Time to identify once and for all the source of my FKC

[covertrussian]

You also might have a loose heatshield. So look around and see if something is loose.

 

VVT looks fine, it's supposed to be ~10* AVCS in those ranges.

[xt2005bonbon]

Trust me. I've been looking for loose stuff for the past 6 years now! It's probably the engine

But who knows. I can check again...

 

And yeah, I know vvt peaks around 10deg for my car (when warmed up). But I was just curious if this value is somehow related (in some equation perhaps) to the final timing value. I guess not now that I think about it.

[covertrussian]

Nope it's an unrelated system, AVCS/VVT is rotating the intake cam changing the mechanical timing in relation to the crank angle, unrelated to ignition timing. But with that said, AVCS will effect how much ignition timing you run for MBT.

[xt2005bonbon]

Right.

[solidxsnake]

It is the 'timing' variable that shows up in btssm. So I would say total timing. And this -5.98 does occur usually above 2-2.5k rpm.

 

I just checked the total timing table and around that load/rpm range, total timing is supposed to be 19-20deg. Interesting that the app shows 15deg...

 

By the way, how about the vvt angle? Is the reported 10deg somehow related to the total timing?

 

Not really. if you have 5° being pulled from your 20° of total timing, you'll be at 15°.

 

Also, keep in mind that the default fields in BtSsm are 1-byte and only out to 1 decimal place, so your timing numbers aren't exactly precise. For instance, where it shows -5.98° of FBKC, it's actually -6 (and -3.84 is -4, -2.11 is -2, etc.) due to the lack of precision in the 1-byte field. As such, the rounding on your actual timing may also not add up exactly to your commanded total timing, and you may not see all of the granularity in your total timing as it adjusts due to the imprecision.

[xt2005bonbon]

OK. So you're saying the total timing reported by btssm already takes into account the fbkc during a knock. I was assuming it was not.

 

Now, if you guys were to check this rpm/load range in your respective tune, I would surmise you have similar total timing in your table there. So, if these values are about right, then I would think that the mixture is too lean then in that range? That is if the knock is real of course (and not due to rattle of any kind). You agree?

[Perscitus]

You'd want to log Total and Learned Timing to get a sense

and face them off against the union of your Base+KCA tables (aka your Total Timing).

 

Total Timing logging param will contain KCA as one of the ingredients, Learned Timing will be a result of FLKC and acts a bit like a trace of your Timing Learn LV values. Similarily to how Fuel Learn A-D etc act as traces of you Fuel adjustments in the LV snapshot.

 

I dont believe FKC (fbkc) is reflected anywhere other than in its own param.

[solidxsnake]

You'd want to log Total and Learned Timing to get a sense

and face them off against the union of your Base+KCA tables (aka your Total Timing).

 

Total Timing logging param will contain KCA as one of the ingredients, Learned Timing will be a result of FLKC and acts a bit like a trace of your Timing Learn LV values. Similarily to how Fuel Learn A-D etc act as traces of you Fuel adjustments in the LV snapshot.

 

I dont believe FKC (fbkc) is reflected anywhere other than in its own param.

 

The built-in "timing" param in BtSsm is the final timing, including any FBKC, FLKC or IAM corrections. You can see it clear-as-day by, when any knock is induced, keeping the car in the same load/RPM cell, and watching the timing increase while the FBKC decreases.

 

At least, this is the case with the '07+ ECU.

[Perscitus]

Yes, good point. I haven't used most of the built-in params in such a long time, I forget the memory addresses their trace back to and which PID that represents.

 

Field descriptions in the RR logger definition put it well.

For example, assume we're logging: IAM, FBKC, FLKC, KCA, Total Timing, Base Timing, Load, RPM, then:

 

IAM-KCA = (KA (max) map value * IAM) + FBKC + FLKC

 

Total Timing = KCA + base timing + timing compensations (IAT, ECT-based etc.)

 

RPM and Load are useful to orient/trace the log with respect to the Total or Base timing or KA additive tables.

[solidxsnake]

Yes, good point. I haven't used most of the built-in params in such a long time, I forget the memory addresses their trace back to and which PID that represents.

 

Field descriptions in the RR logger definition put it well.

For example, assume we're logging: IAM, FBKC, FLKC, KCA, Total Timing, Base Timing, Load, RPM, then:

 

IAM-KCA = (KA (max) map value * IAM) + FBKC + FLKC

 

Total Timing = KCA + base timing + timing compensations (IAT, ECT-based etc.)

 

RPM and Load are useful to orient/trace the log with respect to the Total or Base timing or KA additive tables.

 

I'm the same, I haven't used the built-in definitions in a long time. That said, IIRC, the built-in ignition timing parameter uses the standard OBD-II parameter (i.e. the same one apps like Torque will read), and not the SSM parameter, which is the final instantaneous timing.

[xt2005bonbon]

Thanks guys for the clarifications. I like equations .

[xt2005bonbon]

I drove the car this weekend with the family on a 5hr road trip. That same knock event reoccurred several times (see example below). I can see that it is always the exact same rpm/load range where it happens. I am pretty sure it is not a timing issue. I can hear the noise actually. If I am able to keep the throttle steady with that same load/rpm, I hear like a piston slap type sound.

 

KNOCK DETECTED at 2018-03-03 09:17:16 PM

AFR: 14.6

Boost (psi): 0.60

Boost Target (psi): 2.6

Coolant (degF): 185

FKC (deg): -5.98

FLKC (deg): 0.00

Fuel correct (%): -0.0

Injector DC (%): 14.4

Injector PW (ms): 7.4

Intake (degF): 41

Load (g/rev): 1.19

MAF (g/s): 46.41

Misfire cyl 1: 0

Misfire cyl 2: 0

Misfire cyl 3: 0

Misfire cyl 4: 0

RPM: 2331

Timing (deg): 15.0

VVT int L (deg): 10

VVT int R (deg): 10

*fuel learn A: 6.80

*fuel learn B: 7.70

*fuel learn C: 4.00

*fuel learn D: 0.00

[xt2005bonbon]

I should also add that the ecu learned to pull timing there (-4 deg) during the trip, but quickly 'unlearned' it. I recall seeing this behavior for the past... 6 years!

[covertrussian]

First you can't look at just one row and get the whole picture, you need to look at couple rows before hand.

 

Second, If it's happening at the exact same time consistently it's most likely knock, also when you can hear engine knock, your doing damage.

 

Your tune is calling for 19* of timing in that region. Try reducing it by 5* and see if it's still knocking.

[xt2005bonbon]

Wait, why do you say I am looking at just one row?

 

Sure. If it is real knock, it will damage the engine. But the thing is it's been happening for the past 6 years now (~50k). Don't you think it would have 'blown up' by now?

 

And yeah, I can reduce timing there, and probably will. But now I recall discussing this issue several years back with the tuner. And he said he's been trying to reduce timing around that region in the map, and yet, the knock sensor kept registering the same amount over and over.

[covertrussian]

Since your pasting only one row for us to see. The row right before the knock will show the total timing, which is important since it can be more then Base + KCA map (ie other timing compensations). Sometimes you might have an aggressive IAT based compensation that's causing the knock basically, etc.

 

You don't spend too much time in that area, and ECU's knock control system is correcting for it. But every time you get knock, it's same as taking a hammer to your piston. This is why it's audible.

 

If reducing timing still causes knocking, then yeah I would look at other things causing it to register (rattles or something touching the knock sensor). Heck I've heard of people saying that knock sensor wire tapping on something can induce false knock too.

[xt2005bonbon]

I see.

 

I will try to provide a log. I should also add that I do spend quite some time at that load/rpm range (and around it) because I try to make the drive comfortable for my family (i.e. slowly increasing throttle so as to avoid 'kick in the pants type torque from that vf40')

 

Also. I searched my old email and found some feedback from the tuner on that very issue (where LV showed -6 deg of timing pulled exactly around that range/load and nowhere else).

 

"It seems like the timing is too low, then the ecu pulls some and it gets worse. I added a little less fuel and more timing to see if that would help (in that area). I also sloped the wastegate duty in the low end and made some further improvements on the fueling to improve mpg. Let me know how this change runs. "

[covertrussian]

Actually look at your map and I just looked at mine, perhaps you are running too little timing there. I'm running about 30* of timing at that point.

 

Here is my total timing (Base + KCA). Keep in mind that I'm on a bigger turbo (EVO3 16g) at 14psi, 3" turboback, and 3" CAI. CAI means MAF scale is different and it's running 1* less timing then stock intake wanted.

http://legacygt.com/forums/attachment.php?attachmentid=262756&stc=1&d=1520276095

 

For easier pasting here is my tune: Rename the .pdf to be .hex! Please keep in mind the KCA table!

 

Also your fuel map goes out of closed loop real quickly, compare it to my tune too.

cr_Big16g_MBC_CAI.pdf

[Perscitus]

Yeah xt2005bobon, the knock 'snapshots' are fairly useless (tad less useless than Learned Value snapshots) in getting to the bottom of FBKC, FLKC events and patterns, tracing through timing maps, etc.

 

Which, as you know all too well, is an endless and thankless rabbit hole on these cars.

 

For that you need good resolution (read >40Hz) data logs and time to analyze them, put into perspective, etc.

[xt2005bonbon]

You guys are giving me more homework

 

As a side note, just confirmed we're going to have our third kid.... Getting harder to find time to sit down in front of that PC to analyze data.. And thinking about "Ascent or keeping the OBXT"?

 

 

As a follow-up question to your comment covert: you say I may be running to little timing. Then, why would the ECU pull even more then, if it indeed needs more?

[solidxsnake]

You guys are giving me more homework

 

As a side note, just confirmed we're going to have our third kid.... Getting harder to find time to sit down in front of that PC to analyze data.. And thinking about "Ascent or keeping the OBXT"?

 

 

As a follow-up question to your comment covert: you say I may be running to little timing. Then, why would the ECU pull even more then, if it indeed needs more?

 

The ECU's knock-control strategy assumes that any knock is due to too advanced timing. Running excessively low timing can cause knock (or so I've read), but the ECU's knock control algorithm assumes that if knock is occurring, it's due to too advanced timing for the given fuel.

[xt2005bonbon]

That's what I thought.

[xt2005bonbon]

solidxsnake, what is your current set up on your car? Can't remember if you still run the stock turbo or not. Just curious how your total timing table looks like compared to mine and covertrussian.

[covertrussian]

You guys are giving me more homework

 

As a side note, just confirmed we're going to have our third kid.... Getting harder to find time to sit down in front of that PC to analyze data.. And thinking about "Ascent or keeping the OBXT"?

 

I have a 2nd on the way, my '12 Outback is still big enough, but with 3 I probably will upgrade to a 7 seater myself

 

 

As a follow-up question to your comment covert: you say I may be running to little timing. Then, why would the ECU pull even more then, if it indeed needs more?

 

I mentioned it here. Years ago a pro-tuner told me about the retard knock after I saw it on my turbo Nissan after enabling knock sensor control. Doing a google search it seems like excessive retard can lead to pre-ignition and burns your valves, also can make the header glow cherry red .

 

Maybe it's pre-ignition happening, but since it's at fairly low loads, it's not too damaging (pre-ignition at high loads will put a hole in your piston in no time).

[solidxsnake]

JmP VF46 Custom, GS TMIC, Cobb catted DP and SPT catback. Running between 30-40° of timing in the area you're looking at.