Subaru WRX Forum banner

Accessport Showing MASSIVE Knock Readings After Oil Change: I Need Help ASAP!

15K views 18 replies 4 participants last post by  Jackstand Johnny 
#1 ·
2002 WRX stage 2 Cobb off the shelf 91 oct tune. This hasnt happened before. Yesterday I changed my oil. Same Rotella T6 5W-40 I always use. Immediately after changing my oil I had to leave for my parents house. This was last night. Since the moment I started driving my car after the oil change Ive been seeing really extreme fine knock learning readings. As low as -9.02. This happens consistently at 2750 or so RPMs.

The only thing I did out of the ordinary during the oil change was I started taking apart the factory oil cooler that the oil filter screws into. I did this because it has a leaky o ring that I figured Id try to replace if it wasnt too hard. After realizing that Id have to do a coolant flush I decided against it, but by then I already completely removed the long threaded hollow bolt that holds the oil cooler in place as well as being the threads that the oil filter screws onto.

It all tightened back up fine. No cross threading or anything. IS there a washer or a fitting or something way up inside the oil cooler by the block that I didnt line up or something else thats now rattling around? I think I used a little too much oil too. I used probably 4.85 quarts AFTER already filling the oil filter. Is that enough to cause a problem? Its impossible to read my oil dipstick when its full but it looks overly full. About 10-13mm past the "F" marking on the dipstick. Im going to attempt to post my datalogs from today as I think its getting worse since last night. I think I can only post links to my dropbox and screenshots.

I still have SO much work to do before my protune on July 7th. Now it looks like Ill have to spend my protune money trying to fix this BS! I didnt think it would be the oil change that ruined everything. PLEASE HELP! I need my car back on the road right now, literally! Thanks in advance.

I have no Fing clue how to share these datalogs with yall. Heres a link to the URL of my dropbox folder for the same datalog the screenshots are coming from. The bad knock events happen around frame #300:

https://www.dropbox.com/home/WRX/June 27th 2016?preview=datalog13.csv

Screenshots. So the first screenshot is the start of the datalog and Im only including it so everyone can see what each of the columns represent sch as RPM, Boost, Fine knock etc. The 2nd screenshot is of the actual problem area:





Any thoughts? How can this be directly related to my oil change? I just checked my oil level and its fine. I cant hear any knocks or anything. PLEASE HELP!
 
See less See more
2
#7 ·
You might also include your IAM value when doing your datalogs. It should be a number between 0 and 16 (if it's not 16, something is wrong or it is still learning). IAM gives a good overall health report on how your car thinks it is doing.

From what I can see, there weren't very many knock events in your logs themselves. What you see going up to -9 is a learned value that adjusts timing based on engine load and certain loads where knock had been prevalent in the past. I, unfortunately, don't have much else for you, but maybe someone who knows a bit more on the tuning front can jump in. You might also consider contacting COBB.
 
#8 ·
Sorry, do you mean the D.A.M.? That should be in there. Its been at 16 through all of this. I have 2 theories:

1. I had reset my ecu a week or so ago and used the Vishnu method to trick the ecu into jumping my dam up to 16. Maybe over the past week my ecu has been learning the real parameters and amplified them, showing up as a much larger knock event.

2. I think I overfilled my ojl, but there cant be more than 5 quarts in there. Maybe blowby is causing this?

Thank you very much for your input though. It really does help and matches up exactly with what the members of my Subaru facebook group have told me. I just disconnected my neg battery cable to reset y ecu. Im gonna let it sit for a bit, then reconnect it, let the engine warm up and then drive it home. Ill post up with any updates.

The weird thing is if my FKL is showing past events, theyve NEVER been anywhere close to -9 before. Definitely not recently. Its also still ammystery why this only started happening after the oil change.
 
#9 ·
Ok so I decided to reset my ECU after speaking with an experienced tuner in my FB group. I accidentally did the "Vishnu" method again(couldnt do it intentionally for the longest time). This all took place during a 15 min. ride. First the DAM was at 12. Then I got a feedback knock correction of -2.11. Then a couple minutes later my DAM jumped to 16 because I went into boost at around 2200 RPM. Then for the next several minutes of the ride I saw my fine knock learning bounce up to as much as -3.11 maybe more, but for the majority of the events it jumped only to -2.11 then dropped to -1/xx, -0.70, -0.30 and then 0.00. This is what it always does.

So heres my issue. I reset my ECU a week or so ago when I was working on my exhaust and used the Vishnu method of tricking my ECU. All was well until today. Vishnu Himself said in His writeup that if you use His method, any knock related issues can most definitely become amplified. WEll I had minor knock events before. So maybe using His method amplified my knocks once my ECU actually learned what it needed over the past week? And maybe since I just did the Vishnu method again, everything may seem ok now but in a week or so I can have the same problems all over again.

The only reason why I used the Vishnu method is I had reset my ECU a couple weeks before doing my exhaust for when I installed my Cobb SF sir intake and airbox, but my DAM never got about 13. I got impatient and did the Vishnu method the first time. This time it was an accident. Thoughts? Im thinking its either this or I have a little too much oil and Im getting blowby which is causing knocks, but I checked my compression a few months back and they were all 150-152. Thanks in advance.
 
#10 ·
DAM is cobb - Dynamic Advance Multiplier
IAM is romraider - Ignition Advance Multiplier
its the same thing, just different labels.

the vishnu method has long been debunked.

10-13mm on a subaru dipstick is alot of extra oil. i dont know if it can cause knock, but its not a good thing to be that overfull. what you suggest is a possibility. extra blow-by is lowering your fuel octane enough to pull timing.
 
#11 ·
Ahhh. I thought it was 2.0L vs 2.5L. Makes more sense now.

I think 10-13mm (about 1/3" - 1/2" for us 'Murican folks) would equate to about 1/2 quart high. I say this because the hatch on the dipstick is supposed to represent 1 qt. and it is usually about 1 inch long on cars I've looked at. Therefore, 1/2" = 1/2 quart.

This doesn't sound too incredibly high to me. Higher than ideal, of course. You could try draining out some of the oil and see if your knock goes away.
 
#12 ·
I gotcha. Yeah heres how I used too much oil. I filled the oil filter and installed it, then added 4 quarts, THEN I added about 800ml from a 1 quart bottle. 1 quart = 947ml or something like that, so I added a tiny bit more than 4.8q of oil up top after already filling my oil filter with oil. Thats how much extra oil I used. Nothing crazy.

Oh and I AM Amurrican hahahaha. I use metric because to me its WAY easier than saying 13/16" or whatever. It has always made it much easier for me to much more accurately measure things when fabricating etc and since all my motorcycles and cars over the years are Japanese, I stick to metric. Dont get me wrong, I love America, but its a stubborn asshole for basically being the only place in the world that doesnt use the metric system.:eek:
 
#14 ·
UPDATE: Today at around 230 I reset my ECU with the Accessport. After a few minutes of driving I got a feedback knock reading of -2.11. This only happened once during the entire drive as far as I could tell. Without going into boost the Dam jumped to 16 and stayed there for the hour long drive. The problem is this. 1. The longer I drove the more extreme the fine knock learning values became, ending with -7.xx. Also I drove up a short but steep hill and noticed a big loss of power in 3rd gear(automatic) and with the throttle steady at around 40%, the car kept slowing down like it didnt have the power to make it up the hill. Heres the link the todays datalogs. Check out the last one especially. ITs when the hill thing happened and it also shows my timing dropping all the way to 11 for a while, but no knock readings during that time. Any ideas? Maybe I have a boost leak afterall? https://www.dropbox.com/sh/c593qfixya4iy7t/AAB5mWrsqNJRZ9rADRc-knpua?dl=0
 
#16 ·
Thanks man. Ive never done anything like that before and havent used ATR yet. I heard back from Cob though. Heres the message:

"Your engine is idling very low, about 600 RPM which indicates a mechanical fault. I would look to see if a breather or vacuum hose was accidentally bumped off while changing the oil, causing an air leak. Your oil filler cap o ring seal may have fallen off or got pinched/torn during the oil change so I'd give that a quick look as well.

Sometimes leaks are hard to spot just by looking the engine bay over by eye, even for a seasoned expert mechanic. If you don't see anything, using a smoke tester which performance shops generally have is the proper way to find leaks if they're not easy to spot by looking the engine bay over.

If you disconnected the battery while changing the oil, disconnect both battery terminals, being sure to loosen the nuts on them significantly so you can spread the terminals, shove them all the way down on the battery posts, and hold them down while tightening.

Otherwise it's possible it's coincidence that you just changed your oil and the knock could be from a bad tank of gas. It definitely happens. Get a fresh tank of the best pump fuel available in your area and see if that helps.

It could also be from something else, such as worn spark plugs which just happened to manifest itself now.

I wish you luck with diagnosis and let us know what you find!

-Mike @ COBB"

From this what Im thinking is its possible that when I partially removed my oil cooler and put it back on, maybe I got some debris on the mounting surface between the oil cooler and the block, or maybe I somehow damaged an o ring or pulled a vacuum line off somewhere on my GS air'oil separator. Im gonna take a loot as soon as its not painfully humid and hot out. I also noticed that when I started my car last night it was idling much higher than normal. Hadnt happened before but it was idling at around 1500 or 1600 RPMs. It was steady, not hovering or anything but still...
 
#17 ·
The timing adjustment in the Cobb AP is for when you get bad gas. If you look at the info on it that's what its there for. Reset your ECU using the AP, then go in an retard your timing and see how it goes. Along with looking into a vacuum leak of sorts
 
#19 ·
Wow ok. Thank you very much to the both of you. Sorry it took me so long to reply. Id given up on this for a little while. Have yall done this before? I filled up my tank a few times since last posting but Im still seeing very similar issues. Well, the FKL has settled down a little bit but I filled my tank up again yesterday and now today I saw a spike with my feedback knock correction when it hit -9.xx!!! I cant rememebr if Ive ever seen it get that bad. Now for the most part I usually see -2.11 at the most from my FKL with the occasional spike to around -4.xx, maybe -5...maybe. When I did my compression test I swapped in a brand enw set of +1 colder NGK Iridiums pre-gapped to .028" which is the factory minimum gap spec so I left them alone as to not risk damaging them gapping myself. Thats the only change.

Im wondering if MAYBE Ive been freaking out because my knock gauges have just been showing my DIFFERENT values since installing the intake, airbox and exhaust. Maybe theyre not BAD, just different because of the differences in the nuances of the car as a whole. Things that have caused different false knocks or whatever because of the new parts. Does that make sense? Honestly Id like to think thats what going on...but I doubt it. Just wishful thinking. I still feel like this REALLY needs to be addressed. Im just so burned out chasing this thing that its just kinda crushed my spirit lol. Thanks for your help guys. I think at least some of this can be solved by proper tuning, but I just dont have the skills yet and Im not sure its something as simple as retarding the timing overall as a "blanket" solution since it only happens at very specific and very consistent RPM/load/etc. Ill look into it more though as my due diligence.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top