So I've made some serious headway with my electronic controller. It's obviously been working well enough for me to be driving lots - I've put about 15,000 km on it - but I've never been one to leave well enough alone.
I've got it working REALLY REALLY well now...
EGT is in degrees, Duty is in % of usable range, TPS is 0-255, boost request and actual are in kPa (multiply by 10 for mbar, divide by 100 for bar, divide by about 6.8 for PSI)... Peak boost is 210 kPa / 2.1 bar / 2100 mbar / ~31 PSI
That's a highway onramp run, cruising as slowly as I could get away with in 2nd on a long onramp and then WOT on to the freeway. My 3-4 shift was pretty fast, the log didn't even catch my foot coming off the pedal
I do use some averaging on the pedal plus the logger doesn't show EVERY data point used by the actual control loop.
The freeway has a decent uphill grade here too so this is a pretty solid test of my fueling levels and things. Smoke is limited to a decent puff at tip in and then a light haze that's gone pretty quickly. Judging by my peak EGTs (BARELY over 700C at the end of 4th, at which point I was going more than fast enough) there's still room for some more fuel
I've made some code changes that help make things a bit more responsive over the past little while... I've been somewhat conservative as earlier versions of my code had some issues spiking way over desired boost. Now that things are well under control I can make it a bit more aggressive. I think I'm also going to add a vacuum ball for the 'N75' as it does take a little more time than I'd like to close the vanes fully.
There's a little bit of jumpiness on the 2nd gear part of the run that I can see in my logs is me hitting the top of my max duty cycle map at that RPM, looks like I need to expand the available range there just a tad. Again I was being fairly conservative with a lot of the settings while I got things under control.