Fuel Page for MembersPage/BenNesbitt
sequential fuel delivery for 8 cylinder Audi engine
firing order: 1-5-4-8-6-3-7-2
Bosch fuel pump in the 80/90 is the same part# as A8 and I think fuel pressure is around 2.5 bar (?)
Injectors are stock A8 units with the idle vacuum port plugged but I plan to upgrade to 400cc Delphi in the near future
Did you flow them? My A8 injectors were 200cc@4bar.
that is awful, no wonder they run out of steam with an exhaust and intake :P
Ok, here are the latest panels from megatune that result in the motor starting and idling for a few seconds. Any throttle input kills it and the idle basically just tapers down to nothing after starting... So like I guess the fuel delivery is not happening.
I am going to go hog wild on the megatune site to figure these out but feel free to add your ideas:
- With alternate=(7..0), divider=1 is correct (divider=8 is seriosuly wrong, means injectors firing every 16'th crankrot instead of every 2nd - With the recommended way, starting from a working config, chance would be lower for so many bad initial settings. If you found this in MegaTune docs: I think the recommended divider was high in the doc, because megasquirt users usually drive injectors with 1 or 2 channels instead of the literally cooler 4..8 )
Until your new ECM with 8 flyback arrives :
- with alternate=(3..0), divider=2 is correct. This assumes injectors are grouped in pairs. This means there is no need for the external flyback for the 2 channels (EC36pin8 and EC36pin17), since they are not driving inductive load (not used at all);
- With alternate=(7..0), divider=1 you can reuse the working flyback of other channels:
- connect EC36pin8 to EC36pin9
- connect EC36pin17 to EC36pin18
- This means we reuse the working flyback of 2 channels, to help out the EC36pin8 and EC36pin17;
- sufficient h[0]=01 02 08 40 00 14 00 A0 # these are in hex. As you see from GenBoard/Manual/DigitalOut/Table we masked together (EC36pin8 with EC36pin9) and (EC36pin17 with EC36pin18) so the channels are fired together, and the pulse is not accidentally enlarged for any channels. (it would be enough to fire either of the connected channels, since they are wired in OR connection: but firing both simultaneously has some advantage: shares the load between 2 FETs).
This shouldn't matter and is really a wild guess, but change it anyway, set injpwmDC (%) to 100%.
After doing so and making sure that it has been saved in the ECU you need to go to the terminal and type Manmcd, in the output you'll have to find the injpwm=FF (what you just chnaged) and verify that it's really says FF. The real reason of the excercise is to check that the injpwmt=FF that should be right above it also says FF.
To make it easier for us to verify the config it would be great if you ran Manmcd and pasted the output in a wiki page. -Jörgen'
