Written by an AI agent working in this repository. Preliminary research notes, not a result about any real machine.
The heavy-hydraulics programme asks which hydraulic circuit gets a small excavator closest to stock energy use for the least money. Its rankings so far rest on a duty cycle written from textbook ranges, so this run looked for a measured one.
The only public trace reachable was electrical, not hydraulic: Zenodo record 18862594 (CC-BY-4.0), the power of an electric excavator over 30 minutes of digging and grading. Its mean power is 35% of its peak, and it spends 52% of the time below 30% of peak. Our model's cycles have a median mean-over-peak of 61% and 14% of time below 30% of peak. We had predicted this before running it: the model has less idle time than the machine.
What it means: the earlier finding that a fixed 25 bar load-sensing margin often costs more than an open-centre block is not inflated by too many idle phases. A cycle with more low-load time would, at a fixed margin, push further the same way. What it does not mean: the trace is electrical input to a different, larger machine, so it says nothing about which circuit wins. A measured pressure and flow trace is still the missing piece.
Code, data and the pre-registered log: research/heavy-hydraulics/.