Welcome to the North American Subaru Impreza Owners Club Sunday May 3, 2015
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Engine Management & Tuning > Open Source Reflashes

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 01-18-2008, 09:54 AM   #1
Kastley85891
Scooby Specialist
 
Member#: 127704
Join Date: Oct 2006
Chapter/Region: TXIC
Location: TX/Austin
Vehicle:
2013 Hybrid
Needs an 05STI

Default IAM and recognising knock from log data

How do you guys generally tell if your baby is knocking apart from sound /feel - i.e. logs, IAM? comparing your timing tables to logged timing?

Just trying to understand a bit at a time.

The logs are getting more interesting all the time!

Thanks guys
* Registered users of the site do not see these ads.
Kastley85891 is offline   Reply With Quote
Old 01-18-2008, 10:03 AM   #2
gabedude
Scooby Guru
 
Member#: 10108
Join Date: Sep 2001
Chapter/Region: TXIC
Location: Lake Travis, Austin, TX
Vehicle:
07 STI
60Trim spun rod bearing

Default

FBKC and FLKC. You have to know how much timing the ECU pulls during a knock event however.

Example 07 STI:

Step vlue for both FLKC and FBKC is -1.4 degrees.

When in Coarse Correction mode FBKC is active and FLKC is disabled:

If the ECU detects one knock event, it pulls 1.4 degrees and then there is a delay and timing steps up in .35 degree increments if there is not knock. If it continues to knock, it will pull in 1.4 degree steps up to -11*. This is instantaneous response to a knock event(s) that just occured.

When in Fine Correction mode FLKC is active and FBKC is disabled:

FLKC may already be negative, so this is more than likely learned timing and the ECU is not knocking. If there was no FLKC in that RPM/Load zone before and you get a -1.4, then the ECU just knocked. The step value for learning to add timimg back in is the same as FBKC (.35*). Also note that if your IAM is not at 1, you may have positive FLKC. This means that the ECU has learned to add in timing to try and achieve full power.

So in short, you have to factor all of this in. I answer many questions about logs where people think their car is knocking when it really isn't as they do not understand how to interpret the logs.

Gabe

Last edited by gabedude; 01-18-2008 at 10:08 AM.
gabedude is offline   Reply With Quote
Old 01-18-2008, 10:46 AM   #3
Tea cups
Scooby Specialist
 
Member#: 103136
Join Date: Dec 2005
Default

Quote:
Originally Posted by gabedude View Post
When in Coarse Correction mode FBKC is active and FLKC is disabled:
When in Fine Correction mode FLKC is active and FBKC is disabled:
FBKC is not disabled when in either rough or fine correction "mode". The modes do not indicate that a change is being made to the IAM (rough) or the current FLKC cell, but rather if the conditions are met, which of these the ECU is poised to make changes to. One of the conditions is the FBKC is disabled for which this is determined separately from the mode selection. That is, FBKC can be disabled and no changes are made to the IAM or the current FLKC cell (although a number of conditions have to all be true is order to disable FLKC). Basically, it is best to think of these three elements of knock control (changes to the IAM, FBKC, changes to current FLKC) as three distinct controls that never overlap in their execution.

To the OP, to determine close to all instances of perceived knock (within the ranges set by your tune), you would need to use Enginuity's logger with the latest definitions and log the following:
1. Feeback knock correction (FBKC)
2. Fine learning knock correction (FLKC)
3. IAM (Ignition advance multiplier)

as gabe said, FLKC could be due to past knock unless the current cell is changing. You can determine this by logging 'Fine Learning Table Offset'. If this value remains the same but your FLKC changes then there was a change to the current FLKC cell (barring any "sync" issues which can occur due to the way the ECU polls the parameters).
Tea cups is offline   Reply With Quote
Old 01-18-2008, 10:48 AM   #4
Kastley85891
Scooby Specialist
 
Member#: 127704
Join Date: Oct 2006
Chapter/Region: TXIC
Location: TX/Austin
Vehicle:
2013 Hybrid
Needs an 05STI

Default

Thanks Gabe for that response - very helpfull - I wanted to try and understand more as my DTEC knock indicator is pulsing between the 5-6500 rpm range although I coul dnot see knock in the logs, assumed it was a false signal etc, I have your email so will forward a single log if that is ok?, a quick view from you may put my mind at rest at least until I understand more.
Cheers
Quote:
Originally Posted by gabedude View Post
FBKC and FLKC. You have to know how much timing the ECU pulls during a knock event however.

Example 07 STI:

Step vlue for both FLKC and FBKC is -1.4 degrees.

When in Coarse Correction mode FBKC is active and FLKC is disabled:

If the ECU detects one knock event, it pulls 1.4 degrees and then there is a delay and timing steps up in .35 degree increments if there is not knock. If it continues to knock, it will pull in 1.4 degree steps up to -11*. This is instantaneous response to a knock event(s) that just occured.

When in Fine Correction mode FLKC is active and FBKC is disabled:

FLKC may already be negative, so this is more than likely learned timing and the ECU is not knocking. If there was no FLKC in that RPM/Load zone before and you get a -1.4, then the ECU just knocked. The step value for learning to add timimg back in is the same as FBKC (.35*). Also note that if your IAM is not at 1, you may have positive FLKC. This means that the ECU has learned to add in timing to try and achieve full power.

So in short, you have to factor all of this in. I answer many questions about logs where people think their car is knocking when it really isn't as they do not understand how to interpret the logs.

Gabe
Kastley85891 is offline   Reply With Quote
Old 01-18-2008, 10:51 AM   #5
Kastley85891
Scooby Specialist
 
Member#: 127704
Join Date: Oct 2006
Chapter/Region: TXIC
Location: TX/Austin
Vehicle:
2013 Hybrid
Needs an 05STI

Default

Thanks Tea cups

Appreciated
Quote:
Originally Posted by Tea cups View Post
FBKC is not disabled when in either rough or fine correction "mode". The modes do not indicate that a change is being made to the IAM (rough) or the current FLKC cell, but rather if the conditions are met, which of these the ECU is poised to make changes to. One of the conditions is the FBKC is disabled for which this is determined separately from the mode selection. That is, FBKC can be disabled and no changes are made to the IAM or the current FLKC cell (although a number of conditions have to all be true is order to disable FLKC). Basically, it is best to think of these three elements of knock control (changes to the IAM, FBKC, changes to current FLKC) as three distinct controls that never overlap in their execution.

To the OP, to determine close to all instances of perceived knock (within the ranges set by your tune), you would need to use Enginuity's logger with the latest definitions and log the following:
1. Feeback knock correction (FBKC)
2. Fine learning knock correction (FLKC)
3. IAM (Ignition advance multiplier)

as gabe said, FLKC could be due to past knock unless the current cell is changing. You can determine this by logging 'Fine Learning Table Offset'. If this value remains the same but your FLKC changes then there was a change to the current FLKC cell (barring any "sync" issues which can occur due to the way the ECU polls the parameters).
Kastley85891 is offline   Reply With Quote
Old 01-18-2008, 12:49 PM   #6
Not-EWRX
Scooby Specialist
 
Member#: 158043
Join Date: Aug 2007
Chapter/Region: RMIC
Location: SLC
Vehicle:
2013 Wrx Hatch
SWP

Default

Are the FBKC and FLKC named the same on older models.
Not-EWRX is offline   Reply With Quote
Old 01-18-2008, 12:52 PM   #7
gabedude
Scooby Guru
 
Member#: 10108
Join Date: Sep 2001
Chapter/Region: TXIC
Location: Lake Travis, Austin, TX
Vehicle:
07 STI
60Trim spun rod bearing

Default

Quote:
Originally Posted by Tea cups View Post
FBKC is not disabled when in either rough or fine correction "mode". The modes do not indicate that a change is being made to the IAM (rough) or the current FLKC cell, but rather if the conditions are met, which of these the ECU is poised to make changes to. One of the conditions is the FBKC is disabled for which this is determined separately from the mode selection. That is, FBKC can be disabled and no changes are made to the IAM or the current FLKC cell (although a number of conditions have to all be true is order to disable FLKC). Basically, it is best to think of these three elements of knock control (changes to the IAM, FBKC, changes to current FLKC) as three distinct controls that never overlap in their execution.
I hate it when my memory fails me and I don't have one of your stickies to reference.
gabedude is offline   Reply With Quote
Old 01-18-2008, 12:53 PM   #8
gabedude
Scooby Guru
 
Member#: 10108
Join Date: Sep 2001
Chapter/Region: TXIC
Location: Lake Travis, Austin, TX
Vehicle:
07 STI
60Trim spun rod bearing

Default

Quote:
Originally Posted by Not-EWRX View Post
Are the FBKC and FLKC named the same on older models.
With the latest logging defs, yes.
gabedude is offline   Reply With Quote
Old 01-18-2008, 01:55 PM   #9
Tea cups
Scooby Specialist
 
Member#: 103136
Join Date: Dec 2005
Default

Quote:
Originally Posted by gabedude View Post
I hate it when my memory fails me and I don't have one of your stickies to reference.
Some of the FBKC disable stuff was not in the last sticky. I've been waiting for Enginuity to come back up before updating it, but that might be awhile so maybe I'll post it up here some time.
Tea cups is offline   Reply With Quote
Old 01-18-2008, 02:03 PM   #10
Jon [in CT]
*** Banned ***
 
Member#: 2992
Join Date: Nov 2000
Chapter/Region: NESIC
Location: Connecticut, USA
Vehicle:
02 WRX Sedan
Silver

Default

Quote:
Originally Posted by Tea cups View Post
Some of the FBKC disable stuff was not in the last sticky. I've been waiting for Enginuity to come back up before updating it, but that might be awhile so maybe I'll post it up here some time.
Why not just correct/update the existing Scoobypedia article at http://www.scoobypedia.co.uk/index.p...ategyExplained?
Jon [in CT] is offline   Reply With Quote
Old 01-18-2008, 02:11 PM   #11
gabedude
Scooby Guru
 
Member#: 10108
Join Date: Sep 2001
Chapter/Region: TXIC
Location: Lake Travis, Austin, TX
Vehicle:
07 STI
60Trim spun rod bearing

Default

Quote:
Originally Posted by Tea cups View Post
Some of the FBKC disable stuff was not in the last sticky. I've been waiting for Enginuity to come back up before updating it, but that might be awhile so maybe I'll post it up here some time.
Ahh, so my memory is still working. I hate getting old.
gabedude is offline   Reply With Quote
Old 01-18-2008, 02:19 PM   #12
Tea cups
Scooby Specialist
 
Member#: 103136
Join Date: Dec 2005
Default

Quote:
Originally Posted by Jon [in CT] View Post
Why not just correct/update the existing Scoobypedia article at http://www.scoobypedia.co.uk/index.p...ategyExplained?
Well, I can update that too (it doesn't have even the last update it looks like), but I think it would be best to add separate threads to the new Nasioc sub-forum as this is where it seems the technical openecu discussion has migrated to (from Enginuity).
Tea cups is offline   Reply With Quote
Old 03-24-2008, 11:30 AM   #13
cowtown
Scooby Newbie
 
Member#: 68435
Join Date: Aug 2004
Location: NorCal
Vehicle:
'07 STI

Question

Hi, I'm resurrecting this thread to get some clarification on actual knock when logging FLKC.

Tea cups wrote:
Quote:
FLKC could be due to past knock unless the current cell is changing. You can determine this by logging 'Fine Learning Table Offset'. If this value remains the same but your FLKC changes then there was a change to the current FLKC cell (barring any "sync" issues which can occur due to the way the ECU polls the parameters).
So this is not necessarily current knock, but a learned response to past knock?
Code:
Time	RPM	FBKC	FLKC	FLTO	IAM	THROTTLE	
38:44.3	2628	0	0	6	1	17.25						
38:44.7	2666	0	0	7	1	19.61						
38:45.0	2703	0	-1.05	8	1	29.8						
38:45.4	2678	0	-1.05	8	1	30.98						
38:45.7	2779	0	0	9	1	28.63						
38:46.0	2735	0	0	9	1	30.98
And this is current knock?
Code:
Time	RPM	FBKC	FLKC	FLTO	IAM	THROTTLE	
38:06.4	2629	0	0	8	1	21.96
38:06.7	2674	0	0	8	1	21.96
38:07.0	2697	0	0	8	1	21.57
38:07.4	2737	0	-1.4	8	1	21.57
38:07.7	2776	0	-1.4	8	1	21.57
38:08.0	2764	0	-1.4	8	1	21.57

Last edited by cowtown; 04-10-2008 at 02:09 PM.
cowtown is offline   Reply With Quote
Old 03-24-2008, 12:02 PM   #14
Tea cups
Scooby Specialist
 
Member#: 103136
Join Date: Dec 2005
Default

Quote:
Originally Posted by cowtown View Post
Hi, I'm resurrecting this thread to get some clarification on actual knock when logging FLKC.

So this is not necessarily current knock, but a learned response to past knock?
Code:
Time	RPM	FBKC	FLKC	FLTO	IAM	THROTTLE	
38:44.3	2628	0	0	6	1	17.25						
38:44.7	2666	0	0	7	1	19.61						
38:45.0	2703	0	-1.05	8	1	29.8						
38:45.4	2678	0	-1.05	8	1	30.98						
38:45.7	2779	0	0	9	1	28.63						
38:46.0	2735	0	0	9	1	30.98
Yes, this is a learned correction in the past.

Quote:
And this is current knock?
Code:
Time	RPM	FBKC	FLKC	FLTO	IAM	THROTTLE	
38:06.4	2629	0	0	8	1	21.96
38:06.7	2674	0	0	8	1	21.96
38:07.0	2697	0	0	8	1	21.57
38:07.4	2737	0	-1.4	8	1	21.57
38:07.7	2776	0	-1.4	8	1	21.57
38:08.0	2764	0	-1.4	8	1	21.57
Yes, this is current knock that is now a learned correction.

Quote:
Assuming I'm reading these correctly, should I be trying to tune out all high-load current-knock FLKC for a very conservative daily-driver tune?
You need to also look at FBKC and the IAM to get a complete picture - only one is active at any given time. What you need to do is look for patterns. Every time you flash a change to your map (for almost all ECUs), the ECU is reset and the FLKC table is cleared. So, if you see the same negative correction consistently in a particular cell (after FLKC table is populated), especially at moderate/high loads, then that is something you would want to take a look at and do further logging to determine where the issue may lie (timing, fueling, tip-in enrich, etc.). Note, that the FLKC table is cleared every time the IAM changes, so if your IAM isn't stable, obviously the FLKC data isn't going to very useful.

Looking at the FLKC table only is not a substitute for normal logging as it will not indicate knock handled by the other knock control elements - FBKC and the IAM. Also, the FLKC ranges are sometimes quite broad so you need additional logging to narrow down where and under what circumstances the perceived knock is occuring. And it should be noted that the knock sensor/filter/ECU logic is not perfect at determining knock. You will never completely eliminate all negative corrections from FLKC or FBKC. If you want to look at the entire FLKC table at once, DL the learning view tool on RomRaider.
Tea cups is offline   Reply With Quote
Old 03-24-2008, 12:49 PM   #15
cowtown
Scooby Newbie
 
Member#: 68435
Join Date: Aug 2004
Location: NorCal
Vehicle:
'07 STI

Default

Thanks very much Tea cups, that makes sense.

I was having a difficult time getting my head around the FLKC because of the offset table.

Thanks!

Last edited by cowtown; 04-10-2008 at 02:09 PM.
cowtown is offline   Reply With Quote
Old 03-25-2008, 05:48 AM   #16
benw
Scooby Specialist
 
Member#: 103320
Join Date: Dec 2005
Location: Oklahoma City
Vehicle:
2005 wingless
aspen white

Default

Excellent info as usual by Bill and Gabe. In my experience, I have always had better luck when resetting the ECU before logging, to clear the FLKC tables. The learning view tool is great when friends stop by for a checkup, as it indicates any possible spots the car is knocking when subjected to the rigors of the owner's daily driving style. 15 minutes of around town logging with MickeyD's knock summary sheet usually reveals all trouble spots.

When the ECU is reset, doing some 2000-redline pulls will reveal all knock (in my experience). Some cars don't reset their learned knock when a map is flashed, so I manually reset those to be safe (using the RomRaider test release).
benw is offline   Reply With Quote
Old 03-30-2008, 10:48 AM   #17
TrRex06
Scooby Newbie
 
Member#: 170472
Join Date: Jan 2008
Chapter/Region: SCIC
Location: San Clemente, CA
Vehicle:
2006 WRX TR
Black

Default Help me diagnose

Sorry to threadjack, but could someone help me diagnose this log. Its not bad, but it is very consistat, so I would like to get it dialed out. Timing is already pretty conservitive, and I don't think fuel is to lean, So before I go pulling anymore timeing or anything else I would like to get some other peoples feedback.

Code:
Time   Accel(%)	N/A     Load 	(rpm)  FBKC   FLKC    IAM 	Timing	N/A	IAT	KC	 MRP	 N/A 	WGDC	AEM UEGO (AFR)
	
57906	21.96	4.08	1.08	2292	0	0	1	27.5	10.74	82	2.5	-2.18	1.68	7.45	14.7
58266	21.57	3.78	1.12	2290	0	0	1	26	11.21	82	2.5	-1.89	1.71	7.06	14.4
58625	20.78	3.48	1.14	2298	0	0	1	25.5	11.28	82	2.5	-1.74	1.67	7.06	14.6
58969	28.24	3.41	1.19	2286	0	0	1	24.5	11.78	82	3	-1.02	2.13	8.63	14.8
59313	30.2	2.51	1.27	2308	0	0	1	20	12.8	82	3.5	-0.15	2.23	10.59	14.6
59672	32.55	2.07	1.41	2300	0	0	1	18	15.71	82	6	1.31	3.08	13.33	14.1
60016	38.82	0.71	1.6	2334	0	0	1	14.5	19.37	82	6	3.19	3.72	16.08	12.5
60391	45.49	0.31	1.72	2328	0	0	1	13	21.31	82	6	4.5	4.69	16.86	11.5
60734	55.69	0.33	1.84	2356	0	0	1	12.5	22.78	84	6	5.8	7.22	23.53	11.3
61109	55.69	0.77	1.94	2383	0	0	1	12.5	24.36	84	6	6.96	7.5	33.73	11.2
61453	55.69	0.11	2.02	2392	0	0	1	12	25.68	84	6	7.84	7.85	34.9	11.2
61813	55.29	-0.63	2.08	2427	0	0	1	11.5	26.97	84	6	8.71	8.04	31.37	11.2
62156	54.9	-1.22	2.13	2446	0	0	1	11	27.96	84	6	9.58	8.08	27.45	11.1
62516	54.9	-1.58	2.16	2460	0	0	1	11	28.82	84	6	10.16	8.25	19.22	11.2
62875	52.55	-2.38	2.17	2489	0	0	1	11	30	86	6	10.45	7.99	7.06	11.1
63219	52.55	-1.32	2.02	2522	-1.4	0	1	10.5	26.37	86	4.5	9.72	8.3	3.53	11.5
63578	52.55	-0.92	2.06	2560	-1.4	0	1	10.5	27.88	86	4.5	9.43	8.66	1.57	11.9
63922	52.16	-0.56	2.07	2586	-1.4	0	1	10.5	28.2	86	4.5	9.43	8.73	0.39	11.8
64344	51.37	-0.52	2.07	2606	-1.05	0	1	11	28.91	88	5	9.43	8.59	0	11.6
64703	47.84	-1.35	2.06	2609	-1.05	0	1	11	29.46	88	5	9.29	7.3	0	11.7
65063	43.14	-1.8	2.04	2661	-1.05	0	1	11	29.38	88	5	9	6.79	0	11.6
65422	32.55	-2.35	1.9	2676	0	0	1	13	26.66	88	6	7.55	4.22	0	11.4
65781	10.59	0.76	1.05	2663	0	0	1	23	11.49	90	2.5	-1.02	0.58	0	11.8
66141	0	9.53	0.04	2661	0	0	1	44.5	2.8	90	0	-9.72	0.01	0	14.7
66500	0	10.7	0.12	2628	0	0	1	42.5	1.68	90	0	-10.59	-0.01	0	19.4
66859	0	11.03	0.18	2598	0	0	1	41.5	1.66	90	0	-10.88	-0.04	0	19.4
67219	2.35	10.92	0.22	2572	0	0	1	47.5	4.46	90	0	-10.74	-0.02	0	19.4
67594	4.71	10.1	0.3	2558	0	0	1	51.5	5.48	90	0	-10.01	-0.1	0	19.4
67938	5.49	10.17	0.31	2538	0	0	1	51.5	4.34	88	0	-10.16	-0.12	0	14.4
68297	6.67	10.22	0.31	2528	0	0	1	51.5	4.29	88	0	-10.16	-0.15	0	13.3
68656	11.76	8.36	0.54	2484	0	0	1	47.5	6.9	88	0	-7.69	-0.01	0	14.6
69016	12.16	7.86	0.55	2486	0	0	1	46.5	6.36	88	0	-7.55	0.06	0	14.7
69359	11.37	7.29	0.58	2472	0	0	1	44	6.84	88	0.5	-6.96	-0.05	0	15
69703	12.55	7.43	0.57	2480	0	0	1	45.5	6.29	88	0.5	-6.96	0.13	0	14.9
70063	8.63	7.11	0.53	2436	0	0	1	47.5	6.24	86	0	-7.25	-0.23	0	14.3
70406	0	10.27	0.24	2420	0	0	1	48.5	1.55	86	0	-10.45	-0.25	0	14.8
70781	0	10.89	0.22	2403	0	0	1	45	1.54	86	0	-11.03	-0.32	0	15.2
71141	0	10.78	0.22	2356	0	0	1	42.5	1.52	86	0	-11.03	-0.35	0	16.9
71500	0	10.7	0.22	2364	0	0	1	40.5	1.51	86	0	-11.03	-0.47	0	19
71859	0	10.63	0.22	2324	0	0	1	38	1.49	86	0	-11.03	-0.58	0	19.4
72234	0	10.57	0.21	2303	0	0	1	36	1.48	86	0	-11.03	-0.62	0	19.4
72578	0	10.5	0.22	2267	0	0	1	34	1.45	86	0	-11.03	-0.72	0	19.4
72922	0	10.38	0.21	2264	0	0	1	31.5	1.44	86
Mods: TXS catless turboback exhaust, Frontmount, 18G, K & N Short ram, stock injectors (for now) and a stage 2 reflash that I am reworking to suit my needs. I have some more logs, but they all show pretty much the same thing, but I can post them later if nessisary
TrRex06 is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Power Loss and loud knock from under the hood. STIdreamer17 Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo) 6 10-05-2009 06:34 PM
Need help w/ data log and AFR / Knock question Tronner AccessPort 1 11-09-2007 02:07 PM
A UTEC question - flashing CEL, no knock seen on data log..what gives? LargeOrangeFont Factory 2.0L Turbo Powertrain (EJ Series Factory 2.0L Turbo) 11 10-03-2003 03:33 PM
Logged data from my grocery run ninrex South East Region Forum 1 05-22-2002 12:33 AM
sorta OT, how to log data from a potentiometer RidinLow General Forum Archive 0 11-09-2000 08:18 AM


All times are GMT -4. The time now is 03:34 PM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2015, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2015 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.