gemellocattivo.com

Ok, that should do it.

I decided to make the MAP smoothing a 2D table of rpm and MAP reading so I can play around with different options. Once I get it how I like it I'll see if an equation can replace the table but for now I have a lot of flexibility.

Rather than add a new timer to the coil power relay I used a function that is built into every item....the ability to skip iterations on the thread. I have all the main fuel and spark control on a thread that runs every 7msec. The coil power is on this thread and now comes before the other spark stuff and is initialized to 0 (off) so first pass through it is set to off, then the stuff that turns it on are read so it would turned on next pass, but the item is skipped for 8 iterations/passes or 7x8 56mec. Off for 56 msec which should be plenty log for the control signals to come on and shut the coils off before to main power arrives. The devil is in the details.....

Sounding better....a bit of smoke at the moment though

https://youtu.be/2cOeKkw_0JI

I was wrong on the coil signal...there was an error in my code setting the dwell is 0.

Sounds good!

Thanks...and that is the quiet muffler and the engine not cracking 3500....redline is 9500 and the other mufflers don't do a whole lot of muffling I fear

Wow, complicated stuff.
Good job.

_________________
Wade Williams
Master Ferrari Tech
Restoration Tech
Engine Machinist
Fabricator
Racecar Driver

Everything about this project seems to be right at or just past my limit to comprehend whats going on.......
Heres some MAP data.
gold is crank position 0-720 so every peak is an engine cycle
Red is the signal coming from the MAP senor board zoomed so full scale is 30-50 kPa
Cyan is with the smoothing I added....mostly damped and almost no lag.
Blue is the average of the cylinder MAP readings, very smooth but lots of lag becaseu even sampling at 500hz they don't nessesarily update for a few cycls depending how the timed rate matches or doesn't match crank position.


Heres the lambda data.....I now they were all running and I know the plus I pulled where black...so I'm having trouble believing that readings that show them lean. I'm pretty sure I have a new sensor on the self somewhere, I need to find it and install it.

I have a fuel pressure problem
The cyan line in the bottom graph is fuel pressure, the blackline in the top graph is what the fuel pressure is doing to injector becasue it self correct injection time to compensate for fuel pressure variation....but this is crazy with pressure bouncing between about 7 and 40



Here's another later run where its calmed down a bit and only bouncing 19 to 28 but drifting down the whole run.

Now is it the transducer, pump or regulator?

Or are your huge injectors and comparatively small fuel rail causing the pressure to fluctuate like that?

good question....it could be an ECU issue too as I have code to slow the pump at low demand conditions, but that SHOULDN"T be doing anything but 100% the way its currently setup.



it is a -10 rail which means the correct ID to thread in a -10 fitting so it much larger than the -10 hose feeding it. also this is light load at low rpm....I did pull some older logs and found good pressure at all times. I know I check this way back, I see it yesterday in parts of runs, and every run seems to start with good pressure.....I'm thinking pump.

Bosch (and Denso) put pressure dampners on the supply side of the fuel system to minimise pulsations and spikes. If you can find room on the rails, or even on the lines feeding the rails it might be money well spent.

Bosch put them at the pumps, while Denso put them right on the rails.