First...Lu, it's great to see you back here posting.
HC said:
If one pump was still working, how come the pressure was nearly at the bottom of the scale?
The design of a pressurized oiling system is performed basically along these lines. First, the engine or transmission will have various bearing surfaces and perhaps oil pressure activated devices that need to be supplied. Channels and passages (for the pressurized part of the system) will be incorporated into the design to accommodate these needs.
From these basic requirements, a flow rate of pressurized oil will be derived from either calculations or actual testing on prototype assemblies. The basic flow rate will be derived from the number of bearings, bearing surface sizes and clearances, and will also be affected by expected operating temperatures and oil viscosity. Channels and passages will be sized to accommodate the expected flow rates.
Then a pumping system will be designed to accommodate the expected flow rates. The pumping system has to exceed the expected flow rates by a certain amount for a variety of reasons. Oil, like water, is uncompressible and therefore the pumping system has to have a pressure relief system of some kind (usually a pressure relief valve), to dump the excess capacity back to the sump.
Various criteria are used to determine a design working pressure for the system. One of the reasons that a pressurized system is considered, is because the designer may prefer that the oil not linger at the bearing too long, because it can overheat and break down prematurely. It's better to provide a continuous flow of cooled oil that will stay in good condition longer and provide better lubrication. Working pressure also determines how fast the oil will move through the bearing's various clearances, and in this way working pressure also contributes to the system's overall flow rate.
In an oiling system, volume flow substantially determines pressure. At a given temperature, oil condition, and bearing wear condition, there's a point where once the pump's capacity exceeds the volume required, pressure goes up dramatically until the pressure relief point is reached (due to incompressibility).
Now in aviation, flight safety is a factor in the design, so having two pumps is an excellent solution. However with two pumps, the volume and pressure question gets a little more complicated. Each pump could be designed to exceed the capacity needs of the system, but with two identical pumps running, there's a great deal of excess capacity causing lots of extra oil to be dumped back into the sump through the pressure relief system. This isn't very efficient, and causes both pumps to run at high pressures and stresses.
A case can be made, even with aviation safety requirements in mind, that each pump could be designed to provide most, but not all of the system's volume needs. This would be acceptable, if it's known that the system can run safely for a known period of time without damage at the reduced volume. This safety case can be made, based on the fact that a system running on a surviving pump, will be running at substantially reduced pressures because the surviving pump CANNOT supply the full volume requirement (and pressure goes up dramatically once volume is exceeded) and thus the surviving pump is PROTECTED from failure because its running at substantially reduced pressures and stress. Again, the case can be made that this enhances flight safety in a single pump failure mode.
Nick can elaborate, but since this particular S-92's transmission was almost new, and it's nearly certain that the failed drive shaft reduced the output of the failed pump to virtually nil, the reason the pressure dropped as far as it did, is perhaps because the remaining pump was purposely designed to provide somewhat less than the full flow requirements for the transmission.