RomRaider

Tip-in required is much more dependent on cylinder head temperature than on IAT.
Namely the temperature of the intake wall surface near the intake valves determines the tip-in needed.

Moreover you may use warmed air to improve driveability in winter.

Hi all

Trying to keep this short:

I did E85 conversion on my 2003 EDM 2.0 STi. Based on a healthy RON98 260kw Flywheel HP protune which the car has ran 2 years.
WOT was easy to tune from there for E85 and the car made 295kw@6700rpm Flyhweel 1.50 bar boost. (21.75 psi)
Now I'm in process to tune cold starts and part throttle, temps here in Finland getting to freezing (20°F and below), and ran to a little problem which I think it's Tip-In Enrich related.

When I go quickly from 0% throttle angle to around 50%, rpm 3000-4000 range, the car backfires once just after opening throttle, then clears up and pulls without hesitation.
This happens also below 3000 but you have to look for it to notice, and above 4200 or so its gone. Also if I stab it 100% floor this doesn't happen.
I've tried adding tip-in enrich(goes worse) and zeroing it out (better but not gone). I've been playing with this for a half tank of E85 and coundn't figure it out, so maybe its wise ask from ppl that know better.

If I zero the "Tip-In Enrichment" table is there still some function to additional firing of injectors on throttle angle change? As it can't have negative values
My compensations ("Positive boost error" and "RPM") are also zero. ECT compensation is zero above 80° C. (Ofc My ECT is fully warm when testing 90°C).

Here's a poor phone video. Hard to see RPM but If you drive Subaru you'll know where about it is http://www.youtube.com/watch?v=hv9lVNJMEfI
Good examples of this "backfire" around 0:10, 0:45 and 1:00.

List of mods:
Scooby Clinic SC40 billet turbo (Blouch TD05H-20G based)
External Tial MV-R 44mm wastegate, dumps to downpipe.
Catless 3" Stainless Japspeed (China) turboback exhaust
Custom DIY 76mm big MAF cold air intake, BMC oiled cone Filter. (Yeah I do clean MAF with CRC Maf cleaner regularly, my long term trims are all good within +/-3%)
Bosch EV14 1200cc injectors.
Deatschwerks DW300 pump.
Innovate MTX-L wideband O2.
ECU ID: A4RM000H

Stock: 6mt driveline, intercooler, engine block & internals

Any ideas? I'll probably go out and do some logging/trial/error tune more this weekend to see what CL/OL transitions, AVCS and BPW do at that "backfire" moment.
But I'm more than happy if anyone can point me what to look for.

As promised I tried to find out what's happening but final outcome is: it's still there.

What I tried and tested:

1. AVCS isn't at play on that "backfire" spot, when decelerating avcs is 0 degree advance and isn't fast enough to ramp up when I quickly go ~50% throttle even if engine load goes high enough to ask for it. Even tried with zeroed out AVCS table, no difference.

2.Multiplied whole Tip-In Enrichement table by 2 (or 200%) this way making sure that I wasn't going wrong way.. This widens the problem to even more wide RPM area, very bad at low RPM. Poor throttle change response, stumble, hesitation, smell of unburnt E85 when going wild with throttle at icy parking lot, out of boost mostly.

3. So i ended up with zeroed "Tip-in Enrichment" and compensations at -100% (RPM and positive boost error). With that tune car driveable and not so bad.
But I want to learn and find out what's causing this so here's attached log if you could see something that I can't. I can see from the logs that I go rich on throttle lift off, this could be also part of the problem when lifting and quickly applying throttle again? Would these tables talked here help me: http://www.romraider.com/forum/viewtopic.php?f=25&t=8411? I dont know the adresses for my AR4M000H or how to dig them out.

This is my Tip-In calculated from the RON98 tune: (old injector size) / (new injector size) * (1.3 factor for E85) = 660cc / 1200cc * 1.3 = 0.715 multiplier.
Calculated, no so good

And with these changes its much better:
Zeroed out


edit: Some relevant info about my tune (cannot post whole rom due to commercial "base"):
"300 g/s MAF hack" (0.75 load scale) 76mm DIY MAF, Max flow 250 g/s = 4.48 volts on my scaling.
Injector flow scaling = 550
Injector latency at 14.0v = 0.95 msec.


To be continued..

I've logged it before on my 02 wrx.

I have a question for you all. I got my tip in close while driving. AFRs much closer before I started tweaking it. It used to swing way rich then over shoot the correction and go lean. Now it is only a few points rich.

But I am having a problem with it still. If I rev, the AFR shoot to 10. How do I make tip in not activate or compensate if I were to just rev? Otherwise while driving (with a load) it is fine.

I've noticed a difference between tip-in with little RPM change to tip-in with high RPM change too. Maybe there's some RPM related enrichment table.

My problem is that it tends to swing rich (~12. and then go lean (~16). It does this in open-loop with AVCS off, so it should not be over-compensation or reversion. Any ideas? The only thing I can think of is that it fires the tip-in enrichment early in the throttle opening and then stops. When I log Tip-In Enrichment (last calculated) it shows that it will not change consecutively more than three times, even if the throttle changes another ~10%.

I have tried changing all of the counter thresholds and increasing the cumulative throttle threshold, but it still does not seem to fire the tip-in routine more than three consecutive times. Anything I'm missing here? The attached log is in closed-loop, but it does the same thing in open loop. It goes rich, then lean.

What injectors do you have? And describe how your TGVs are modified if they are.

I see the same thing with my chopped stock injectors. My theory is that because there isn't much of a spray pattern but rather a straight beam of fuel coming out, the tip-in squirt goes into the combustion chamber without wetting the walls much. I think the ECU is expecting the squirt to only have some of the fuel go in at first but the rest to 'dissolve' off the walls later.

Now if you tell me you have good injectors in and not my hackjob ones, then my theory goes out the window.

My injectors are Deatschwerks 750s (modified blues). TGVs are deleted. I think you might be on to something with your theory. I played with my applied reset counters more (set them 1 less than the applied disable) and was able to get the TIE to fire almost constantly during throttle change. Once I got that to work, it started going way too rich. I'm currently messing with the reset counter to see if I can optimize the timing to space them out just right.

I have spent a lot of time adjusting the additional pulse width like everyone says and in my case that does not appear to be very effective. I'm now finding that changing the counters (disable and reset) is MUCH more effective than changing the additional pulse width. I'm just using the additional pulse width as calculated by ratioing stock values with my injector scalar. I'll confirm on my way to work tomorrow, but a disable counter of 7 (stock 06 STI) with a reset of 14 might just be as good as it gets.

I wish I had a stock car handy for comparison, but I agree with your theory. If I run the counters like stock, it sprays a few times and then waits even though the throttle could still be changing another 30%. It appears to me the stock setup expects the fuel spray to be effective over a longer duration (~2x as long) than I'm seeing.

Yep, you have the ~same s*** injectors as I do. Although, mine are even worse because they have no screen on them. I've tuned yours before and they were almost as bad as mine.

Here are 4 graphs showing the behavior. The top 2 are OL (A/F Correction = 0) and the bottom 2 are CL. Notice they both have that same rich initial dip. I can accept that happening in CL if it overcorrects or something but the AF sensor shows the same exact trend even in OL.


I played around with increasing the Tip-In DACT table before and went as high as 16 on the former from a stock of 4. It definitely seemed to keep injecting during longer periods of constant throttle change. I guess I never really tried decreasing the Reset Counters. Good idea. I was also going to try halving that table from a stock value of 29.

What should I log to find out what cell the tip in enrichment table is using? I tried logging tip in throttle, that never got above 8% even though I was stabbing the throttle more than 8%, the table goes all the way up to 30%. I never see above 8%

Does anyone else see when logging throttle tip in %, no matter how big the throttle blip is, it is very small changes? Because of this, I have no idea what throttle cell the tip in table is being used..

i believe that part of the issue is the speed of the logger and the duration of the tip in register.

ie, it spikes higher, but you have to catch it at just the right moment... similar to knock sensor output.

Has this been figured out yet, I still have tip in problems with ID1000's ,apexi intake and fmic.

anyone?