Feature wish list
Turbo Mini Forums

There is now an initial code version that supports injection timing (with respect to the start of the injection pulse) with a 3-D table, injector staging with either 2 tables or with a single table and a ratio for any injector mix of size and impedance, and mid-pulse timing with a single advance value (with one advance value for unstaged and another for staged injection). This code version has sacrificed ignition capability (for ease of implementation) and, potentially, does not provide very good accuracy of injection timing during certain situations such as engine startup and very rapid acceleration.

With that in mind could you all provide the features/behaviours that you wish the code would have and in which priority they should be tackled.

Some of the features I think should implemented, in no particular order, are:
- ignition control
- better injection timing accuracy
- complete open valve injection mode where the injectors are pulsed on the open valve for the inside cylinders as well as the outside cylinders instead of only for the outside cylinders
- single overlap injection mode where a single injector pulse is done to provide both the inside and outside cylinders with fuel. the injector pulse starts when the inside cylinder intake valve is open and finishes when the outside cylinder intake is open.
- a mix of the 2 previous modes depending on rpm, load, pulse width ?
- automatic adjustment of the injection timing using feedback of 2 O2 sensors
- fuel cut-off if injection pulse width is larger than the available injection window
- good documentation of the code features and how to use them

So add to this list or create another one and try to give which priority you think the features should have.

Regards,
Jean

http://www.jbperf.com/

When you say fuel 'cut off' do you mean it just won't accept any increase over the stated injector duration - no matter what is in the fuel table?
Cutting off (switching off) the injector could be pretty dodgy for a turbochared engine being set up on a RR.

The unit i'll be playing with is going to have the 'outer limits' set up using two O2 sensors like Marcel did - and that then has pre-defined 'limits' stored. There will be a function that allows a 'countdown' towards these limits - basically showing how much injector duration % you have left. It'll also ativate an LED at 95% duration (in the pre-defined injection window) so you have a visual 'alarm' that you're in need of bigger injetors or more fuel pressure.

I imagine that isn't going to be easy to set up inr MS however? Does it have auxillary outputs that can be used (say whilst in dyno mode) that show this (maybe use the shift light or other output)?



I am tempted to say that in respet of the 'open valve' and 'overlap injection' modes - I feel that is getting a little too far ahead at this time.

You've already set up a 'fixed' (but manually adjustable) injetion 'start' point that operates like a normal injector pulse - and one that uses the advance map to 'move' the start point dynamically.

You've also fixed up a 'symetrical' injection duration about a 'fixed' (but manually adjustable) injetion 'start' point.

To my mind - those now need to be validated - ideally with the use of the MPi fuel injectors.

Edited by TurboDave16V on 7th Jan, 2006.

Having a LED activated when the pulsewidth reaches a certain limit would be quite easy to implement in the code and there are auxiliary outputs on the MSII. It would be more difficult and computation intensive to have that as percent of duty cycle but still not too bad.

By fuel cut off I meant switching off the injectors as is done for rev limiting. This basically turns off the engine until a lower rpm is reached. I'm not familiar enough with how an RR session is run to see the problem but this method of rev limiting is used frequently. I think that the alternative of having a cylinder go lean at high load and high boost would be much more serious. What kind of rev limiting would you be using?

As for the different modes, I was just listing the different possibilities. I don't think they are on the top of the priority list either. And I do agree that the modes already implemented need to be validated. The most important features in the list in my opinion are the first 2 and the last 2.

Jean

http://www.jbperf.com/

Another point I should add is that some of the computation limits could be done in Megatune and presented in the real-time display. However, that means that a manual operation would be needed for protecting the engine.

http://www.jbperf.com/

Since you asked me to put this in here for you...
How about dual VE tables for a single injector to address the pressure differential between the 2 cylinders in a single port introduced my the ram charging effect of the second to fire cylinder.
I hope that description makes sense...

-Tyler
DO NOT TOUCH MY HORNS OF DOOM!

I was flicking through some old mags over the weekend and came across an article by dave walker where he was using diferent timing for the inner and outer cylinders on a CVH, aparently it worked to an extent, with the mini head I get the impression that this give a better result than the CVH.

The reason they don't work is the feedback I believe - or maybe the sampling rate of the unit?
DTA has tried both ways of doing it, both work when holding rpms on a dyno - or increasing rpm's slowly - but try dynamically accelerating from 1st to 4th, and it's a turd. Nothing more.

TBH - I'll be happy for a basic MS with A-series controlability. That's all i need.