killerpenguin21 Posted October 9, 2012 Share Posted October 9, 2012 let me preface by saying i am not new to tuning, i am the sole builder/tuner of my track toy evo. ive wrapped my head around most of the subaru ecu workings, but not the finer details...and thus went for an accessport so i dont try and cloud my head with a whole new tuning style or something. apologies for the GIANT first real post. 2005 GT, 90k miles, AVO dry panel filter, new NGK plugs at time of flashing to AP stage 1 (500 miles since flash). LV done after about 250 miles of highway driving. http://i17.photobucket.com/albums/b78/killerpenguin21/LearningView_SS_9-24-2012100635PM.jpg 100 miles later http://i17.photobucket.com/albums/b78/killerpenguin21/LearningView89900miles.jpg about another 100 miles later http://i17.photobucket.com/albums/b78/killerpenguin21/LearningView90kmiles.jpg Boost plots http://i17.photobucket.com/albums/b78/killerpenguin21/log2boosttables.jpg now, my daily commute is about 3 miles and the car barely gets fully warm, so my low end trims are constantly all over but within limits, not really worried. my concern is the D trim obviously, since thats a lot of fuel to be pulled out under power if it truly does effect closed loop. but i cant for the life of me duplicate that airflow not under boost and in closed loop. ran some logs through virtual dyno and the car comes up at 220whp/240ish ft/lbs, with ZERO knock anywhere. but a boost error of 2.5ish psi makes me think something might be going on...but it could also just be the stock turbo falling off based on where it starts to get high. Does this look like a leak somewhere, or just a stock performing MAF that might need to be scaled? Link to comment Share on other sites More sharing options...
G5892 Posted October 9, 2012 Share Posted October 9, 2012 My trims move around a lot also. Car runs fine so I didnt pay any mind to it. Subscribed in case anyone has an idea of whats going on. Link to comment Share on other sites More sharing options...
NSFW Posted October 9, 2012 Share Posted October 9, 2012 Interesting, I wouldn't have guessed that a panel filter would make that much difference, but I never ran one. The 40+ range is hard to stay in for any length of time because it's right on the edge of closed-loop conditions. You can get there with gentle acceleration, but since you will be accelerating, you end up well beyond the speed limit after a couple/few seconds, so it's hard collect data from that MAF range. If you can find a big hill, that helps. I've used a mountain pass to stay in the 40-60 range for extended periods. The -7% correction can cause your mixture to be 7% leaner than the fuel table calls for, if your MAF scaling is accurate. But maybe your MAF scaling is actually 7% too high, and those two errors cancel out. There's really no way to know what your AFR is without actually measuring it. If you're not knocking, then it's not really a big deal, but you should log it periodically to ensure that you don't start knocking when the weather changes or something like that. I have a hunch that the target boost table is just calling for more boost than the turbo is capable of making, and the boost error isn't really a problem. Some people just like to tune optimistically in an effort to minimize lag. Try plotting those curves over RPM rather than over time, and then compare to other people's boost-over-time curves for stage-1 setups. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 9, 2012 Author Share Posted October 9, 2012 ive seen various reports about the factory maf scaling...seems like it wasnt every really the greatest? it seems that i couldnt get it to knock. tried low rpm 4th gear starting the pulls and nothing , so it seems that if i cant get it to knock like that the afr's are probably decent. subaru ring lands just concern me. when i get some time off, i need to finish my new ownership maintenance and redo the main vac lines. if the accordion rubber elbow to the throttle body a common place for splits/vac leaks? that gave me a huge headache on my last car... Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 9, 2012 Author Share Posted October 9, 2012 i think my next project is to log the maf with the romraider maf tool and see if it recommends some crazy corrections or not...and try and convicnce myself its not sketchy without a wideband...just dont plan on modding this car that much. Link to comment Share on other sites More sharing options...
fahr_side Posted October 9, 2012 Share Posted October 9, 2012 I'd ditch the panel filter and fit a new stock one. As NSFW mentions, it's hard to stay long in that range, especially if you have zeroed out the CL/OL delays. When you think about it, this range is the one where you're least likely to have problems with leaks because it's likely that MRP is going to be very close to zero... not much reason for anything past the turbo to leak at that point, and if it were a leak somewhere between MAF sensor and turbo inlet you'd be in positive trims, not negative. Good though that you're concerned about this, as many people don't realize the D trims are carried over into open loop. The other comment that the stock MAF scaling is suspect is dead-on... it's quite rich in most cases. Subaru does this for safety... you have a big margin for error if there's a leak in the intake, the injectors are dirty or whatever. Just note that if you correct the MAF scaling the side-effect is that the ECU sees it as lower load, and that means timing will be advanced accordingly. I can't really tell you if the boost targets are reasonable or not without the actual values on the graph. Note that they need to be set high enough that they are reachable when the car is loaded to the max and pointed up a hill, but not so high that turbo dynamics gets into a bitch fit trying to reach them when on a flat road and just the driver is in the car. Perhaps try the worst-case scenario with some full-size people in the car, and log a pull uphill to see what the boost error looks like then. Please log not only WGDC, but also turbo dynamics integral and proportional so we can see what's really going on. OTOH you could take the fact that you have no knock to mean you're running pig-rich as Subaru intended and there's nothing to worry about. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
NSFW Posted October 9, 2012 Share Posted October 9, 2012 You'd need a wideband O2 sensor to really know for sure what's going on, but IMO if it's not knocking it's probably just fine as-is. But I gotta say, if you've already built and tuned an Evo, then it's a no-brainer to give your LGT a catless up-pipe, a high-flow-cat downpipe and a tune. The power-per-dollar is great, and it should be plenty reliable if you just check on it periodically to make sure knock is minimal. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 9, 2012 Author Share Posted October 9, 2012 I'd ditch the panel filter and fit a new stock one. As NSFW mentions, it's hard to stay long in that range, especially if you have zeroed out the CL/OL delays. Good though that you're concerned about this, as many people don't realize the D trims are carried over into open loop. The other comment that the stock MAF scaling is suspect is dead-on... it's quite rich in most cases. Subaru does this for safety... you have a big margin for error if there's a leak in the intake, the injectors are dirty or whatever. Just note that if you correct the MAF scaling the side-effect is that the ECU sees it as lower load, and that means timing will be advanced accordingly. but not so high that turbo dynamics gets into a bitch fit trying to reach them when on a flat road and just the driver is in the car. Perhaps try the worst-case scenario with some full-size people in the car, and log a pull uphill to see what the boost error looks like then. Please log not only WGDC, but also turbo dynamics integral and proportional so we can see what's really going on. i dont really see a panel filter in the stock air box really messing with anything though? the previous owner put k&n in it so im crossing my fingers he didnt oil down the maf. need to find some time to clean it. I was just reading on rescaling the maf and the effect on load this morning actually. the maf plotting in rom raider looks pretty neat, so ill at least be able to see where its at, ive never dealt directly with maf scaling, the evo is a KV maf, and they rarely seem to need scaling. I have not zero'd out the CL/OL delay, but i didnt look at what cobb sets it too, but i know they change it, as there a/f curve vs. stock is the main reason i knew i needed a tune. The turbo dynamics stuff outside of target boost and wgdc i have not read into too much. i think based on how close my actual boost readings are to target, that i dont have an issue there. You'd need a wideband O2 sensor to really know for sure what's going on, but IMO if it's not knocking it's probably just fine as-is. But I gotta say, if you've already built and tuned an Evo, then it's a no-brainer to give your LGT a catless up-pipe, a high-flow-cat downpipe and a tune. The power-per-dollar is great, and it should be plenty reliable if you just check on it periodically to make sure knock is minimal. trying to make a fast car on street tires within rime attack class rules lightens your wallet pretty quickly...im really attempting to not mod this car lol. we all know the story...well i want a dp, but if i get a dp i might as well get a wideband bung on it, and if i have a wideband then i can up boost and...and...and... Link to comment Share on other sites More sharing options...
fahr_side Posted October 9, 2012 Share Posted October 9, 2012 The stock filter is not restrictive until around 400whp. If a same-sized panel filter flows more air at a given pressure drop it will also filter less well... Yes, your MAF might be oily from a previous K&N install but that usually messes up the other trim ranges too. I'm a big fan of zero CL/OL delay. It makes fueling far more predictable, improves driveabilty and has only a small impact on economy. Same with flattening the tree-hugger spike in AVCS timing at low rpm / load. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 10, 2012 Author Share Posted October 10, 2012 The stock filter is not restrictive until around 400whp. If a same-sized panel filter flows more air at a given pressure drop it will also filter less well... I'm a big fan of zero CL/OL delay. It makes fueling far more predictable, improves driveabilty and has only a small impact on economy. Same with flattening the tree-hugger spike in AVCS timing at low rpm / load. i only bought the avo since i wanted to replace whatever was in there in my round of maintenance and what not when i bought the car...thought it had decent reviews, didnt get it for any gains...that being said i dont see how the filter would mess anything up on a completely stock, but i may be wrong? ill take a look through access tuner after work tonight and see if i can find anything in regards to the CL/OL, i know ill be able to see the AVCS map. Link to comment Share on other sites More sharing options...
bnguyenbb6 Posted October 10, 2012 Share Posted October 10, 2012 Just curious, how many times did you fill up between LVs? Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 10, 2012 Author Share Posted October 10, 2012 about 1 and a 1/2. flashed the new map, drove 250 miles from back from my parents house, refilled and been driving since (i only have a 2 mile commute through the city, which is part of why my fuel trims go whacky). 1st tank was citgo 93 2nd tank shell 93 Link to comment Share on other sites More sharing options...
fahr_side Posted October 10, 2012 Share Posted October 10, 2012 In the Opensource definitions there's a section of tables devoted to CL/OL delay with all sorts of counters, thresholds and so on. All we have to do is set the first table to zero and the rest are shut down. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 10, 2012 Author Share Posted October 10, 2012 yeah...so minor oversight. I stupidly missed the fact that the AP sub002 locks the ecu...dammit. i dislike accesstuner (mainly because i havent really learned it yet). i cant get ATR to open a stock rom so i can easily compare... anyway, stage 1 map has the main cl/ol delay map fully zeroed. im interested in comparing the avcs tables, but its a pain since RR and ATR use different units on various tables... lame. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 11, 2012 Author Share Posted October 11, 2012 ran the maf tool today. 20 miles of driving. stop and go for 2 miles from my apartmet to the highway and then 40-60mph for 20 minutes or so. heres the data/interpolation http://i17.photobucket.com/albums/b78/killerpenguin21/mafinterpolation.jpg and heres the calculated changes to the maf table... http://i17.photobucket.com/albums/b78/killerpenguin21/maftablechanges.jpg 5% seems like a lot to me, but this is also the first time ive attempted maf scaling...thoughts? Link to comment Share on other sites More sharing options...
fahr_side Posted October 12, 2012 Share Posted October 12, 2012 You always get a rich dip going into boost, so I'd take that 5% with a pinch of salt. My advice is the scale open loop first with Bad Noodle's Excel sheet, but you need a wideband for that. Once OL is scaled nicely, run the CL scaling tool in RR logger, then blend the two areas together manually until the C and D trims are near zero. Note that on MAF cars you usually will run rich just going into boost since MAFv rises faster than MRP. Since fueling is calculated by load, which is calculated from airflow, you dip rich until MRP catches up. The alternative would be to scale the MAF under hard acceleration, but then you'd run lean while accelerating gently or cruising at high speed. Not good. You want to scale the C and D trim ranges while accelerating up a gentle hill while being as gentle as possible with the gas pedal to avoid getting into boost. One useful trick is to disable boost control completely by either zeroing out the WGDC or connecting the wastegate actuator directly to the compressor outlet. Always log total timing, plus FLKC and FBKC when changing MAF scaling, plus IAM. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 12, 2012 Author Share Posted October 12, 2012 yeah, i know about tip in etc. the dip in that interpolation line is because it logged some idle at stop lights. i ignored that when applying it to the scaling. so basically it sounds like i should go re log that drive timing, flkc etc, then apply the suggested scaling and see what happens? Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 15, 2012 Author Share Posted October 15, 2012 things have gotten more interesting.... http://i17.photobucket.com/albums/b78/killerpenguin21/learningview90035miles.jpg Link to comment Share on other sites More sharing options...
ripemeat Posted October 15, 2012 Share Posted October 15, 2012 get a wideband before you touch anything for that 40+ trim. Link to comment Share on other sites More sharing options...
fahr_side Posted October 16, 2012 Share Posted October 16, 2012 There are two tools you can use in a pinch if a WBO2 is out of the question. 1. Move the stock front O2 sensor into the downpipe just after the turbo. 2. Log the heater resistance on the front O2 sensor. While the sensor is hot and the current through the sensor is close to zero, the heater resistance reflects EGT with reasonable accuracy. In general though... leave OL MAF scaling alone if you have no WBO2. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
killerpenguin21 Posted October 16, 2012 Author Share Posted October 16, 2012 yeah im planning on it. 98% of my driving will be outside of d range, so not worried about that right now. if i can get more than 1 day off from work at a time i might run out and grab a stock filter just to see what happens... Link to comment Share on other sites More sharing options...
fahr_side Posted October 16, 2012 Share Posted October 16, 2012 Your A and B trims look much better now. That 1.40 FLKC at what, under 2k6rpm and 1.10 load.... could be noise. You usually see real knock at around 4k0rpm which is roughly peak load, or 6k0rpm which is just sensitive. Obligatory '[URL="http://legacygt.com/forums/showthread.php/2008-gh8-238668.html?t=238668"]build thread[/URL]' Increased capacity to 2.7 liters, still turbo, but no longer need spark plugs. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.