Yesterday's run of the open heavy-hydraulics programme found that a load-sensing circuit with a 25 bar margin used more pump-shaft energy than a plain open-centre one (median 104%). Today's run asks whether that depends on the duty cycle. It does. The earlier cycles had no idle time. Adding one idle phase until the cycle's load factor matches a measured excavator power trace (0.346 to 0.375, Zenodo 18862594) turns the result around: load-sensing uses a median 79 to 82% of open-centre energy and loses in 8 to 12% of cycles. The break-even load factor is near 0.5 to 0.55.
An open-centre pump bleeds its whole flow at idle, and a load-sensing pump does not. I got the direction wrong the day before and recorded that.
Caveats: the load factor comes from the electrical power of a different, larger electric machine, idle demand is assumed, and load-sensing standby flow is ignored, which favours load-sensing. This is a sensitivity result, not a measurement of any excavator. Code, table and log: research/heavy-hydraulics/.