Discussion and questions related to the course Practical Dyno Tuning
Hi, how come If you try to calculate a fuel percentage change based on Target Lambda error on these ECUs, it never seems to work out?
Target Lambda 1.0, measured Lambda 0.95. I subtract 5% from the VE value in the VE table, but the Lambda goes very lean. Does this have something to do with the way Link calculates fuel delivery? Thanks
Can you give us an example of the change to the VE table? What were the values before and after you subtracted 5% fuel. Are you sure the engine was at steady state so the measured lambda was correct?
At low engine speeds (i.e. idle), it can take 1 second or more for the combustion result to be reflected in the measured lambda reading. So the throttle, and RPM need to be steady for a second or more to make sure you are applying the measured lambda to the correct cell in the VE table. As the engine speed increases this time delay is reduced, so it's only a few tenths of a second at 4000 RPM typically.
When turning from data logs this is something to be aware of. On the dyno, you need to be patient and wait for the lambda to be steady before making a change, particularly at low RPMs.
It shouldn't go lean if the ECU setup is correct. There's a couple of caveats here though - First of all you need accurate dead time data for the injectors because the 5% trim is only applied to the effective pulse width rather than the effective pulse width plus dead time so if the dead time isn't correct there's not much chance of the actual fuel delivery being changed by the correct amount. The other parameter that needs to be tuned is the fuel charge cooling coefficient. You can check the help fil on this parameter to see how it should be dialled in or watch the following webinnar - https://www.hpacademy.com/previous-webinars/how-to-tune-the-fuel-charge-cooling-coefficient-in-a-link-g4/