Input please on my Learning View

[hadvw]

Looking at the other thread, your have a BNR16G... and I'm assuming the Stage 2 supporting mods? (DP, UP)?

 

If so, I'm wondering why your tuner didn't rescale the learning tables. A final column of 1.60+ really doesn't tell us a whole lot and covers a pretty huuuuuuuuge range. A typical stage 2 doing about 18PSI will have a cap of about 3 load, sometimes going higher in cold temperatures, as high as 3.20.

 

edit: Just realized that the above doesn't really help a whole lot with your idle/stall issues, but still somewhat important to know. You have learned knock all over the place that would be worth looking into.

 

I'm just getting started with logging/LV, so I have no clue what you mean by rescaling the learning tables - is that something that should have been done? This is with APv1, so is it possible it's not doable with that software (Protuner)?

 

How do I look into learned knock? Tomorrow I'm planning to clean the MAF/IAT, and possibly do a boost leak test. Should I reset the ECU after, drive for a while and re-run LV? Or, what parameters should I log? I just set up RR, so hopefully I'm ready to go with that..

[iNVAR]

Those things are "fixable" by using the Cobb Tuning software. Also "fixable" by using RR but obviously you can't open Cobb maps in RR to fix it. You'd have to open up the map and transcribe the tables by hand into RR and then unmarry the AP, and then flash the newly created map you transcribed everything into.

 

Then you could get rid of the AP completely if that's what you want.

 

By rescaling the tables, I mean that you final load range spans a pretty huge gap. It doesn't hurt anything, but when you get learned knock in that last column, it'll pull timing from an RPM and load range that spans a pretty large area, which isn't necessary, and also doesn't tell you more specifically where your knock is occurring from a quick glance.

 

All you know at this point is that you're knocking at the 4000-5800+ range, in 1.60+ area.

 

That could just as easily mean you have knock right at 1.61, or it could be at 2.50 load.

 

Load, BTW, is grams/rev as you can see. It indicates how much air you are flowing per revolution of your engine. The more air you are flowing per rev, the harder your engine is working. Mass Airflow is measured in grams/second. Engine speed is measured in Rotations Per Minute. Divide this by 60 to get the number of rotations per second. Take your airflow (g/s) and divide by this new number (RPS) and now you have g/rev.

 

Most people at Stage 2 and above rescale their tables so the ranges cover a bit more, and consequently, that final column's range doesn't cover such a large gap.

 

For example, mine looks like this:

0-1.00

1-1.50

1.50-2.00

2.00-2.50

2.50+

 

It's up to you if you want to do that.

 

As far as looking into learned knock, that's what you just posted.. You posted a LearningView, which shows your knock as well as your learned fuel trims.

 

If you're cleaning your MAF and IAT, I would suggest resetting the ECU and see how things go. Remember, when you reset the ECU, you lose all the learned knock and fuel trims. Your engine idle will be lumpy for a bit, and your IAM (upper right of the LV screenshot) will drop down to something below 1 temporarily. This will all go back to normal soon, but it'll take a bit.

 

As far as the problem, I noticed in the other thread you said you replaced your panel with an AVO panel. Try swapping it back to a stock paper filter. I had an AVO panel filter, and I'm not 100% sure but I *think* it may have caused me problems as well.

[seabass07]

Well I guess I'll cross post as well. The avo would explain some of the knock, especially if you are already tuned lean. It will lean out the afr just a bit and can cause knock. An AFE filter caused consistent high rpm/load knock on an otherwise stock setup for me.

 

As for the stalling, your fuel trims don't seem off enough to look like a vacuum leak that would cause stalling, but that's still the first place I'd look if I had a rough idle or stalling.

[BigTDogg MA]

thread jackers!!!!

[bmx045]

wat

[hadvw]

Ok, a new LV update after (IIRC) several months of driving. Not sure exactly when I last reset the ECU. Went back to OEM style paper filter after getting some knock with the AVO.

 

Still very new at this. Knock appears to be gone (yay!). However, AF Learning has changed. For the better, for worse, or not worth worrying about? (see previous image about 10 posts back)

 

Used FreeSSM to up my idle +200 rpm w/ and w/o AC. Hopefully that will help the stalling.

[seabass07]

Looks like a leak. Any update on messing with the transfer clutch in freessm?

[iNVAR]

The LV looks clean, maybe a tiny leak somewhere. Could also be that your MAF isn't scaled that great, but it looks like your car's handling whatever the issue is fine, at least with regards to learned knock.

 

Do a few pulls in 3rd to redline and log it to see if you're getting FBKC. FLKC only paints part of the picture.

 

If everything looks okay, maybe just clean your MAF and leave things alone. Many people prefer their AF trims be +/- 5% since more or less than 5 would indicate a leak of some sort usually.

 

Up to you.

[hadvw]

Looks like a leak. Any update on messing with the transfer clutch in freessm?

 

On subaruoutback.org, you said post-turbo leak, due to -ve trims? Maybe TMIC? I have bulletproofed OEM (one side). I need a boost leak tester, or maybe just pick up some carb cleaner. What about the large +ve trim at low range (idle?)?

 

As far as the transfer clutch, same as you, all I can adjust on my '05 LGT is center differential clutch pressure, not Correction of Transfer Duty Ratio as ntippet has available in his '04 OB.

[hadvw]

The LV looks clean, maybe a tiny leak somewhere. Could also be that your MAF isn't scaled that great, but it looks like your car's handling whatever the issue is fine, at least with regards to learned knock.

 

Do a few pulls in 3rd to redline and log it to see if you're getting FBKC. FLKC only paints part of the picture.

 

If everything looks okay, maybe just clean your MAF and leave things alone. Many people prefer their AF trims be +/- 5% since more or less than 5 would indicate a leak of some sort usually.

 

Up to you.

 

Thanks. Recently cleaned the MAF (around the time I switched back to a paper filter).

 

btw, was doing some logging - would battery voltage dropping as low as 12.8 at times (at idle) when turning lights + AC on/off indicate a problem? My tuner noticed that at best, my alternator's putting out 14.0v, and he usually sees 14.4-14.7.

 

Also, what's stock idle on our cars? In gear (holding brake), last night, saw as low as around 625 or so. Bumped it +200 - seems to work. Hopefully it also applies while car is cold.

[seabass07]

On subaruoutback.org, you said post-turbo leak, due to -ve trims? Maybe TMIC? I have bulletproofed OEM (one side). I need a boost leak tester, or maybe just pick up some carb cleaner. What about the large +ve trim at low range (idle?)?

 

As far as the transfer clutch, same as you, all I can adjust on my '05 LGT is center differential clutch pressure, not Correction of Transfer Duty Ratio as ntippet has available in his '04 OB.

 

The positive trim means it's seeing more air than it is metering at idle. The idle one swings pretty wildly though. If you've gone back to a paper filter and have a clean maf sensor, it's time to look for leaks.

 

Idle speed at operating temp is set to 750rpm.

 

The transfer clutch pressure should be the same as the duty ratio correction. The 5eat tcu sees everything in pressure instead of duty ratio. Not sure why, but the adjustments should give the same results as on the 4eat.

[hadvw]

The transfer clutch pressure should be the same as the duty ratio correction. The 5eat tcu sees everything in pressure instead of duty ratio. Not sure why, but the adjustments should give the same results as on the 4eat.

 

Ah - so just bump by 8% and try it?

[BigTDogg MA]

From 9/6:

 

From 9/9:

 

 

 

From Today:

 

 

 

Boost leak test was fine, new plugs, what should I be looking at? I did a 3rd gear pull to log some info, should I post an image of that as well?

[ripemeat]

ewwww

 

still a leak somewhere it looks, small, at idle, running leaner than before, which can also cause a misfire if its bad enough

[BigTDogg MA]

What makes you say at idle?

[ripemeat]

0-5.60g/s is idle airflow range.

[BigTDogg MA]

OK, so what does that top left box of stuff mean?

 

A/F Learning #1 % stored?

[ripemeat]

those are learned fuel trims, positive numbers meaning the ecu has learned to add that % amount of fuel because the afr's sensed were leaner than 14.7(post maf vac leak), and the opposite for negative values (learned to pull that amount of fuel based on richer than 14.7 afr). typically +/-5 % is appropriate and stock maps sway negative on that spectrum.

[Spec B]

And to expand on this - the adjustment made to the last box (40.00+) is a critical piece to the safety of the tune. Whatever correction is in that box gets blindly applied to the high-end maf scale when you go WOT. So if there is -5% correction it will lean out the mixture (by around 5%) under high RPM/Load. A larger than - 5% adjustment is probably fine on a stock tune but that type of adjustment on an aggressive protune (withan already lean top end) can cause knocking and even pop your motor.

 

Always keep a close eye on that last box - anything more than -3% correction gets my attention.

[BigTDogg MA]

Makes sense. I'll hunt down the leak as soon as I get time. Thanks dudes. Carry on hijackers.

[BigTDogg MA]

So I brought my car to Motive Auto this weekend for an oil change and to have him chase down this P0301. Results are as follows:

 

Cylinder 1 - Compression/Injector/Piston/Valves/Coi Pack/Plug - All OK. Cyl 1 compression 130psi.

 

Boost leak test was solid, leakdown test was great. Only thing he found was a crappy positive terminal clamp, which he replaced.

 

WTF else could this be? The car dies on occasion when coming to a stop after the reset of the ECU, but after a day or so it's fine. I'm >< this close to disabling the P0301 in the tune. My battery is old, I don't know how old, or when it was last replaced, but the light only comes on after starting and the car takes about 6-8 cranks to get up in the morning. I'm replacing it before the winter, so I'll at least wait till then before doing anything with the tune.

 

Help?

[seabass07]

Clogged injector maybe?

 

The LV doesn't look like there's any leak. The idle trims sway wildly. I started a thread way back because it went from like -5 to +10 in a day without me even touching anything under the hood. IMO, ignore it or change the fuel trim ranges so that they are relevant, for example 0-10;10-20;20-30;40+.

 

The engine nearly dying at idle after a reset is normal. Mine has never actually stalled though. It idles very low and stumbles. It should go away pretty quickly though.

[BigTDogg MA]

I believe he checked the injectors impedance only. They are original with 130K miles on them, so it's possible that one is fouled. I'll dump some Techron in next fill up and see what happens. Maybe some Uplon as well.

[BigTDogg MA]

Alberto mentioned it could also be my intake manifold gaskets, as there were issues with early models and Subaru changed the gasket later. Thoughts?

[hadvw]

Alberto mentioned it could also be my intake manifold gaskets, as there were issues with early models and Subaru changed the gasket later. Thoughts?

 

Hmm. I believe that's with '07s, when they switched from "side feed" to "top feed" injectors and the early '07s had that problem, but don't take my word for it - my memory's a little fuzzy due to lack of sleep..