Welcome to the North American Subaru Impreza Owners Club Thursday August 21, 2014
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 > Factory 2.0L Turbo Powertrain

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 06-27-2002, 01:50 PM   #1
Ryan23
Scooby Specialist
 
Member#: 15116
Join Date: Feb 2002
Chapter/Region: MAIC
Location: Newburg, MD
Vehicle:
2000 Ver6 STi Type-R
EJ257/GT35R

Default Knock Correction and timing (long!)

I figured I would start a new thread being that the Delta Dash thread is pretty packed. Here's my insights into the wonderful world of knock correction based on what MY (Important word here!) ECU is doing.

First off, some basic observations for the uninitiated. Positive knock correction (as read in Delta Dash) appears to be a good thing. Usually seen on acceleration (with no knock), it adds + number counts which looks to be added timing in degrees for a given RPM. It could be a ratio, I haven't seen any concrete evidence. Negative knock numbers indicates timing getting pulled based on a couple different factors. The first and most obvious, knock is heard at a given RPM by the knock sensor. The ECU looks to handle this in a couple different ways. Sometimes, when knock is heard, the ECU reacts with pulled timing and negative knock correction immediately. Other times, it appears to squirrel this info away in it's "Knock Table" for later use. I think the big factor here on it's decision making is frequency. Couple of knock events or more, immediate reaction....Single event, nothing done initially but it remembers it. My car, historically, sees knock events anywhere between 700~1800 RPM. These are not real knock events rather random noise that doesn't get filtered by the sensor. In my "test pool" of 4 WRX's, they all exhibited some knock sensor activity at idle. Some worse than others. I will get on average 3 events in 2 minute of idling whereas I have seen one car that had a dozen. I'll get one at the occasional stoplight, idle in traffic, ETC. The question commonly asked is, "Does the ECU know that it isn't real knock?" Based on what I have seen, No. My car usually sees negative knock correction (sporadically) between off idle and 2K RPMS. It doesn't correlate with a specific knock event, just historical ones. Now, if a little timing is yanked between 1K and 2K what's the big deal? Probably nothing. I launch at the strip around 5K and by the time I'm moving I am pulled down to about 4~4.5K. I never seen that lower RPM range except maybe leaving a stoplight from a lazy start. However, I don't have enough data yet to see if a little timing missing around 2K equals a little timing missing around 5K.
What I have also noticed is the occasional sporadic single spike which looks to be noise. I've seen them everwhere, 2.5K RPM @ a steady cruising speed, 3K, ETC. The engine isn't under boost, not being taxed, level ground ETC. The ECU hears these too and I have seen some negative correction over time based on one single event.
Moving right along, I have also noticed that the ECU is cautious in the 5K~6K RPM. It doesn't look to add too much positive correction in that range and is very quick to yank it back. The interesting correlation I have made here (Once again, with my ECU and MAYBE yours) is that this decision hinges on what the ECU sees from it's MAP sensor. Fortunately, my MAP is clamped to only see up to 12.2 PSI. Maybe clamped is the wrong word, it's more of a boost ratio. IE: 16.5 "Real" PSI = 12.2 MAP sensor PSI. 15 PSI "Real" = 11.2 MAP PSI and so forth. I have seen many instances (I log ALL the time) where .2 ~ .4 PSI makes ALL the difference in the world of timing and knock correction at 5K. If I set my EBC to give me 16.5 PSI, I see, on average, 20~26 Degrees of total advance (with NO knock events). Bring the boost back down to 15 PSi and I'll see up to 31 degrees in @ 6K all other factors pretty much equal. I don't have any webspace to post my logs for everyone to view, but it's pretty consistent. What I am looking to do now is 1st.) Relocate my knock sensor to a less "noisy" spot. One that will still hear det but not any silly background noise in that same freq range and 2nd.) I'm looking to drop my "MAP clamp" down to 10.5~11 PSi at 16.5 "real" PSI and see what timing does. Long term knock correction by the way appears to be going up as well at the lower boost setting. I see on average, 6~7 Positive correction @ 15 PSI whereas I use to see 3 ~ 4 positive correction at 16.5 PSI (with some negative knock correction thrown in for good measure). Remember, this is with no knock events being heard either for the whole run. 10 degrees of timing up top would definately account for some car to car variances! I also am going to log the ECU by itself (no unichip) and see if it reacts the same @ lower and higher boost. By the way, my gas stays pretty consistent, 93 Octane from the local Exxon.
If you have any interesting observations you could bring to the table, I would love to hear it. I think once my car gets some consistent timing down I would really be able to appreciate my mods a lot more .

P.S. I know someone will ask, ........ many vendors sell Delta Dash. I got mine from TurboXS.

Ryan

- TXS Stage 2 Unichip
- TXS Turboback
- Blitz SBC-ID
- Vishnu Pulley
- Vishnu Upipe

In my "fast closet":

- STi "Pink" Injectors
- MRT Front Mount
- Walbro 255 LPH Pump

On the way:

- IHI VF34
and hopefully soon, a full featured, competitively priced, engine management "product" that fixes all those problems noted above
* Registered users of the site do not see these ads.
Ryan23 is offline   Reply With Quote
Old 06-27-2002, 02:00 PM   #2
codean
Scooby Specialist
 
Member#: 10186
Join Date: Sep 2001
Chapter/Region: MWSOC
Location: some kind of bmw tech...
Default

Hmmm..

I have never seen a knock event at low rpm's on my delta dash

I do have sti motor and tranny mounts, I wonder if that would make any diffrence?


The mods I have are:

Turboxs turbo back
vishnu up-pipe
SBC-ID
STI motor mounts
STI tranny mount
K&N filter
Colder NGK's spark plugs
codean is offline   Reply With Quote
Old 06-27-2002, 02:37 PM   #3
RiftsWRX
Scooby Guru
 
Member#: 6124
Join Date: Apr 2001
Chapter/Region: MWSOC
Location: Glendale Hts, IL, USA
Vehicle:
2000 NFR AP1 S2000
'07 Honda FIT sport (5MT)

Default

Hmmmm... Riddle me this bat man.... at 15 PSI, what is your A/F sensor #1 value at ~3900 RPM... and what is it at 16.5 PSI...

Jorge (RiftsWRX)
www.ProjectWRX.com
RiftsWRX is offline   Reply With Quote
Old 06-27-2002, 03:04 PM   #4
TurboRex
Scooby Specialist
 
Member#: 8635
Join Date: Jul 2001
Location: Texas
Vehicle:
04 STI, 07 FXT

Default

How are you altering your map signal? I thought the unichip controlled this without any user tuning options.

Your thoughts on raising boost lowers timing is not entirely correct from my findings. I think the ecu bases timing on maf. Thus, by reducing boost you lower maf and thus the engine runs a more aggressive timing curve.

I am curious to see what effect lowering the map signal will do. I would guess no change to timing, but your car will run leaner. I am about to install an mrt fcd so we can compare notes.

Greg
TurboRex is offline   Reply With Quote
Old 06-27-2002, 03:39 PM   #5
SnortWagon
Scooby Newbie
 
Member#: 14124
Join Date: Jan 2002
Location: Long Lake, MN
Vehicle:
2002 WRX Wagon
Silver

Default

Ryan23,

Thanks for the post. Although I don't have any sophisticated tools to log data, I have certainly had my car pull timing (based on knock that I couldn't hear), and then, mysteriously return the timing to 'normal' later on down the road (about 400 miles later). Maybe with your logs and others helping, we can figure out how and when the ECU decides to restore timing to some base map within the ECU.

I am always wondering if my butt dyno is accurate or not - probably not - but my car certainly runs stronger some times than other, and it's not always gas or weather dependent.
SnortWagon is offline   Reply With Quote
Old 06-27-2002, 08:33 PM   #6
Twelvz
Scooby Specialist
 
Member#: 11302
Join Date: Oct 2001
Location: VA Beach VA + Cleveland!
Vehicle:
11 GT-R
10.x @130+

Default

Quote:
Originally posted by codean

I have never seen a knock event at low rpm's on my delta dash
Ryan did a quick "idle" data log of 3 of our cars over this past weekend. We all had it. We all have various mods, though noone has any STI mounts.
Twelvz is offline   Reply With Quote
Old 06-27-2002, 08:41 PM   #7
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 Twelvz


Ryan did a quick "idle" data log of 3 of our cars over this past weekend. We all had it. We all have various mods, though noone has any STI mounts.
Oh oh. Sounds like Ryan23 might be abusing his EcuTek software license.
Jon [in CT] is offline   Reply With Quote
Old 06-28-2002, 12:12 AM   #8
mlambert
Scooby Guru
 
Member#: 13831
Join Date: Jan 2002
Location: 99 problems but a bitch aint 1
Default

hey ryan im curious about jorges question, if you have the logs could you answer it. might be onto something!
mlambert is offline   Reply With Quote
Old 06-28-2002, 10:09 AM   #9
Ryan23
Scooby Specialist
 
Member#: 15116
Join Date: Feb 2002
Chapter/Region: MAIC
Location: Newburg, MD
Vehicle:
2000 Ver6 STi Type-R
EJ257/GT35R

Default

Sorry about the delay, been busy at work!

Jorge - I took a quick look at the logs last night and I showed .76 ( < 11.1) on A/F gauge 1 during my 16.5 psi run. During the 15 PSI run I showed .76 @ WOT with a couple small corrections to .8 . We're still in the 11.xx range of A/F ratio, so I'm not sweating it. EGT's appear fine according to my Westach probe (4" from cylinder).

Codean - You may be on to something with that. I have often wondered if that would make a difference as our engines do have a lot of "play" from the factory. I'm sure you've watched one shut down before! Always triggers a knock event for me.

Turborex - It's not user adjustable unless you get down to the harness and do an "adjustment". I spoke with Mark @ TXS and he is getting me the "resistor value chart" curve. Basically, it appears to be a resistor in Parallel that pulls the map voltage down. Nothing you average hobbyist can't do (I'm a certified Navy Micro-Minature Circuit Repair tech, no sweat for me!)

Snort - No doubt, I have felt the same thing. It's blazingly obvious now with DD. Ya know, sometimes ignorance is bliss.....

Jon - I didn't make any copies and I only have one hardware key anyways. If DD will work with another person's ECU, why not? I did a quick scan of Ecuteks' website and all I saw was:
Quote:
If you wish to use the software with your own car, or a single model of car, support for a single range will be fine.
Anyways, as far as timing be MAF dependent only, that I'm not seeing. I'm ASSuming that the MAF signal is altered in a linear fashion to compensate for fueling but I'm not 100% sure on that. I wish I had a place to put my logs up online, but I'll try to explain in text format:

All Data from the same run @ 16.5 PSI (Launch @ 5K)

WOT, 5399 RPM in 1st gear:

MAF = 110786
MAP = 10.7
Timing = 18
Knock correction = +2
A/F #1 = .76

WOT, 5482 in 2nd Gear,

MAF = 110564
MAP = 11.2
Timing = 18
Knock correction = +2
A/F #1 = .76

WOT, 5380 in 3rd gear,

MAF = 111239
MAP = 11.6
Timing = 13.5
Knock Correction = 0
A/F #1 = .76

WOT, 5419 in 4th gear,

MAF = 111691
MAP = 11.8
Timing = 11
Knock correction = -4
A/F #1 = .76

Here above, MAF looks it COULD play a part, but notice boost has gone up at the MAP also. Now look below and explain this:


All Data from the same run above, WOT in 3rd

RPM = 6102
MAF = 117740
MAP = 10.7
Timing = 19
Knock correction = 0
A/F #1 = .76

RPM = 6178
MAF = 118185
MAP = 10.6
Timing = 23.5
Knock correction = +4
A/F #1 = .76

RPM = 6367
MAF = 119385
MAP = 10.3
Timing = 25.5
Knock Correction = +4
A/F #1 = .76

RPM = 6482
MAF = 120820
MAP = 10.2
Timing = 26
Knock correction = +4.5
A/F #1 = .76

RPM = 6631
MAF = 122255
MAP = 10.3
Timing = 26
Knock correction = +4.5
A/F #1 = .76

RPM = 6708
MAF = 122491
MAP = 10.2
Timing = 26
Knock correction = +4.5
A/F #1 = .76

RPM = 6849
MAF = 124169
MAP = 10.2
Timing = 26
Knock correction = +4.5
A/F #1 = .76

There, MAF has gone up but boost has gone down, look at the transition to + knock correction. Granted the whole 5K range was @ 0 knock correction but I think this was due to boost in the 11.xx range coupled with what the ECU thinks is the engines "sensitive" zone (more Likely to Det zone, in other words).

Blah, another monster post and I must get back to work. I'll post the comparison's to the above with my 15 PSI run. Whatcha think? Counterpoint? I'm fumbling around in the dark like most here, so if your seeing something I'm not, let me know.


Ryan
Ryan23 is offline   Reply With Quote
Old 06-28-2002, 10:28 AM   #10
RiftsWRX
Scooby Guru
 
Member#: 6124
Join Date: Apr 2001
Chapter/Region: MWSOC
Location: Glendale Hts, IL, USA
Vehicle:
2000 NFR AP1 S2000
'07 Honda FIT sport (5MT)

Default

.76 SPECIFICALLY at 3900?

I want to know THAT spot.....

Jorge (RiftsWRX)
www.ProjectWRX.com
RiftsWRX is offline   Reply With Quote
Old 06-28-2002, 10:51 AM   #11
gryphon
*** Banned ***
 
Member#: 11807
Join Date: Oct 2001
Location: Lewisville, TX
Vehicle:
2002 WRX
Black

Default

Ryan23,
I don't have STi Motor or tranny mounts and I'm seeing plenty of knock events at idle. Add me to the list

I'll play around with some of my values (non-Unichip filtered) and see what correlation I can come up with between knock timing and MAF / MAP. My question is, should this be done on an ECU reset to "remove" that Knock Table so that the timing you are seeing is based more upon (again, ASSuming ), MAF/MAP and not previous knock or "conducive to knock" events?
gryphon is offline   Reply With Quote
Old 06-28-2002, 10:57 AM   #12
Ryan23
Scooby Specialist
 
Member#: 15116
Join Date: Feb 2002
Chapter/Region: MAIC
Location: Newburg, MD
Vehicle:
2000 Ver6 STi Type-R
EJ257/GT35R

Default

Quote:
Originally posted by RiftsWRX
.76 SPECIFICALLY at 3900?

I want to know THAT spot.....

Jorge (RiftsWRX)
www.ProjectWRX.com
16.5 PSI, Yes .76 @ 3870 and 3958. I don't have a log of exactly 3900 At 15 PSI, I'll let you know in a couple hours when I get my laptop (15 PSI logs are on there and not on my desktop).


Ryan
Ryan23 is offline   Reply With Quote
Old 06-28-2002, 11:03 AM   #13
TurboRex
Scooby Specialist
 
Member#: 8635
Join Date: Jul 2001
Location: Texas
Vehicle:
04 STI, 07 FXT

Default

I belive rpm is the main factor in timing. As rpms increase, so does timing. The next factor is maf. If maf is more at the same rpm, then timing will be higher (a/f can be a factor if the ecu thinks the car is too lean then it will pull timing).

We could test this theory if people compare maf readings at the same boost level and what the timing is at that rpm. I would think a completely stock wrx running 15psi boost should have more timing than one with a turbo back exhaust. You need to remember too that unichip users have had timing manipulated so it would not be a good subject for this test.

A would suggest timing at redline or 6500 rpm where it has less chance to be manipulated by a/f, etc.

Greg
TurboRex is offline   Reply With Quote
Old 06-28-2002, 04:09 PM   #14
Ryan23
Scooby Specialist
 
Member#: 15116
Join Date: Feb 2002
Chapter/Region: MAIC
Location: Newburg, MD
Vehicle:
2000 Ver6 STi Type-R
EJ257/GT35R

Default

Quote:
Originally posted by RiftsWRX
.76 SPECIFICALLY at 3900?

I want to know THAT spot.....

Jorge (RiftsWRX)
www.ProjectWRX.com
Jorge,
The closest I have to 3900 is 4047 RPM @ 15PSI, I'm still at .76. At 3566 I'm at .76 also. Interpolating the difference still gives me .76 I'll do another run tonight to see if I can catch 3900 on WOT.

Ryan
Ryan23 is offline   Reply With Quote
Old 06-28-2002, 07:18 PM   #15
codean
Scooby Specialist
 
Member#: 10186
Join Date: Sep 2001
Chapter/Region: MWSOC
Location: some kind of bmw tech...
Default

forget about what I said.............I got one knock event every time I did a 5 min idle test.
codean 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
How come I Can't see feedback knock correction, fine learning knock correction? AraSh Open Source Reflashes 23 01-08-2014 07:04 PM
Knock Correction and Oil?? mpggomd Factory 2.0L Turbo Powertrain 3 07-31-2006 06:02 PM
Correlations between Knock Lite and negative knock correction xFactor Engine Management & Tuning 1 01-12-2006 07:36 PM
Timing, knock correction, and other questions root Engine Management & Tuning 18 07-26-2004 06:24 PM
Timing and knock correction woes Egan Factory 2.0L Turbo Powertrain 10 10-31-2003 08:06 PM


All times are GMT -4. The time now is 03:11 PM.


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