Learning to tune my car for Autocross and NHA HillClimbs

[powermizer]

I figured this as good of place as any for asking what are the max numbers we should see on the LNF? On injector pulse width? Injector duty cycle? I haven't pushed these too far, so if it can be known it might help all who don't know. Thanks

[powermizer]

Injector duty cycle should be no more than 80% to 85%? Pulse width is figured out by RPM and how many pulses the injector receives per cycle.

[mkriebs]

IDC in the LNF appears to backfire around 35%-40%, IIRC. I pushed my car a bit on E85 and that's what I noticed.

John, can I PM you about a few things? I know you're not one to check those often, but I would like to pick your brains on a few things. Thanks!

[cobaltssoverbooster]

Now I believe this particular thread hasn't evolved to this stage of tuning yet and injector timing is a very advanced function that should be taught after general tuning knowledge has been acquired, so I will keep this shorter in an attempt to keep questions about deeper functions of the ecu to the threads that have already been started on this topic. Please post in the injector threads to prevent confusing the OP so he can be taught at his own speed.

This little tid bit here will pertain to everyone as its can help define what is watched on the table primary setup in the scanner software. Most of us don't watch pulse width as much on GDI. Pulse Width (Percentage on-time) isn't as important as measured spray time (ms). The window of injector spray is a changing event with GDI, therefor 100% is constantly being redefined. What you want to watch is injector ms on time. Per what I have run across (which isn't every setup, keep that in mind please) is the average cap seems to be about 7.6ms of on time before the injectors start having issues. The maximum rail pressure I have hit was 2800 psi and at that point I noticed the injectors starting to have a hard time opening. Basically I commanded more pressure and my logs showed fuel trims starting to add fuel instead of remove it; all due to the injector struggling to open which requires more spray time to get the achieved value.

You can calculate the window of injector spray by referencing the opening angle and converting the spray time in ms into an angle reference based on rpm speed. So 1 rpm will move 6* in one second. At 1000 rpm the crank moves 6000* per second. To get ms to be an angle, divide by 1000 so: 1rpm is .006*/ms and 1000 rpm is 6*/ms. knowing this time based info you can convert each rpm range into how many degrees the injector is spraying per actual on time and compare it to any other crank events of your desire.

Full Example:
Data from log: 7000 rpm / 7.6ms actual on time / 360* Activation point in cam overlap / and last ignition advance of 19*

7000 X 6 = 42,000 deg/second now convert to ms angle value - 42,000 / 1000 = 42 deg/ms
now since the injector sprayed for 7.6 ms we can find the amount of degrees it was spraying for. 7.6 ms X 42 deg/ms = 319.2 degrees of crank rotation the injector was on for.
now a full 4 cycle rotation is 720 crank degrees with 0 & 720 sharing the same point which is TDC Compression. If you fired the ignition at "0" degrees then it would fire the plug at 0/720.
with this in mind the injector fired at 360* - 319.2* (which we have calculated it to be firing for) leaves us at 40.8* left before TDC Compression. This also means at 19* ignition advance we still have 21.8* of crank rotation before the spark plug fires AFTER the injector has finished spraying fuel. This is why in my injector thread I have debunked blowing the spark out by spraying fuel during the ignition event on factory bolt on applications. If you change to exotic fuels or swap turbochargers this event becomes more of an actuality and thus secondary injection systems are used to supply extra fuel efficiently without leaving the risk of injector/spark overlap.
In the picture I have attached you can see how the injector runs out of usable room as the Starting point of injection shifts. I left the data plot kind of crude as to make it all fit on the screen.

Myself and MikeM have both posted Injector finding on here. I explained how injector/spark blowout wasn't possible and how to track your cams against the injector data and Mike came in like I was hoping someone would and showed how mapping your cam events can further shed light on making efficient power as cam event to injector spray time can have big gains in numerous ways for all GDI platforms.

[powermizer]

Thank you for that info. I can see why its important to time it all right. I guess I should have said just for one who is watching a data chart. I'm glad I didn't now! lol Thanks again.

[TBSteck]

I have put the stock tune back in (with a write entire) and the car / engine still isn't running right? The last two stock tune logs are attached. Remember this car is catless now. Nothing is the tune has been touched and when I just pulled the DTC codes I only saw the P0036 - HO2S Heater Control Circuit (Bank 1, Sensor 2) code listed twice. The first one with a (SES) (Old) (History) (Current) (Immature) after the code description.

The catless 1 file was with me driving and the catless 2 was my wife driving with me looking at the real time logs and I could feel the car miss almost everything the Bst Des DC (%) would spike, this was reported using the TBS LNF 150603.cfg & VCM Scanner.cfg files upload in an above post.

Any help or insight or what I should check would be deeply appreciated.

Thanks to everyone that looks at this.

[TBSteck]

I came up with the idea that maybe the WB O2 sensor was bad, so I changed it. My new logs are attached, do they look better?

If you don't mind, let me know what you think. Remember this is the stock tune and my car is catless and the NB O2 isn't even connected.

[JimmyS]

Do you still feel the miss or not? If not, work on dialing in your fuel trims with the stock tune and keep on tuning.

[TBSteck]

No, if it did I didn't feel it like before that is why I think it is OK. Thanks for taking a look.