ECMlink ECMLink is kicking my ass! Help please

[WOTDSM]

Cool man. Just keep in mind you'll have to redo the SD VE Adjust tool for cruise conditions since we've just made cruise 4% richer.

 

[1990AWD]

Alright, I've been playing around with it for a while, here's a log of me driving down the road and back with a 3rd gear pull in there and idling at the beginning and end. Tell me what you think.

 

[1990AWD]

Another log from the drive home from the shop where my car was, about a 35km drive. My LC-1 and AFRest are way off, worse than before I think. What'd I do wrong and what do I need to do to change it? Also, it seems to sometimes go into open loop and stay there for no reason, why is that?

 

[snowborder714]

Your SD table needs adjusted to get your AFRatioEst and wideband readings closer. You can use the built-in tools (right click on graph > one of the SD VE Adjust options) or do it by hand based on load and RPM.

Looks like your SD table could use a bit of adjustment at idle. Seems like you're a little high on the CombinedFT value.

Your pull doesn't look bad, but again, SD adjustments are needed. You're also getting some knock which should be addressed.

As an SD user, you can pretty much ignore BoostEst and it's relation to your MAP sensor reading. You'll want to focus on AFRatioEst and the wideband to get the SD table dialed in.


Random request - can you change the "Display as" values for your MAP sensor to MAP and your LC-1 to Wideband? Makes viewing the log easier for me

 

[1990AWD]

Okay, so basically get the SD values dialed in.

Changed the names for you. haha

 

[snowborder714]

Yup, you want to get them dialed in at idle first (looks like you're pretty close already), then cruise, then WOT. For idle and cruise, you can either adjust them by hand (using Track datalog for the SD table) or using the built-in tools. For WOT, I'd recommend using LinkTools if you aren't already.

 

[1990AWD]

Ok, and so when adjusting my SD values, I'm just looking to get wideband and aftest to be lined up, right?

 

[snowborder714]

You've got it. There's other things to consider (SDAirflowPerRev and CombinedFT at idle, for example), but that's the basis.

 

[1990AWD]

Okay, thanks Brian. I'll do my best and spend most of tomorrow tuning.

 

[MY1GDSM]

Brain is right. Though I wouldn't get too caught up on some what the numbers "should be" if the car is running good an the trims get in line.

From looking at that log if you were running more boost you would be having quite a few more issues.
You are anywhere from 5% to about 12-13% lean across the board as soon as you are into Boost.
I was going to post in here yesterday... But got caught up yesterday with some other things.

But, I really wouldn't even bother with keeping boost Est lined up. There is a good thread on here with quite a bit of info on why from some of us here as well as some more other info. From a thread another OP started asking about them lining up(or not). Anyway, It's nice to have things lined up in the 5-6k range to get some of the estimates giving descend info. But, I would just concern yourself with getting you AFR to match your Targets in DA Above all else.

It's not to hard I'd just look at the Load cells that you are hitting across the lean part and Add about 5% to the VE maps above and across that and then another 5% in the areas that are leanest. I don't ever use Linktools or some of these other "Auto" Features. But it's not very hard to run a trace with link and do it yourself if you can do some math and estimation in your head quickly.(IE Keeping it simple... Say you just hit a cell of 100 and you are 5% lean still, Just change it to 105 et cetra) I'd do that an Work out to interpolate things a bit farther as well as smooth out an extend to an extent some of the High spots on the edge of what you have been hitting in the map. So you don't end up 12% lean in a higher load cells as being that lean at 20+psi is going to piss the car of a lot more then at around 15psi.

You can if things are off an you really want them to match you can adjust things globally(like said) to get them lined up Rather or if you start getting some bit high/low VE numbers.

But, if it is just "for the sake of doing it" and not something like you hitting the wrong load cells(not really as relevant with SD) Then I wouldn't bother as you may just cause more issues and spend more time re adjusting your VE map to deal with issues.

 

[snowborder714]

But it's not very hard to run a trace with link and do it yourself if you can do some math and estimation in your head quickly.(IE Keeping it simple... Say you just hit a cell of 100 and you are 5% lean still, Just change it to 105 et cetra) I'd do that an Work out to interpolate things a bit farther as well as smooth out an extend to an extent some of the High spots on the edge of what you have been hitting in the map. So you don't end up 12% lean in a higher load cells as being that lean at 20+psi is going to piss the car of a lot more then at around 15psi.

The only issue with this if you really want to get technical with how the ECU works is that it doesn't just read from one cell. It pulls from 4 cells at once, and doing the math in your head to figure out how much of a percentage it's pulling from each of those 4 cells is very complicated. This is where LinkTools comes into play as it does all this complex math for you.

But you can still get a fairly good tune by not worrying about this. Just wanted to make you aware of what's going on in the background.

 

[MY1GDSM]

Brain I'm fully aware of how link handles things. As I said you would need to interpolate things as well... .

I don't use any of the tools because to me it just takes away from fully being involved in what is going on.

Not everyone may be able to do things as I do I suppose.
But it just make more sense to me this way.

I'm sure you have seen as much as me people using these kinds of tools only to have a completely Fubar VE map and I just would recommend people to start at a least get an Idea of doing things this way rather than using a calculator to spit number out to them. Only to not fully get what is going on or understand why the car has an issue... .

 

[snowborder714]

You're correct that you still need to know what the tool is doing and fully understand that process so that you're just using the tool as a way to decrease the time spent on manual tasks. If you're using the tool without the knowledge of what it's doing and why, you're better off not doing it at all until you learn.

 

[1990AWD]

Well guys, here's my latest log. Just been adjusting the VE table like you said.

 

[1990AWD]

Here's a couple more logs. The first was me driving around, I'm still having a tough time trying to get it smooth between idling and WOT/boost. Just such a big gap in the VE table.

The second log is a third gear pull, actually quite good, a little lean yet though. I adjusted the cells along the pull so hopefully they line up even a little better, but I don't have time to go try that yet.

 

[1990AWD]

Here's another log of a third gear pull. It's actually looking pretty damn good, just needs some more refining.

Any tips though on how to get a smoother drive between idle and WOT? It's just that its seems like such a quick increase in VE numbers.

 

[1990AWD]

So I have some questions.

The first log I have here is just me driving through traffic. If you look closely at any point where I totally lift off the throttle or just step back on it a little, my AFRs are out to lunch. What do I have to change to get them proper? Why does the AFRest not go to full lean when I let off and no fuel is even entering the engine?

Second log is another third gear pull, my AFRs are getting pretty close, but any more help and advice would be appreciated.

Bottom line, my idle is good, my WOT is getting there, but I'm still not satisfied with everything in-between, mainly my throttle response.

 

[gofer]

So I have some questions.

The first log I have here is just me driving through traffic. If you look closely at any point where I totally lift off the throttle or just step back on it a little, my AFRs are out to lunch. What do I have to change to get them proper? Why does the AFRest not go to full lean when I let off and no fuel is even entering the engine?

Second log is another third gear pull, my AFRs are getting pretty close, but any more help and advice would be appreciated.

Bottom line, my idle is good, my WOT is getting there, but I'm still not satisfied with everything in-between, mainly my throttle response.

What are you using to control boost and why are you running narrowband simulation when your logging your WB through your EGR input?

Even with coasting fuel cut (off the throttle and decel) and a WB readings of 22:1 your AFREst will show around stoich because it pulls the AFREst data from your OLMax/MinOct DA table.

At first your BoostEst and Omni readings threw me off a bit, they were consistently 5psi off throughout the pull when usually they track close to about 5.5k if your WB matches your AFREst. For whatever reason your engine displacement setting in BoostEst was 2.7l so moving it to 2.0l made them track close. That being said, I'm impressed by the airflow numbers you're getting out of the 14b on such low boost, kind of leaves me scratching my head to be honest.

With the mods you've got and the airflow numbers you're logging I'd be extremely happy with that tune. It looks like you could lean out the 6.5k/7k - 29.4/31.2 VE cells a few % to clean up that AFR at the end of the pull. I'd also play with timing advance during spool-up as well as leaner spool-up AFR's and see if you can't get it to spool faster. You could also bring timing back faster after peak torque, knock permitting of course.

 

[1990AWD]

What are you using to control boost and why are you running narrowband simulation when your logging your WB through your EGR input?

Boost control is just a mbc, and my LC-1 setup and inputs was all set-up two years ago. That's what I had understood to do at the time and have assumed that was correct.

Even with coasting fuel cut (off the throttle and decel) and a WB readings of 22:1 your AFREst will show around stoich because it pulls the AFREst data from your OLMax/MinOct DA table.

Ah, okay, simple enough.

At first your BoostEst and Omni readings threw me off a bit, they were consistently 5psi off throughout the pull when usually they track close to about 5.5k if your WB matches your AFREst. For whatever reason your engine displacement setting in BoostEst was 2.7l so moving it to 2.0l made them track close. That being said, I'm impressed by the airflow numbers you're getting out of the 14b on such low boost, kind of leaves me scratching my head to be honest.

I've been told to not worry about BoostEst but rather just the AFRest/Wideband, or no?

Airflow wise, I have no clue how to know if my numbers are accurate or not. I just assume as I play with the VE table that it would change accordingly.

With the mods you've got and the airflow numbers you're logging I'd be extremely happy with that tune. It looks like you could lean out the 6.5k/7k - 29.4/31.2 VE cells a few % to clean up that AFR at the end of the pull. I'd also play with timing advance during spool-up as well as leaner spool-up AFR's and see if you can't get it to spool faster. You could also bring timing back faster after peak torque, knock permitting of course.

Okay so leaner and more timing will help spool, I'll try that.

Thanks, I'm just trying to always keep learning.

 

[1990AWD]

Car is running really good right now, AFRs are near perfect in WOT and car pulls like crazy, may even go this Friday to the track.

 

[Bud92gsx]

Awsome news.I wish I knew what all these numbers mean.

 

[WOTDSM]

Boost control is just a mbc, and my LC-1 setup and inputs was all set-up two years ago. That's what I had understood to do at the time and have assumed that was correct.



Ah, okay, simple enough.



I've been told to not worry about BoostEst but rather just the AFRest/Wideband, or no?

Airflow wise, I have no clue how to know if my numbers are accurate or not. I just assume as I play with the VE table that it would change accordingly.



Okay so leaner and more timing will help spool, I'll try that.

Thanks, I'm just trying to always keep learning.

Less timing helps spool up. Like antilag, the closer the spark happens to TDC, the faster the turbo will spool.

Also, have you done a BLT? Seems like your boost tapers off pretty bad. I usually like to see less than 1psi change throughout the pul

 

[MY1GDSM]

Less timing helps spool up. the closer the spark happens to TDC, the faster the turbo will spool.

BUT, Just because your are spooling the turbo up sooner... . Doesn't mean you are making more power

 

[WOTDSM]

BUT, Just because your are spooling the turbo up sooner... . Doesn't mean you are making more power

Precisely.

I was going to say on the turbo that the OP is running I would leave the timing alone. I would be more worried about spooling a bigger turbo such as an hx40 or s256. You'd probably lose more power by retarding the spool up timing on a turbo that's smaller than a 20G.

 

[1990AWD]

Ah, I always assumed more timing helped spool. Oh well, its not very laggy anyway.