Welcome to the North American Subaru Impreza Owners Club Wednesday April 16, 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 > Engine Management & Tuning > Commercial ECU 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 01-01-2007, 07:51 PM   #1
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default Why is the stock KC map so lumpy?

Here's what the stock KC (ignition correction) map looks like on the WRX. It appears to have remained the same over the years.



Here's my smoothed out KC map (I adjusted the base timing map so that the total timing, assuming IAM or DA of 16, remains close to the same).



The KC maps from the STi are not quite as bad as the WRX maps, but they are a bit tougher to interpret since there are 3 of them.

The base timing map is a bit lumpy, but not always the opposite of the KC map so the overall timing in places would appear to jump a bit.

Some guesses I have:

1. Areas more prone to knock are given higher KC values so that if the IAM drops, more timing is pulled from those areas.

2. Areas less prone to knock are given lower KC values so less timing is pulled if the IAM goes down. This is evident in the low load areas of the map where KC is 0.

3. Subaru engineers pulled KC where knock occurred. 4400 rpm load sites seem to have a few degrees less timing than 4800 and 4000 rpm load sites. What makes 4400 such a knock prone rpm?

4. Some of the lumpiness I think could be attributed to the VE of the engine - They seem to add KC where the engine makes a lot of torque and pull KC where it doesn't. But then add some in at high RPM perhaps to help safeguard the engine against high RPM knock if the IAM is lowered.

From what I've seen, it also appears that people have a few different methods of tuning the timing maps. Some completely revamp them and smooth them out significantly (saying that with a consistent KC maps makes it easy to tell where the ECU has pulled timing), while some basically leave it alone and only make small changes to account for knock or lack there of.

What's everyone else's theory?
* Registered users of the site do not see these ads.
drees is offline   Reply With Quote
Old 01-01-2007, 10:57 PM   #2
Freon
Scooby Specialist
 
Member#: 88322
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Indianapolis, IN
Vehicle:
2009 BMW 135i

Default

Well, the base maps are very smooth. With a smooth map and a bumpy map added, you still end up with a result that is bumpy. So strictly #1 and #2 don't really explain why the final timing used after all the compensations and such is kinda bumpy. One would expect final timing to be somewhat smooth. It really isn't as bad as the KC maps make it look, but the KC and base maps only somewhat "mesh" or seem to be complementary. My point here is you may not need to know why the KC maps are bumpy, but possibly a more interesting subject is why final timing is still somewhat bumpy.

Now looking at how the actual KC code operates, I can tell you #1 and #2 are roughly true. The IAM is multiplied by the KC map values, so when IAM drops by a certain amount, your absolute loss (in degrees) in final timing used will be larger if you have a larger portion KC and less base timing. Keep in mind dropping from 15 to 13 degrees may have a larger effect than some other area on the map that drops from 26 to 24. Keep that in mind building your base and KC maps. I don't think the way it operates says that you should always build in larger KC values and less base if you are worried about knock. You could have the engine pull too much timing causing other problems, like boost creep/overboost when your root problem an overly aggressive wastegate map. So don't just run out and throw on tons of KC and little base timing.

Example, you could build timing with base of 18 and KC of 3 for a total of 21. If you determined this spot may be a problem or prone to knock, you could change it to 15 + 6. If your IAM drops to half max, you will drop to 18 degrees total timing instead of 19.5. This may or may not be desirable for you. There are other factors that can come in as well, but this portion of the timing logic does work this way and you should tune it knowing this.

As for 4400rpm, it could be a resonant spot for the intake and exhaust. Maybe the effective air load is higher there. The calculated engine load really doesn't take into account how well the engine is breathing. Even if the MAF is perfect, there is no way to know exactly how much fresh air goes out the exhaust valve on in/exh valve overlap, nor how much exhaust gas stays in the cylinder between cycles. It certainly changes with RPM, load, back pressure, and for those with AVCS, cam timing. There could be all sorts of oddities due to the exhaust and intake manifolds, heads, and cams here. Maybe even injector phase timing which no one has even identified yet as tunable.

I almost wonder if some of the larger values you see in the lower loads are literally on the hot side to "test" the gas volatility if you will, where knock can be picked up rather than find out at a higher load. Just a guess, though. It is really kinda useless until you know your gas well to start, and can correlate gas volatility at lower loads and RPM to higher areas.

I don't think without an engine dyno and thousands of hours an aftermarket tuner could ever intelligently justify a map as bumpy as the OEM. There are going to be some spots that aren't 100%. Usually when people tune they take the limited time they do have and really work on the WOT areas, then hit what they can on the rest and smooth it out. There is only so much you can do.

You can get whatever you want just right as long as that is what you focus on. If you were to focus on every individual load point on a map and optimize them one at a time I'm sure you can get there. That's quite a battle and one that may not be practical. The smaller and more isolated the bump or dip in the map, the harder it is going to be to explain. I wouldn't lose too much sleep over it.
Freon is offline   Reply With Quote
Old 01-02-2007, 01:54 AM   #3
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default

Well, I wouldn't say that the base timing maps are all that smooth. For kicks, here's what the stock base timing + KC map looks like.



The area between 4000-6000 rpm and 3-4 load just doesn't make sense to me. For example, why does 4400@3.88 have ~2 more degrees timing than the load cells one higher and lower than it. Some of those bumps/dips in the stock timing map must lead to that "jerky acceleration" feeling that you often get with a stock map.

So I guess one question is this: Assuming you had the map tuned well enough so that if you did encounter a bad tank of gas does the knock sensitivity of the engine change that differently over different load/rpm ranges to cause you to jump KC that much?

BTW, this discussion so far hasn't really touched the ignition fine tuning that the ECU does as well... I also wish I had a tool to read the learned ignition timing from the ECU, sure would make tuning easier than pouring through logs.
drees is offline   Reply With Quote
Old 01-02-2007, 09:24 AM   #4
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

drees, i don't really have an answer for you.

i CAN tell you that i am a "flat KC map" kind of guy.

i can also tell you that timing is lowest at VE peak, and low rpm.

ALL of my timing map "shape" choices are goverened by those two rules, and they haven't let me down yet.
ride5000 is online now   Reply With Quote
Old 01-02-2007, 02:44 PM   #5
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default

Quote:
Originally Posted by ride5000 View Post
i CAN tell you that i am a "flat KC map" kind of guy.

i can also tell you that timing is lowest at VE peak, and low rpm.

ALL of my timing map "shape" choices are goverened by those two rules, and they haven't let me down yet.
Yep and yep. Seeing some of your maps posted in the "so what's your max engine load?" and "for reference, my latest ign. map" threads gave me the inspiration to try the flat KC map, primarily to help make tuning easier. We'll see how it works out for me.

Another OE base timing oddity: Anyone notice that the for 05 they increased timing in the bottom load column from 10.94 to 23.95? For emissions perhaps? Seems like a drastic change.



Edit: Between 04-05, they also pulled a bunch of timing, in the .80-1.72 load columns from 1600-3200 RPM and between 03-04 pulled a bunch of timing between 2.0-3.48 load and 2000-4400 rpm... See the plot below. The area in blue is where Subaru changed base timing between 04-05 and in purple where Subaru changed KC timing between 03-04.


Last edited by drees; 01-02-2007 at 03:20 PM.
drees is offline   Reply With Quote
Old 01-02-2007, 05:53 PM   #6
R4ND0M_AX3
Scooby Newbie
 
Member#: 67483
Join Date: Aug 2004
Location: Sacramento, CA
Vehicle:
2005 WRX STi

Default

Is it really so bad to have lumpy timing? It is what it is right?

On a side note. Do you think when Subaru did the tuning they simulated the worst conditions (hi ambient temp, and crap gas) to tune the base timing then worked on the KC tables with ideal conditions?
R4ND0M_AX3 is offline   Reply With Quote
Old 01-02-2007, 06:04 PM   #7
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default

Quote:
Originally Posted by R4ND0M_AX3 View Post
Is it really so bad to have lumpy timing? It is what it is right?
It's not bad if that's what's required to reach MBT or the threshold before knock... But is that what Subaru is doing?
drees is offline   Reply With Quote
Old 01-02-2007, 06:33 PM   #8
Freon
Scooby Specialist
 
Member#: 88322
Join Date: Jun 2005
Chapter/Region: MWSOC
Location: Indianapolis, IN
Vehicle:
2009 BMW 135i

Default

Quote:
Originally Posted by R4ND0M_AX3 View Post
Is it really so bad to have lumpy timing? It is what it is right?
Well, lumpy is right if you have been given specific reason. If you can actually test and show that an odd lump is "better" via some sort of measure of output, then sure, it is "right". But such a finely adjusted map implies you need a lot of testing and data to actually know it is better. This is how you tune or design any system. It's not guesswork.

This is a battle of how much time and money you can spend to develop the tune. You might start with a good general trend based on best guesses and theory on how the engine works, then spend time actually tuning by experiment.

I imagine the finest, or bumpiest aspects of the tune are due to engine dyno time at Subaru. I doubt any, or few, of the tuning community will ever be able to spend the kind of time or money that Subaru does coming up with something with such fine details in the maps.

Practically it will be hard for any of us tuning our own cars to come up with something so crazy as the stock KC maps. I don't think you cannot use the knowledge of how the KC works to design a KC map that makes sense. I don't think KC should just be flat.

Quote:
On a side note. Do you think when Subaru did the tuning they simulated the worst conditions (hi ambient temp, and crap gas) to tune the base timing then worked on the KC tables with ideal conditions?
There are many compensations to deal with better and worse conditions. There should be leeway. I imagine in the most extreme (most ideal or least ideal, especially when concerning gas quality) they spent less time tuning.

From a manufacturer's standpoint, they want to spend the most time and money on the most common situations. Or, they want to spend R&D for best ROI. Least risk of damage over warranty life (or beyond), competitive enough on power, and just inside requirements for emissions. The more extreme the situation (i.e. 100 octane gas, or 150F ambient temps, or 86 octane gas) the less time they are going to test. They may just use simple and crude safeties to handle very low quality situations. Examples vary from high det fuel maps, then more drastic is boost control disable, then even more drastic, crude, and poorly performing is limp mode. If the first measure, the fuel map, kicks over, the car doesn't run like complete crap. It still makes power, but less, and fuel economy doesn't take a turd or anything. Boost disable is going to be darn noticable, but still isn't anything like limp mode. But, no one should actually drive in limp. It really does mean something is wrong.

Their goals are going to be different than ours, as the tuning community is not going to worry about emissions much if at all, and probably less on fuel economy, and at least a good measure less on reliability. The aftermarket is most concerned with power and response, and a bit of drivability. Yanking your catalytic converters with a large exhaust and a tune alone can pick up quite a bit of power and response by trading off just a few aspects that are different between the OEM and an aftermarket tune.
Freon is offline   Reply With Quote
Old 01-02-2007, 07:25 PM   #9
TMessick
Scooby Newbie
 
Member#: 882
Join Date: Feb 2000
Location: Michigan
Vehicle:
2004 WRX Wagon
very blue

Default

^^ Ya, based on my time doing OEM tuning (no, not for Subaru), Freon's got it about right. Typically, we get a timing curve from the base engine dyno folks who have a "spec" engine in terms of cam centerlines, compression ratio, etc. and spend lots of time dialing in the perfect spark for that particular point on that particular day, on that particular engine... The "good" engine dyno folks repeat several times, but at the end of the day, you sometimes get a spark curve that looks like the Himalayas. In my experience, physics really doesn't like discontinuities, and we usually end up having issues in vehicle if we run a spark curve that's too lumpy (surge, weird transient response, transient knock, etc.) I usually end up doing a bit of "smoothing" on the base dyno spark data, which pisses off the dyno engineers but in my opinion comes out with a smoother cal that drives a bit better. Also, Freon is spot-on in terms of trying to focus your production development/verification on optimizing the high-percentage driving conditions. Chances are the spark timing is a bit off at -20F, 9000 ft above sea level, on 85 octane. Think emissions temps where fuel economy, base power, and emissions are measured (50-80 degF) and things should be pretty optimized. Don't get me wrong, we try to test for all possible conditions, but sometimes, the fringe conditions just need to be "good enough". At least on the projects I've worked on however, there's not much power/tuning left on the table at these "nominal" conditions though (I'm not saying there aren't tradeoffs being made like the OL/CL delay stuff, but we typically push spark and fueling as far as "safe" to get the big advertised power numbers and advertised fuel economy)

The other thing you need to consider is that in a production environment, the tolerance stackups for things like cam centerline, compression ratio, knock sensor sensitivity, etc. can be pretty scary. The stock cal needs to cover all production vehicles. Compromises are made to "best-fit" things.

Oh yeah, I also suspect/reccomend that the knock correction map be not flat. Ideally, it would be a (mostly) smooth curve with larger numbers at the areas where you are more knock sensitive and/or critical. Typically, I'd expect zeros in the LH side and generally increasing numbers with MAF. Low RPM is typically more knock-challenged than high RPM, but there is usually a "hump" in the RPM trend based on engine manifold/cam tuning.

Good luck trying to sort that out without cylinder pressure transducers or a dyno though. Best "guess" would be to push the envelope using the "flat" knock correction map on say 93 octane and come up with a "perfect" spark surface. Then repeat on 87 octane and come up with a "perfect" low octane surface. The surfaces will be the same for some portion and the delta will give you an indication of relative knock sensitivity. Probably not worth the time and effort though since most of us don't change octane or general driving conditions enough to need that perfect sensitivity....

Your Mileage May Vary (and your MPG label will suck a lot more in 2008)
TMessick is offline   Reply With Quote
Old 01-02-2007, 07:32 PM   #10
TMessick
Scooby Newbie
 
Member#: 882
Join Date: Feb 2000
Location: Michigan
Vehicle:
2004 WRX Wagon
very blue

Default

Quote:
Originally Posted by drees View Post
Another OE base timing oddity: Anyone notice that the for 05 they increased timing in the bottom load column from 10.94 to 23.95? For emissions perhaps? Seems like a drastic change.
This seems like a decel (i.e. nearly closed pedal) type of MAF. In the past, I've been forced to increase timing in this area either due to high exhaust temperatures (if someone drives down a long hill with their foot barely on the pedal, things can get hot and still stay closed loop) OR due to exhaust noise/"pop" complaints on decel (kind of a very subtle exhaust pop/burble sound due to poor combustion quality and the charge actually finishing burning during the exhaust stroke). No idea if this is why Subaru did this for the same reasons though.
TMessick is offline   Reply With Quote
Old 01-02-2007, 07:46 PM   #11
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default

Quote:
Originally Posted by TMessick View Post
This seems like a decel (i.e. nearly closed pedal) type of MAF.
It not only affects decel, but idle timing, too, I think. At least my 03 idles around 11-12 degrees. Anyone with an 05 WRX note timing at idle?
drees is offline   Reply With Quote
Old 01-03-2007, 08:37 AM   #12
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by drees View Post
It not only affects decel, but idle timing, too, I think. At least my 03 idles around 11-12 degrees. Anyone with an 05 WRX note timing at idle?
drees, the idle timing map is distinct from the base ign (+kc) maps.

i confirmed this when i had to raise idle advance due to running leaner closed loop afr targets up around 16:1. the slower burn rate necessitated more advance. however, adjusting the main ign maps had no effect.

unfortunately, though i know they exist, i don't know where the idle ign map is on my ecu rev (and xmlwrite doesn't either, apparently) so i had to "fudge" it by globally raising an ECT compensation for idle timing.

in the end, it did the trick. i raised it 4* which brought idle advance it up to ~16*, and idle quality improved.
ride5000 is online now   Reply With Quote
Old 01-04-2007, 04:37 AM   #13
drees
Scooby Specialist
 
Member#: 25905
Join Date: Oct 2002
Chapter/Region: SCIC
Location: San Diego, CA
Vehicle:
2003 Silver Wagon
Tuned with Enginuity

Default

Ah, I did not realize that. I'm also slowing raising up my closed loop afr targets (at 15.0:1 right now). I think we're using the same ECU rev, A4TC400L.

I was thinking about the KC timing drop in the midrange from 03-04 I mentioned earlier and realized they must have reduced timing there to help account for the extended closed loop delay they added in 04.

Anyway, so far my car has been responding well to the flat KC map along with the other timing and boost tweaks I've been making. But then since my IAM stays at 16 (or close to it) the effects of a lumpy KC map don't come into play.
drees is offline   Reply With Quote
Old 01-04-2007, 08:03 AM   #14
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by drees View Post
Ah, I did not realize that. I'm also slowing raising up my closed loop afr targets (at 15.0:1 right now). I think we're using the same ECU rev, A4TC400L.
yep, that's the one.

Quote:
I was thinking about the KC timing drop in the midrange from 03-04 I mentioned earlier and realized they must have reduced timing there to help account for the extended closed loop delay they added in 04.
that makes sense since the delay is only about fuel. since the extended delay causes comparitively leaner afrs it would stand to reason that they had to reduce overall advance in the region that typically would have run richer in previous years.

Quote:
Anyway, so far my car has been responding well to the flat KC map along with the other timing and boost tweaks I've been making. But then since my IAM stays at 16 (or close to it) the effects of a lumpy KC map don't come into play.
that's a big difference between our DIY maps and the maps that FHI had to come up with... the variations from car to car are eliminated, and we are probably FAR more dilligent than your average joe about putting only the best gasoline we can find in the tank.

i think the most compelling reason to flatten out the KC map is so that logs can be interpreted easier, and the shape of the basemap + KC is much easier to visualize. neither of these factors come into play with the "average joe."
ride5000 is online now   Reply With Quote
Old 01-04-2007, 12:51 PM   #15
Smoby
Scooby Specialist
 
Member#: 97642
Join Date: Oct 2005
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2002 Wrx Sedan
MBP

Default

I am a little iffy about tuning with the stock knock sensor. I was taught that changing parts on your motor will change the noise frequency the sensor is tuned for and give false readings.
Smoby is offline   Reply With Quote
Old 01-04-2007, 01:05 PM   #16
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by Smoby View Post
I am a little iffy about tuning with the stock knock sensor. I was taught that changing parts on your motor will change the noise frequency the sensor is tuned for and give false readings.
well considering that no tuners i know replace the oem knock sensor (and nobody even suggests that it's a good idea to do so), i'd say you don't have too much to worry about.
ride5000 is online now   Reply With Quote
Old 01-04-2007, 02:10 PM   #17
Smoby
Scooby Specialist
 
Member#: 97642
Join Date: Oct 2005
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2002 Wrx Sedan
MBP

Default

I am not talking about replacing, just changing what frequency is actual knock.
Smoby is offline   Reply With Quote
Old 01-04-2007, 02:13 PM   #18
Smoby
Scooby Specialist
 
Member#: 97642
Join Date: Oct 2005
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2002 Wrx Sedan
MBP

Default

This is why you need a dyno to get the most of timing. There is really no way to tell what MBT is. You have to get your motor to knock before you know it is too much. That is what is so iffy with me, I dont like the idea of purposly making your motor knock.
Smoby is offline   Reply With Quote
Old 01-04-2007, 03:12 PM   #19
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by Smoby View Post
This is why you need a dyno to get the most of timing. There is really no way to tell what MBT is. You have to get your motor to knock before you know it is too much. That is what is so iffy with me, I dont like the idea of purposly making your motor knock.
well, you'd better get used to the idea, because that's the way the oem ecu is running your car.
ride5000 is online now   Reply With Quote
Old 01-04-2007, 03:13 PM   #20
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by Smoby View Post
I am not talking about replacing, just changing what frequency is actual knock.
by far the largest determinator of knock freqency is cylinder bore diameter. if you don't change that, you don't change the frequency.
ride5000 is online now   Reply With Quote
Old 01-04-2007, 07:31 PM   #21
TMessick
Scooby Newbie
 
Member#: 882
Join Date: Feb 2000
Location: Michigan
Vehicle:
2004 WRX Wagon
very blue

Default

The OEM knock systems I've worked with have all been adaptive. It's true that various engine mods can change the background (i.e. non-knocking) noise signature of the engine, but a "good" knock control system will adapt and still function. Even if it didn't, assuming th proper signal frequency was chosen for the knock sensor, you'll have a really strong signal to noise ratio for any significant knock anyway and it'll be picked up. Howeverm there are some condiitons (like at very high revs) where the knock sensor saturates on background noise and you can't detect the added noise from knock. Mostly, I'd be worries about significant changes that effect the mechanical noise. I'm thinking like a gear timing drive for a V8 or vastly different cams or high clearance pistons or something. I don't think the "normal" range of mods most people make will shift the noise signature enough to matter.

Also, although the cylinder bore diameter DOES determine the actual frequency of the knock, it doesn't take into account signal to noise ratio. The knock may be happening at 4kHz, but there may be some other background noise there that makes your signal to noise crap, so you have a clearer signal at a resonance frequency -- say 16kHz -- where the backround noise doesn't exist.
TMessick is offline   Reply With Quote
Old 01-05-2007, 01:31 AM   #22
Gethin
Scooby Specialist
 
Member#: 2998
Join Date: Nov 2000
Chapter/Region: VIC
Location: BC
Vehicle:
2002 WRX
WRB

Default

I read somewhere det freq = 1,800,000 / (Pi * bore_mm)

I used software hooked up to my stock knock sensor via a DIY detcan setup and the formula seems to be accurate. http://www.spectraplus.com/

I use a flattened out KC map as well and it's peachy.
Gethin is offline   Reply With Quote
Old 01-05-2007, 08:17 AM   #23
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

Quote:
Originally Posted by TMessick View Post
Also, although the cylinder bore diameter DOES determine the actual frequency of the knock, it doesn't take into account signal to noise ratio. The knock may be happening at 4kHz, but there may be some other background noise there that makes your signal to noise crap, so you have a clearer signal at a resonance frequency -- say 16kHz -- where the backround noise doesn't exist.
to go along with this, txs has stated that the utec has a high pass filter applied to the knock sensor circuit. the tuner pro also has a high pass filter applied to its audio/headphone output. they've determined the HPF provides better signal to noise ratio.

while the simple HPF is probably not as "tuned" to the engine frequency-wise as the oem ecu's filter, it has the tremendous benefit of being able to be adjusted sensitivity-wise, by rpm range.

the first thing i did once i had detcans was to determine the noise floor of the engine, and set the utec's sensitivity accordingly. now, having been custom tuned for my particular engine, even though it is technically more crude than a more complex bandpass filter, it nevertheless picks up knock that the oem ecu does not. (as far as i'm concerned, any engine noise that decreases with timing retard is knock.)

perhaps some day the open ecu gurus will find the addresses that determine additional parameters of the oem ecu knock detection--for now they only provide for rpm-based range of knock "listening."
ride5000 is online now   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
why is the stock tire iron such a dump? svek Tire & Wheel 12 05-09-2005 01:30 PM
Why is the Flash Gordon DVD so expensive? Lightsped Off-Topic 8 10-27-2003 05:11 PM
Why does the stock system suck so much???? Hassassin Tri-State Area Forum 24 04-25-2003 01:13 PM
Why is the WRX's Slalom Speed so slow? sajohnson Brakes, Steering & Suspension 39 04-24-2002 08:42 AM


All times are GMT -4. The time now is 09:43 AM.


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.