Welcome to the North American Subaru Impreza Owners Club Monday October 22, 2018
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Go Back   NASIOC > NASIOC Technical > Engine Management & Tuning

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 12-17-2017, 09:42 PM   #1451
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by Tryptofan View Post
Did you try adjusting/richening your AFR's yet? I would work on that before I touched timing. Fueling is a lot more difficult for the ECU to control precisely than timing - if you're right on the edge of being too lean, you might have one run that looks fine, and another that knocks, just because your AFR drifts .1 or .2 richer or leaner.

You should be changing timing in degrees, not %, right?

In my understanding, each cell on the table phases linearly into the next cell. So, going left to right (ignoring any change in RPM), if you had only 3 load columns set at 1, 1.8, and 2.6 containing values of 2, 4, and 6, anything in between those values is linearly interpolated. So, at a load of 1.4 and constant RPM, the value would be 3, and at load 2.2, the value would be 5.

The value can never be greater than the highest value or less than the lowest value, so if you go higher than a load of 2.6, the value the ECU uses will always be 6. Likewise, going lower than 1 would always be a value of 2

The same holds true for RPM. If you have rows with 2400 RPM, 4400 RPM, and 6400 RPM, and each row has constant values (2, 4, and 6, let's say), it gets linearly interpolated again between rows. value = 3 at 3400 RPM, value = 5 at 5500 RPM, but the value is never higher than 6 once you go over 6400 RPM.
Thx clears things up and is what I thought. Im horrible at math, so I used the floating point across the entire 2600 row and added 3.00, which resulted in numbers I can't remember at this point. I think the max timing reduction, 2400 at 2.60 load, was 12 * -. 35 = 4. 2.

So, at 4.2 being pulled 2400-4000 2.6 load and above I had no improvement. I then took 5* out via the AP and there is an improvement, but not much.

I'm wondering, since this knock is mostly occurring on cylinder 1, if I need to check the plug.

If it was false know I don't think I'd see any improvements, correct?
* Registered users of the site do not see these ads.
PDXREALTOR is offline   Reply With Quote
Sponsored Links
* Registered users of the site do not see these ads.
Old 12-17-2017, 09:47 PM   #1452
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
Thx clears things up and is what I thought. Im horrible at math, so I used the floating point across the entire 2600 row and added 3.00, which resulted in numbers I can't remember at this point. I think the max timing reduction, 2400 at 2.60 load, was 12 * -. 35 = 4. 2.

So, at 4.2 being pulled 2400-4000 2.6 load and above I had no improvement. I then took 5* out via the AP and there is an improvement, but not much.

I'm wondering, since this knock is mostly occurring on cylinder 1, if I need to check the plug.

If it was false know I don't think I'd see any improvements, correct?
If you pulled that much timing and saw no improvement then I'd lean towards phantom knock. 5* should make real knock go away regardless of the slightly lean AFR. But before we move on, try enrichening your open loop fueling by 0.5 Afr ie if the table value was 11 AFR, then make it 10.5. If it still shows no improvement then I'd be confident it's phantom knock.

And yes the ecu will continue to use the values in the last column for any loads that extend past the max value in the header column.
ShinjiML is offline   Reply With Quote
Old 12-17-2017, 09:47 PM   #1453
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by Tryptofan View Post
Did you try adjusting/richening your AFR's yet? I would work on that before I touched timing. Fueling is a lot more difficult for the ECU to control precisely than timing - if you're right on the edge of being too lean, you might have one run that looks fine, and another that knocks, just because your AFR drifts .1 or .2 richer or leaner.
No. I'll be honest, the fueling table scares me.

I don't understand how it works and I've yet to spend time researching it. One thing I've always wondered is why the ol fueling is always changing in the logs if it's a static number from the olpf table. Another thing is, my wb02 and my Pol fuel table don't match up.
PDXREALTOR is offline   Reply With Quote
Old 12-17-2017, 09:54 PM   #1454
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by ShinjiML View Post
If you pulled that much timing and saw no improvement then I'd lean towards phantom knock. 5* should make real knock go away regardless of the slightly lean AFR. But before we move on, try enrichening your open loop fueling by 0.5 Afr ie if the table value was 11 AFR, then make it 10.5. If it still shows no improvement then I'd be confident it's phantom knock.

And yes the ecu will continue to use the values in the last column for any loads that extend past the max value in the header column.
When I look at olf primary table I see a lot of numbers. How do I know which one is the 'table value'?

The thing that makes me think it's real is the improvement I see with massive timing reduction, even thought it's small. And, a couole of large knocks stand out. One - 7 I could feel the timing being pulled. Guess I'd feel it if it were phantom and the ecu didn't know the difference.

The other time, recently, was a - 8 and I swear I heard that one. However I'm on solid mounts both engine and tranny so... There's a lot of noise going on.

I'm not opposed to reducing fuel. I just need to know how to do it and understand to some extent what else is being affected.

I understand basic tuning, it's the tables and how they all flow that is new and confusing to me. But, I catch on quick.
PDXREALTOR is offline   Reply With Quote
Old 12-17-2017, 10:09 PM   #1455
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
When I look at olf primary table I see a lot of numbers. How do I know which one is the 'table value'?

The thing that makes me think it's real is the improvement I see with massive timing reduction, even thought it's small. And, a couole of large knocks stand out. One - 7 I could feel the timing being pulled. Guess I'd feel it if it were phantom and the ecu didn't know the difference.

The other time, recently, was a - 8 and I swear I heard that one. However I'm on solid mounts both engine and tranny so... There's a lot of noise going on.

I'm not opposed to reducing fuel. I just need to know how to do it and understand to some extent what else is being affected.

I understand basic tuning, it's the tables and how they all flow that is new and confusing to me. But, I catch on quick.
If you were previously knocking -3* but after pulling -5* you now see -2*, to me that could very well be phantom knock. There's no reason the ecu pulls -3* and it's OK, but if you pull -5* it's still not ok and continues to knock just as much or almost as much.

As for fueling, in the open loop fueling table primary table, reduce 0.5 from the 2.5+ g/revs columns and 4k+ rpm rows, so it should be rectangle covering the lower right portion of the fuel map. Don't need to touch the "high knock" or "low dam" fueling tables or any tables of that sort. Those should already be very rich as is.
ShinjiML is offline   Reply With Quote
Old 12-17-2017, 10:16 PM   #1456
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
No. I'll be honest, the fueling table scares me.

I don't understand how it works and I've yet to spend time researching it. One thing I've always wondered is why the ol fueling is always changing in the logs if it's a static number from the olpf table. Another thing is, my wb02 and my Pol fuel table don't match up.
The fueling table is a very rough guideline for what AFR you will actually see. Open loop does not have any direct feedback mechanism to apply corrections, which is why you will see the fueling change quite a bit with weather and atmospheric changes.

It basically tells the ECU how long or how short of a pulse-width to use in order to achieve the desired AFR in the table. The ECU uses your maf scaling/ve table and the injector scalar and latency to determine how long of an ipw to use. However, it doesn't actually know if it is hitting those targets so in that sense, it's operating blindly. That's why it's called open loop.

At the end of the day, what the wideband sees is what really matters to tuners.

By contrast, the DIT cars can actually run full closed loop, including during wot.
ShinjiML is offline   Reply With Quote
Old 12-17-2017, 10:24 PM   #1457
Tryptofan
Scooby Newbie
 
Member#: 216924
Join Date: Jul 2009
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2009 WRX STi
White

Default

Quote:
Originally Posted by PDXREALTOR View Post
When I look at olf primary table I see a lot of numbers. How do I know which one is the 'table value'?
The numbers in the OLF-Primary table are the table values. It's just like a table in Excel, basically.

The ECU looks up what the AFR should be when the car is in open-loop mode based on this table. So for example, if you're at 3800 RPM, and calculated load is 3.0g/rev, the ECU looks at the OLF-Primary table at the specific cell corresponding to those numbers. Say for example you have an AFR value of 11.1 in that cell - that means the ECU will try to target that AFR based off what information you give it about the injector size (set with the injector scale parameter), and the mass of air flowing into the engine (based either on the MAF sensor, or on air temperature and manifold air pressure if you're on speed density). EDIT: looks like Shinji beat me to this explanation

I'd also suggest logging commanded fuel final from now on in addition to your wideband O2 sensor so you start learning about fueling. When you're in open loop fueling mode, Comm Fuel Final is what the ECU thinks the AFR will be, based off your primary open loop fueling table value, the injector scalar value, and the calculated mass flow of air it's given. If comm fuel final is significantly different from your wideband O2 values (> 0.5 AFR points), it will be something to look into in the future.

For now, for safety, do what Shinji suggested, and subtract 0.5 from all values in the OLF-Primary table (just select all the values in the table, and hit the subtract key on the keyboard until you've removed 0.4 or 0.5 points). If your knock still doesn't go away, then that suggests phantom knock like he said, which is another issue entirely.

Last edited by Tryptofan; 12-17-2017 at 10:33 PM.
Tryptofan is offline   Reply With Quote
Old 12-18-2017, 12:21 AM   #1458
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

I ended up with 4.2* of timing removed in 4th gear only, in the 2400 RPM column @ 1, 1.5, 2, 2.5 loads. I did so proportionally, so the 2 column had 3.15* taken out, etc.

After seeing high knock numbers I then took 5* out of the AP in ADDITION to the numbers already taken out of 4th gear via ATR, and saw high numbers again. I have it all logged. I just need time to sort through it and get the garbage out.

I also logged the commanded fuel, and commanded fuel final, as well as TPS and ACCELL POS... Turns out the CF Final is matching up, what I think based on your guys input, pretty well. Some numbers are almost spot on, others are just inside or outside the .5 number mentioned by Trypto. ACCELL is def not pegged when the DBW goes WOT (did I say that right?)

I just swung by home to grab some stuff I needed before running my kid across town. I'm going to see if I can quickly create a map that will remove the fuel, but not sure I'll have time.

Logs forthcoming..... I can't even begin to explain how much this input you have given me has helped. And I can't say thanks enough! Haha. and I haven't even got to really sit down and go through the last several posts!!! Can't wait for some time to do that. Likely all night and then a long day of work after sleepless evening!

EDIT - wait until you see the TD Error logging.
PDXREALTOR is offline   Reply With Quote
Old 12-18-2017, 04:02 AM   #1459
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Ok - I added between .45 -.5 of fuel to 3600-7600 rpms and 2.35-3.60 (max) load.

I added back in the timing when I added the fuel, so this was just a fuel change.

No go.

Here is a link to a google folder with 2 4th gear pulls and one data 60 mile highway datalog including only the kc events sorted by time. If the other files sourced in onedrive (original excel files) worked better let me know.

https://drive.google.com/drive/folde...Z9?usp=sharing

Thoughts on making the same .5 changes to the fuel table at a load of 1.75-3.60 and RPM of 2800-4k? This seems to be where the knock is starting, yes?

I did not have time to sort the maps with the pulled timing. Those maps also show the accelerator pedal and throttle position in great length. , as Shinji mentioned above I don't think that's the issue here. There's about 100 miles of logs in those. I will do so tomorrow after work, at the same time I sit down to study the previous posts from the turbo charger info forward.

Still curious on knock count rising consistently on cylinder 1. Wouldn't it be impossible for phantom knock to always report on the same cylinder?

I have fuel rails sitting on the counter - just waiting for the coming time off. Would one injector or spark plug (say from cylinder number 1?) set off these large knock events. A dribbler of an injector, or a bad spark plug? What about putting in colder plugs all around? Low rpm high load problems. *sigh*

It's odd how I can do a pull and get very little -1.4 knock, then go to do another pull a mile down the highway and end up pulling out cause Im' in the -6 KC range.

On the plus side I was introduced to not just one, but two Ford Focus STs today. First time I've ever met one. It was fun, they have a nice looking front end. Also met an Audi that was probably a bit disappointed after our little visit. A lot of that point is - I can hammer the crap out my car and walk away with a -1.4 KC at the most. It's only this low RPM stuff that's giving me fits.

Last edited by PDXREALTOR; 12-18-2017 at 04:11 AM.
PDXREALTOR is offline   Reply With Quote
Old 12-18-2017, 03:11 PM   #1460
Tryptofan
Scooby Newbie
 
Member#: 216924
Join Date: Jul 2009
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2009 WRX STi
White

Default

You're getting the throttle plate position and drive-by-wire stuff a little more it sounds like. It's doing exactly what it's supposed to if you see throttle position at 100% even with accelerator position below 100%.

Read about what TD Boost Error is, and how turbo dynamics works on subarus in general before you start worrying about it. After you do that, you'll realize TD Boost Error being high or low doesn't mean anything bad all on its own.

Quote:
Originally Posted by PDXREALTOR View Post
Ok - I added between .45 -.5 of fuel to 3600-7600 rpms and 2.35-3.60 (max) load.

Your comm fuel final is way off from your sensor readings in the low knock log - commanded AFR is 10.3 to 10.5 in the high RPM range, but your actual AFR is significantly leaner than that. Not great - that says to me whoever tuned your AFR table had trouble getting open loop fueling values to match with your O2 sensor, and took a shortcut to fix the problem.

You need to add that extra fuel everywhere you're seeing knock - if you see knock in the spool up region, you should be adding it there too (not just really high load and over 3600 RPM). For now, while you have this problem, you want to see your O2 sensor reading 11 to 11.2:1 at every RPM value over 3100 RPM and every load value over 1.5g/rev. This won't give you the best spool up, but it will give you the best protection against knock.


Here's what I do when the data is hard to look at like this. Make graphs comparing your "good" run, and your "bad" run in Excel, if you know how. Your 4th gear datasets are perfect for this. Here's a screen capture - RPM is on the x axis in every plot.



In the case of your high knock log, I see the following:

AFR is leaner by as much as 0.5 right where you have your big knock event. It's much leaner in general over 3500 RPM.

Load is higher in the high knock run. So is boost. Timing is much lower, so that suggests adjusting timing isn't fixing your problem.

If it were me, I would do these two things:

1. Set boost targets lower - 22 PSI. You're probably less likely to knock there while you do pulls.

2. Work on getting your AFR down to 11 to 11.2:1 at all RPM and load values above 3100 RPM and 1.5g/rev. You can always change them later.

If knock doesn't go away then, there's a decent chance it's phantom, and then you have mechanical problems to work on (but at least you aren't eating up pistons/bearings while you diagnose your problems!).

In my opinion, you need to stop playing with timing until you do those things. It isn't helping the problem at all.

Last edited by Tryptofan; 12-18-2017 at 03:17 PM.
Tryptofan is offline   Reply With Quote
Old 12-18-2017, 05:01 PM   #1461
Tryptofan
Scooby Newbie
 
Member#: 216924
Join Date: Jul 2009
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2009 WRX STi
White

Default

Quote:
Originally Posted by PDXREALTOR View Post
Still curious on knock count rising consistently on cylinder 1. Wouldn't it be impossible for phantom knock to always report on the same cylinder?
.
Forgot to address this. If you really are only seeing your knock counts only on cylinder 1 (and you have datalogs that prove this, not just a perception or a feeling that it's only cylinder 1), it could be a lot of things. It could be phantom knock from something loose near that cylinder banging around (best case scenario). It could also be a lot of bad things.

It depends on what you're willing and able to do. You can pull the plug and look at it and see if you see anything wrong with it (take pictures because someone else may see something you don't, and you won't want to pull it again - also replace it with a new one). Replace the coil pack on that plug for fun if you have a fresh one on hand and see if that fixes it, takes very little time to do. You could have a bad injector causing that cylinder to get a lean mixture (which could explain your AFR not matching comm fuel final and explain the knock). You could have something cracked and rattling around or letting oil into the cylinder - though you'd probably see smoke if that were the case.
Tryptofan is offline   Reply With Quote
Old 12-19-2017, 06:08 AM   #1462
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Trypto - a continued thanks for these posts. It's 2:30 am here, but I wanted to update since I have big news! I read these posts you wrote on my mobile today while working. It actually sealed the deal of what I did tonight that completely solved my knock problem. I couldn't believe that this tune was so bad timing wouldn't fix it, and I was about to lower the fuel map. More info than I can share in the open forum - PM if you want the details - but it just didn't make sense in the big picture.

It actually hit me this morning and I started to review some emails while on the pooper. lol. They were between my ETUNER and myself. I was working with him a few months ago when all of a sudden the 4th gear pulls went bananas with knock!

At that time I was in the transition of getting a Cobb 20g, and FMIC, etc etc. When I got the 20g I ended up having a local shop install and dyno tune it. It never ever knocked like that, but I only have record of a few 4th gear pulls, and it wasn't on my car long before I ripped the motor out for forged pistons and tons of other 'safety' mods since the power the Cobb 20g got me was over 380 and unexpected. It was also pig rich to help protect the ringlands at hp output of over 380.

I decided to get forged pistons, and did a bunch of other upgrades when I did those, back in early November. That is when oodles of knock started happening again, in 4th and 5th gears. The day after the forged piston/FP green dyno tune. I thought I had a bad tank of gas, then I added extra torco to the mix above what I was tuned for on the dyno. Eventually I ended up here in this conversation with you and Shinji.

So short version for TL;DR - Etuning with a tuner, add an aftermarket part, oodles of knock in 4th gear. Change turbo and Dyno tuning with local tuner on new turbo, no knock problem running pig rich. Upgrade to, forged pistons, FP green, etc, oodles of knock in 4th gear again.

There was one aftermarket part that followed this journey - a lightweight crank pulley. I took it off tonight, and put on the stocker. Problem, for the most part, solved. I had one instance of high knock on a 4th gear pull that I backed out of.

Here are several 4th gear logs, a couple third gear logs, and three highway driving logs of which I ran the drill to try and induce that partial throttle knock over and over without success.

This has to be false knock due to vibrations. The OEM pulley absorbs them almost entirely vs. the LW pulley which doesn't absorb ****. The only thing that throws me is the super pig rich Cobb 20g tune did not exhibit the 4th gear high knock behavior. Perhaps the super rich tune caused the motor to vibrate differently and that's what made that tune not knock. What do you guys think? Is it possible the motor revs just faster enough with the lightweight pulley to cause issues on some tunes?

Check out these logs. https://drive.google.com/drive/folde...ud?usp=sharing

I spent hours searching about lightweight crank pulleys and knock, and never found any concrete evidence of it so I ended up writing it off. Boy that thing cost me hundreds in gas - forget about the time!

I have one more long day tomorrow, then I am going to read through these informative posts for education. I hope you guys will help me with my map/tuning questions still. I have the Cobb tuning guide I'm going to read, and another guide from NASIOC, tuning 101 or something like that, that I'm going to read too.

I'm also looking at EFI university info and considering starting with some of their courses.

The fueling is still not matching my WB02, but my WBO2 matches the dyno WBO2 so as one of you said - the tuner likely doesn't care above that actual POL table when he's got tow WBO2s in front of him. Gotta figure out how I'm going to fix that since this week I'm putting on fuel rails!

Oh man..... you have no idea how relieved I am.

Last edited by PDXREALTOR; 12-19-2017 at 06:23 AM.
PDXREALTOR is offline   Reply With Quote
Old 12-19-2017, 07:10 AM   #1463
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
Trypto - a continued thanks for these posts. It's 2:30 am here, but I wanted to update since I have big news! I read these posts you wrote on my mobile today while working. It actually sealed the deal of what I did tonight that completely solved my knock problem. I couldn't believe that this tune was so bad timing wouldn't fix it, and I was about to lower the fuel map. More info than I can share in the open forum - PM if you want the details - but it just didn't make sense in the big picture.

It actually hit me this morning and I started to review some emails while on the pooper. lol. They were between my ETUNER and myself. I was working with him a few months ago when all of a sudden the 4th gear pulls went bananas with knock!

At that time I was in the transition of getting a Cobb 20g, and FMIC, etc etc. When I got the 20g I ended up having a local shop install and dyno tune it. It never ever knocked like that, but I only have record of a few 4th gear pulls, and it wasn't on my car long before I ripped the motor out for forged pistons and tons of other 'safety' mods since the power the Cobb 20g got me was over 380 and unexpected. It was also pig rich to help protect the ringlands at hp output of over 380.

I decided to get forged pistons, and did a bunch of other upgrades when I did those, back in early November. That is when oodles of knock started happening again, in 4th and 5th gears. The day after the forged piston/FP green dyno tune. I thought I had a bad tank of gas, then I added extra torco to the mix above what I was tuned for on the dyno. Eventually I ended up here in this conversation with you and Shinji.

So short version for TL;DR - Etuning with a tuner, add an aftermarket part, oodles of knock in 4th gear. Change turbo and Dyno tuning with local tuner on new turbo, no knock problem running pig rich. Upgrade to, forged pistons, FP green, etc, oodles of knock in 4th gear again.

There was one aftermarket part that followed this journey - a lightweight crank pulley. I took it off tonight, and put on the stocker. Problem, for the most part, solved. I had one instance of high knock on a 4th gear pull that I backed out of.

Here are several 4th gear logs, a couple third gear logs, and three highway driving logs of which I ran the drill to try and induce that partial throttle knock over and over without success.

This has to be false knock due to vibrations. The OEM pulley absorbs them almost entirely vs. the LW pulley which doesn't absorb ****. The only thing that throws me is the super pig rich Cobb 20g tune did not exhibit the 4th gear high knock behavior. Perhaps the super rich tune caused the motor to vibrate differently and that's what made that tune not knock. What do you guys think? Is it possible the motor revs just faster enough with the lightweight pulley to cause issues on some tunes?

Check out these logs. https://drive.google.com/drive/folde...ud?usp=sharing

I spent hours searching about lightweight crank pulleys and knock, and never found any concrete evidence of it so I ended up writing it off. Boy that thing cost me hundreds in gas - forget about the time!

I have one more long day tomorrow, then I am going to read through these informative posts for education. I hope you guys will help me with my map/tuning questions still. I have the Cobb tuning guide I'm going to read, and another guide from NASIOC, tuning 101 or something like that, that I'm going to read too.

I'm also looking at EFI university info and considering starting with some of their courses.

The fueling is still not matching my WB02, but my WBO2 matches the dyno WBO2 so as one of you said - the tuner likely doesn't care above that actual POL table when he's got tow WBO2s in front of him. Gotta figure out how I'm going to fix that since this week I'm putting on fuel rails!

Oh man..... you have no idea how relieved I am.
The logs certainly look better. However, it looks like there's still a bit of noise there. Whatever knock you had around 5300 RPMs appears to be learning away for the most part, but still showed up as fresh -1.4 on some of your new logs. You're also still pulling some timing in the same 3600 RPM.

At this point, I'm fairly certain what you're seeing is Phantom knock, especially with how it only shows up in those certain RPM areas, but I would still recommend fixing the fueling as tryptofan had mentioned a few times.

After that, I would recommend pulling 3 degrees of timing and lowering boost by 3 PSI and see if it still Knocks in those areas. This should be more than enough to eliminate any real knock, but if it continues to do so, then you can move on to tuning the knock sensor. Forged engines tend to be pretty noisy compared to oem and it's common to have to tune the knock sensor.

Also, the other thing that makes me uncomfortable is the fact that your wastegate duty is set to 100% under full throttle. I had asked you before if you were running a hybrid boost setup because if you aren't, then I would really be concerned about tuning boost in this fashion.
ShinjiML is offline   Reply With Quote
Old 12-19-2017, 08:02 AM   #1464
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by ShinjiML View Post
The logs certainly look better. However, it looks like there's still a bit of noise there. Whatever knock you had around 5300 RPMs appears to be learning away for the most part, but still showed up as fresh -1.4 on some of your new logs. You're also still pulling some timing in the same 3600 RPM.

At this point, I'm fairly certain what you're seeing is Phantom knock, especially with how it only shows up in those certain RPM areas, but I would still recommend fixing the fueling as tryptofan had mentioned a few times.

After that, I would recommend pulling 3 degrees of timing and lowering boost by 3 PSI and see if it still Knocks in those areas. This should be more than enough to eliminate any real knock, but if it continues to do so, then you can move on to tuning the knock sensor. Forged engines tend to be pretty noisy compared to oem and it's common to have to tune the knock sensor.

Also, the other thing that makes me uncomfortable is the fact that your wastegate duty is set to 100% under full throttle. I had asked you before if you were running a hybrid boost setup because if you aren't, then I would really be concerned about tuning boost in this fashion.
Hey - Yes! First thing on my list to learn/do is to dial in that fuel table to match the WB02 in my car.

I also like your idea on dialing back things to see if it's phantom or real. I have mixed feelings on tuning the knock sensor completely out, but also think I 've read there's ways to turn certain frequencies out, in which case, I'd be interested. I just don't want to end up with real knock one day and have the knock sensor not report it.

For kicks I just ordered a kartboy crank pulley with the condition I can return it should it cause issues. I don't want to drop fluid damper type money on a pulley, but, it's tempting.

It's 5am here. *sigh* All I can do is think about out the ideal way to get Torco out of a 5 gallon pale into a suitable 1/2 gallon container, and how to store it in my trunk, which leads to a trunk organizer. I have issues. lol.

EDIT** apologies I thought I answered your hybrid boost question earlier. Yes, I am running a MBC in parallel. I have the max boost limit column setup to what is was boosting off the dyno tune - just have the 3rd line disconnected and the, as you've seeen, WGDC jacked up to 100 in the WOT column, and post WGDC comp table. I've also triple verified the MBC is actually controlling max boost.

Last edited by PDXREALTOR; 12-19-2017 at 08:07 AM.
PDXREALTOR is offline   Reply With Quote
Old 12-19-2017, 08:06 AM   #1465
lap682
Scooby Newbie
 
Member#: 476145
Join Date: Oct 2017
Default

update on my 2018 STi with insane feedback knock and 0 Dam, I met up with STimikey last night for a road tune and now the car is 100% The car now has power in areas where none of the other OTS maps had it and now its butter smooth!
lap682 is offline   Reply With Quote
Old 12-19-2017, 08:08 AM   #1466
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by lap682 View Post
update on my 2018 STi with insane feedback knock and 0 Dam, I met up with STimikey last night for a road tune and now the car is 100% The car now has power in areas where none of the other OTS maps had it and now its butter smooth!
What was the cause of your knock? Glad it's sorted!
PDXREALTOR is offline   Reply With Quote
Old 12-19-2017, 08:15 AM   #1467
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
Hey - Yes! First thing on my list to learn/do is to dial in that fuel table to match the WB02 in my car.

I also like your idea on dialing back things to see if it's phantom or real. I have mixed feelings on tuning the knock sensor completely out, but also think I 've read there's ways to turn certain frequencies out, in which case, I'd be interested. I just don't want to end up with real knock one day and have the knock sensor not report it.

For kicks I just ordered a kartboy crank pulley with the condition I can return it should it cause issues. I don't want to drop fluid damper type money on a pulley, but, it's tempting.

It's 5am here. *sigh* All I can do is think about out the ideal way to get Torco out of a 5 gallon pale into a suitable 1/2 gallon container, and how to store it in my trunk, which leads to a trunk organizer. I have issues. lol.

EDIT** apologies I thought I answered your hybrid boost question earlier. Yes, I am running a MBC in parallel. I have the max boost limit column setup to what is was boosting off the dyno tune - just have the 3rd line disconnected and the, as you've seeen, WGDC jacked up to 100 in the WOT column, and post WGDC comp table. I've also triple verified the MBC is actually controlling max boost.
100% WGDC in columns of full boost is correct for a hybrid boost setup. This basically hands boost control over to the MBC regardless of what your boost targets are (assuming your initial and max WGDC are both set to 100% for the MBC boost columns). This means you can easily use the MBC to turn down boost to verify your phantom knock.

The knock sensor will still report real knock, it's just systematically desensitizing it to not jump at phantom knock in the rpm areas of concern. Of course if you'd rather be on the safe side, you could leave it as is and just deal with the minor phantom knock, as it's not too bad at this point.
ShinjiML is offline   Reply With Quote
Old 12-19-2017, 08:22 AM   #1468
PDXREALTOR
Scooby Specialist
 
Member#: 468393
Join Date: May 2017
Location: PDX OREGON
Vehicle:
2017 STI

Default

Quote:
Originally Posted by ShinjiML View Post
100% WGDC in columns of full boost is correct for a hybrid boost setup. This basically hands boost control over to the MBC regardless of what your boost targets are (assuming your initial and max WGDC are both set to 100% for the MBC boost columns). This means you can easily use the MBC to turn down boost to verify your phantom knock.

The knock sensor will still report real knock, it's just systematically desensitizing it to not jump at phantom knock in the rpm areas of concern. Of course if you'd rather be on the safe side, you could leave it as is and just deal with the minor phantom knock, as it's not too bad at this point.
1st on the list is the fuel table, and installing the new rails I have sitting on my kitchen table.

Then I must read the cobb info about the knock sensor, and come here with my (hopefully) educated questions.

Good point on turning down the MBC for checking on the knock! I was thinking of hitting the MAX WGDC table to limit boost real quickly. But, with the MBC, and global timing on the AP I can do it all remotely. Nice suggestion!

Question - does it hurt anything to have the max boost column set to, say, 30 psi, if the MBC is in control, and the boost/fuel cut is set to the correct level? I scaled that column, but would prefer to just enter 27 in and leave the MBC as it is now. I did it the way I did because I wasn't sure if it wold affect any boost limit compensations that would reduce boost. However, now that I think about it isn't that is done by reducing the WDGC? I need to sleep.
PDXREALTOR is offline   Reply With Quote
Old 12-19-2017, 08:32 AM   #1469
lap682
Scooby Newbie
 
Member#: 476145
Join Date: Oct 2017
Default

Quote:
Originally Posted by PDXREALTOR View Post
What was the cause of your knock? Glad it's sorted!
Cobb OTS maps just didn't agree with my car, once we loaded his base map dam stayed at one and ZERO feedback knock.
lap682 is offline   Reply With Quote
Old 12-19-2017, 10:23 AM   #1470
ShinjiML
Scooby Specialist
 
Member#: 257231
Join Date: Sep 2010
Chapter/Region: Tri-State
Location: Hudson Valley, NY
Vehicle:
2011 STI ShinjiTuned
DGM

Default

Quote:
Originally Posted by PDXREALTOR View Post
1st on the list is the fuel table, and installing the new rails I have sitting on my kitchen table.

Then I must read the cobb info about the knock sensor, and come here with my (hopefully) educated questions.

Good point on turning down the MBC for checking on the knock! I was thinking of hitting the MAX WGDC table to limit boost real quickly. But, with the MBC, and global timing on the AP I can do it all remotely. Nice suggestion!

Question - does it hurt anything to have the max boost column set to, say, 30 psi, if the MBC is in control, and the boost/fuel cut is set to the correct level? I scaled that column, but would prefer to just enter 27 in and leave the MBC as it is now. I did it the way I did because I wasn't sure if it wold affect any boost limit compensations that would reduce boost. However, now that I think about it isn't that is done by reducing the WDGC? I need to sleep.
Nope setting the boost in the map won't do anything if WGDC is 100% and passing control to the MBC. The MBC will limit the boost.
ShinjiML is offline   Reply With Quote
Old 12-19-2017, 12:59 PM   #1471
Tryptofan
Scooby Newbie
 
Member#: 216924
Join Date: Jul 2009
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2009 WRX STi
White

Default

Quote:
Originally Posted by ShinjiML View Post
The logs certainly look better. However, it looks like there's still a bit of noise there. Whatever knock you had around 5300 RPMs appears to be learning away for the most part, but still showed up as fresh -1.4 on some of your new logs. You're also still pulling some timing in the same 3600 RPM.
You still have a knock event I would consider somewhat concerning (-2.8 and -3.15 degrees of learning) over 5k RPM in your "Long Drive" log, where you've got 25.5 PSI of boost and a lean AFR of 11.7:1. So once you fix the AFR values and do like Shinji and I suggest (crank your boost down ~2-3 PSI, pull a little timing to keep things safe), you can stay knock free and enjoy the car more. Just keep logging it periodically to make sure the problem is gone.

It does look way better in spool up - no more massive -7, -9 feedback knock events. Maybe your ECU was already picking up some noise from the aftermarket pulley and running fairly lean was adding enough noise on top of that to put it over the edge - not sure. But as long as it stays gone, I'd be pretty happy knowing it was just phantom knock in that area.

Taking a course like EFI 101 would probably be a good thing to do. I'm not sure how deep they go but I've heard good things about it.
Tryptofan is offline   Reply With Quote
Old 12-19-2017, 02:33 PM   #1472
02wagonowner
Scooby Newbie
 
Member#: 479423
Join Date: Dec 2017
Location: Reno, NV
Vehicle:
2002 Impreza WRX
Silver

Default

Some logs from this morning. Newly loaded stage 2 map OTS 91 octane(same problems before tune but was afraid stock tune couldn't handle exhaust and engine so added tune), turbo back, JDM 2.0 with AVCS not connected 60k miles, stock intake, Grimmspeed BCS, 91 octane.

https://drive.google.com/open?id=1sT...4OeiNIULuNFLtc

Under load the car is hitting a "wall" and power is completely lost and it is sputtering BAD. It revs up fine in neutral, and no obvious mechanical noises of concern that I hear. Some things I note in the log is the massive drop of IAM and then it comes back up to 16. Lots of Feedback Knock Learning at throttle. And also AF correction and learning seem to have issues. At idle car seems to hold steady fairly well but it will every 20 seconds or so "pop" a bit and RPM will swing a little as will the AFR's...

I replaced spark plugs, MAF, and coils as I thought it had to be a spark problem seeing how many people have that issue. I now have no idea what is wrong. I have read about bad knock sensor, o2, camshaft, but I really don't know what to look for now. I really thought vac leak but idle vacuum looks correct from what I read. Now I am thinking possibly fuel pressure regulator or maybe injector scaling is off or something and it's running below 11.02 (which I read is lowest stock sensor reads?). Getting colder now the problem has seemed to gotten worse. Used to be like 4,000/4,500RPM in warmer temps but now with air at 20F I can get it to stutter down to below 3,000 if that helps at all too.

If anyone has any suggestions I am lost. No CEL's either on or flashing during stuttering. The reason RPM's don't go higher on WOT or really any decent load, is because it's so bad is just stutters and stops accelerating. After though it drives normal at lower throttle cruising.

I appreciate all help.
02wagonowner is offline   Reply With Quote
Old 12-19-2017, 04:00 PM   #1473
wrx_b666st
Scooby Newbie
 
Member#: 472604
Join Date: Aug 2017
Default 2016 Wrx Fa20dit

Hey guys. Here are my two most recent logs from 3rd gear.

My car is a 2016 WRX with an IAG stage 3 short block, muffler deletes, drop in K&N air filter and DW65c fuel pump.

Any feedback would be nice seeing as how the cars power seems to be hit or miss. 1st and 2nd gear feel fine but 3rd on up just feel sluggish and I keep seeing a LOT of Feedback knock at anything under 3k at WOT.

https://drive.google.com/file/d/1o3X...ew?usp=sharing

https://drive.google.com/file/d/14X8...ew?usp=sharing
wrx_b666st is offline   Reply With Quote
Old 12-19-2017, 10:03 PM   #1474
Tryptofan
Scooby Newbie
 
Member#: 216924
Join Date: Jul 2009
Chapter/Region: SWIC
Location: Phoenix, AZ
Vehicle:
2009 WRX STi
White

Default

Quote:
Originally Posted by 02wagonowner View Post
Some logs from this morning. Newly loaded stage 2 map OTS 91 octane(same problems before tune but was afraid stock tune couldn't handle exhaust and engine so added tune), turbo back, JDM 2.0 with AVCS not connected 60k miles, stock intake, Grimmspeed BCS, 91 octane.

https://drive.google.com/open?id=1sT...4OeiNIULuNFLtc

Under load the car is hitting a "wall" and power is completely lost and it is sputtering BAD.
What exactly is this "OTS" stage 2 map set up for? Is it exactly for your combination of modifications - a JDM 2.0 liter WRX(?) motor (is the JDM turbo for this motor the exact same as the stock USDM turbo - td04 I'm assuming?), stock intake, GS BCS, aftermarket downpipe(???). Which injectors are you using, and which injectors is the map set up for - the JDM injectors, or USDM injectors?

Another really important question - is your "OTS" map meant to function without AVCS??? That alone could could be a big contributor to bad idling, misfires, etc.

It sounds to me like you have a car that is not at all set up for a generic map meant for a simple set of modifications.

Also, your stock O2 sensor can't be used as a wideband for purposes of monitoring the tune. You can really only use it to monitor the car under low boost conditions.

Last edited by Tryptofan; 12-19-2017 at 10:15 PM.
Tryptofan is offline   Reply With Quote
Old 12-19-2017, 11:51 PM   #1475
02wagonowner
Scooby Newbie
 
Member#: 479423
Join Date: Dec 2017
Location: Reno, NV
Vehicle:
2002 Impreza WRX
Silver

Default

Quote:
Originally Posted by Tryptofan View Post
What exactly is this "OTS" stage 2 map set up for? Is it exactly for your combination of modifications - a JDM 2.0 liter WRX(?) motor (is the JDM turbo for this motor the exact same as the stock USDM turbo - td04 I'm assuming?), stock intake, GS BCS, aftermarket downpipe(???). Which injectors are you using, and which injectors is the map set up for - the JDM injectors, or USDM injectors?

Another really important question - is your "OTS" map meant to function without AVCS??? That alone could could be a big contributor to bad idling, misfires, etc.

It sounds to me like you have a car that is not at all set up for a generic map meant for a simple set of modifications.

Also, your stock O2 sensor can't be used as a wideband for purposes of monitoring the tune. You can really only use it to monitor the car under low boost conditions.
It's supposed to be set up for an 02 WRX with stage 2 modifications, td04, including the grimmspeed and turbo back(so yes down pipe) but not the jdm. I have read a ton of threads about jdm WRX motors running just fine on usdm maps minus avcs so figured I'd start here and fine tune. I only brought up injectors as a potential but I've never read that as being a problem between the 2 motors.

As stated, however, the car exhibited the same symptoms before and after tune...I was really hoping that the logs would tell more than I can. Did you look at these or just the post? They show some interesting AF learning/correction and quite a bit if feedback and dropping of IAM. Just not sure what can cause this.
02wagonowner 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

All times are GMT -4. The time now is 12:37 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2018 Axivo Inc.
Copyright ©1999 - 2017, North American Subaru Impreza Owners Club, Inc.