86's behavior(tuning pov)
 

Just needed to vent before I drive my car into a building. I recall even when I was completely stock this car was sensitive to heat. I live in Miami so humidity and heat are you neighbors, you see them one minute then they're gone. I recall the car behaving like some oem turbo cars, happy when cold then not so much when hot, this surprised me coming from such low hp NA car.

Im now many moons from being stock and the behavior is still present. I feel very content with my tuner but this car has a damn personality. Please keep in mind most of my examples are a bit exaggerated to illustrate a picture or explain a behavior.
When it's hot my car usually struggles or lacks acceleration, I have to rev it up to get it moving, the acceleration isn't smooth at times. My morning drives to work which are still far from cool 80F, the car accelerate forward with easy, simply applying my foot to the gas pedal and still under positive pressure from the turbo(under o psi on boost gauge) the car is pushing me back in my seat moving along.

I'm sure most could say something's off with my tune based on my examples but I've had the chance to compare and is not tune related. I'm sure to a degree something could be tweaked but nothing a regular dyno tune or road tune would solve, unless hundreds of hours were to be spent tuning(still may not solve it :lol:). I don't know if the motor is just sensitive if the stock ecu's logic is not so smooth. I don't really know how to explain or if what I say is making any sense.

If I go WOT from 2.5k it feels like I'm pulling trailer home to compared to going WOT after 3.5k or higher. I know the timing and everything is completely different and also under 3k I don't have any boost built. I would expect the feel at 5k to be the same, coming from 2.5k or 4k WOT. Similar weather conditions give me some time fine timing correction and then the next run would add timing, why does this damn thing fluctuate so much. The crappy part is that I feel all these little changes. I also understand gasoline varies from station to station, still.

Guess any experiences similar or completely different would be appreciated to see if anyone else feels the same way about their ride or if I just need to stop being so dramatic. Nothing really to solve, just hopped after spending so much cash the darn thing would be more appreciative. Yes i do know it's not a person and it can't thank me for changing the oil too often and trying to care for it as if it were my first born. Still just pissed.


Maybe this with the clutch acting up and not wanting to let 1st gear in is contributing to me hating this car, for now.

Tune related.

Pick up an ECUTek programming kit so you can datalog the vehicle. The investment in that cable on an F/I vehicle like yours is worth it, especially if you have to troubleshoot anything like this.

Oh yeah I'm a logging freak, i think I drive my tuner retarded with how annoying I could be just to trying to point out things in a tune I think are causing issues when I fact I don't really know squat.

Just now went for a drive, cool car, not after being driven for hours in traffic and still get fine timing correction up to 3 counts when loading the car.

At times when I try to power out of a tight turn I feel the car just bogs and am even though I hear and see the boost increasing, I ain't moving much.

Well what is going on in the log during the situations where performance is poor vs when the behavior is excellent?

There's certainly a lot to look at, but start with overall timing (not just the corrections), AFR target & actual AFR (nice if you had a real wideband), VTC?

Appreciate your feedback. I do have a wideband. My AFR idle and cruise sits at an average of 14.2-14.6. Sometimes it goes up to 15 or it could drop to 13 but goes right back to 14's. I know my LTFT doesn't go beyond -+3.9 on a regular city drive(I checked this morning lol)

I'll try to take a log in the morning when things are usually nice and peppy as opposed to when I'm getting home after work which is when the car feels sluggish, after being driven in traffic for about 45mins.

Forgot to mention my WOT AFR usually goes from 11.8 to 11.3 as rpms increase to redline. Did a 5th gear pull to check recently and it was at 11.5, this is running 10.5psi.


The VTC part is where you got, me, is this cam timing? Side note, did a log from about 2.5k RPMs and the timing did stand out to me.. it start at 8.. climbs to about 15 or so by redline, going off my head so it could be +-2 degrees or so, I'll double check.


Is there a parameter that shows target AFR and actual, well I know the actual is the front 02 which isn't great but my Custom map O was calibrated to log the wideband AFR.

Closed loop isn't of concern -- the problem you have is under load. VTC is cam timing. You need to log the situations where the issue crops up, and if it's intermittent, compare to what it looks like when everything is OK.

Actual AFR (wideband), cam timing, ignition timing (actual), load, manifold pressure just to name a few to start with.

So city driving isn't' closed loop? I know it isn't so simple, but im feeling the hesitation, bogging during mild acceleration not really when im going WOT.. but then again this is all going over my head. Anyway you'd be able to see one of my logs? Driving up a little expressway ramp, slight incline, I give it gas at low rpm's and bog.. of course, this isn't all the time, normally given this same scenario the car just pulls forward as boost is building, very progressively.


Thanks a lot for your help, this may not solve anything until my tuner checks stuff out but definitely helps see things from a different perspective.


Does close loop switch to open when under X load? I thought it was usually going WOT.

Equivalence ratio commanded is the commanded AFR.

When your AFR drops out of stoic targets, you're typically no longer in closed loop, it depends on how hte map is setup sometimes.

Your bogging is happening during mild acceleration events, which I'm assuming is happening either before boost builds, or during light amounts of boost -- hard to say without seeing the data as I'm just going off your description.

Yes I can view logs.

You're absolutely right, the bogging happens just under boost and while building it(less than 5psi or so) Just yesterday I felt the weirdest thing, went wot and felt the hesitation(already at full boost) till about 5.5k or so, then a surge of power, sort of like vtec. That had never happened or at least not that I had noticed at least.

Data of this behavior should help in troubleshooting and directing you towards a solution.

Hesitation like that would usually be either yoyoing timing (lots of correction happening) or AFRs going lean. I wonder whether the lower end could be due to load limits or similar.

I noticed the yoyo timing.. I have a log I did today and I get correction up to -3 and then addition, not back to back but fairly close in the event. I'll see if I could post some logs or something to show what im trying to explain as my tuning vocabulary is very limited. I didn't understand the part about load limits.


What puzzles me is that is not constant or always, just at times. That's why I initially thought it was a temperature thing.


Your help is very appreciated.

What are your IATs when all of this is happening? You could have knock correction plus high IAT correction, this could all be causing high egts. My knowledge is limited but I was logging similar performance issues when running crap 91 with high IATs

Timing map is suspect if you're pulling 3 degrees under load.

Also, what port injectors are you running and how have they been balanced with the DI system? When I switched injectors, it took me a while to sort out the areas where the PI system turned on, but operated in a non-linear range - have you logged PI and DI pulses and to see if the hesitation occurs with switchover?

Im running stock fuel system for now, seems to be ok for 10psi of 93 oct. Never really logged the PI and DI pulses.

My IAT were 98F max, low 93F it must have been around 90F outside and i dont have any source of cold intake.

Trying to attach a few logs I've taken.. see if anyone could check out. These are all city drives with the "morning" one having a lot of idle time as i sat in traffic and the rest have some mild throttle applied.

load limit tables look like this

http://www.ft86club.com/forums/attac...1&d=1434676602

Just to be sure, this happens with A/C off? Maybe you said and I missed it...

Stock, A/C definitely sucks away power. :)

You've got DI and PI pulses on the log - if you're stock, you shouldn't have too many transitional issues.

Your logged IAM = 0.2!! and still has (-ve) corrections. You'll need a better tune or fuel. With that IAM value, you'll be running much more of the Base timing table A (as defined in RomRaider) than Base Timing B (which is what most of the tuners adjust as it affects timing when IAM = 1.0).

yep just saw that iam looks pretty ordinary if he is running correct fuel for tune.

side issue do you know the parameters the ecu uses to determine the base timing b to a switching or transition.

i know it runs base b plus knock correction max a when iam is 1

so at what point does the transition occurr to base a plus i assume knock correction max a to determine overall base timing plus offsets due temp iat ect ?

Grossly, what I recall from looking at the disassembly, the IAM acts as an interpolation value between tables B and A:

Base timing = IAM*(Base timing B) + (1-IAM)*(Base Timing A)

IAM*KCA is added to this and then corrections are made to get a final timing value

***
Ignore above...

I was thinking of Primary OL fuel table which progressively adds in fuel from the OL additive table based on IAM.

Just had a look at the dissassembly again and Base Timing tables do not have an obvious IAM modification. The Base Timing A and B tables in A01G are interpolated based on a value stored at FFF8ADB8 - this logs at 1.0 during normal operation (full Base_Timing_B map), and 0.0 (full Base_Timing_A) at startup and shutdown.

Was this done on a dyno? From those logs and how super low the timing is, one can make the assumption you're either running 87 octane fuel or there's something in your engine bay making a ton of racket and making the knock sensor go crazy. The timing map is in various spots makes one wonder. The IAM being super low is a result of the knock sensor feedback -- although it may not mean much as you can tune the timing maps to shift very little if none at all due to the IAM.

You're also seeing stoich AFR's while making boost, and you really should be seeing some enrichment when boost comes in -- the leaner AFR in boost will cause cylinder temps to shoot up which can cause some detonation.

Considering you have wideband, why aren't you running full closed loop?
Enrichment on throttle as well as boost should be there as mentioned.

I will not be loading your car at all with IAM that low, putt around and make sure you have fuel side sorted by getting better fuel to eliminate it from the equation. If it's still happening, you need to get your tuner to do something.

Base A is when the AVCS isn't operational, so when the car is in it's warmup phase ;)

If it's ECUtek, could there be an issue with the hybrid SD? This would be impossible to determine by logs alone though as you would need to see how it's setup.

FWIW, I haven't looked at the logs as it seems you have that covered above.

here is an easier to read log from yommy

http://datazap.me/u/steve99/log-1440...-35-39&solo=33



side issue :-)
so your saying base timing A only get used at startup or limp mode and never used after that, ie their is no interpolation from B to A when IAM drops

ECUtek has the table labelled Base Timing (AVCS Off) so one would assume that's it's purpose.

This was with AC on, its pretty hot lately but the difference isn't really that bad with it off or on



Lol just when I thought it wasn't really a tune situation. Not sure why things were setup this way but its clear its tune related. I'm using 93 oct and usually get it from the same station. This is a remote tune, I wish I could have done a dyno tune. I don't really have many options down here, any advise would be greatly appreciated. I'm on ecutek at the moment.




That's the thing, I was noticing the IAM going a little crazy and that fluctuation is what I probably felt and attribute it to the heat. I wanted to go full time closed loop and that's the reason we started calibrate the WB to read on ecutek custom map O, my tuner was working on it but we never got back to it. Some say its not good in case the wideband fails but then again I think a simple(maybe not so simple) failsafe to prevent from a bad reading or too lean of a condition would be enough. The stock 02 could also go bad just as likely as the wideband in my opinion.



I sent my tuner these same logs and hopefully he's able to pick up what most of you guys are saying. I'd like to give him a chance as I never revised my tune after changing to my GTX28 from a smaller precision. Not really sure how much that could be at fault but im hoping its the case.

I really appreciate your help and feedback if you'd like me to log any parameters and events please let me know. I really just want this thing to drive normal but mostly safe. I'm surprised how the car feels so good at times, like I had mentioned this is probably when all is nice cool which points back to my ignorant assumption it was all a "heat" thing.

Tagging Mike to see if he could shed some light on the subject @moto-mike

car might also have developed some hardware issue ,intake or exhaust leak or sensor fault.

Wouldn't a sensor trigger a cel, I've checked for boost leaks and nothing so car, also checked for vacuum leaks but don't think as I get about 22" vacuum. Let see when mike revises the logs if he finds anything.

We'll chime in here and it is being also followed up with email. The original tune from prior was dead smooth AFR/boost curves and IAM generally at 1, with next to no knock. The turbo was changed and now the datalog that I just looked at looks totally different. We'll have to get to the bottom of this to figure out exactly what's up, but I'd guess we may be seeing some hotter air.

In regards to full time closed loop, we disagree with some of the other voices here regarding that application on the BRZ/FR-S. While some OEMs may use a similar type of strategy, they have failsafes built in that this ECU does not. So if the sensor drifts your AFR will also and you'll never know it. Frankly it is a easier way to tune in bulk since you just let the custom map set the AFR, but the easier way is not always the best way. It is indeed much harder to set the correct MAF scaling and temp compensations manually. Also we leave 6 additional custom maps open when we leave that feature out, which enables us to implement the custom TC, flex fuel support, electronic boost control, smooth rev limit, etc...

You changed turbo without a retune?

Thanks for the closed loop info on the 86.
Forgot about the original ecu limitations as I run standalones on my FI cars. Not sure if ecutek allows change of digital inputs like they do with analog inputs, otherwise a good wideband controller with a fail safe trigger output can be used in a similar fashion to overboost ptotection?

Shouldn't expect a different turbo to work with an old tune... You should have put this info in the top post. Did the tune work fine with your old turbo?

I've compared an SBD "Greddy knock-off" tune with minimal knock living further north than me in QLD where things are really hot. I've got the Greddy T518Z turbo and in the 2800-3800 RPM range, I have to run 6-8 degrees less timing simply because I'm getting higher pressures or hotter air a lot sooner.

Post #30 mentions my turbo changing,it's a long way from #1 I admit. . As stated I wasn't sure it would really make a big difference but again I was wrong. I thought I had mentioned it on the first post. Not pointing fingers as I never doubted my tuner and made that clear. I guess a turbo change would make that difference, we'll see when I get a revised map.

I think that's my current situation as the gtx is pushing air a lot quicker than my old precision.

omg....