As you said the first log looks fine. I can tell you its very hard for me to get 1-2-3 to knock on my car. So I would assume what you are seeing there is the reduced load of lower gears removing the possiblity of knock.

Somehow between the first one labeled datalog 17 and the second one labeled datalog 19 you started pulling dam and got a bunch of knock learned. as you had that -11 learning before that log 19.

The fact that your wot pulls look pretty decent move me away from thinking it is gas, or somethign simple like that.

The fact that you are seeing 2 knock counts in one line of log when you have 11 degrees pulled already(line 454) makes me think it isnt a timing map issues, as obviously pulling timing isnt helping alot if it needs to go so far as to pull 14 degrees. There isn't enough data in the logs for me to really take a guess what the actual issue is. You should log calc. load just so the logs are easier to look at, and ignition timing to see what the final timing table looks like. Logging knock sum cylnder is just going to give you an ulser and nightmares about 4 blowing up.

What map are you actually running? The perrin stage 1 map? Do you have mods not supported by it? Put the stock air filter back in and try the cobb map imo.

 

Thanks for taking a look. I am running the perrin stage 1 map with the stock air filter as they i forgot to ship their filter with the AP...just contacted them about that. Also I have tried the Cobb map with very similar results. This whole thing is baffling to me.

I will remove the Knock sum logs and add in timing and calc. load. There are no mods to this car at all besides the AP.

I will go out and do a couple of back to back 3rd gear pulls and another cruising log. I am trying to keep this car from committing suicide! LOL

 

Im assuming your running the Perrin 93 octane map?

Mine knocked like crazy and they revised it based on my longs and its just as bad. Im not pulling 14 degrees like you but I would see 4 degrees and it would taper off for a couple lines.

I would use the advice above and switch to a Cobb map or even drop down to a 91 octane map.

 

Yes its the Perrin 93 map that was already revised once for knock. Problem is that the Cobb map knocks in similar ways as well (which is what scares the hell out of me)

Ya Socko has a good suggestion, try flashing back to a OTS Cobb map that is appropriate for your application and log again to determine if it is caused by the Perrin map given to you or something else.

Yeah I did that and it still knocks but I will try it again and post the results with the same parameters.

I would flash back to stock on a fresh tank and log. Your second log needs to be like the first. At this point at least you have a warranty.

Im going to sound like a dummy but how to I flash back to stock and still be able to datalog? Sorry for the dumb question.

 

Ya Socko has a good suggestion, try flashing back to a OTS Cobb map that is appropriate for your application and log again to determine if it is caused by the Perrin map given to you or something else.

 

I would flash back to stock on a fresh tank and log. Your second log needs to be like the first. At this point at least you have a warranty.

 

This, maybe not necessarily flash back to stock, but reset the ecu let the dam settle and then proceed to get new logs. Knock learning shows that it has detected knock before, and now reducing timing in those load/rpm cells to compensate for it.

There was a section that it did knock correct that i am concerned about, in cells 710-730. It went up to -6. As far as the knock in 5th and 6th under slight pedal acceleration, this is referred to as "tip in knock". Our cars are notorious for it and anything under 4* is really acceptable bc the load is so low.

Come back with better logs w/o the sum tables, and include load, maf reading, and injector duty cycle. I hope this helps!

 

If you have atr you can revert a cobb map to stock and save it and flash it. Otherwise i ahve no idea where you can get the stock map. PM me your email and i can send you one, I have atr on my work puter here.

PS I dont know if this can work. as atr possibly saves mapes that only my ap can use. You will liekly want to put in your request for atr right away and jsut make a stock map yourself if you dont have it already.

 

This almost seems like it could be a knock sensor thing, maybe it is faulty or picking up noise from something else.

Pulling 14 degrees of timing at 2500rpm with no load is ridiculous and doesn't make any sense.

Im interested in what happened at line 718, it goes from -2.45 fine knock straight to 11 something all while still pulling 5 degree of feedback knock. Not to mention his dam is down to .8 which means he is pulling timing across the whole map already.

 

Nice to be able to log knock sum on all 4 cylinders. I am jealous.

I really don't think it is much to worry about. In the cold I have battled to death with low load knock like that, to no avail. I find in the colder weather it seems that possibly drivetrain noise causes more false knock.

I find that I raise the FLKC and FBKC load thresholds up a bit. Your ECU probably doesn't pay attention to knock below about .5 load anyways, so a lot of the knock sum incrementation is a non-issue. I ended up raising it up to about .9 or 1.0 load, and this has solved my problems for years.

In short I would leave it alone till you get a pro-tune. Or you can change a few things yourself with ATR.

 

I really don't think it is much to worry about. In the cold I have battled to death with low load knock like that, to no avail. I find in the colder weather it seems that possibly drivetrain noise causes more false knock.

Yeah, sometimes low load low rpm knock that I see doesn't make any sense at all. And I think I was seeing less knock with all-season tires compared to a set of AD08 under low temperatures. Super sensitive knock sensor that picks up noise through the rough ride of Yoko AD08? My WOTs are usually clean and smooth. It's just tip in and lift of knock that bugs me. And it seems to disappear when the ambient temp goes up... The weird thing is that I haven't seen it last winter when I was protuned for the first time and I see it now on the same map. I tried adding some 100 oct but that doesn't seem to be helping a lot. My tuner thinks that I may have some minor crack in a ringland that's causing it (in-cylinder blow-by). But the absolute compression test numbers are all 140+ with y #4 being low by ~7psi. So, if there is some blow by than adding higher octane gas should help but it's not... Well, Subaru engines are weird!

 

Just went out and did some more logging, Reset the ECU the flashed the Cobb 93 stage 1 map for the First two pulls.



Pull One is 3rd gear 2500rpm to 6700, to fourth to 6500 (looks clean to me)

https://spreadsheets.google.com/ccc...c?key=0Ahm9CwAF7Er8dEd5c1BKdkxmSm5scTVsTjY2cmtwR2c&authkey=CM6WsOEL&hl=en#gid=0
Pull Two 6th gear 2200rpm just rolling on throttle again (much cleaner than before)

https://spreadsheets0.google.com/cc...authkey=COC2-6MI&hl=en&key=tKvSTfjGDN9OP--006WL3-g&hl=en&authkey=COC2-6MI#gid=0

Now I flashed back to the Map that Perrin Revised 93 oct stage 1

Pull from 2500rpm in third, through till 6700 or so, then 4th to about 6500

https://spreadsheets1.google.com/cc...authkey=CJagy9cM&hl=en&key=t_O6Vcy2a8t5Gj66T_cSb6g&hl=en&authkey=CJagy9cM#gid=0

Pull two 2200 rpm 6th gear rolling on the throttle then off a little

https://spreadsheets2.google.com/cc...authkey=CIiMue0M&hl=en&key=tkt2Zzal814yMnDlLT4Z80A&hl=en&authkey=CIiMue0M#gid=0

Now I flashed to the Cobb Stage one 91 oct map and did the same pulls

Pull One 3rd from 2500 to 6500 or so

https://spreadsheets3.google.com/cc...authkey=CKSa38kP&hl=en&key=tt4dlxQBZ1nEfStWC7MooMg&hl=en&authkey=CKSa38kP#gid=0

Pull two 6th gear rolling on off again

https://spreadsheets4.google.com/cc...authkey=CIqJl5cG&hl=en&key=tDS83geCRM6SOg9ahTOoK5g&hl=en&authkey=CIqJl5cG#gid=0

Pull three 91 octane stage 1 3rd @2500rpm to 4th to 6500rpm

https://spreadsheets5.google.com/ccc?hl=en&key=tWMRyR5AWVJamMvSx_7GF5g&hl=en#gid=0
I dont know, I am starting to think the Perrin map may just suck. Also something weird happened as I was driving back home, was sitting at about 65mph in 6th and hit a small to medium bump on an overpass, at about that time my feedback knock jump to -7.0 pretty much instantly, was in the same gear, no change in throttle position, or road condition (i.e. not uphill or downhill) Not sure what to make of that?

BTW thanks for all of the help so far, its nice to know that there is a group of people out there willing to take time out of their day to help me from blowing this thing up! :tup:

 

they are all locked

The perrin map is tuned for a filter, if you dont have it and they did maf adjustments for it it coudl cause some weird stuff to happen.

 

Sorry about that, unlocked them all forgot to when I loaded them. As for the Perrin thing...thats what I was thinking but wasnt sure.

Thanks for your help.

 

Still cant open them?

 

Ok they should be all fixed...sorry i ****ed that up and also somehow had them out of order....been a long day already!

 

Yeah the wot knock looks much better in the cobb maps than in the perrin map. could be relative to not having the filter though. The 2ish load knock is still really serious though imo.

I can tell you i have never sustained 2 load ever lol. put putting along is .9-1.6 generally any much more than that generally tends to be on its way to 3.0+. I did rebuild my throttle maps though so that may have something to do with it. I think the years of partial throttle boost nightmares back in the day with my wrx has possibly just got me out of the habbit of driving like that.

It is possible you are lugging the hell out of the car causing the knock? I can tell you getting that knock out of 6th before 3000 is a real ball bust. I only delt with it wot and it was still very rough. Like i said i dont load the car in that manner at low rpms. If you are concerned about it, just dont go there imo, light throttle, wot or downshift. If you want to make the car knock, in my experiance its easiest to do in 5-6 below 3000. I think that is a big part of what you are seeing here.

Your inj dc looks pretty much like mine, way tighter than i would like 5200 plus. But it is what it is. maf peaking at mid 280's is inline with the stock stage 1 map.

That sustained 2 load with more or less no rpm change is nutty to me. The fact that it is inj. dc in teh 80's is possibly more crazy. I am at a loss for what actually happening there as I have never seen anything like that in my logs ever, and I think i have about 80MB of logs now. Though i am interested to try it on the way home tonight, haha. It is my beliefe that if i hit 2.0 load i will be above 3k pretty quickly though.

 

Well looks like I will be having it pro tuned sooner rather than later, as for the lugging the gf does that crap all the time, I did it once to see if I could replicate the knock and that may have been what you saw either way I just won't stick around the 2k- 2600 range for to long...now I can tell the gf I have to drive more aggressively in order to save the car. I am just glad that the wot runs look clean. I think I am going to keep it on the 91 Oct. Tune until I can get it to the tuners. Now I just have to find a good tuner in the va dc md area.

 

While that lugging may not be good for the motor, I can easily tune out the knock you get in that range.

 

Pull One is 3rd gear 2500rpm to 6700, to fourth to 6500 (looks clean to me)

https://spreadsheets.google.com/ccc...c?key=0Ahm9CwAF7Er8dEd5c1BKdkxmSm5scTVsTjY2cmtwR2c&authkey=CM6WsOEL&hl=en#gid=0


Clean, yes.


Pull Two 6th gear 2200rpm just rolling on throttle again (much cleaner than before)

https://spreadsheets0.google.com/cc...authkey=COC2-6MI&hl=en&key=tKvSTfjGDN9OP--006WL3-g&hl=en&authkey=COC2-6MI#gid=0


The knock here should be fixxed, not a HUGE deal, but any tuner can fix it easily. FWIW usually once the turbo has spooled (or at least load is steady) the knock would show up as FLKC, but since you just did a flash it seems the ECU might still be in rough correction mode, in which case it used FBKC.



Now I flashed back to the Map that Perrin Revised 93 oct stage 1

Pull from 2500rpm in third, through till 6700 or so, then 4th to about 6500

https://spreadsheets1.google.com/cc...authkey=CJagy9cM&hl=en&key=t_O6Vcy2a8t5Gj66T_cSb6g&hl=en&authkey=CJagy9cM#gid=0


This knock should also be tuned out. Looks like too much timing to me.



Pull two 2200 rpm 6th gear rolling on the throttle then off a little

https://spreadsheets2.google.com/cc...authkey=CIiMue0M&hl=en&key=tkt2Zzal814yMnDlLT4Z80A&hl=en&authkey=CIiMue0M#gid=0


Should be tuned out.



Now I flashed to the Cobb Stage one 91 oct map and did the same pulls

Pull One 3rd from 2500 to 6500 or so

https://spreadsheets3.google.com/cc...authkey=CKSa38kP&hl=en&key=tt4dlxQBZ1nEfStWC7MooMg&hl=en&authkey=CKSa38kP#gid=0

Clean


Pull two 6th gear rolling on off again

https://spreadsheets4.google.com/cc...authkey=CIqJl5cG&hl=en&key=tDS83geCRM6SOg9ahTOoK5g&hl=en&authkey=CIqJl5cG#gid=0


Needs to be tuned out. Can you log target AFR there? I am curious what it is, as the timing seems pretty close, maybe it needs to be a bit richer there?


Pull three 91 octane stage 1 3rd @2500rpm to 4th to 6500rpm

https://spreadsheets5.google.com/ccc?hl=en&key=tWMRyR5AWVJamMvSx_7GF5g&hl=en#gid=0


Shift knock, less then 3 counts (only one) don't even worry about it.


I dont know, I am starting to think the Perrin map may just suck. Also something weird happened as I was driving back home, was sitting at about 65mph in 6th and hit a small to medium bump on an overpass, at about that time my feedback knock jump to -7.0 pretty much instantly, was in the same gear, no change in throttle position, or road condition (i.e. not uphill or downhill) Not sure what to make of that?


Our knock sensor (just like any other car's) is not perfect. That is why the ECU has lots of different parameters it uses to know how to interpret it. W/o going into a ton of detail, I tried to help you understand what is going on.

 

https://spreadsheets.google.com/ccc...com/ccc?key=0AnyzysZN4mSkdHdqcllRLXdQVTVuNk9VLXo1Z1Ficmc&hl=en&authkey=CMjKjaEG

Tried to hold close to 2ish load in 6th, car feels funny at partial throttle boost. But it doesn't feel like its lugging.

That is running the cobb 93 stage 1 map which at that load/rpm range is straight ots. I don't log inj dc generally cause it isn't very relivant to me, but would have been interesting to see if it was so high. damnit!

 

I know what you mean...mine didnt feel like it was lugging either....it was weird. I have driven manuals my whole life (hate automatics..one of the reasons I didnt get an EVO) Seriously though, thanks for all your help and all!

Going to have it protuned fairly soon, until then I will just try not to blow it up! LOL

 

LittleBlueGT and all, could you give me an idea (if you have time) on why this is happening:

https://spreadsheets.google.com/ccc...om/ccc?key=0As1hVi0lfy5VdHl6bXg5WU9KRlNucHJtZlo4TWJPT1E&hl=en&authkey=CJDnmPMK#

And I never actually saw -11 feedback corerction on this map but somehow it's showing in Learned KC. And I monitor my car almost everyday and every drive for the last few month.

Took this log today. Sorry for highjacking but thought that it's relevant to OPs topic

UPDATE: longer version of today's log (mostly 5th and 6th gear) and some 2nd at the end: https://spreadsheets.google.com/ccc...com/ccc?key=0As1hVi0lfy5VdEttd1JnTnEtbUExSDBLSFJUUmxqcWc&hl=en&authkey=CObuyuQE

My car is Stage 2 protuned for ACN91 and currently runs a mixture of 2 gallons (per tank) of 100 oct and ACN91.

 

To me that looks like real knock, absolutely.

At 2600 rpm I usually call for about 17 degrees timing on a similar set-up. Your car is running a TON more then that, even after 11 degrees of FLKC it is still above that.

What tune again? Something sure doesn't seem right there.

 

He is on to something here. My cobb 93 map is running 11-17 degrees of timing in the 2000-3000 1.9-2.1 load range. You are runnign 25-30. Something is fishy here.

You should have access to atr by now if you signed up to get it yesterday. Check your timing maps. They should be the same as mine in the 2ish load range I have only putzed with mine from 2.73 load up.

Map 1 is primary timing 2 is DA 3 is total added up. 2011 sti 93 stage 1

https://spreadsheets.google.com/ccc...com/ccc?key=0AnyzysZN4mSkdGhlSWQzUk1GR2p2YTVxalZGVlJDOWc&hl=en&authkey=COPd2c0D

 

This post directed at me or the other poster? Reason I ask is after looking at my logs I can only seem to find one instance of the timing being that high under load and that was with my Perrin map. Can't see it in the others? Maybe I just don't know what I am looking at.
Adam

 

ah crap, damn thread hijackers

 

Sorry

 

I paid $100 for an ots map from a well respected tuner on these boards. I threw it away in a week or 2 cause i saw regular -4.2 knock, and it wasn't atr fixable. I think you need to consider what you got happening there and if you still think its worth it to run your custom map over a cobb ots map which is fairly easy to fix yourself.

It is very likely if you have a custom map it has custom avcs tuning which can change timing maps pretty dramaticly. So comparing to the cobb map isnt going to be very useful.

For what its worth. I see -1.4 knock events and learned knock in the cell that goes from .9-1.6 2400-3300 rpm on and off now and again. So it is possible its a common place to see issues. I don't understand exactly where the issue is and i dont want to make sweeping timing table changes, and havent spent the time to compile all my knock events in that range and try to narrow them down to a couple cells to modify as i am not too concerned with that little bit of low load knock. Your issue on the other hand likely needs addressing. Your timing does look simmilar to mine but avcs changes could make that timing impracticle for your custom map.

 

I paid $100 for an ots map from a well respected tuner on these boards. I threw it away in a week or 2 cause i saw regular -4.2 knock, and it wasn't atr fixable. I think you need to consider what you got happening there and if you still think its worth it to run your custom map over a cobb ots map which is fairly easy to fix yourself.

Yes, and that frustrates me. I pay money to some reputable local shop (and it's significantly more than $100) and still need to deal with it myself I was trying to avoid that.

It is very likely if you have a custom map it has custom avcs tuning which can change timing maps pretty dramaticly. So comparing to the cobb map isnt going to be very useful.

Is there a way to "hack" a locked map from a tuner just to see what's inside?

Your issue on the other hand likely needs addressing. Your timing does look simmilar to mine but avcs changes could make that timing impracticle for your custom map.

And it's mostly happening when it's colder... I still have an idea that some of that knock (not all of it but some) is not real. I don't know but maybe I'll just swap a knock sensor just to be sure that it's working properly.

 

Holy wwrx, i wish i was at your altitude, you hold boost like crazy. At +3500ft and 5000rpm ive already tapered to 14.5psi and by 6000rpm im at 12.7psi.

You may want to consider running the LWG map if you aren't already. You are overboosting by over a psi in the lower rev range.

At your current temperature you are hitting over 100% injector duty so be careful while it is colder. In summer that should drop to 90's though.

For what its worth i always see learned knock at boost transition (1.1calc load) and also at 5psi as you see. This comes and goes and shouldnt be detrimental to our engines.

 

You may want to consider running the LWG map if you aren't already. You are overboosting by over a psi in the lower rev range.

Overboosting to 8 psi when the map calls for 4 was nothing to do with WGDC. The WG opens up at about 8 or 9 psi w/ 0% WGDC, so it doesn't matter what WGDC your map has, his boost will be the same at the rpm/elevation/temp/throttle position/gear, etc.....

 

Just edited post two up, in case you missed it.

 

the hwg map has higher wgdc set throughout. So if your only tapering on the back end you are going to spike on the front end with the hwg map. Which could obviously have serious negative consequences.

Is it tapering off in 5th and 6th? cause it will taper off early in 3rd, its how the map is setup.

You could dl atr and set it up however you want.

 

Ok i was just wondering if you ran the HWG map it could cause your turbo to run out of steam before it would naturally.

Im running the hwg map and i just hit 18psi sometimes but most of the time 17.5 psi or so. I taper to like 14.5 by 5000rpm while most people are still running 16 or more psi. In the end it probably just comes down to my altitude i guess. /threadjack

 

+1 To resetting the ECU and letting your car "relearn" your driving habits. This is what happened to my car, I wasn't hitting target boost (21psi) and only getting up to about 9-10psi. Turns out I was just low on coolant and my DAM dropped to below 0.2 to prevent the engine from blowing up, which aka means, wouldn't let me hit boost. I also heard that the Perrin maps are prone to knock so check out the Cobb maps. I suggest protune though.

 

My perrin map was ok for knock but it ran a bit richer i think.

It wasn't long before my tailpipes were completely black. I also smelt more fuel in the oil on that tune, it was alot colder while i was running that tune so that probably had more of an affect than the tune.

 

Just to clarify something. While logging my feedback knock showed -2.5 in 5th gear about 3k 10lbs of boost, at one point it was -4.0 in 6th gear about 3.5k 12lbs of boost. After a while I didn't get any knock in 6th but -2.5 in 5th gear. Is the -2.5 the correction of timing or what does it mean(aka bad or ok). This is a temporary tune until I can find a tuner in New Mexico, just curious if I should avoid boosting or just avoiding these situations that are causing the knock.

 

That depends, is the knock coming as the load is sharply changing? Post up the log.