S-AFC Troubleshooting

5805 Views 32 Replies 4 Participants Last post by Warrtalon, Apr 21, 2005

Warrtalon

Discussion Starter · Apr 19, 2005

I installed my S-AFC and thought it was working, but I'm seeing no change in LTFT at 1k idle when moving the low-throttle 1k S-AFC settings up and down. I'm afraid that maybe my pink/orange MAS wire splice wasn't done properly and is not actually modifying the A/F. Moving the 1k low-throttle settings up and down 20% made no change in idle characteristics or in exhaust smell (never smelled rich).

Does anyone know of any troubleshooting methods to determine if the S-AFC is actually modifying the MAS signal?

PS. I sent a much longer explanation directly to turbolarry just in case.

1 - 20 of 33 Posts

turbolarry

· #2 · Apr 19, 2005

Got your PM, and I hope you don't mind that I copy some of it here for others to reference in the future too;

Warrtalon said:
This is where the question arises. When I got to the part where the logger was connected and reading the RPM/LTFT/STFT, I noticed that my LTFT was 1 at idle (850 rpm). I expected it to be off by more, but I guess this is a good thing, since you say to adjust the AFC until the LTFT is within +/- 3. At 2k rpm, the LTFT was still 1, and at 3k rpm (still sitting still), the LTFT was 4. So, I tried adjusting the S-AFC to get the LTFT to 0 at each RPM, but no matter how much I changed my S-AFC settings, the LTFT would _never_ change. I was waiting several minutes to make sure, but nothing was happening. I went as far as +/- 20% at 1k rpm on the low-throttle setting, but it didn't seem to have any affect. I figured the LTFT would change, or at the very least, my idle would become rough, increased, or decreased...strangely, nothing noticeable happened. This leads me to believe that my S-AFC is set up properly except for the MAS signal. I know that I have the ping/orange wires setup properly, but I suppose it's possible that the connections aren't 100%.

Do you have any suggestions on how to ensure that the S-AFC is actually modifying the MAS signal? My power, ground, RPM, TPS, and air flow signals all seem to be right on, so I'm pretty sure I did everything right. I just want to make sure my S-AFC is actually modifying the MAS signal before I move on with my tuning. Bear in mind that I used some conservative settings for the 5-7k rpm range (-3% increasing to -9%), since I know that I'm running extremely rich after 5200 rpm (dyno chart with A/F dropping below 10). The results of this SEEMED to make a NOTICEABLE difference in power, especially on the top end of each gear. The car used to flatten out in the 5k range, then pull near 7k, but now it just goes from 4-7500 without missing a beat. This could all be my imagination, but I was pretty sure my S-AFC settings were making a difference.

It's more than likely it's intecepting the signal and altering it just fine. If it wasn't, your car would be running horribly. Just at idle, with no load, it hasn't reached the point of changing the LTFT. Dive it around for a day or two and it will probaly change.
If you really want to see it's effects and make sure it's working;
Set your logger to monitor front O2's (important for this). Turn the car on and let it idle.
On the AFC, slowly work your way lean, you'll start to hear misfires (popping sounds). A good point is probably where things just start to pop a little but are pretty close to "smooth" (rich). Watch the O2 sensor voltage. Normally the voltage will swing back and forth between .2 and .8 volts at idle. As you adjust leaner, the ECU will compensate back richer. Slowly click down leaner, you will then see the O2 voltage come back up slowly. This is the ECU re adjusting. Keep going down until the ECU can no longer bring the voltage back up to .2 to .8volts.

That's the old way of figuring out your low settings by only monitoring O2 voltage, but it works and will show that your AFC is working.
http://www.roadraceengineering.com/newafc.htm

See less See more

Warrtalon

Discussion Starter · #3 · Apr 19, 2005

I am now pretty sure that the AFC is working, because when I modify it up and down, the STFT changes accordingly. The thing I'm confused about now is whether I'm supposed to be adjusting just the LTFT reading or if I'm stupposed to always add the STFT and LTFT and try to get that sum to equal 0. For example, at 1k rpm, my LTFT is at 1, but my STFT fluctuates from -4 to -1. So, I leaned it 1%, which made the STFT fluctuate from -2 to 0. If I add this to the LTFT, I am centered right around 0. Is this correct, or am I just supposed to make the LTFT drop to 0? Strangely, after a long while of messing with it, my LTFT eventually dropped to -1, but I'm not sure why.

Now, if I am supposed to always use the sum of the FTs, then I'm really confused at 2k rpm. At 2k rpm idle, my LTFT is 4, but my STFT is as high as 10. I have to RICHEN almost 20% just to get the STFT down to -4 so as to even out the +4 LTFT and get a sum of 0. I don't understand why I would need -1% at 1k, but then +20% at 2k. Then, at 3k, I have to set it at about +6% to even out. With such a crazy change in numbers over the 3 rpm ranges, I have no idea what figures to carry forward to 7k (as suggested in the guide). I could just go and try to tune the hi-throttle, but I really wanted to get the fuel trims set perfectly before doing so.

Any suggestions? Oh, also, the LTFT is completely different at day and night.

See less See more

turbolarry

· #4 · Apr 19, 2005

Warrtalon said:
The thing I'm confused about now is whether I'm supposed to be adjusting just the LTFT reading or if I'm stupposed to always add the STFT and LTFT and try to get that sum to equal 0. For example, at 1k rpm, my LTFT is at 1, but my STFT fluctuates from -4 to -1. So, I leaned it 1%, which made the STFT fluctuate from -2 to 0. If I add this to the LTFT, I am centered right around 0. Is this correct, or am I just supposed to make the LTFT drop to 0? Strangely, after a long while of messing with it, my LTFT eventually dropped to -1, but I'm not sure why.

Your doing it right. STFT's will fluxuate. First just try and keep your LTFT's within +/-3%. When they stay within that range for a period of time, leave it. Your STFT's should be closer. Are you still on stock injectors?

Warrtalon said:
Now, if I am supposed to always use the sum of the FTs, then I'm really confused at 2k rpm. At 2k rpm idle, my LTFT is 4, but my STFT is as high as 10. I have to RICHEN almost 20% just to get the STFT down to -4 so as to even out the +4 LTFT and get a sum of 0. I don't understand why I would need -1% at 1k, but then +20% at 2k. Then, at 3k, I have to set it at about +6% to even out. With such a crazy change in numbers over the 3 rpm ranges, I have no idea what figures to carry forward to 7k (as suggested in the guide). I could just go and try to tune the hi-throttle, but I really wanted to get the fuel trims set perfectly before doing so.

If at 2K your getting a LTFT of 4 and a STFT as high as 10, it's just the ECU saying; "I regularly need to add 4% more fuel than normal." "At that one moment I had to add 10% to keep things at 14.7."
Target the LTFT's first. If the ECU says +4%, help it out and give it 3% at that RPM. Do a lttle at a time and watch the changes. Do some lite cruising at that RPM and watch for changes. It took me three days to get mine within +/-3%.
Trust your logger. If at 2K it says add 10%, add 10%. For some reason I had to richen it up a lot too at 2K.
As far as what to carry forward to 7K, use your same setting that's at 4K RPM

Warrtalon said:
Oh, also, the LTFT is completely different at day and night.

Mine does the same thing. Temperature plays a roll. Keep playing with the settings and shoot for that middle ground.

Warrtalon

Discussion Starter · #5 · Apr 20, 2005

Well, now I'm back to thinking my S-AFC isn't doing anything. I had begun to think that my STFTs were changing according to my S-AFC, but after playing with it for a long time, I'm not so sure. I can put the S-AFC at +40% or -40% at 1k rpm at idle and not feel even the SLIGHTEST change in the sound/speed/smoothness of the idle. The idle should change noticeably as I lean and richen, but there is no change. Likewise, I can do a long 3rd gear pull with my S-AFC set to 0% at every rpm in hi-throttle and get what look like regular o2/timing numbers across the board. Then, I can set my S-AFC with some leaner settings after 5k rpm and not see a single change in the o2s or timing.

So, I've removed my patch harness and am inspecting the wiring, but from what I can see, I did it exactly according to your guide (except the rpm pin), and exactly according to the DSM vfaq. Everything else on the S-AFC seems to work except for the actual fuel modification. I even heat shrinked these connections, so it shouldn't be a matter of having a poor connection. Oh well.

Hey, any idea why my O2s don't read very rich after 5k rpm in 3rd? When I dyno'd, my AFR literally dropped off the chart (<10) after 5900 rpm after starting its descent at 5200rpm. I thought this indicated extreme richness, but then my O2s are mostly .91v from 4500-6500, then go to .93 from there to redline. I thought .91-.93 was on the lean side, whereas I expected to read rich. Any ideas?

See less See more

turbolarry

· #6 · Apr 20, 2005

At idle (1K), does your O2 voltage cycle normally at -40%?

Warrtalon

Discussion Starter · #7 · Apr 20, 2005

Yes, no matter how far I change the fuel up or down at idle, the o2 always cycles between .2 and .8. AFTER the ECU adjusts for a few seconds. Last night, I re-checked everything, then re-installed the AFC, reset the ECU, initalized then reset the AFC settings, and then cranked it up. As it was warming up, if I leaned out the fuel at 1k rpm, the car would start to stutter and pop (what I expected), then if I went rich, it would smooth out and raise the RPM.

Ok, so this told me that the S-AFC was DEFINITELY working at first, but then once the car got fully warmed up, I found that I could go all the way to -50 without the car idle stuttering or popping, so I got really confused. That's when I tried going from one extreme to the other quickly. Well, when I went from -50 all the way up to +20, the idle change noticeably and the O2s were spiked. I then did the opposite thing...I went from +20 all the way back down to -50, and the o2s bottomed out while the car nearly stalled. What happened each time, though, is that the ECU corrected itself and brought the idle back to normal each time as well as the o2s. Now, how is this possible? The tuning guides I've read have said that you should go lean until the exhaust starts to pop and start tuning from there, but my car keeps correcting itself on the fly. It takes a few seconds, but the idle always goes back to normal, and the o2s always return to the normal cycle.

I then did a 2nd gear pull (simply can't afford to do a 3rd gear pull anywhere late at night) with the S-AFC set to 0% across the board, and I finally saw what I expected: .95-.98 from around 4500-7500rpm. I then leaned out the settings from 5-7k and ran again...this time everything was .91-.93 in succession except for one dip from .93 to .91 then back to .93. I decided to richen slightly at this rpm (~6200) to compensate. Timing also rose gradually into the high teens until it maxed at 20 at 7200rpm (sounds good). It was midnight, and I was happy to at least see the S-AFC doing something, so I called it a night. The readings could be very different on a 3rd gear pull, but I just couldn't do it yet.

Strangely, my Karman peaked at 1589 during both runs, so I don't know if I'm getting any benefit or not. When I first put on the S-AFC and gave it similar settings, I was hitting 1612 (7400rpm) in 2nd, 1632 in 3rd (7400rpm), and 1659 in 4th (6900rpm). Last night was considerably warmer than when I got those initial readings, but the main issue is that the peak air flow didn't change despite the better o2 readings with leaner settings. Oh well, I guess I will only know on a dyno...

See less See more

ldstang50

· #8 · Apr 20, 2005

Here is what I can think of about your start up and why the S-afc works then doesnt' work. You have the safc to start 'changing' fuel at 1000rpm. When you first start up the car, she idles around 1300 or so rpm, then once warmed up, she is at 800rpm. When on initial start up, the idle is within the safc parameters to change the fuel settings, but once warmed up and idling normal, the rpms fall below where the safc monitors. Thats my take on that.
Quick question, why do you have the safc set to start thinking at 1000rpm? That is really low. That takes out from the rpms you can tune at. I have mine set at 2500, 3500, 4000, 4500, 5000, 6000, 7000, 7500. This allows me to fine tune when my turbo starts spooling at 3300rpm up to full boost when it hits at 4000rpm and i can fine tune the high rpm when i have the most rpms, and air flowing in. To me the 1000rpm is pointless since you aren't at boost down there and the computer monitors and adjusts everything just fine.

Warrtalon

Discussion Starter · #9 · Apr 20, 2005

ldstang50 said:
Here is what I can think of about your start up and why the S-afc works then doesnt' work. You have the safc to start 'changing' fuel at 1000rpm. When you first start up the car, she idles around 1300 or so rpm, then once warmed up, she is at 800rpm. When on initial start up, the idle is within the safc parameters to change the fuel settings, but once warmed up and idling normal, the rpms fall below where the safc monitors. Thats my take on that.

Not a bad idea, but I didn't start messing with it until the RPMs had settled to below 1k. Even though I idle at 850, the 1k setting still adjusts the fuel input for 850...it doesn't just work at that exact rpm.

ldstang50 said:
Quick question, why do you have the safc set to start thinking at 1000rpm? That is really low. That takes out from the rpms you can tune at. I have mine set at 2500, 3500, 4000, 4500, 5000, 6000, 7000, 7500. This allows me to fine tune when my turbo starts spooling at 3300rpm up to full boost when it hits at 4000rpm and i can fine tune the high rpm when i have the most rpms, and air flowing in. To me the 1000rpm is pointless since you aren't at boost down there and the computer monitors and adjusts everything just fine.

Well, I'm talking about my low-throttle settings. You are supposed to initially setup the S-AFC by tuning the low-throttle. You do this by getting the LTFT right at 0 for 1k, 2k, 3k rpm, then you extrapolate for the later rpms. Then, for hi-throttle, you take these base settings from low throttle then adjust accordingly. I'm just trying to get this initial tuning part done, but it's not working out. Luckily, though, my LTFTs as of last night were 0 for 1k, 2k, and 3k rpm. I know what you mean by changing the ne-points for higher rpm so that you have more tuning. I have already changed my hi-throttle points for this exact reason, as I agree that it gives me more tuning points in the rpm ranges that need it most.

turbolarry

· #10 · Apr 20, 2005

Warrtalon said:
The tuning guides I've read have said that you should go lean until the exhaust starts to pop and start tuning from there, but my car keeps correcting itself on the fly. It takes a few seconds, but the idle always goes back to normal, and the o2s always return to the normal cycle....

The guides were writen with larger than stock injectors in mind. Are you still on stock injectors?
If your are, try leaving the low settings at 0%, let the ECU handle the closed loop stuff, and just concentrate on high throttle.

ldstang50 raises a good point that I never thought of and I think this is a great idea. :thumb:
If you are on stock injectors, why not move the Ne points to a higher range? What about;
3K, 4K, 4.5K, 5K, 5.5K, 6K, 6.5K, 7K
For people with an AFC, stock injectors, and targeting high throttle tuning this is a great idea.
ldstang50 do you run like this?

ldstang50

· #11 · Apr 20, 2005

Warrtalon said:
Not a bad idea, but I didn't start messing with it until the RPMs had settled to below 1k. Even though I idle at 850, the 1k setting still adjusts the fuel input for 850...it doesn't just work at that exact rpm..

Right. The afc will make 'calculatated' adjustements when you are between your ne points. Ex. between 3000 rpm and 4000 rpm with your 3000 set at +17% and your 4000 set at 20%, the afc will automatically make the adjustments for all rpms between the two and adjust the fuel accordingly.

Warrtalon said:
Well, I'm talking about my low-throttle settings. You are supposed to initially setup the S-AFC by tuning the low-throttle. You do this by getting the LTFT right at 0 for 1k, 2k, 3k rpm, then you extrapolate for the later rpms. Then, for hi-throttle, you take these base settings from low throttle then adjust accordingly. I'm just trying to get this initial tuning part done, but it's not working out. Luckily, though, my LTFTs as of last night were 0 for 1k, 2k, and 3k rpm. I know what you mean by changing the ne-points for higher rpm so that you have more tuning. I have already changed my hi-throttle points for this exact reason, as I agree that it gives me more tuning points in the rpm ranges that need it most.

Right, I don't know what mine is set at as I haven't touched it in a while, plus my car is torn apart at this moment. You are running stock injectors and stock turbo correct? Do you have a boost controller? I know with mine, I was initially tuned to lean out the fuel, I was wondering if you were doing the same.

turbolarry

· #12 · Apr 20, 2005

Warrtalon said:
Well, I'm talking about my low-throttle settings. You are supposed to initially setup the S-AFC by tuning the low-throttle. You do this by getting the LTFT right at 0 for 1k, 2k, 3k rpm, then you extrapolate for the later rpms. Then, for hi-throttle, you take these base settings from low throttle then adjust accordingly. I'm just trying to get this initial tuning part done, but it's not working out. Luckily, though, my LTFTs as of last night were 0 for 1k, 2k, and 3k rpm. I know what you mean by changing the ne-points for higher rpm so that you have more tuning. I have already changed my hi-throttle points for this exact reason, as I agree that it gives me more tuning points in the rpm ranges that need it most.

If your on stock injectors let the ECU handle the low stuff and use ldstang50's idea. If I was running stock injetors I'd give it a try.

ldstang50

· #13 · Apr 20, 2005

turbolarry said:
For people with an AFC, stock injectors, and targeting high throttle tuning this is a great idea.
ldstang50 do you run like this?

Larry
When I first put in my s-afc i already had 700cc, 255lph, intake, mbc, exhaust and dynoflash, so my settings are pretty irrelevant.
I did however set my s-afc to compensate at higher rpms as the dynoflash takes care of my idle and low rpm running.
I did very similar tuning methods but i didn't have an o2 volt gauge.
The car itself is set to run a little rich just to keep it on the safe side from the factory. However at higher rpms because the boost tapers a little bit, mitsu leaned out the fuel so the car was closer to 11.5. But if you are running an MBC and it holds, thats one area (above 6000 rpm) where the s-afc would really help. I'm sure running 21psi at 6500 rpm vs 16psi you are pushing more air so the air fuel will be different.

turbolarry

· #14 · Apr 20, 2005

ldstang50 said:
Larry
When I first put in my s-afc i already had 700cc, 255lph, intake, mbc, exhaust and dynoflash, so my settings are pretty irrelevant.
I did however set my s-afc to compensate at higher rpms as the dynoflash takes care of my idle and low rpm running.
I did very similar tuning methods but i didn't have an o2 volt gauge.
The car itself is set to run a little rich just to keep it on the safe side from the factory. However at higher rpms because the boost tapers a little bit, mitsu leaned out the fuel so the car was closer to 11.5. But if you are running an MBC and it holds, thats one area (above 6000 rpm) where the s-afc would really help. I'm sure running 21psi at 6500 rpm vs 16psi you are pushing more air so the air fuel will be different.

Your settings maybe, but the basic idea is great. :thumb: You don't understand, but for years it's been drummed into my head;
AFC + bigger injectors + 1K, 2K, 3K...
That's how it's always been for me so to hear this idea, it's like WOW! Let the ECU handle the low stuff, shove the Ne points into the higher RPM's, and concentrate on high throttle tuning. I think that a lot of Evo people will adopt this.
Do you mind if I put this in the tuning guide? A little paragraph for AFC owners on stock injectors.
:thumb:

ldstang50

· #15 · Apr 20, 2005

turbolarry said:
Your settings maybe, but the basic idea is great. :thumb: You don't understand, but for years it's been drummed into my head;
AFC + bigger injectors + 1K, 2K, 3K...
That's how it's always been for me so to hear this idea, it's like WOW! Let the ECU handle the low stuff, shove the Ne points into the higher RPM's, and concentrate on high throttle tuning. I think that a lot of Evo people will adopt this.
Do you mind if I put this in the tuning guide? A little paragraph for AFC owners on stock injectors.
:thumb:

Ya thats fine.
I was the same way when first started with the afc. The NE points vary from car to car. The 05s might need to have different NE points because of the 10.5 hotside. It also spools up little quicker too so that would mean you would want your NE points lower. Still at 2500 rpm for your first NE, that gives you plenty of range to tune with.
Have you heard of the blue wire mod?

ldstang50

· #16 · Apr 20, 2005

turbolarry said:
If your on stock injectors let the ECU handle the low stuff and use ldstang50's idea. If I was running stock injetors I'd give it a try.

Even with my 700cc the car idled fine, it just ran like crap.
I'm not sure how big of injectors are too big to handle for the stock ecu, but I'm pretty sure when you are getting the 850cc and above range, you will have an EMS or another form of standalone to run those monsters.

Warrtalon

Discussion Starter · #17 · Apr 21, 2005

turbolarry said:
The guides were writen with larger than stock injectors in mind. Are you still on stock injectors?
If your are, try leaving the low settings at 0%, let the ECU handle the closed loop stuff, and just concentrate on high throttle.

Ugh, really? I never knew your whole tuning guide was meant for larger injectors. Damn, no wonder...the ECU is easily keeping the 1-3k idle LTFT at 0 all the time, so I actually have all my low-throttle settings at 0%.

turbolarry said:
ldstang50 raises a good point that I never thought of and I think this is a great idea. :thumb:
If you are on stock injectors, why not move the Ne points to a higher range? What about;
3K, 4K, 4.5K, 5K, 5.5K, 6K, 6.5K, 7K
For people with an AFC, stock injectors, and targeting high throttle tuning this is a great idea.
ldstang50 do you run like this?

Yeah, I had already sort of done this by scrunching my NE-points between 4500-7k while leaving the 1-2k spots for the tuning guide. Now that I have a wonderful fuel trim, I can re-do my ne-points again to go from 4-7500. Despite the 10.5cm hotside, I don't really get any action until almost 4k. Remember, I have 0 mods other than this AFC, and the only reason I have it on the car is because I had it left over from my previous DSMs. I was VERY curious to see how much hp I could release from Mitsu's prison of fuel after 5200 rpm.

My stock dyno run showed a significant dropoff in AFR at 5200 rpm, where it went from ~11.5 to 11.0, then at 5800, it fell off a cliff and went below 10. This is where I've targetted my S-AFC changes so as to get those AFR readings back to 11.5 all the way to 7500. I won't be able to get to the dyno again until June, but until then, I'm trying to use the logger to tune my o2 to .93v at every rpm. At the moment, it looks good...I got 5k-7200 all at .93v at the moment, so I'm happy. Now, I'd like to see how much more hp I can find in stock form (e.g. no exhaust, mbc, intake, fuel pump).

See less See more

turbolarry

· #18 · Apr 21, 2005

ldstang50 said:
Have you heard of the blue wire mod?

Ya.
On the AFC we do not use the blue wire, but it can be spliced into the O2 sensor. From there you go into Etc. (the AFC's top screen) choose sensor check and you should be able to some what monitor O2 voltage (scroll down);
http://www.roadraceengineering.com/newafcsetting.htm
It is generally believed that the "blue wire mod" degenerates the O2 sensor's signal to the ECU. I have never tried it so I can't say that it does. But I don't think anyone has concrete evidence either.
http://www.dsmtuners.com/forums/sho...t=blue+wire+mod

Warrtalon said:
At the moment, it looks good...I got 5k-7200 all at .93v at the moment, so I'm happy. Now, I'd like to see how much more hp I can find in stock form (e.g. no exhaust, mbc, intake, fuel pump).

:thumb:

Warrtalon

Discussion Starter · #19 · Apr 21, 2005

turbolarry said:
It is generally believed that the "blue wire mod" degenerates the O2 sensor's signal to the ECU. I have never tried it so I can't say that it does. But I don't think anyone has concrete evidence either.

I used to do the blue wire mod on my DSMs, but I didn't bother on the Evo, due to the rumors. I have the logger to track O2s, so I don't need to anyway.

ldstang50

· #20 · Apr 21, 2005

Here is my dyno after my most recent round of mods. You can see where the torque and hp numbers meet. Where the hp numbers really start to climb is where the turbo is starting to spool. This is on the stock turbo, stock hotside, stock exhaust mani, stock intercooler and few other things stock.
You can see the air fuel curve how smooth it is, this is a Dynoflash here.
Where the turbo starts spooling is where I have my afc fine tuned at. This should give you a rough estimate for turbo spool. Remember I was running 30psi at the time also

See less See more

1 - 20 of 33 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.