2008 USDM Legacy GT, 5EAT
EJ20X longblock, with avcs delete kit (stock cam, but modified with a delete kit)
93 octane
2008 LGT upper/tgvs/rails/injectors/intercooler etc. everything above the heads is OEM/USDM
VF52 with wastegate port entry and exit ported, VF52 restrictor pill, stock BCS
*stock turbo back exhaust
2014 WRX Intake
Air pump delete, baro sensor from top of pump intact
so a basic "stage 2" (downpipe) and full tune.
the only problem we've had so far is that it's not reaching target boost, which is ~15 psi. I worked up the wastegate duty until I felt it was not being realistic, and then upped the boost controller frequency to 20hz, and worked up from near the bottom again. and at 74 to 80% duty, it's only getting 13.5 to 12ish psi of boost, give or take
either the wastegate flapper now sees much more exhaust gas pressure and is easier to open, and the boost controller cannot vent enough flow to decrease pressure against the wastegate actuator, or the stock exhaust cat-back with 220,000 miles is clogged and limiting to 170 to 200 g/s range
note that from 2008 to 2014, the same engine bay fuel system setup was used on the LGT and WRX, and that the 2014 WRX was the most updated per cylinder injector compensations in an OEM calibration, which I used in this. that's how I know. I looked at the fuel trim corrections early on when using the OEM per cylinder injector comps, and looked at the 2014 WRX OEM tables, and figured it out. I went online and found the OEM injector feed parts and compared them visually. same/same.
cruise (tgv closed) avcs and timing are modeled from OEM 2014 WRX, with correction for displacement and compression ratio difference
off-cruise (tgv open) avcs and timing are modeled after a 2007 STi which I tuned extensively, and is corrected for displacement and compression ratio difference
I have another sort of example I'm working on, but haven't gotten any logs back from the owner and he's got some sort of problem with his upstream as he's getting a P014E code. it's an ej20x in a 2012 STi, with the exhaust avcs being utilized. it's using the same cruise avcs and timing, with 0 commanded exhaust avcs as here, but then the intake/exhaust avcs + timing are derived from the 2011 JDM S206 (aggressive overlap) for TGV open, compensated for compression ratio difference of course. I guess it drives fantastic, but I was hoping for datalogs. He didn't seem compelled to message me at all until a couple days after I messaged him about 10 days after it was flashed. success, I guess?
in any case, if you want to take your OEM calibration and modify it for a different displacement and/or compression ratio, use boyle's law for correction factors against avcs and timing airmass axis. huzzah.
if it's general consensus that the dash/logged parameter for (mpg est.) is about 10% off, then 38 to 40 mpg average is still 34 to 35 mpg.
we'll eventually figure out the boost control problem, but the rest of the tables are good, so pay no mind to copying wastegate duty/target boost. DBW tables IIRC are from a 2008 STi.
| Attachments: |
|
File comment: closed loop, open loop, and open loop throttle/pedal min afr tables closed loop of course is 14.7+offset, so a value of 0.441 means it's requesting 15.141 when up to temp in that cell both tables are based off a 2D MAF g/s to lambda (gas AFR of 14.7 = 1 lambda), which is determined by displacement, compression ratio, and fuel grade. because of this, it makes MAP*RPM by IDC and MAF volts by IDC much much smoother. also the RPM and load points are chosen to maximize the resolution for where each table is useful. you don't need low low/low rpm on the open loop table because it wouldn't be transiting to open loop anyways resolution maxxing fueling.PNG [ 111.38 KiB | Viewed 581 times ] |
|
File comment: intake avcs and injector timing (key component for mileage) avcs.PNG [ 133.87 KiB | Viewed 581 times ] |
|
File comment: average of all data excluding lines with vehicle speed being 0 and/or mpg estimate being 0 mpg2.png [ 13.05 KiB | Viewed 581 times ] |
|
File comment: average of all cells for mpg estimate, including all data with vehicle speed zero and mpg estimate of zero mpg.png [ 47.26 KiB | Viewed 581 times ] |
File comment: file from datalog
MP-Stage2-EJ20x-08LGT-v9.zip [424.09 KiB]
Downloaded 88 times |
|
File comment: 19.5 minute datalog
log03032024.csv [1.38 MiB]
Downloaded 34 times |