Data Logs! Post them here.

[solidONE]

Originally Posted by Tor

More?

http://datazap.me/u/tor/tor-2322-0?l...4703-4281-4286

Looks good! Between 3000~4000 rpm and 4500~5800 rpm could probably take a bit more under the same conditions. Though any warmer you'll likely encounter more knock. I'd probably add half a degree here and there do some pulls convert them to v-dyno graphs then post said graphs on facepage boasting about the gains I just made.

[Tor]

Originally Posted by solidONE

http://datazap.me/u/tor/tor-2322-0?l...4703-4281-4286

Looks good! Between 3000~4000 rpm and 4500~5800 rpm could probably take a bit more under the same conditions. Though any warmer you'll likely encounter more knock.

Here is my interpretation (disregarding the FLKC below 4000 where I didn't add timing):

http://datazap.me/u/tor/2322-formula...1285-1758-2016

Pull 3: -0.65 at 6600
Pull 4: -0.65 at 6000 (clears the knock at 6600 from previous pull)
Pull 5: No knock at (partly clears the previous -0.65 at 6000 with -0.3).
Pull 6: Clears all previous knock, -0.65 at 7100 (no additional timing added there since the previous flash)
Pull 7: -0.65 at 6800.

So there is random non-repeated knock at 6000, 6600, 6800, and 7100. So you are probably right that I hit a sensible limit there already and could add a bit more at 5200 and 5600. I think I will stop here though. I suppose it's probably very diminishing returns after this. 2 deg more timing in the 4400 to 6600 band is a noticeable difference already though.

Regarding higher IAT, I already accounted for that before adding timing. I raised the Timing IAT Activation to 100% (-100% depending on definitions) at 6000 and 7000 rpm above load 1 (and will now do so for 4000 and 5000 too). I also changed Timing compensation IAT to start to reduce timing at 25 deg C (77 F) with -0.7 and at 30 C (86 F) with -1.05. It's getting colder here now, but next time I see those temps I would rather increase these compensations if required than remove timing. My reasoning is to have as much as possible available at lower temps and let these tables take care of knock at higher temps.

My own take on what made this increase possible is that I now have an AFR curve that follows AFR command without large deviations. My guess is that the more stable one can keep the conditions in the combustion chamber, the less likely it will knock.

I'd probably add half a degree here and there do some pulls convert them to v-dyno graphs then post said graphs on facepage boasting about the gains I just made.

I know these posts come across a bit like bragging. And to be honest I am actually also a bit proud of the results I have gotten with this tune so far. I started out with someone else's OTS tunes and have 36 tables changed with significant changes like:

"86inches" mathematically smoothed MAF scaling, Engine Load Compensation major adjustments (my favorite change! ), MAF IAT compensation that keeps AFR and LTFT constant with changing temps, previous required Load Limits removed/changed (no longer required with corrected ELC), POL remodeled inspired from the stock table which enables more timing up top, Timing reshaped and increased. And a lot of other changes, some of them taken from both the US and EU MY17 tunes.

So I am not afraid to call this tune my own now and, yes, to brag a bit and share my enjoyment. In addition to making these changes to my own car, I helped 2 other European guys with the Gruppe-S header to achieve similar results. We have a small facebook chat going and they have been helpful not only with trying a lot of my changes out on their cars too but also for exchanging thoughts. Their tunes also each have a custom 86inches MAF and ELC table specifically for their cars. I was surprised about the variation between engine/sensors between cars with otherwise almost identical mods, speaking in favor of getting these adjustments done.

I'm actually thinking about putting my stock header back on and go through the motions once again with the stock setup to see what gains can be made with adjusting MAF, ELC and load limit there too. I think the stock ELC is wrong when AVCS are changed and that is why some have problems with LTFT. High/wrong/divergent fuel errors at different manifold pressures, so one day you hit one cell more often and get a certain LTFT, the next day you drive differently (e.g. aggressively/sedate) and hit another cell more with a completely different fuel error and get a completely different LTFT.

Another option I would consider since I am starting to feel confident enough about this would be to take on remote tuning someone else's car. So if anyone has an engine with stock intake/exhaust and is interested in a really customized tune and is willing to do the required logging/flashing, feel free to send me a PM. This is hours of work for me which I would do for free for the fun and experience I would gain from doing it. As I am not a pro there will probably be steps forwards and backward (but all would be done in a safe manner) so anyone wanting to do this needs patience and respect that I have a job that occasionally demands pretty much of my time too.

If it proves to offer gains on a stock car, going forward I could imagine myself providing this service to anyone interested for a modest fee for my time. No one seems to be offering a truly customized service with open source tunes, which I think is a shame as not everyone has the opportunity to invest the required time/energy to get things to run as close to perfect as possible. And in the world we live in it doesn't make sense that someone without a real interest in the subject needs to learn all this stuff and waste valuable time they could spend on their real interests if they could pay someone a reasonable fee to do it. On the other hand, it doesn't make sense either, that those who do take the time to learn all this stuff tune one car (their own) only to forget most of the stuff again as time passes.

The problem as I see it is, that there is really no one who offers such a service because, I suppose, the time required to do this right doesn't match up with the cost of the time of a pro tuner. Send the customer a base tune, do some minor adjustments and done. On to the next task...

Given the individuality of each car, I also don't understand the reasoning of pro-Ecutek tuners and their locking down of the tunes. A truly customised tune can't just be copied over to another car and expected to run equally well. It could easily lead to the speculation that pro-tuners actually just copy over/sends a base tune and do minor "hacks" to get it to run well on the customer's car.

Probably there are exceptions to pro-tuners, but this was my big beef with getting my car pro-tuned in the first place - I would never let anyone do changes to my ECU were I can't see what was done.

Here an example a MAF scaling made on a dyno by a guy who tunes cars professionally for money! Personally I wouldn't let him touch my car with a 10 foot pole:

Sorry for the long post that got somewhat off track.



P.s.
It's very easy to explain why the above MAF scaling is an incredibly bad idea:
The engine only needs that bump in the scale at manifold pressures above 0.9 bar (WOT). Below 0.9 bar, closed loop territory, the above bump cause huge fuel errors and the ECU to set very high LTFT. Hence the required bump should have been compensated using engine load compensation and the MAF scaling keept smooth.

P.p.s.
I don't mention any tuners specifically but if this is exceeding any rules of the forum let me know and I'll edit my post.

[solidONE]

Originally Posted by Tor

I know these posts come across a bit like bragging. And to be honest I am actually also a bit proud of the results I have gotten with this tune so far.

Bro.. that comment was tongue-in-cheek, half joke. I'd seriously do this, and have done this.

Also, the FBKC you are seeing below 4000rpm looks like tip in knock.

http://datazap.me/u/tor/tor-2322-0?l...4848-4199-4012

[Tor]

Originally Posted by solidONE

Bro.. that comment was tongue-in-cheek, half joke. I'd seriously do this, and have done this.

Also, the FBKC you are seeing below 4000rpm looks like tip in knock.

http://datazap.me/u/tor/tor-2322-0?l...4848-4199-4012

Yes yes, I know it was a joke... But I basically did the same just here and not on Facebook a few posts back.

I noticed the FBKC as well. I think there may be too much timing at 2000 and 2200 rpm. After correcting load limits I adjusted the timing in those cells and was probably too greedy. I saw it in previous logs too but mostly just -0.7, so I didn't bother to correct it. E.g. here it's a while after tip in:
http://datazap.me/u/tor/tor-2321-pul...5&mark=326-331

I suppose some of it may also tie in with my other issue a few posts back and be caused by CL/OL delay. E.g. here:
http://datazap.me/u/tor/tor-2322-0?l...4409-4848-4849

My plan is to reduce the timers in the next flash in any case (Throttle A counter threshold to 0). But otherwise, it doesn't bother me too much as I think it's not an issue for normal driving (i.e. I never floor the car at less than 2000 rpm except for logging). In the second example where it switches twice between OL-CL-OL, it does bother me since at 3000 rpm it's more likely to occur in normal driving.

If it doesn't help I would like to come back and ask for some help what else I can adjust to fix it.

[solidONE]

Originally Posted by Tor

Yes yes, I know it was a joke... But I basically did the same just here and not on Facebook a few posts back.

I noticed the FBKC as well. I think there may be too much timing at 2000 and 2200 rpm. After correcting load limits I adjusted the timing in those cells and was probably too greedy. I saw it in previous logs too but mostly just -0.7, so I didn't bother to correct it. E.g. here it's a while after tip in:
http://datazap.me/u/tor/tor-2321-pul...5&mark=326-331

I suppose some of it may also tie in with my other issue a few posts back and be caused by CL/OL delay. E.g. here:
http://datazap.me/u/tor/tor-2322-0?l...4409-4848-4849

My plan is to reduce the timers in the next flash in any case (Throttle A counter threshold to 0). But otherwise, it doesn't bother me too much as I think it's not an issue for normal driving (i.e. I never floor the car at less than 2000 rpm except for logging). In the second example where it switches twice between OL-CL-OL, it does bother me since at 3000 rpm it's more likely to occur in normal driving.

If it doesn't help I would like to come back and ask for some help what else I can adjust to fix it.

I trust you've already applied additional transient ignition retard to your tables? You were starting your pulls in that log at or past 3K so it souldnt have anything to do with your settings at 2000-2200. Not a huge deal but I'd address it in the transient tables if you haven't done so already. Especially since your already in there making changes and reflashing.

As far as the transition timers go my current tune has them set at 31. I don't remember if that was changed from factory settings but it's been at 31 for a while. Along with my maf intake temp compensation tables, which appear to be vastly different from what everyone else that have bothered to change this table is using, my LTFT in OL have been not perfect but satisfactory. ( "satisfactory" as in "I gave up messing with it") There is another way to keep LTFT at 0% in OL that Kodename47 has been applying in his boosted application. You can ask him if you're curious what he has done.

Lets see the results with the additional ignition advance!

[Tor]

Originally Posted by solidONE

I trust you've already applied additional transient ignition retard to your tables? You were starting your pulls in that log at or past 3K so it souldnt have anything to do with your settings at 2000-2200. Not a huge deal but I'd address it in the transient tables if you haven't done so already. Especially since your already in there making changes and reflashing.

As far as the transition timers go my current tune has them set at 31. I don't remember if that was changed from factory settings but it's been at 31 for a while. Along with my maf intake temp compensation tables, which appear to be vastly different from what everyone else that have bothered to change this table is using, my LTFT in OL have been not perfect but satisfactory. ( "satisfactory" as in "I gave up messing with it") There is another way to keep LTFT at 0% in OL that Kodename47 has been applying in his boosted application. You can ask him if you're curious what he has done.

Lets see the results with the additional ignition advance!

I copied/modified to fit the Transient Ignition Retard from MY17 tune. In the whole load 0.8 it has -0.078, same as A01G. I'm having a hard time figuring out when these tables are active? Also, there is the "Minimum CFP Delta", it's different for the MY17 cars, anyone knows what's doing?

Changing the CL delay period to 1/3 with 208 (stock settings 625) and Throttle A counter Threshold to 0 (stock 313) didn't seem to do a whole lot. I think I'll change them back since it doesn't seem to have any effect.

I still need to work on MAF IAT comp for low temps. I didn't have the opportunity so far because it's only starting to get colder now. But between 20 and 35 deg IAT it seems to work well. Did you make a thread about your changes? I'd be curious to see your changes.

And as for Virtual Dyno, haha, it's truly disappointing again. Or I was just stupid. With the added timing I get less again, or it's due to the conditions here at the moment (rain/wet), or maybe it's because I forgot to turn off the A/C (on for defogging). I know it's said to kick out under full throttle, but I have yet to see proof of mine actually doing that - I think there are more conditions that have to be met before it kicks out.

But to be honest I think it's exceeding the accuracy at this point. Maybe the last results weren't correct either (too high). Anyway, on damp streets today with A/C on:



http://datazap.me/u/tor/tor-233-250-...ata=22&solo=21

If there is any truth to this, the red line (current tune), seems flatter from 4800 to 6400 rpm. This is where I added the most timing. All 3 pulls are done on the exact same stretch of road.

Butt dyno says it's moving well and on a damp street, it sure wheel spin in second and sometimes in 3rd too (Yokohama AD08R with a lot of heat cycles).

Wheelspin with more than 100 km/h in "the slow 86" (traction control on):



2 days ago I got a ride in a WRX STI on the Nürburgring GP circuit in absolutely pouring rain soaking wet track with semi slicks. The grip of that thing is ridiculous! It was actually only barely noticeable that it was raining. It actually made me think about selling the 86 and join the dark AWD side because damn it was fun to overtake every single car on the track in a 15 min stint.

[Nobunnyhere.86]

http://datazap.me/u/nobunnyhere86/ru...a=1-3-11-12-14

http://datazap.me/u/nobunnyhere86/runs?log=1&data=1

http://datazap.me/u/nobunnyhere86/runs?log=2&data=1

This is a "40C bin" 3.0 stage 2+ 91 beta tune from shiv


May someone take a look at these and tell me if they look okay? I am still in the process on learning how to read these properly so at times I have no clue what exactly I am looking for and what numbers would be normal vs what I am showing.. Thanks !

[Wayno]

Do you mean F40C ?

Originally Posted by Nobunnyhere.86

http://datazap.me/u/nobunnyhere86/ru...a=1-3-11-12-14

http://datazap.me/u/nobunnyhere86/runs?log=1&data=1

http://datazap.me/u/nobunnyhere86/runs?log=2&data=1

This is a "40C bin" 3.0 stage 2+ 91 beta tune from shiv


May someone take a look at these and tell me if they look okay? I am still in the process on learning how to read these properly so at times I have no clue what exactly I am looking for and what numbers would be normal vs what I am showing.. Thanks !

[Wayno]

Originally Posted by Nobunnyhere.86

I am still in the process on learning how to read these properly

This is what to look at. AFR should be around 11.7 - 11.8

http://datazap.me/u/nobunnyhere86/ru...-2-9-13&solo=2

[Nobunnyhere.86]

Originally Posted by Wayno

This is what to look at. AFR should be around 11.7 - 11.8

http://datazap.me/u/nobunnyhere86/ru...-2-9-13&solo=2

looking at some of my pulls AFR goes up to 20 at times and averages around 14.0 area at normal driving, so I am guessing that's not good right?? would that cause me to pull time as well ?

[Breadman]

Originally Posted by Nobunnyhere.86

looking at some of my pulls AFR goes up to 20 at times and averages around 14.0 area at normal driving, so I am guessing that's not good right?? would that cause me to pull time as well ?

I have the same richness you do with the v3 but I'm a 17. Apparently it's not bad but it's just wasting fuel and losing a bit of power. I might flash back to stock

Sent from my Nexus 6 using Tapatalk

[steve99]

Originally Posted by Nobunnyhere.86

looking at some of my pulls AFR goes up to 20 at times and averages around 14.0 area at normal driving, so I am guessing that's not good right?? would that cause me to pull time as well ?

20 afr is normal when you have zero throttle injectors shut off

wayno is talking about full throttle afr it should be 11.8 or so

see here
http://www.ft86club.com/forums/showthread.php?t=69871
theirs a bunch of pretty colored links in my signature below have a look their as well

You running pretty rich 11.2 or likely richer as thats rich limit of sensor reading. looks like some maf rescaling needed

you have some knock, you need to log IAM ( drop out KC Learned) as i suspect its less than 1 as you has positive FLKC in some logs

[solidONE]

Originally Posted by Tor

I still need to work on MAF IAT comp for low temps. I didn't have the opportunity so far because it's only starting to get colder now. But between 20 and 35 deg IAT it seems to work well. Did you make a thread about your changes? I'd be curious to see your changes.

And as for Virtual Dyno, haha, it's truly disappointing again. Or I was just stupid. With the added timing I get less again, or it's due to the conditions here at the moment (rain/wet), or maybe it's because I forgot to turn off the A/C (on for defogging). I know it's said to kick out under full throttle, but I have yet to see proof of mine actually doing that - I think there are more conditions that have to be met before it kicks out.

But to be honest I think it's exceeding the accuracy at this point. Maybe the last results weren't correct either (too high). Anyway, on damp streets today with A/C on:



http://datazap.me/u/tor/tor-233-250-...ata=22&solo=21

If there is any truth to this, the red line (current tune), seems flatter from 4800 to 6400 rpm. This is where I added the most timing. All 3 pulls are done on the exact same stretch of road.

Butt dyno says it's moving well and on a damp street, it sure wheel spin in second and sometimes in 3rd too (Yokohama AD08R with a lot of heat cycles).

Wheelspin with more than 100 km/h in "the slow 86" (traction control on):



2 days ago I got a ride in a WRX STI on the Nürburgring GP circuit in absolutely pouring rain soaking wet track with semi slicks. The grip of that thing is ridiculous! It was actually only barely noticeable that it was raining. It actually made me think about selling the 86 and join the dark AWD side because damn it was fun to overtake every single car on the track in a 15 min stint.

At the last column for transient retard I have it a bit higher than the stock MY17 (-2.891 vs 0.078), but everything under 0.8 loads I have much less retard than the stock MY17 table. I tried the MY17 table and mine seemed to work better. Less retard under 0.8 loads and more retard at and over 0.8 loads.

I never started a thread or never posted in detail what I did with my MAF temp compensation.. because reasons. I did PM several guys including steve, kodename and wayno my current tables and that was how I found out that some of them are using a vastly different approach. I dialed my maf to 68*f in full Di then full Pi, came up with a injector flow scaling that works for my Maf dialed to 68*f (64~72*f) and dialed it some more. Then I adjusted the compensations according to the established MAF scale and voila! "satisfactory" results! LMAO

PM me your email I'll send you my current tune and you can dissect it for tables you might want to use or try out.

Also I think you should wait till its dry to do pulls to convert to V-dyno. It doesn't make sense that you are loosing power while running more aggressive ignition timing. I'm sure you understand how v-dyno comes up with its numbers from data logs. If its accelerating faster, as it should with more power, its going to produce higher numbers in the application. It's not magic.

[Tor]

Originally Posted by solidONE

At the last column for transient retard I have it a bit higher than the stock MY17 (-2.891 vs 0.078), but everything under 0.8 loads I have much less retard than the stock MY17 table. I tried the MY17 table and mine seemed to work better. Less retard under 0.8 loads and more retard at and over 0.8 loads.

I never started a thread or never posted in detail what I did with my MAF temp compensation.. because reasons. I did PM several guys including steve, kodename and wayno my current tables and that was how I found out that some of them are using a vastly different approach. I dialed my maf to 68*f in full Di then full Pi, came up with a injector flow scaling that works for my Maf dialed to 68*f (64~72*f) and dialed it some more. Then I adjusted the compensations according to the established MAF scale and voila! "satisfactory" results! LMAO

PM me your email I'll send you my current tune and you can dissect it for tables you might want to use or try out.

I gave up on "the perfect result too". Satisfactory is a good goal, or robust so it at least runs the same with varying conditions. Mine started to set higher negative LTFT. I imagine it could be due to the change in CL timers or MAF IAT comp due to it getting colder. But nevertheless, even when it runs a bit leaner, in the colder temperatures it doesn't knock more than before. So I suppose I could label that as satisfactory too, although I'm not entirely happy.

Here are my logs from today:
A few pulls for VD:
http://datazap.me/u/tor/tor-222-330-...zoom=3404-6629
And me beating on it in uphill in 4th:
http://datazap.me/u/tor/tor-222-330-...zoom=7277-7502

I'd love to exchange tunes, I'll send you a PM. Thanks!

Also I think you should wait till its dry to do pulls to convert to V-dyno. It doesn't make sense that you are loosing power while running more aggressive ignition timing. I'm sure you understand how v-dyno comes up with its numbers from data logs. If its accelerating faster, as it should with more power, its going to produce higher numbers in the application. It's not magic.

I didn't expect a damp road to make much difference, but it could be there is a tiny bit of slip, although it should make the curve fluctuating? To be honest I am quite sure my A/C doesn't kick out at full throttle. I have several experiences to support that.
1) 3 tracklogs on the same day. 2 without A/C, one with. The one with A/C on looks absolutely horrible with regards to knock compared to the two others.
2) Another track day at 30 deg C (86 F) ambient. For a 50 mins session I had the A/C on the whole time. Half of it the cabin was nice and cool, the other half the A/C started to blow hot air only. There was no real kicking in and out. Second half the A/C just stopped delivering any cool air. This was my first track day and I didn't have an oil cooler. I never experienced since that the A/C pumps hot air, neither did I see such high oil temps again (135 C / 275 F ) since I installed an oil cooler immediately afterward.

Anyway, as mentioned, I did another log today. Dry road, A/C off. I am back with the previous results. I can't tell much difference to the other logs where I already added the bulk of the timing. I think I am messuring the imperfections of the road at this point rather than the performance changes. Still, compared to the old timing there is a noticeable gain:



The 177 hp from the old timing is in line with most of my previous logs. If I really got 183-185 hp, I think that's probably pretty good for the cheap Gruppe-S UEL header. I would love to take the car on a real dyno at this point. I would probably be disappointed though. European dynos are not as optimistic as many of the US dyno charts that are being posted here from various header manufacturers and pro tuners. My bet is 180 wheel and 215 crank. If I get 220 crank, I'll be very very happy (stock car was messured 159/189). When I see claims of 240 crank hp with a header, tune and E85, I just laugh "US dyno".