitting on the ground for over a week?  That is, assuming it's a fuel line that is leaking. What test on the ground would prove that possibly there is an internal leak in the boost pump or that it will soon fail?

Sent from my iPhone; Art Jensen