An excerpt from "How Bill Lear Built the Model 23"
www.aerotalk.com/index.cfm?action=view&ForumID=15
Another series of events provide another glimpse into the working of Bill Lear. The early Model 23 aircraft had seven electric fuel pumps. Two in each wing, one in each tip tank and one in the fuselage fuel tank. From day one they had a high failure rate. In spite of the fact they were of new manufacture, the high failure rate was known in the market place and did not help the image.
Electric fuel pumps were Bill's area of expertise, airborne electro-mechanical components. Months went by, all manner of testing took place but no one could figure out why the pumps were failing. Finally on one of his trips to the West Coast he learned of a jet pump system used in military aircraft that had no moving parts. It relied on excess fuel volume produced by the engine driven fuel pump that was routed in a small line to the fuel tank and the jet pump. It would pick up more fuel and return to the engine in a bigger line.
Bill learned the Ronson Company (yes, the same company that made cigarette lighters for your coffee table) made jet pumps for the military. He contacted them, asked for a jet pump for testing and a quote for a production order. They complied with both requests. The pump worked fine...the price would be about $4,000 each. An outrageous price for an item with no moving parts. It took less than a week for Bill, the engineers and machinist to design a small casting, a screen to keep debris out of the pump, and two standard AN hydraulic fittings, one modified on a lathe to convert it to a nozzle. The cost in production, about $200 each. Four of the seven electric pumps could be replaced by Bill's jet pumps.
Post script to the story:
About the time Bill was getting the jet pump designed, somebody figured out why the electric pumps were failing. The brushes were backed by a single spring. There was a resonant frequency in the airplane that would excite the brush...cause it to vibrate...cause it to arc and burn. The fix was simple. Back the brush with a second spring. Now no single resonant frequency could excite the brush. The jet pumps stayed.