Getting fuel trims dialed in with safc/logger..

[Sublime_06]

Okay I got an SAFC-1 and a pocketlogger and I am trying to go ahead and get my fuel trims zero'd out. The only problems is I am not sure if I am going about it properly. I will give you guys idea of how I am doing it and you can corrent me if I am wrong.

I have read a lot of threads and I am still unclear.. I read the definitive piggy back guide to tuning to please dont waste your time posting a link to it.

I understand how to go about setting up the high/low throttle points, the ne points, ect...

From what I gathered reading through a few threads, I am supposed to free rev to the specific rpms as defined in the SAFC. So for example, the 1000rpm point I rev the car to 1000rpm hold it there and watch the ltft and the stft. When I did this for the first time my long term fuel trim was +12% and the stft was around +5%.. So to zero out my trims am I supposed to increase fuel on that specific rmp point until the STFT reads -12%?.. or 0?.. or what? And then repeat for all the other rpm points up to 4k, and then carry that 4k percentage onto the last few ne points?

Thanks guys for any input. Sorry to ask a question thats been asked a few times over.

Cheers,
Chris.

 

[spyderturbo007]

From what I gathered reading through a few threads, I am supposed to free rev to the specific rpms as defined in the SAFC. So for example, the 1000rpm point I rev the car to 1000rpm hold it there and watch the ltft and the stft. When I did this for the first time my long term fuel trim was +12% and the stft was around +5%.. So to zero out my trims am I supposed to increase fuel on that specific rmp point until the STFT reads -12%?.. or 0?.. or what?

You'll want to add or remove fuel with the SAFC until the sum of the STFT and LTFT are as close to zero as you can get. Anything from -5% to +5% is fine.

And then repeat for all the other rpm points up to 4k, and then carry that 4k percentage onto the last few ne points?

Correct. Only because you'll probably never be in closed loop above 4k, so using your 4k point for everything else will be fine.

Thanks guys for any input. Sorry to ask a question thats been asked a few times over.

No problem. Thanks for searching first.

 

[Sublime_06]

Well I went for a cruise today, just found a nice open stretch of road and set the cruise. I started with the 2k and 3k rpm points. My ltft started around 12% at both points and the stft was around 6%. I adjusted the safc until my stft for both the 2k and 3k rpm points was around -12% to -10% give or take.. I just cruised around at those rpms for a while and after driving around both the ltft and the stft started adjusting themselves basically in sync with one another. So now I have my stft at -2% and my ltft at 2%..

Can I assume that I am doing something right? . If so I guess I didn't realize how long it took for the ltft to start to adjust itself.


Thanks again,
Chris.

 

[JayRolla]

Perfect. Leave it like that and drive for a day then go check them. Also at this point I would clear the fuel trims before driving for a day so they can reset and adjust to where you have it tuned.

 

[JayRolla]

Yeap

 

[Sublime_06]

Haha that last post was pretty dumb I already know how to reset the ECU.. stupid quesiton. .

Anyways, I was thinking about getting a Burn2 chip burner or just ordering a chip from Keydiver, since I have a 95 eeprom ecu. Either way, is it a good idea to get a wideband whether I decide to get a chip or not?

Would the safc and wideband be good enough for now? I know I'll eventually have to get a chip once I get bigger injectors.

Thanks,
Chris.

 

[Sublime_06]

One more thing that I am a little confused about. When you say -10% across the board do you mean for the hi or lo settings?

I purchased it 3 years ago and the safc was already set-up. I haven't done anything to the car execept for a full 3" turbo back exhaust. The settings that were plugged in were as follows.

lo-settings.

1k +10%, 2k +10%, 3k +8%, 4k +6%, 4.5k-7k +3%.

hi-settings

1k -10%, 2k -10%, 3k -6%, 4k -3%, 4.5k 0%, 5k +3%, 6k +3%, 7k +3%

I've had my datalogger for maybe a year now and never really used it for tuning. Basically just to clear DTC from my EGR system.. . So I thought I would check out some tuning guides and see how close my trims were after reading through them.

Anyways so after check the trims and trying to zero them out I ended up with the following numbers.

lo-settings

1k +30%, 2k +25%, 3k, +15%, 4k +5%, 4.5k-7k +5%

hi-settings are all the same, i haven't touched those yet.

So do these lo-settings seem really high? Because I keep seeing people mention -10% across the board(which my hi-settings are, more or less), but my lo-settings are way up on the plus side.

Thanks again for all the input.

Cheers,
Chris.

 

[Sublime_06]

The 550's were in the car when I purchased it. I just meant as far as doing mods since I purchased the car. But ya, I guess my settings aren't odd at all really. I was just looking at a few sample settings on the RRE site and one of them is as high as +39 for the 1k rpm lo setting. So I'll just continue doing what I'm doing to get the fuel trims zero'd out. Then I guess I'll start doing some 3rd gear WOT pulls and see what kind of results I get. Maybe you could give me some input when I post those results.

Thanks again for the timely response.

 

[Sublime_06]

As far as being extremely positve such as those figures from RRE's site, from my understanding I do not see any purpose of a positve compensation for a larger injector. I would like to hear some opinions on this though.

Well I'm not sure either but the only way I can get my fuel trims to zero out is by going positive in the lo-settings. :shrug: If I go more negative then the trims get even further apart..

 

[spyderturbo007]

Once the ECU reads the adjusted value (because it really cannot tell the difference anyway) it creates a set of maps for timing and fuel.

Actually, that isn't correct. The ECU's maps are defined and do not change. As sensor readings change, the ECU adjust fuel delivery based on those set maps. All you are doing is molesting the airflow reading which in turn forces the ECU onto a different map, or a different section of the same map.

If you are out there doing some tuning and make some adjustments, you are going to wait for the trims to change. Now you get them right and come out the next day; it's colder out, maybe raining, it's going to throw the whole thing off again and have to start the process over.

The ECU does a very good job of compensating for environmental changes all by itself. Although fuel trims may change a little based on changing environmenal changes, I'm never had to re-tune just because it got colder outside.

After it establishes a pattern, it creates a map. That there are multitudes of values that effect this ( like O2 sensor voltage, engine temp, ect…) but that is the basic principle of it.

Once again, the ECU doesn't create maps, they are predefined.

Based on what your profile says, you have a simple, straight- forward setup. Having 550’s and a front-mount, you should be somewhere around (-10%) across the board, O2 voltage around .96V to .98V, and should be able to safely run 18-19 psi of boost with pump gas.

Your math is a little off here. Mathematically, 550cc injectors are 18% larger than stock, so in a perfect world, your SAFC corrections to hit the target AFR should be -18%.

If you are seeing the ECU pull timing a certain points and it’s a pattern throughout the RPM range, you know it’s compensating for detonation. Good luck man, let us know if you need more help.

Not true. Just because timing gets pulled, doesn't necessarily mean you are knocking. IAT's above 86 degrees and the ECU pulls one degree of timing. Coolant temp gets too high, and you lose another degree. Additionally, as you move up in airflow values, the ECU will move you onto a different section of the map. The higher the airflow, the lower the target timing. So, you can get a timing drop just from moving to a higher airflow value. You really need to look at the timing map and corresponding airflow value, coolant temp and IAT to determine target timing.

Well I'm not sure either but the only way I can get my fuel trims to zero out is by going positive in the lo-settings. :shrug: If I go more negative then the trims get even further apart..

I'm sure it has to do with the enviorment you are in. Maybe something about air humidity, ect...

If you are running larger injectors and have to go positive on the SAFC, there is something wrong. There is no way that injecting 18% more fuel and going 30% positive on the SAFC is normal, nor is it caused by environmental conditions. You need to look at boost leaks, vacuum leaks, shot O2 sensor, etc.

 

[JayRolla]

I agree, you should have to pull a little fuel to get those fuel trims inline. I would zero out the safc and see where the trims go after resetting them. Car should run real rich with it zeroed out.

 

[Sublime_06]

I did a boost leak test and it is solid. Not a single leak. Vacuum is good too.

All my settings on the lo-settings are positive, and nearly all my settings on the hi-settings are negative. The car runs perfect this way..

I dont understand why all the examples given on the RRE site have mostly all the cars with 550cc injectors positive on the lo-settings and negative on the hi-settings.. Your saying that with 550cc injectors I should not have to go positive on my lo-settings?.. These are RRE sample settings with a car that has 550cc injectors...

Low Setting:
1K +39 2K +13 3K + 6 4K +3 5K +3 5.5K + 3 6K +3 7K +3
High Setting:
1K -10 2K -10 3K -10 4K -10 5K -3 5.5K +6 6K +8 7K +8
TH-Point
Lo 30% Hi 80%

 

[JayRolla]

Usually yes. Think about it. You went from 450's too 550's. Roughly 15% more fuel so to get the exact same tune you should just pull 15% from lo and hi. Now the stock afr is around 9.5:1 so pulling more than the 15% is normal at WOT since you want to run around 10.9:1 on pump gas.

But do what your fuel trims need to get close to zero or 100%. I would invest in a WBO2 for tuning WOT. It will take all this guessing game out.

 

[Sublime_06]

Yah I just purchased an AEM Uego it should be here sometime early nextweek.

 

[Sublime_06]

I do have an 95 eeprom ecu.. maybe I should just get a chip already, . Does anyone know if the .bin file that Ceddy posted in the 2g ecu eprom tuning guide will work on a 95 gst? I know his was for a gsx.. not sure if they are interchangeable.

Thanks,
Chris.

 

[->PrOjEcTGS<-]

I wouldn't go leaning it out too much until you get your wideband hooked up. Narrowband sensors are only good at detecting 14.7:1 AFR's. Anything above or below that and it's useless. .98 could be 9:1 while .96 could be 12:1.

Looks like you're seeing some knock by the looks of your timing in the 4500rpm area.

About your lo map needing to be so high. Look at your maf and see if it's been hacked by anyone in the past. The more air that's bypassed around the sensor the leaner it will run and therefore you'll have to richen it up. This has more of an impact at idle. If it was hacked it was probably the previous owner's way of leaning out the tune before they bought the SAFC.

 

[Sublime_06]

Alright, I wasn't going to go too low on the afc, it was pretty high to start with -10% across the board. From my understanding the base for 550cc is -15/-18%?.

I actually haven't looked at the MAF. I was reading through some threads and thought that maybe that was the reason. I guess I'll go take a look at it. Is it okay if the MAF is missing any of the honeycombs?

 

[->PrOjEcTGS<-]

It's not recommended to remove them because they straighten the flow of air so that the sensor can accurately read the air coming into the sensor. I'd leave them in there. If their already out. Then your car will generally run leaner because of it. You should be able to tune around that with the SAFC though.

 

[Sublime_06]

Well I went out, made a few more adjustments and here are my new results. Overall I think its pretty close.

Pull a little fuel at 4k rpm, add a little fuel at 5.4-5.5k rpm, pull a little more fuel at around 6.2k rpm.

What you guys think?.. Man can't wait to get that wideband in.

You must be logged in to view this image or video.

 

[clockworktoy]

To answer your question as to why most of the examples given on road race's site are positive it is because the honeycombs are removed and the adjustment screw is flush on most of them. Im assuming your mass is hacked as well and that is why you have positive settings as well.

 

[spyderturbo007]

Timing looks pretty good in that log but without an airflow value, we can't tell why the timing is being pulled. Next time you make a pull if you add airflow, IAT and coolant temp to the list of logged values so we can correlate actual timing to the ECU's target timing.

 

[Sublime_06]

Hey man I can do that. Wish I got more than 11samples/sec.. Boourns.