Data Logs! Post them here.

[Vin]

Originally Posted by steve99

looks fine.


if you were rearly kean you could lean out the section over 7100 rpm where it still runs a little rich. but its all fine

Thanks, I'm wondering if I really should do it, don't think the current 'little rich' will hurt that much, also struggling with how the maf scale should look like that, compared to stock Wayno 0% I've changed:

3.711V -1%
3.906V -1%
4.062V -2%
4.297V -2%
4.492V -2%
4.727V -2%
5.000V -2%

The 7100 - redline area is 4.05V - 4.13V, so I could change it to:

3.711V -1%
3.906V -1%
4.062V -4%
4.297V -4%
4.492V -4%
4.727V -4%
5.000V -4%

But won't that give a strange looking maf scale? Also any idea why my LTFT WOT is not 0% ? Is that purely because the 7100 - redline area runs a bit rich?

[Tor]

Originally Posted by Vin

But won't that give a strange looking maf scale? Also any idea why my LTFT WOT is not 0% ? Is that purely because the 7100 - redline area runs a bit rich?

LTFT is set in closed loop. The reason it's not zero is because the MAF scale is "generic" and not scaled specifically for your engine and fuel combination. It doesn't matter much at all though as long as LTFT remains fairly constant.

1) When the car is in closed loop, fueling is continuously adjusted with feedback from the O2 sensor. So whatever trim is applied doesn't matter.

2) In open loop, only one thing matters and that is the actual AFR. That is mainly a combination of your LTFT (set in closed loop) and the MAF scaling above 3.2 volts. Since you can't adjust LTFT directly you are left to adjusting the MAF to give you a sensible AFR.

This is the easy way to go about it as opposed to do a full MAF scaling. The downside is you have to wait for the ECU to learn the trims after flashing the car and between each adjustments.

Doing a full MAF adjustment, you would adjust the lower part of the MAF scale so it will set lower LTFT. It's a hassle and probably won't be worth the trouble and has more potential to mess things up.

LTFT vary anyway and are supposed to do so within reason. So do some logs to check that it remains in the same ballpark with intervals. If you have wild swings (and thus wildly swinging AFR), the might be something wrong with the lower part of the MAF which would require adjustments.

I personally wouldn't be too bothered with is going rich above 7k. Realistically how much time does your engine spend at those rpm? And if anything it might add some protection against knock. It might cost a bit of power but at 7k rpm you are on the way down on the power curve anyway and supposed to shift gear.

Disclaimer:
At least that is how I understand things and am always prepared to be corrected and learn from the more knowledgeable.

[Vin]

Originally Posted by Tor

LTFT is set in closed loop. The reason it's not zero is because the MAF scale is "generic" and not scaled specifically for your engine and fuel combination. It doesn't matter much at all though as long as LTFT remains fairly constant.

1) When the car is in closed loop, fueling is continuously adjusted with feedback from the O2 sensor. So whatever trim is applied doesn't matter.

2) In open loop, only one thing matters and that is the actual AFR. That is mainly a combination of your LTFT (set in closed loop) and the MAF scaling above 3.2 volts. Since you can't adjust LTFT directly you are left to adjusting the MAF to give you a sensible AFR.

This is the easy way to go about it as opposed to do a full MAF scaling. The downside is you have to wait for the ECU to learn the trims after flashing the car and between each adjustments.

Doing a full MAF adjustment, you would adjust the lower part of the MAF scale so it will set lower LTFT. It's a hassle and probably won't be worth the trouble and has more potential to mess things up.

LTFT vary anyway and are supposed to do so within reason. So do some logs to check that it remains in the same ballpark with intervals. If you have wild swings (and thus wildly swinging AFR), the might be something wrong with the lower part of the MAF which would require adjustments.

I personally wouldn't be too bothered with is going rich above 7k. Realistically how much time does your engine spend at those rpm? And if anything it might add some protection against knock. It might cost a bit of power but at 7k rpm you are on the way down on the power curve anyway and supposed to shift gear.

Disclaimer:
At least that is how I understand things and am always prepared to be corrected and learn from the more knowledgeable.

Thank you very much for your input, also recognize the nice shape of my AFR 100% same as yours around the 3 - 4.5k area (before you started correcting it). I won't go leaner above 7k, like you said it's beyond max power anyway and it's indeed a small safety measure

[Rezi]

@Tor
I'm running nearly the same MAF-Scale like Vin (Above 3.8v he is running about 2% leaner, under 3.2v it's the same.)
Mine doesn't go rich above 7000rpm, it holds the AFR pretty constant. LTFT is at 0%. Above 700km since flash. It is also reading 0.2 higher because of the OFT (AFR & Commanded AFR). Last time we where at around 11.5AFR which i didnt really liked, the engine likes leaner better. Also it had more FLKC when it run rich than "lean".

http://datazap.me/u/rezi/stg2-lean-v...4&zoom=451-571

The only difference between Vin and me is, he runs the UEL avcs, myself the EL avcs (stage 2, modified oem header).

[denso]

Hi guys. Please have a look at my latest WOT log.
Custom turbo kit with stock O2 scale (changed now) and about 7psi. 102ROZ fuel.

IAM is placed at 0,8 at my ROM.

http://datazap.me/u/denso/06042017?log=0&data=4

[steve99]

Originally Posted by denso

Hi guys. Please have a look at my latest WOT log.
Custom turbo kit with stock O2 scale (changed now) and about 7psi. 102ROZ fuel.

IAM is placed at 0,8 at my ROM.

http://datazap.me/u/denso/06042017?log=0&data=4

I cannot access log, did you tick the public box ?

Also with the iam did you set iam initial in tune to. O.8 or is it set to 1 and it dropped to 0.8 during driving ?

[denso]

Sorry Steve, its public now.

IAM is set at 0.8. I can send you my ROM if you want.

[phrosty]

Originally Posted by Tor

at 7k rpm you are on the way down on the power curve anyway and supposed to shift gear.

Originally Posted by Vin

like you said it's beyond max power anyway

I might be misunderstanding, but in case I'm not -- you should not shift once you're out of the power band: our gearing is setup so you will always make max wheel torque by shifting as close to redline as you dare.

[steve99]

Originally Posted by denso

Sorry Steve, its public now.

IAM is set at 0.8. I can send you my ROM if you want.

I cannot see true measured afr as looks like your using stock o2 scale, maybe use the extended scaling so it will read dowb to 10 afr.

Looks like you disavbled ltft ?

Your ricght on the limit of the stock maf sensor 5v, maybe use the wrx\forester turbo maf sensor it reads 30% higher flow and is plug and play comatable, jyst need rescale to suit your intake

The provlem with setting iam initial to 0.8 is that each time car is turned off\on iam goes back to 0.8. so if your on dyno with short runs reflash etc the iam is always likely at 0.8. Then when you get on road if the ecu doesnt se much knock it will start to drive iam higher and you will get knock as iam increases.

You may be better setting iam initial to 1 then tune for minium oe no knock. Then ecu will only reduce timing on knock, it wont increase it into the knock area

This is what pretty will every subby tuner does

[Tor]

Originally Posted by phrosty

I might be misunderstanding, but in case I'm not -- you should not shift once you're out of the power band: our gearing is setup so you will always make max wheel torque by shifting as close to redline as you dare.

No, I didn't mean we should shift at peak torque. What I tried to say is that above 7k power is dropping off a lot anyway, so I doubt is noticeable if AFR is e.g. 11.3 or 11.7. AND somewhere in that in that area, you doing your shifting anyway.

[Vin]

So I went from pretty good logs to logs with lots of flkc during cruise in 6th and 5th without going wot. It actually makes iam drop sometimes, anybody can help me figur this out?

It's all when going in 6th then just applying some gentle throttle around 135 kmh:

http://datazap.me/u/vvanrooij/1225-f...og=0&data=1-11

http://datazap.me/u/vvanrooij/1225-2...c?log=0&data=1

Here it actually dropped IAM. I was logging for misfires at that time so that's why the logging parameters changed:

http://datazap.me/u/vvanrooij/1225-f...g=0&data=1-7-8

And my 'perfect' log from two weeks ago

http://datazap.me/u/vvanrooij/1225-1...-2-after-200km

All on wayno 122.5 Uel 100, all on same fuel, I did refill the tank today but that didnt change things (same BP98, different pump)

Thanks in advance

[Tor]

Originally Posted by Vin

So I went from pretty good logs to logs with lots of flkc during cruise in 6th and 5th without going wot. It actually makes iam drop sometimes, anybody can help me figur this out?

It's all when going in 6th then just applying some gentle throttle around 135 kmh:

http://datazap.me/u/vvanrooij/1225-f...og=0&data=1-11

http://datazap.me/u/vvanrooij/1225-2...c?log=0&data=1

Here it actually dropped IAM. I was logging for misfires at that time so that's why the logging parameters changed:

http://datazap.me/u/vvanrooij/1225-f...g=0&data=1-7-8

And my 'perfect' log from two weeks ago

http://datazap.me/u/vvanrooij/1225-1...-2-after-200km

All on wayno 122.5 Uel 100, all on same fuel, I did refill the tank today but that didnt change things (same BP98, different pump)

Thanks in advance

First of all. This is a good quote (about airplane engines):
http://www.contactmagazine.com/Issue...ineBasics.html

One thing to understand is that detonation is not necessarily destructive. Many engines run under light levels of detonation, even moderate levels. Some engines can sustain very long periods of heavy detonation without incurring any damage. If you've driven a car that has a lot of spark advance on the freeway, you'll hear it pinging. It can run that way for thousands and thousands of miles. Detonation is not necessarily destructive. It's not an optimum situation but it is not a guaranteed instant failure.

Our engines have a very high compression ratio of 12.5:1. So it's most likely designed to be running on the edge of knocking or lightly knocking all the time.

Fact is that on a stock tune and normal gasoline, the anti knock mechanisms of the ECU is active a lot. What we see registred in the logs is not knock, but the ECU making corrections to timing.

There is always more FLKC in higher gears. Wheter IAM drops or just FLKC is being applied is set in the programming of the ECU. Some outdated but still informative information is available here:
http://www.romraider.com/forum/topic1840.html

IAM is more likely to drop if having highish FLKC at lower rpm. On the other hand you can have huge FLKC at high rpm with IAM staying at 1. That must be part of the way Subaru have chosen to program the logic.

What I want to say with the above is that I think there is too much focus on being "knock free" on this forum.

What you see in the logs is the ECU doing its work and protecting your engine. Basically you could run it like that without any problems all day long.

It comes down to optimisation. If you can remove a slight amount of timing, and that way prevent the ECU from applying a bigger correction you end up with a net gain in performance.

That IAM drops is not ideal, because it removes everywhere even where it's not needed (IAM = rough correction as described in the previous link). But it's not the end of the world and doesn't mean your engine is about to explode. It just means that a certain logic in the ECU was triggered.

If it happens consistently it would be a good idea to remove some timing at the areas where the drop happens to increase performance. I wouldn't change anything, though, based on it happening once. Maybe it was a one time occurance and it wont happen again or will only happen rarely. Making changes based on that would mean you lose out power in all the circumstances where it doesn't happen.

Same goes for FLKC. FLKC is a temporary table that is stored on the ECU. If you have the same value in the same area all the time, it probably means the ECU is just applying the same correction stored in the table, not that your engine knocks all the time. If you see e.g. -1 is the logs consistenly at the same points in the log and removing 0.5 deg yourself would prevent the ECU from removing 1, you would end up with a net gain of +0.5.

But it would have to be consistent all the time. A log without random FLKC here and there is probably not "perfect" since, in reality, it means you could be running more timing.

With regards to your log, (said in with Crocodile Dundee voice): That's not FLKC - this is FLKC:

http://datazap.me/u/tor/tor-2141-tra...ark=12271-3913

Note that the ECU drops the IAM in an area where a measly -0.95 FLKC is registered at approx. 4800 rpm. But on the other hands happily runs at IAM 1 with -4.47 FLKC at 7100 rpm later on (the two markings).

The above is a track log. If I was to remove timing based on that I would detune the engine tremendously when driven on the street. I'd rather let the ECU do its work when on track, or as a compromise remove only a slight amount of timing so the ECU removes less on track but the majority is still available when driven on the street. (Also the A/C was on by accident for this particular log if that could have played a part, in the other logs it looked a lot better, but that's a different story:
http://datazap.me/u/tor/tor-2141-tra...oom=3684-12836 )

The same would be the case if making a log of driving aggressively on the autobahn. You would probably have a ton of FLKC. Which I wouldn't bother to correct because it would remove performance in lower gears.

Just let the ECU do its work in those situations and keep an eye on it and see what happens over a longer period and if there are any rpm/load areas where the ECU removes timing consistently.

[Vin]

Thanks Tor, for this elaborate answer. On one hand I agree, but in the end the goal should be to have no repeated FLKC that results in a IAM drop.

I actually found my engine cover (the plastic one on top of the intake manifold) to be very loose, the front clip was broken and the whole thing could move a lot, maybe this triggered the knock sensor. I reflashed the rom (to clear the FLKC table) and logged again after 60 km, unlike the previous three days I know have only 0.6x flkc in the same area. I took two logs during two 20 min drives. For me this is fine.

I actually have a dyno planned in two weeks, will report on the result of the 100 STG2 tune and Gruppe-S header on EU BP 98 fuel.

[trvth]

I'm pretty sure my E85 setup is working fine, but I'd be grateful if you guys could give this log a quick look:

http://datazap.me/u/trvth/e85p23b?lo...=1-2-5-9-12-14

It's a log of E85 pulls in 2nd and 3rd after some cruising around

Also -- I'm curious -- is there anything in the logs that one could use to estimate the percentage of ethanol? My supplier tells me they're currently getting E75 delivered, with E85 once the weather is reliably warmer. I expect more power with true E85 ... can I see it in the logs?