Discussion and questions related to the course Practical Dyno Tuning
Any suggestion how to get OEM injector data of a rather older car? A Nissan Micra k11 cg13de in this case. Or any way to get around this problem?
You might be able to send them in to a shop to do the testing, but most people these days just replace them with injectors that have known data.
Thanks for the reply.
The thing is that it's a 1.3lt N/A engine, revving up to around 8000, making an optimistic 130bhp. So the necessary injectors are rather small and practically not available as after market. For instance, ID725 will work at around 30% duty cycle at rpm limit and no ID dealer is offerring them.On top of this, budget is an issue and large amount of OEM spares are available.
If you aren't planning to change injectors, why don't you just ignore characterizing them and "bake" any errors into your tune? Pick any exiting injector with similar flow potential and get tuning.
If you do want to test them, it is necessary to have an ECU (or engine simulator) that can enable injectors with a specific frequency/duty cycle and you spray them into a graduated cylinder and measure the flow volume at different voltages, fuel pressures and duty cycles, for a known period of time (say 10 - 60 seconds). From that raw data you can produce the necessary injector info you may need for your ECU.
Note that if you are charging someone for your time to do this, characterized injectors may be much cheaper. If your time is free -- then this is a good weekend project. A little googling will find videos and websites showing you the process that others have followed.
Thanks for the hints!
I think these days a little too much emphasis is put on how critical injector data is. Of course if you're dealing with a VE based fuel model then injector data is more important, however as David mentions, you can still perform a tune but you'll be baking in any error into the main fuel table. You need to understand that years ago we had ECUs with no characterisation data at all and often aspects such as injector dead time were simply ignored. Don't get me wrong, where possible I'll always work with injectors where I have good data, but if you don't have this data you can still tune the engine. The only caveat I'd add here is where you get into reflashing late model ford and GM ECUs (obviously not the focus of your question but I'll add for the sake of completeness). In these ECUs you really do need proper data if you're swapping injectors, and just guessing is going to end up with a calibration that's a mess and horrible driveability.
Hi Andre, thanks for your input. Baking some error in is not a big issue and I am well aware of the past "no injector data" days, since my first tuning attempt was on a ancient EMS Stinger v2, but I gave it up due to lack of solid information.
The target I am after though, is a reliable tuning for a gr. A rally car that could ideally adapt to voltage changes that could keep the car from getting back to service due to a faulty alternator for instance.
Will at least the close loop AFR control still work with generic injector data?
If the closed loop AFR controller gains are tuned well and your other settings are close enough (VE table, air temp compensation, etc), it should be fine. You might be able to find some surrogate injector data to plug in by Googling around. For example, the era it was made will help... if it's the old EV1 style for example you can expect longer lag times.