Could someone please take a look at these logs?

[ssbtech]

Ok, I lied. One more question.

 

I was talking to a mechanic yesterday and he suggested that the O2 sensor be replaced if I'm smelling sulphur as that could indicate that it's running rich.

 

Yay/Nay?

[fahr_side]

 

I'm still betting on false knock.

 

The stakes on this bet are a piston. Keep this in mind as you roll the dice.

 

Uneven power delivery is usually either boost bouncing up and down from poor tuning or timing bouncing up and down as you pass through the load/rpm cells that have large knock correction values stored.

[ssbtech]

Well it's on the factory tune now, which admittedly wasn't great either.

 

How much would the timing have to change to feel it? I don't know how to interpret the "Ignition total timing" column in the log.

 

Interestingly in the last log, it was applying FBKC (fancy name for "pulled timing"?) in the 2130 to 3808 range, then 0 from 3868 to 5022 RPM, then a little bit to 5300.

 

Compare that to the log on Stage 1 where it was applying correction for the entire pull.

 

Just trying to work out all the possibilities here for when I take it to the dealer in a couple of weeks. When I had the injector replaced, the tech looked at the -8* FBKC on my log sheet and said "yeh, that's normal". I just wish there was someone local who could look at the logs, look at the symptoms and go through the possibilities like you guys can.

 

 

http://images.ridemonkey.com/index.php?size=full&src=http%3A%2F%2Fsureiscute.com%2Fimages%2Fcutepictures%2FI_Have_No_Idea_What_I_m_Doing.jpg

[ssbtech]

The stakes on this bet are a piston. Keep this in mind as you roll the dice.

 

 

Maybe the startup squeel is a piston scraping the wall

[seabass07]

First thing you need to do is read the knock sticky at the top of the tuning section. That should clear up the questions you have about FBKC (feedback) vs FLKC (fine learning).

 

You can probably feel the 1.4deg of timing being pulled if you are doing a pull. It feels kind of like a misfire.

[ssbtech]

I never feel the timing being pulled during a log, the acceleration is perfectly smooth.

 

It only gets jerky when I add throttle in 5th gear as the boost starts to build.

[seabass07]

I think that's where your knock is. At least from looking at the LV you posted, that's what it looks like.

[ssbtech]

Sigh.. another P0302 this morning.

 

That should be another $180 at the dealer to swap coils. Any miniscule handling benefits from the low center of gravity of the boxer engine are quickly being eroded by an engine that's almost impossible to work on.

[ssbtech]

RomRaider is having problems now. It's showing some things just fine (RPM for example) but when I enable logging for the roughness monitor it craps out saying something like "expected 80, received 00" and cycles through "connecting, reading data, connecting, reading data.."

 

My Learning View today successfully read the CEL code, but I was surprised to see that the knock chart was all zeros. Do you think that might have been having trouble reading too?

[ssbtech]

Interesting..

 

According to: http://www.romraider.com/forum/viewtopic.php?p=73898#p73898

 

Differences between A2WC522N (my 05 Legacy GT's stock ROM) and 522S (Outback XT) are:

 

I have the Legacy GT ROM on my Outback XT...

 

I wonder if RR isn't reading things correctly.

[ssbtech]

Two LVs from today. One showing zero knock anywhere, the second showing a little. Still thinks my car is a Legacy GT. CEL code still present but the CEL light went away after a few drive cycles.

 

Also attached is a screenshot of RomRaider showing the error. Works fine until I enable the roughness monitors then errors out.

 

[SeeeeeYa]

All in all, looks pretty good.

[ssbtech]

Well, of course for all the knock on the log

 

And if I could just figure out why RR won't work properly now with the new ECU ROM.

[iNVAR]

You're trying to log way too much stuff. Reduce your FBKC and FLKC to 1-byte, remove the manifold pressure (4-byte) that you have in there. Why do you have two logged? You only need the sea level one. Reduce target boost to 1-byte. Get rid of the rear O2 sensor. Use Engine Load (1-byte).

[ssbtech]

I didn't know what the differencewas between 1 byte and 4 byte.

 

Regardless, RR had no problem logging all that plus the 4 roughness monitors before. Now it's getting an "invalid header" error (you can see in bottom left corner).

[iNVAR]

It's finicky. Just remove some crap.

[ssbtech]

Hmm.. ok. Just odd that before the dealer flash it worked, now it doesn't.

 

Thanks.

[ssbtech]

And "finicky" isn't the word I want to hear should I use this to ever flash the ECU

[ssbtech]

rr_system.log is 33,000 lines long with no date/timestamps. I think I'll delete it and try agian tomorrow.

 

Some results on the romraider forum related to "Invalid header" but no real resolution.

[fahr_side]

Just reduce the clutter as Invar suggests.

[ssbtech]

I'll try just the roughness options with nothing else and see if it can read those.

[ssbtech]

You're over-thinking this issue with the pre-turbo cat. There are a number of reasons it could be blocked. Fact is that healthy or not it slows spool and costs you power and efficiency. They do sometimes fail dramatically and destroy a turbo with shrapnel. Just get rid of it already.

iNVAR is probably correct and you have a leak somewhere. If you're pegged at your maximum WGDC and still falling short of your boost targets you can look for the following problems:

 

1. Exhaust leak pre-turbo. You fail to build enough pressure in the exhaust manifold to drive the turbine to create the boost you're targeting and could hit before.

 

2. Air leak post-turbo. You're spinning the turbo hard enough but the air bleeds out of the intake system and doesn't pressurize the intake manifold. The boost control system sees the problem but doesn't understand the cause. The turbo gets pushed to limits set by the boost control system (maximum WGDC) to try and hit the target and still fails.

As a side-effect the air that's been drawn through the MAF sensor is still being fueled for, even the air is leaking out before it reaches the cylinders. Now you're spraying fuel to burn air that isn't there, so you run rich. Of course without a WBO2 sensor you can't see this because the dumb OE O2 sensor in the exhaust manifold is basically blind under boost.

 

3. Your pre-turbo cat is so clogged that it's drastically reducing exhaust flow to the turbo. This causes late spool (see how your boost error does not drop below 1psi until 3,426rpm? On a stock turbo car you should see peak boost by 2k0rpm!). It also causes the boost control system to wastegate much less, in order to try to reach target, meaning WGDC climbs to compensate for the lack of exhaust flow / energy.

A clogged cat can also prevent all of your exhaust gas escaping from the cylinders, which dilutes the incoming mixture and increases exhaust gas temperature. That in turn can cause knock, even though your AFR is correct (any leaks notwithstanding).

 

Do a leak test on the intake system to cross #2 off your list.

I'm recommending you just ditch the pre-turbo cat because it will save you headache down the road if it disintegrates, it will fix issue #3 if clogged, and swapping it / replacing the gaskets assures you on issue #1.

 

It is possible that your clogged cat is causing a leak at say the manifold / up-pipe joint. It doesn't have to leak much to smell it in the cabin and a small leak will delay spool badly but still allow you somewhere close to target boost at higher revs. I've seen this many times.

 

In your position I'd get a proper intake leak test done first. If that comes up healthy replace the up-pipe and downpipe at the same time. Catted or uncatted DP is your choice. Then get a pro tune for stage 2. If you're in a full-on smog nazi state and afraid of the man, or don't want to change parts / bribe people for inspection, then either gut your up-pipe (difficult) or get an oem catless one. Then go get a pro tune for stage 1.

 

 

I just wanted to acknowledge this as one of the most informative posts here and will serve as the basis of my notes for the dealer.

 

I should add that I've been noticing a bit of a rough idle lately, not terrible but enough to make the car shudder a bit.

[ssbtech]

And I'm wondering if I should get a wideband 02 sensor to see if the sulphur smell is from running rich or is just a normal catalytic by-product. Do they replace the factory O2 sensor and tie into the ECU or is is it standalone item with its own gauge?

[fahr_side]

You're welcome.

 

Most WBO2 kits are standalone, can output to data-loggers, and may or may not include a gauge. While they can often output a narrowband signal, the factory O2 sensor in the LGT is a quasi-wideband and can't be emulated by any of the WBO2 kits I'm aware of.

 

Note however you can read AFR down to about 11.2 with reasonable accuracy while out of boost. You can even move the stock sensor to the downpipe and read to 11.2 in boost.

[ssbtech]

I've noticed something - the power assist for the brakes is provided by engine vacuum, correct? When you've shut off the engine, there's still residual vacuum that will let you "floor" the brake pedal quite easily then it gets stiff. This usually hangs around until the next time I start the engine the next day.

 

I've noticed recently that I'll get in the car the next morning and there is no residual assist for the brake pedal. Could that be indicative of a vacuum leak? Or am I just being silly?