Quote:
Originally Posted by Luckrider
I have kind of skipped the last several (alright probably 10 or more posts), but @ jamesm, I have seen you post more than once that you don't touch the DI (which you repeat here). Do you think there is merit to @ mad_sb's method of scaling DI and PI since it is know that the stock PI is a limit at some point for fueling the FA20 (granted, that is with e85 and some boost). Has anyone even determined what the effective limit of the direct injectors is? |
I've read that they flow around 400cc/min but I can't remember where. I do think that raising di pressure targets could increase their capacity, but I don't think that's why he does it. I think his idea is to get fueling error on the di side as low as possible, which I've considered in the past and just never tried. You reach a point of diminishing returns around 3-5% error across the board in terms of low speed drivability, which is where low fueling error is very important.
Another reason that we employ different methods is that every car I've tuned has been turbocharged (cars don't seem to stay NA for long around here lol). When the maf and injectors are replaced at once, I need to use the di in its stock configuration as something to scale the maf against (just up and down for the larger maf housing, not fine tuning the closed loop portion, that's done later once the fueling systems are recombined). If I were tuning a car with a stock airbox and maf, that would be my approach. Keep the stock maf scaling and tune di and port injection both to it. The idea is that you can only solve for one variable at a time, and being that the di is the only known quantity in an aftermarket turbo install, I don't touch it.
I will definitely adopt his approach when I get a chance to tune a car with a stock airbox. It seems that you could ultimately achieve lower fueling error that way, though I'm not sure it'll be noticeable as <1% is very doable on the turbo setups with my method. It's two different solutions to different problems if I'm understanding it correctly. I'm sure we'll be collaborating on even better methods in the future. Were all just kinda figuring this out, and no one told me how to do it my way, I just came up with it by chasing my tail long enough and trying different things. There is no real 'right' IMHO, it's what works for your situation. In the end it's low error you're after, and if you achieve that you've done it right.