For the sake of completeness and thanking everybody for their help, I wanted to bring everybody up to speed on what we've done to improve fuel economy and where things stand now.
First, we installed the Flow Force injectors that had been recommended. Second, the valve timing was way off as had been suggested so the shop re-baselined that. Third, the fuel pressure at the rail was lower than we'd realized so that was updated in the Megasquirt tune. Fourth, we modified the cold start curves to be a little less aggressive and burn less fuel. Fifth, we leaned out the AFR targets in non-boosted areas of the VE table.
So we made all these changes and retuned. The result is about 30 mpg on the highway, about 20 in stop-and-go traffic, and 23 in general city driving. This is a massive improvement over the 15-17 I had before. I'm hoping for even more improvement in the city fuel mileage once the weather gets warmer. I figure the higher energy density of summer gas paired with less warmup enrichment time should reduce fuel consumption.
The new fuel injectors really smoothed out the idle so I'd say that was a worthwhile addition. We suspect that my old RC 550 injectors might have been gummed up. But while that's possible, I'd run a lot of fuel injector cleaner through them over the time they were running my supercharger so I find that hard to believe. Another possibility is that the Megasquirt had trouble running them for some reason. And I suppose a final issue would have been the lower-than-expected pressure at the fuel rail. In any case, we're running much better now.
After making all these changes I had some stalling issues on warm restart, especially when running the headlights, which imposed a substantial electrical load. I noticed it was running in open loop idle mode so I switched to closed loop mode. I also set up the idle VE table and set the mixture to ~14.6. Now the engine is idling much better, it's able to handle different electrical loads, and the AFRs are more steady.
I think I'm still seeing some lean conditions after warm restart. I found the thread at Very lean on hot restarts
, which suggests modifying the ASE taper to extend the time that ASE runs. I'm going to make sure the current tune is stable and observe the ASE behavior for a while. I'll play with the settings in that other thread if it turns out there's a problem.
I want to thank everybody again for all the help. I really appreciate how we all contributed to solve this multi-faceted problem.