All good points. Please see my thoughts.
There are two ways that a "feedback carburetor" can work. Evidently that system works by relying on the base carb settings being lean, and then adding more fuel. The other way - the traditional one -relies on the base carb settings being rich, and then bleeding air in strategically (usually this was done into the metering circuits or emulsion tubes back in the day - not simply bypassing the throttle).
Agree. There were a few obscure 'feedback' carburetors that aimed for the target AFR and attempted to adjust lean or rich, but nothing of any success.
The nature of my real job automatically makes me think FMEA. What happens if the automatic trimming system fails. If the system relies on base carb settings being lean, and aux fuel delivery fails (either due to loss of fuel supply or loss of injector signal or who knows what), then it reverts to base carb settings ... lean. The traditional way of doing this reverts to base carb settings ... rich. Seems to me that the traditional (air bleed) method is safer, especially if you are using this in a somewhat high performance application.
When thinking FMEA, it seems that EFI would be the last option one would choose. The automatic trimming system on the K&N unit is basically a 'mini-EFI' with an ECU. With fewer sensors needed it is less subject to failure than a regular EFI system. If an EFI system fails, the vehicle is basically stranded. Not so with the ECI unit and a carburetor. And fwiw, it's important to recognize the degree of "lean" that an ECI user would set the carb. The
most I've leaned a carb in testing is 1. in AFR, so if target is 13.0, I've leaned to 14.0. At this number, the car runs fine, other than an occasional light surge at certain throttle positions with no increase in water temp. ECI "influence" is minimal, therefore a system failure is certainly not catastrophic. So if I have a street rod and the ECI fails, I drive home on the carb-only. If I have an EFI failure, I'm calling a tow truck. And if I have a drag car and experience an ECI failure, I richen the carb and continue. If I have an EFI failure, I'm trailered until proper diagnosis and replacement.
Feedback carbs way back in the early days of 3-way catalytic converters basically just took the oxygen sensor signal, processed it a little bit (electronics were not very sophisticated back then), and applied that to the solenoid(s) for the auxiliary air bleeds. The base carb settings were such that at wide open throttle, the engine got the slightly-rich mixture that it needed even with the air bleed in effect. With cold engine, before the oxygen sensor gave a signal, it just ran at the base carb settings, slightly rich. If the circuits or solenoids failed in the other direction (commanding it lean), the air bleed had little enough effect that the engine would still run (although perhaps a bit rough).
Like you describe here, the base carb settings at WOT are set for a slightly rich mixture and the ECI is not in use. There is a programmable cold start feature on the ECI that works well. You lean the idle circuit slightly and program the cold start based on duty cycle and time. Once the O2 sensor is warmed and operating, the target AFR for that RPM range is in effect.
Feedback carburetors were not a thing done for "high performance". It was strictly for emissions. The early approach of using EGR for NOx reduction and air-injection and an oxidizing catalyst for CO and HC had reached its limits. Three-way catalysts were needed, and they needed the engine to run very close to stoichiometric.
Yessir. I have not experimented with the ECI using an 'emissions' mindset, but it would be fun. It will never match the OEM/EFI systems for emissions, but it would be an interesting project to spend a little time on.
Feedback carbs were from the era of sub-200-horsepower Corvettes - the dark ages of emission-strangulation. Or my buddy Al's mid-eighties Dodge van with a 95 hp emissions-strangled single-barrel-carb slant six.
And 4-cylinder Chevy LUV trucks!
I had a couple of vehicles from that era. The vacuum hose routing diagram looked like a bowl of spaghetti.
Yessir. I 'converted' more than a few to non-feedback and the top of the engine compartment cleaned up nicely with the spaghetti gone.
There is no valid reason why you would want to do something like this in this day and age.
The ECI is targeted at those who want to continue running a carburetor. For street rods it's period-correctness, reliability and simplicity, for racers it's performance, reliability and simplicity. Cost savings is a factor in both. It is not intended to replace EFI for the intended EFI market.
