Changes by last author:
Added:
|
After looking many detailed InputTrigger/TriggerLog logs, I think it would be possible to tune ignition advance with free-revving
This means without Actually, this is a kind of GenBoard/LoggerIntegration/DedicatedLogger/VirtualDyno. This eliminates the problem associated with * unknown gear * drag * elevation * bumps * changes in drivetrain efficiency (changing oil-temp) ---- high RPMdot problem This uses the own theta of the engine, measuring the change in RPM precisely. InputTrigger/TriggerLog shows that even energy of individual events can be measured in * free-revving * with 60-2 (maybe 36-1 too?) * only crankwheel (cam is sloppy) * if fuel (VE, j[] and l[] tables) already tuned (WideBand rulez) Timing from predetermined tooth - or simple "coil type" trigger This is a standard method. GenBoard firmware is migrating to the superior "timing from the last possible tooth" method (AdvancedIgnition). Without "timing from the last possible tooth", if the ignition is tuned by free reving it'll probably end up with a lot of advance. The reason for this is that the engine has already became faster, but it still uses the higher period data from the times of lower RPM, so the ignadv will be faster. Low frequency of calculations This would effect tuning, but not as badly as the timing from predetermined tooth. This would also result in slightly (really small effect) higher ignadv values in most of the map. Reason: the engine uses the ignadv from a lower RPM applied to a higher RPM, so the bigger advance will get into the lower RPM loadsite of the map. ---- Travelling across the map Of course only those loadsites can be tuned, that are met. * for every ignadv-map candidate * rev up with 4..5 different TPS * keeping the engine at constant TPS will NOT result in constant kPa. Actually kPa will decrease with raising RPM ---- Required Software to: * log enough details * analyzes the logged data * preferrably suggest next map to try (so it need not be built manually) |