Data Logs! Post them here.

[Bergen23]

Originally Posted by Koa

They just got to the D00C tunes, and perhaps one newer than that.. hoping its soon.. for now try and stay out of areas where that crazy fine learning knock is tipping in

Interesting. I'm on a D00D tune.

[Bergen23]

Originally Posted by steve99

did you update template file manually ?

Yes I did

[steve99]

Originally Posted by Bergen23

Yes I did

ok definitly looks like oft software logging issue chase up the openflash guys

[JB86'd]

Originally Posted by steve99

@JB86'd will be Open loop maf scaling off, 11.25 is rich limit of o2 sensor scale so it may even be richer. do you have stock intake ??

however if you correct maf scaling and hence AFR, check for knock especially if your on 91 octane, you may have to pull some timing as the rich mix will buy you knock resistance.

Both @Phantobe and I are on socal 91 oct. I have stock intake with an AFE dryflow filter. It's an OTS tune so I don't expect perfection, but if I'm running richer than 11.25 that's super rich afaik. I already had to pull some timing due to FLKC in high rpm/high load. This gas is killing me

[Phantobe]

Our 91 sure does suck lol.

On an unrelated note... @steve99

I've been running the same custom tune & running 91 Chevron, where I pulled a little bit of timing to account for knock, here are some WOT pulls.

http://datazap.me/u/phantobe/91chevr...og=0&data=1-14

http://datazap.me/u/phantobe/91-chev...og=0&data=1-14

Pretty much no timing pulled in those first two logs everything looks good.

Then more recently yesterday I decided to do a pull & I'm getting heeps of knock o.o

http://datazap.me/u/phantobe/91chevr...e?log=0&data=1

I know my IAM is alittle below 1.0, but its not going down it keeps going up this morning it was at .99. Is it possible some of these knock events could be fake? It doesn't seem right that I go from no knock conditions, then to where the ECU is pulling 3 degrees of timing while my IAM is traveling back towards 1.0.

[Kodename47]

Originally Posted by Phantobe

Our 91 sure does suck lol.

On an unrelated note... @steve99

I've been running the same custom tune & running 91 Chevron, where I pulled a little bit of timing to account for knock, here are some WOT pulls.

http://datazap.me/u/phantobe/91chevr...og=0&data=1-14

http://datazap.me/u/phantobe/91-chev...og=0&data=1-14

Pretty much no timing pulled in those first two logs everything looks good.

Then more recently yesterday I decided to do a pull & I'm getting heeps of knock o.o

http://datazap.me/u/phantobe/91chevr...e?log=0&data=1

I know my IAM is alittle below 1.0, but its not going down it keeps going up this morning it was at .99. Is it possible some of these knock events could be fake? It doesn't seem right that I go from no knock conditions, then to where the ECU is pulling 3 degrees of timing while my IAM is traveling back towards 1.0.

You have to understand that IAM and FLKC are independent but working towards the same end point. IAM automatically starts increasing whenever you are above a set load threshold and there has not been any IAM change for a set delay period. I will also point out that you want to be looking at FLKC at the point where it reduces. On your chart you can see it slowly returning towards 0 which is again a on a delay timer, this is the ECU trying to return timing. It's only if it drops again or the steps seem to stop going back towards 0 that further knock is seen. Your definite issue areas seem to be at 3500 and 4800.

Whether any knock is real or false is another debate and it's completely down to the sensitivity of the sensor and the knock thresholds set per cylinder. I really wish we had the ability to see those. However without det cans then we can't be sure, so you have to assume that the ECU is pulling timing due to genuine knock. The knock control on these engines seem to be a little more advanced than older Subarus, if only that the knock sensitivity is set a little high. Then again, I'd rather it was that way.

As to why it's happening, any big air temperature changes? If it's got much colder or much hotter then that could be a factor. Also your LTFT are negative meaning that fuel is being pulled. Compare the AFRs to before, if it's a little leaner then this may be the cause especially if you're on 91. Also check for changes in load, you may be interpolating in different cells or just different values.

[Celadrielas]

http://datazap.me/u/celadrielas/a01c-cel-1134

This is the current tune I am running. Notes: Fixed exhaust leak, still have cut snorkle ( ), AEM intake - with extra hardware chip plugged in still, stock exhaust, 91 OCT - Quick Trip.

http://datazap.me/u/celadrielas/cel-a01c-1135-0 -- Still having issues with this tune (like I do). All same stats as above in car set up, IAM drop to .88 -- Triggered CEL once but I've done several pulls and they all read the same. no CEL since.

with v.1.13.5, I wanted to put back some of the timing which was taken out to see if I could keep it from knocking and have that little extra too.

[Phantobe]

Originally Posted by Kodename47

You have to understand that IAM and FLKC are independent but working towards the same end point. IAM automatically starts increasing whenever you are above a set load threshold and there has not been any IAM change for a set delay period. I will also point out that you want to be looking at FLKC at the point where it reduces. On your chart you can see it slowly returning towards 0 which is again a on a delay timer, this is the ECU trying to return timing. It's only if it drops again or the steps seem to stop going back towards 0 that further knock is seen. Your definite issue areas seem to be at 3500 and 4800.

Whether any knock is real or false is another debate and it's completely down to the sensitivity of the sensor and the knock thresholds set per cylinder. I really wish we had the ability to see those. However without det cans then we can't be sure, so you have to assume that the ECU is pulling timing due to genuine knock. The knock control on these engines seem to be a little more advanced than older Subarus, if only that the knock sensitivity is set a little high. Then again, I'd rather it was that way.

As to why it's happening, any big air temperature changes? If it's got much colder or much hotter then that could be a factor. Also your LTFT are negative meaning that fuel is being pulled. Compare the AFRs to before, if it's a little leaner then this may be the cause especially if you're on 91. Also check for changes in load, you may be interpolating in different cells or just different values.

Thanks for the info!

As to why its happening, thats why I'm kinda boggled. The temperature generally is in between 60-80 pretty frequently, no sudden drops in climate. I never seen my AFR running lean on a WOT pull its always pretty rich. Guess I'll monitor it to see if the knock reduces by itself or I'll just pull some timing in the tables.

[Bergen23]

Tried another log today after updating everything for the 3rd time. Still not logging correctly

http://www.datazap.me/u/colekmiecik/...og=0&data=8-11

EDIT: the guys at OpenFlash got back to me. They said beyond updating OFM (which I have the most recent one available) and updating the template (which I've also already done) they have no other suggestions on how to fix the issue.

[steve99]

Originally Posted by Celadrielas

http://datazap.me/u/celadrielas/a01c-cel-1134

This is the current tune I am running. Notes: Fixed exhaust leak, still have cut snorkle ( ), AEM intake - with extra hardware chip plugged in still, stock exhaust, 91 OCT - Quick Trip.

http://datazap.me/u/celadrielas/cel-a01c-1135-0 -- Still having issues with this tune (like I do). All same stats as above in car set up, IAM drop to .88 -- Triggered CEL once but I've done several pulls and they all read the same. no CEL since.

with v.1.13.5, I wanted to put back some of the timing which was taken out to see if I could keep it from knocking and have that little extra too.

Mate i know you not going to like this ,but i think its all due to the AEM intake and the hardware chip thingo that probably adjusts maf readings. Just been doing some scaling on those intakes and they are a pain , suggest you go back yo stock intake with a drop in filter and sell the aem to someone.
Or buy grimspeed intake they are very close to stock scaling and you wont have these dramas

[steve99]

Originally Posted by Bergen23

Tried another log today after updating everything for the 3rd time. Still not logging correctly

http://www.datazap.me/u/colekmiecik/...og=0&data=8-11

EDIT: the guys at OpenFlash got back to me. They said beyond updating OFM (which I have the most recent one available) and updating the template (which I've also already done) they have no other suggestions on how to fix the issue.

just a thought are you on the latest ofm as per the downlads site ?

if so the just log KC LEARNED instead of say short term fuel trim and we can derive the IAM and FLKC from that Will also keed to know what tune your on and if you could post up you KNOCK CORRECTION MAX A table as kc learned should match knock correction max a table if iam is 1 and no flkc in your log.

[sachu]

Originally Posted by steve99

If the Knock correction max A table values are zero then IAM will not be able to pull any timing in event of crappy fuel or some other issue causing knock.
you would only have flkc retard available about 5 degrees.

You cannot "leave the IAM at 1" the ecu software calculates when to reduce IAM, you cannot stop this unless your going to re-code the ecu software.

IAM can only retard timing if their are values in the knock correction max a table

I believe that the values are relivant to the rpm/load areas where knock is more prone on the engine ie the values are higher in arround 3000 as opposed to 7000 and probably a percentage of the overall timing, likely some engineer spent much time working these out so when the iam drops to say 0.5 the ecu pulls timing differently in different rpm/load ranges.



Engineers at subaru put that system in place and I would say they know far more about ecu's and knock than most especially on their engines.

The internal working of the ecu software are extremely complex and I don't claim to know anything more than the very basics

Yes you could do it but your effectively defeating one of the ECU's main safety features in relation to knock correction.

Some tuners smooth the values in the Knock correction max A table or the Base timing tables but have not seen any that remove knock correction values with these cars.

Apparently Group N Rally WRX STI ECU did remove knock control but not something you would want on a street car.
http://www.romraider.com/forum/topic360.html

What advantage do you see ?

maybe you should read up here
http://www.romraider.com/RomRaider/H...ndKnockControl

http://www.romraider.com/forum/viewt...hp?f=25&t=1840

Hi Steve,

Seriously appreciate your responses man. so damn informative.

I wasn't clear in my question i suppose. I meant resetting IAm to 1.0 from the default 0.7 Not hard coding it to 1.0 so it never corrects.

Suppose the question was reset it to 1, but leave all advance timing table as zeroes and go off of the main timing table as trusted timing and have just fine knock correction as the main fallback. Is that feasible?

[steve99]

Originally Posted by sachu

Hi Steve,

Seriously appreciate your responses man. so damn informative.

I wasn't clear in my question i suppose. I meant resetting IAm to 1.0 from the default 0.7 Not hard coding it to 1.0 so it never corrects.

Suppose the question was reset it to 1, but leave all advance timing table as zeroes and go off of the main timing table as trusted timing and have just fine knock correction as the main fallback. Is that feasible?

Most tuners set iam to 1 so they can do dyno pull straight away and not have to run or drive car waiting for ecu to set iam to one.

otherwise if you set it initial to 0.7 the timing keeps changing as iam climbs to 1 assuming theire is minimal knock, most tuners tune for minimal knock on the fuel your going to use.

[Phantobe]

Damn this tuning stuff is starting to become frustrating lol.

91 Chevron - Tuned with some timing pulled
Everything looks great for like 2 weeks IAM hardly drops below 1.0. Now all the sudden another week later im getting 3 degrees of timing pulled & IAM drops to .49!

@JB86'd since we both are running the same shitty petrol, have you experienced anything like this? I feel like I can't win lol.