*** Knock, do you have any?

That makes sense. Thanks!

Fish, your post give me hope. My IAM has NEVER dropped, and if it did I would need medication. I log almost everything, and on the road to my present state of never seeing knock of any kind I've occasionally encountered a 1.4 this or a x.x that... but NEVER have I EVER seen a reduced IAM in more than six years of tuning. I do not believe in "false knock." Each and every knock event I could pinpoint the "why" after analyzing my tune and my logs. There is no single cell in my timing map that has not been addressed. It has taken years.

 

So when you post this, and you are not concurrently posting about increased oil use, I take that as encouragement that my paranoia about knock is keeping at bay the dreaded thought I have made a fatal mistake in the course of my tuning. So, if you can "get away" with this scenario... I figure I'm still good.

 

But it still makes me cringe. I want to say, "fix it!"

I've had this issue for a number of years. I don't want to turn this thread into yet another of mine. All seems to indicate it's false knock or at least I frickin' hope so. Several tuners have looked at it. Essentially for some nonsensical reason the car will every now and then register -4 FLKC at "lol" load and drop iam. When it does I want to drive it off a cliff.

I've had this issue for a number of years. I don't want to turn this thread into yet another of mine. All seems to indicate it's false knock or at least I frickin' hope so. Several tuners have looked at it. Essentially for some nonsensical reason the car will every now and then register -4 FLKC at "lol" load and drop iam. When it does I want to drive it off a cliff.

 

Then change the IAM load parameters, can't remember what it is called off hand, but it is the part that says what loads the ECU can make changes to the IAM. On my old LGT with my clutch I had to have it not change IAM unless the load was above about 2.0 or so.

Just today while logging, for no reason flkc registered a -9. It will do this at random a few times and IAM will drop. Virtually all these events take place at 20% or less throttle, load at 0.9-1.2 and at around 2.5-3.2K rpm, under negative boost. It's gotta be noise but I don't like the idea of making the ecu ignore it. Not yet. Just annoying is all. And it seems to be prevalent and happen if I drive the autotragic in manual mode.

My IAM drops about once every 2-3 months and it always crawls back up. Some times faster than others. And it will still register those obnoxious flkc corrections.

Did you log boost error? The ECU will pull timing via knock correction when it sees overboost, even if the knock sensor hears nothing. This can happen even if MRP is negative, as long it's higher than target. Solution: revise boost targets above the highest value you ever log in this area (load and rpm below where you could actually spool).

Good suggestion. I have not been monitoring boost at part throttle. However in the area where the flkc was active, knock sum did jump from 3 to 10 and then 15 iirc

The only systems the ECU seems to have at it's disposal to protect the engine from overboost and potential knock are FBKC and FLKC. Of course your issue could be noise, but I have fixed all sorts of low load 'knock' issues by improving the boost targets.

You can always tell you have a noise or overboost issue because no matter how much timing you remove from the base map, the correction values never go away. If you're in that loop of 'I keep cutting and cutting it and it's still too short!', you know the problem isn't likely to be real knock.

Interesting about the target boost value. That being said, I don't think Bill has ever written anything like that in his ECU analysis area in RR. Although I do learn new things everyday. FWIW everytime I have seen FLKC go up, it was evidence by an increase in knock sum.

 

Also, if you raise the load value for IAM evaluation, you don't stop the ECU from reacting, you just stop it from using IAM as a way to react, it can still use FBKC or FLKC.

I cant be sure but i have yet to see the logic mentioned above in any of the earlier lgt roms but i have seen it in wrx/sti's. I have seen the discussions on it on the RR forums and tgey are calling it pre emptive correction basically for when the ecu thinks it may encounter knock because of a set of limits being triggered at once. I have actually been looking for it to happen in any of the later lgt roms as well but most times i see fbkc it has a direct correlation with something explainable. Always nice to see you posting lblgt.

 

Dave

I don't have access to any early LGT models to be able to comment. WRX/STi/FXT 32-bit models up all exhibit this behavior and I've even seen timing pulled a few ms before boost error goes into negative numbers... pre-emptive timing reduction.

On the Legacies I only have experience with '10 and up models. Boost control is so damn good on those I've never seen overboost, nor timing correction associated with it.

I don't have access to any early LGT models to be able to comment. WRX/STi/FXT 32-bit models up all exhibit this behavior and I've even seen timing pulled a few ms before boost error goes into negative numbers... pre-emptive timing reduction.

On the Legacies I only have experience with '10 and up models. Boost control is so damn good on those I've never seen overboost, nor timing correction associated with it.

 

I guess its one of those things they opted to leave out from the 05-09 lgt logic but again im not saying its not there, just saying i havent seen it personally where as i have seen it on the wrx/stis. I know it was really irritating seeing fbkc and trying to tune it out, come to find out the ecu wasnt even seeing knock, it just thought the "situation" could possibly induce knock so it was going to be proactive and pull timing.

 

I would love to play with more 10+ lgt's but unfortunatly i have only had access to one locally and it had odd issues that pointed tword something mechanical. It seems there are not a lot of 2010+ lgts running around here modding.

 

I did see decent gains over the stock cobb maps but really would like to get my hands on a few more to gain a base for comparisons. Any local with 2010+ lgts want to do some testing?

 

Sorry to stray OT

 

Dave

I wish I had access to '05~'09 cars too...

 

Life.

I can confirm boost has been ... problematic on my car. She loves to build boost quite quickly. I have a Perrin EBCS that has been sitting on shelf for 2 years now along with a WB02 sensor and a 15PSI actuator. It's been delayed by a combination of "life happens" and "let's not throw more factors into an unknown and then try to figure out WTF is going on. However, at this point, I think the installation of at least the EBCS and WB02 is warranted. The PSI actuator can wait even though I don't think it would muddy the waters.

A stiffer actuator just makes boost control more tricky to tune.

Then it will have to wait until we nail a good tune.

I cant be sure but i have yet to see the logic mentioned above in any of the earlier lgt roms but i have seen it in wrx/sti's. I have seen the discussions on it on the RR forums and tgey are calling it pre emptive correction basically for when the ecu thinks it may encounter knock because of a set of limits being triggered at once. I have actually been looking for it to happen in any of the later lgt roms as well but most times i see fbkc it has a direct correlation with something explainable. Always nice to see you posting lblgt.

 

Dave

 

 

Interesting, I haven't been tuning a lot lately due to building a house (and a really really nice garage). But I may have a new STI on order by the winter.

interesting, i haven't been tuning a lot lately due to building a house (and a really really nice garage). But i may have a new sti on order by the winter.

 

!!

I did not want to start a new thread since a couple replies can probably settle it.

I think I got closer to figuring out my "IAM won't stay at 1.0" woes.

 

Recently I had a tiny vacuum line pop off, car went in limp mode, dumped so much fuel the inside of my TMIC was moist. Despite this, LV still showed -1.4 at low load, per datalogging around 0.8-1.2 or so, right around 1800-2400rpm.

 

What I have noticed is this. Pretty much no matter the circumstances, still see FLKC pulled in that area. We've tried to tune it out with the assumption it is true knock lacking evidence to the contrary. I've noticed that the amount of pulled timing depends on load in that area as well as gear. For example in 2nd gear it pulls about 0.3, 3rd gear and up it pulls usually 0.7 to 1.4, 2.8

 

What has been happening (I have to dig for logs because I've caught it in the act a couple times), sometimes it goes off the deep end and pulls 4 degrees of timing or even more and eventually pulls IAM.

What I have had happen a couple winters is IAM get all the way down to 0.0

My guess is, now knowing the above, is I probably happened to stay at constant throttle in that area, the sensor picked up what it thought was knock, and continued to pull timing then IAM as much as it could.

 

Does all this make sense, is my reasoning sound? Does this give me a good enough reason to tune out those areas of FLKC?

If you're staying at throttle and it's that low of a load and rpm, and it knocks the IAM all the way down I would suspect bad grounds or engine noise, or bad gas. You're talking closed loop areas there so fueling shouldn't be a problem(unless your maf scaling is off and your LTFT are well in the positives in the respected g/s range). If timing was the problem, the knock sensor shouldn't be picking up more events after it pulled a degree or two already.

How about posting a good screenshot of your Primary Ignition table.

 

How about det cans, to help you sort out false vs real knock. Or a knock light.

 

And why, after all this time on these forums, do you have any vacuum line capable of "popping off?? That is something that should have been addressed years ago....

 

Also, post that LV that shows pulled timing ONLY in the 0.8-1.2 load area.

 

Provide your mods, if any, and tune data, etc. You have data scattered everywhere, maybe a new thread WOULD be a good idea. Nothing in your post provides anything that can help you.

That moment, seeing IAM at 0.0 was epic. For a moment I thought "wait, I must have the wrong parameter selected to be shown on the AP". Nope!

My first hint this being noise was that rich condition and the engine still hearing knock.

Also, we mirrored the pulled timing on the cold side the same as on the hot side.

So, the colder it gets, more timing pulled. I ran like this for an entire winter, didn't make a difference. IAM has only a handful of times dropped below .875, most of the time it sticks at 1.0 but every now and then it does drop and sometimes takes forever to climb back up.

3 tuners have looked at it, none have been able to tune it out.

No FLK or FBK whatsoever outside of what I mentioned above, this has been going on for 3 years now, I think I am going to rule it out as noise and will tune it so that IAM will not be adjusted for FLKC heard under 1.5 load

 

LE: I've secured all lines, except for the one that popped off either I overlooked, or when I had the TGV deletes it did not get zip-tied back as it was before.

Sounds like this isn't going to be a straight answer so I don't want to derail the thread again

I'll just have to work with the local tuner on this.

Edited November 12, 2012 by fishbone

And why, after all this time on these forums, do you have any vacuum line capable of "popping off?? That is something that should have been addressed years ago....

 

I had a vacuum line pop off in the middle of a full throttle run. it was the little damn blue T under the intercooler and I had to put little zip ties on until I could get worm gear clamps

Mine was the one towards the front of the car under the manifold. Couldn't find it at first either.

to be honest, if you pull a couple degree in your timing table and it still knocks there and sometimes dropping the iam with it, disregard it, do you really think you're running >4* too much timing there? I highly doubt it, and 3 other tuners probably would've noticed. I'd forget it assuming the rest of your tune is kosher- fueling and WOT.