Hi, KBMB.
In terms of principle of MD, MsMars is correct, that you should always target that your test piece must be as much as possible be situated in the center of aperture of metal detector (which is the middle). In terms of where to place, here is what I recommend:
Validate the MD setting using the worst case scenario. Meaning which of your product specification which you may think will affect your detection (Product effect,e.g. which has the most moisture content, which ingredient may be "conductive", packaging etc) and false rejection.
Once your settings is established, passed the product with test piece in the aperture (estimate) and over the product (should you have smaller product that the middle of aperture this is much easier), make tests that is statistically sound for you to compare data.
If there is no significant difference between the signal of different test piece location, document it and in your hourly monitoring, you can have the confidence that the performance is the same even if you place it in the middle of middle detector.
If there is significance difference (meaning the test piece in the middle have very small difference between the product without contaminant), I would suggest you monitor always inside (as there is significant product effect).