Hello everybody, this is my first post on this very good forum.

/*START OF BORING PART :lol: */
My name is Cédric and I'm 24. Cars is one of my passions. I'm working for a big automotive spare parts manufacturer as engineer but not for ecu...

I would like to understand a lot of things about motor management for my knowledge and maybe after to be able to start modifying maps on my own car (S3 TFSI) :angel: ( or maybe first on the car of my girlfriend corsa CDTI!!!!!).

/*END OF BORING PART*/

1) My first question is about timing injector maps: As I have seen in a tuto which describe maps for a TDI the timing injection is given in angle of the crank shaft. Why this map is not given in milliseconds or microseconds? As we know the pressure value of the fuel, the time in milliseconds or microseconds should give us exactly the same quantity regardless of the rotation speed?

2) Why do we need to specify a N75 map whereas we have a turbo pressure map? In my mind the N75 valve is slaved to the turbo pressure map and the ecu should adapt the value of the N75 to always be able to respect the specified value given by the turbo pressure map... :cry: .

Thank you very much for your attention ;) .

For 1), opinions might vary but I think its easier for an engineer to visualise the injection window when everything is given in degrees - durations in the time domain might seem to make more sense, however a static period of time of injection duration or advance can mean completely different things in different situations therefore I'd imagine it makes more sense to tune using the engines cycle as reference.

For 2, you have to understand that vnt turbos are not on, off - they can vary infinitely between vanes fully open, and vanes fully shut. To give the ecu some idea of where to put the vanes in a particular situation, the n75 map offers a value which should be not far off the mark. In some situations the ecu will adapt this value in realtime, depending on the ecu's own settings and the driving situation, so the value in the n75 map isnt what you always get, but the more accurate the map is set, the better in general your boost control will be because the ecu's own adaptation will be small.

Just 2p from a diesel tuners perspective.

Thank you Dakta,

For the first point I think as you said that is purely subjective and I stay on my opinion that is more simple with time in seconds than in degrees :mrgreen: .

For the second point, I'm not ok with you: it's totally feasible to regulate the value of the turbo pressure using the N75 without map for this one. If we have the required pressure value it's possible to set the value (in %) of the N75 valve using a algorithm (PID for example).

At the beginning, I understood that this map was maybe to limit: values to never exeed for the N75..

For the first point I think as you said that is purely subjective and I stay on my opinion that is more simple with time in seconds than in degrees :mrgreen: .

Well Cerdric, you are free to your opinion, unfortunately it changes little - degrees are a more appropriate unit in a lot of cases so it gets used, so anyone manipulating the values will use it too - so that's pretty much that.

For the second point, I'm not ok with you: it's totally feasible to regulate the value of the turbo pressure using the N75 without map for this one. If we have the required pressure value it's possible to set the value (in %) of the N75 valve using a algorithm (PID for example).

I know this, because I've done it.

I developed a micro that used PID to regulate boost, I also wrote another peice of firmware that used openloop (i,e an n75 style map) and used pid to adjust these values as I described previously.

The latter was better. I won't go into too many details, but there's more to boost control than making a turbo hit x psi, it's engine-flow management which controls both intake and exhaust pressures, for which tight, closed loop control of the intake pressure is not always the most effective approach in a given moment. The priorities of an engine management system are as dynamic as the boost the turbo generates.

Now, you don't have to accept the information presented, but engineers do things for a reason, and as I said before, that's pretty much that. Whilst you may not agree with what I have said, you should find my response somewhat reliable in terms of facts regardless.

the N75 map are used for "instant" reaction of pedal (could avoid spikes), for example, your turbo map at 4000 rpm are 1bar of boost, when you release the pedal the boost will drop, when you accelerate again the vnt "will start" from position that produce max boost (since boost are under 1 bar). At 4000 rpm turbo will produce a huge pressure very fast (spike) and the valve n75 will not reduce fast as pressure increase (ecu pid algorithm). With n75 map the ecu know for 4000 the position of vnt "will start" from the middle (50% for example) the spike will be smaller or none.

Hey. I have done such a geometry controller as well, and I could only effectively control the boost spikes by using a Boost Map which would be a reference to the "close loop" controller, but also in addition using an "open loop" kind of map (like the N75 map, but in my case I was using an electric actuator) just to first position the vanes, like an offset to the controller. If I tried to use only the "close loop" controller as Cedric suggested, I could never set the PID parameters to cope with all different situations, resulting in serious oscillation in some situations while it worked ok in others, specially when you floor it and that damned boost spike hits so fast :D

Both systems (open and closed loop) are necessary for reasonable control of boost.

Basic control using both open and closed loop control blended together can be written in surprisingly few lines of code, so it is an approach I would recommend based on my own experiences.

Closed loop for stability, open loop for response.

Unfortunately the OP was unable to benefit from this knowledge, but it does make interesting discussion.