Lets talk Aux fueling.. Shotgun or PI?

LamboLover

Corporal
Apr 6, 2017
238
242
0
Ride
Everything
I wonder about ignition cut followed by DME controlled batch PI cut and engine load cut on all cylinders. If you cut the PI too quickly whilst engine load was still higher than DI alone could cope with you could induce a lean on other cylinders or would the HPFP not crash that fast (thinking time to drain the inlet manifold vs the fuel rail)? What circumstances are we covering? Is it just misfires or are there other circumstances?
 

doublespaces

Administrator
Oct 18, 2016
9,303
4,331
0
AZ
Ride
2009 E93 335i
If you cut the PI too quickly whilst engine load was still higher than DI alone could cope with you could induce a lean on other cylinders or would the HPFP not crash that fast (thinking time to drain the inlet manifold vs the fuel rail)?

Assuming we don't explore sequential control due to complexity, this is a really good question. But is it worthwhile to cut the fuel entirely? Would this be like giving up on the entire pull after a single misfire event?
 

martymil

Major General
Sep 6, 2017
3,331
1,907
0
Down Under
Ride
S65 1m
I've also had people tell me that the AIC controller is not ideal in part throttle situations, spraying fuel when you're merely holding a gear and feathering in and out of boost. I've never ran PI on this car so I can't say first hand. But I still think there is a need for a DME controlled PI controller. Needs to have per bank fuel trims taken into account or something even better than that.

In the ideal situation you would need dme controlled sequential pi and shut down the cylinder completly that is misbehaving that's including spark and di
 
  • Like
Reactions: doublespaces

LamboLover

Corporal
Apr 6, 2017
238
242
0
Ride
Everything
Complexity vs gains perhaps. In my initial compact OBD design I only have space and thermal performance for one 9A low side injector driver. However, a device that plugged into the same OBD pins (but was not constrained by OBD enclosure size) and therefore also has CAN and crank position signal could do full sequential with DME control by having six 2A low side injector drivers. The main thing is working out what and when to feed them with.
 

doublespaces

Administrator
Oct 18, 2016
9,303
4,331
0
AZ
Ride
2009 E93 335i
Complexity vs gains perhaps. In my initial compact OBD design I only have space and thermal performance for one 9A low side injector driver. However, a device that plugged into the same OBD pins (but was not constrained by OBD enclosure size) and therefore also has CAN and crank position signal could do full sequential with DME control by having six 2A low side injector drivers. The main thing is working out what and when to feed them with.
To be done properly, perhaps this this means more electronics in the ECU box? I think people would much prefer to buy something like this instead of the ancient AIC. Just a matter of development time/effort.
 

LamboLover

Corporal
Apr 6, 2017
238
242
0
Ride
Everything
I wondered, because of the extra unused CAN bus on the later ECUs, but when I realised that a crank signal was at the OBD port, the latency is not so crucial and all the CAN bus has to carry is either individual cylinder pulse widths, or a global pulse width and a mask.