Solution’s Advantages and Impact
Maintainability, Reliability, Repairability. During the period from 1969 to 1970, Drenth gathered reliability statistics on controllable pitch propellers in marine applications. Studying 657 installations in service, all of which running for a total of 5,270,000 operating hours, 32 chance failures were noted – producing a mean-time-between failure of 165,000 hours and a reliability service period of 4000 hours. For a 95 % confidence level, the upper and lower limits of the calculated MTBF of 165,000 hours are 270,000 hours and 120,000 hours. Drenth further analyzed the breakdown of failures and found that 26% of failures were related to hydraulics.
Hydraulics has such a high probability of failure that it is standard practice to have a spring, embedded with the pitch actuator, that will restore propeller/fan blade pitch in the event of loss of hydraulic pressure. EMCP uses a lead screw (alternatively, ball or roller screw) that will stay in position in the event of electric failure. Depending on tuning of friction in the screw’s nut, EMCP will stay in position once its nut arrives at the commanded setpoint and in event of failure. But for the supply of electric power, EMCP actuation is entirely contained within the fan/propeller and is otherwise invulnerable to ancillary system failures like hydraulics lines being cut or failing, clogged filters, etc.
Regular maintenance for EMCP might be as simple as removing the fan/propeller’s front cover and spraying lubricant into the actuating mechanism immediately available after having removed the cover.
A fan can be differentiated from a propeller along this dimension, in that a roller or ball screw will be more appropriate for a propeller rather than the lead screw we have included in our fan design. Both of these remove backlash and significantly improve the lifetime of the mechanism. A lead screw is appropriate for a fan since backlash is only experienced when the fan’s air flow is reversed and since the load from the fan on the screw will be negligible (150N for our Phase I design).