The fluid enters the filter and, after factoring inertia, attempts to take a right angle turn to the interior of the filter. It's going to attempt to take a 90° turn to the shortest path "out". Now it's not going to do that and will have shearing of flows that are reduced as it (basically) has no good gosh darn reason to make the trip down to the bottom of the filter (the long path to the outlet). If it was big enough, it would be a "still well".
Now if the pressure differential is high enough, the bypass valve will open. If it's on the open end, the bypassed, unfiltered, flow just takes the short dip in an out. If the bypass is at the far end, then it bleeds off at the far end where this "silt" may be accumulated.
Quote:
Once the engine is started, and the system is under pressure,
For the most part, the only time a bypass works is during start up or during "transitional" conditions.
Let's say you've got large/long oil galleries that empty when you shut down. When you start you'll have little to provide "back pressure" to the engine side of the media. If your pump is in relief due to too high a visc ..trying to move too high a volume from standing still to whatever rate it wants to pump it, you're going to see an increased differential across the media. It's going to be a combination of no back pressure and reduced volume at an attenuated pressure. If that differential is high enough, the bypass will open.
If your oil pump is not in relief, once the engine is fully enveloped, there's virtually no differential. As the filter loads it will elongate and increase the transitions in terms of duration and amplitude ...but it's not all that big of a deal, imo. Now flat shifting @ 7000 rpm? Starting @ -10 w/15w-40 with some time on the filter
(or some Euro-Alloy with 135 peak pressure and 15w-50 and a 10 quart sump that never gets warm - sure).