5 Liter Eater

As we up our injector sizes there is pretty much a guessing game as to what the new fuel pressure should be or how much to take out with an AFC. I am proposing this method to find the point at which your new injectors act like the stockers out of boost to help out drivability and gas mileage.

What do you guys think of this process for finding out how close you fuel setup is to stock while in closed loop so that the LTFT (long term fuel trim) isn't too out of wack:

1- Start the car and run it to temp, good and hot so the O2 is accurate
2- Shut it off and reset the ECU
3- Set your VAFC settings and/or fuel pressure (do this quickly so the temp stays up and O2 is still semi accurate)
4- Restart car and look at A/F gauge and see which way it goes first, lean or rich, and how long it takes to snap out of that condition. Typically it will go rich since the ECU has reset but it shouldn't stay there too long.
5- Go back to step 2 until when you start it up the A/F is pretty much stoich or goes rich briefly until it snaps back lean.

If your FPR rises 1:1 then the stock map should be correct throughout closed loop/non-boosted operation right?

This process should eliminate any collected fuel trim the ECU has. I just don't know that my logic is right about after an ECU reset the A/F should snap back and forth right away. Any input is welcome.

5 Liter Eater

TTT

DSF

I get what your saying. I really don't have a comment, but just thought I'd share my experience. I'm using the AFC hack with DSM 450's for fueling, along with the stock FPR and stock pump. I have A/F, EGT, and injector pulse gauges (in addition to others). Normally I have my VAFC set at -40% for lo throttle and -30% for high (I'm using the MAP--->TPS mod on my VAFC as well). Just for curiosity sake, while cruising on the highway I was changing my fuel calibration from -45% to -30% and it made no changes on my A/F, EGT, or pulse monitor. Read into that what you will.

EE_Chris

I like your logic.

However, isn't a stock ECU calibrated for clean emmisions? And isn't the ST/LTFT parameters meant to keep the system in a state of clean emmisions?

I say yes.

What does this mean to your original Q....well, if you use the fuel trim parameter to see how close you are to stock a setting...aren't you really running a little too lean?

And I say yes to that one also.

If only we could program a richer target A/F ratio into the ECU, so that it uses fuel trim to keep it around that ratio...I think that would work like a champ....but let me know when that happens. (And that 'idea' kinda sounds like an auto-tune of sorts)

But then again, I've read some theory that for OBD-II, the ECU is programmed to give the Catalytic Converter the needed a/f ratio for effecient operation....which isn't conducive to making power.

All in all, a sound idea....just doesn't seem feasible since the stock ECU wants to keep emmisions clean (from my 'limited' knowledge of ECU operation strategy).


[Modified by EE_Chris, 1:59 PM 4/18/2002]

MiraiZ

You should always calibrate the injectors electronically...go Stand alone...anyone disagree?

DSF

The only way I could see hacking a target a/f ratio in the ecu would be to add a constant resitence to the O2 sensor. That way the ecu would think in't running leaner than it is and provide more fuel. This would only effect closed loop of course. The ecu definitely likes leaner closed loop/cruising a/f ratios. At a constant 75-80mph (@3800rpms) my egt readings would stabilize around 1250 degrees F. No matter what I did to the VAFC, they wouldn't come down. Of course I didn't go as far as running 0% correction. Normal in-town traffic my egt's range from 750-850*F and WOT again around 1250*F (a little on the rich side).

Oh, and of course a "real" standalone system is always a good thing. That's why I'll be running Hondata on my Civic. The CRX is my daily beater. Hey the car was free


[Modified by DSF, 9:19 AM 4/18/2002]

5 Liter Eater

Well, for more power you'd get the wide throttle map tuned on a dyno. What this method is supposed to do is get the bigger injectors as close to the stockers in closed loop to avoid large LTFT values.

I'm not trying to hack at the target closed loop A/F ratio. I'm just trying to come as close to it as possible without the ECU having to correct a lot.

I could see the A/F not changing while changing the VAFC while cruising because the STFT will always want to keep a stoich mixture in closed loop, although I would expect to see it hang out in the rich area more than the lean area until it figured that out. If it has to do this all the time then that correction carries over into the LTFT. Does LTFT effect the pulsewidth in open loop? I think it does. Like if your VAFC is set too rich in the narrow throttle so the ECU learns it has to take out a lot of pulsewidth to stay stoich, then isn't it going to carry that LTFT over into the open loop map? If so then you could dyno soon after an ECU reset and tune it well but after a week or so your LTFT would hose up your perfectly tuned map.

I really wish there were a scantool available that you could drive around with to tell you what your STFT and LTFT are.


[Modified by 5 Liter Eater, 9:48 AM 4/18/2002]

5 Liter Eater

This excerpt is from a C5 Vette forum: http://www.c5forum.com/diy/ltft.php and seems to indicate that STFT has an effect on open loop LTFT.

Long Term Fuel Trim Description
The long termm fuel trim is a matrix of cells arranged by RPM and MAP. Each cell of the long term fuel trim is a register like the short term fuel trim. As the engine operating conditions change, the PCM will switch from cell to cell to determine what long term fuel trim factor to use in the base pulse width equation.

While in any given cell, the PCM also monitors the short term fuel trim. If the short term fuel trim is far enough from 0%, the PCM will change the long term fuel trim value. Once the LTFT value is changed, it should force the STFT back toward 0%. If the mixture is still not correct (as judged by the HO2S), the STFT will continue to have a large deviation from the ideal 0%. In this case, the LTFT will continue to change until the STFT becomes balanced. Both the STFT and LTFT have limits which vary by calibration. If the mixture is off enough so that LTFT reaches the limit of its control and still cannot correct the condition, the STFT would also go to its limit of control in the same direction. If the mixture is still not corrected by both STFT and LTFT at their extreme values, a Fuel Trim Diagnostic Trouble Code (DTC) will likely result. When the PCM determines that the LTFT is out of the operating range, the following DTCs will set:

Here is another from some random website:http://www.gatewaycleanair.com/mecha...2_3/air2_3.htm

Diagnosing Contaminated Ford MAF Sensors
Air flow measurement is a fundamental requirement for any fuel injection system that calculates injection time based on mass airflow readings. The powertrain control module (PCM) determines the amount of fuel delivered by computing a desired air/fuel mixture (A/F) ratio, for example; 14.7/1. If the PCM calculated that there were 14.7 lb/minute of air mass flow, then the fuel injectors would have to deliver 1 lb/minute of fuel mass flow. Precise airflow measurement is critical for proper fuel control.

When the mass air flow (MAF) sensor gets contaminated, it skews the information so that the MAF overestimates airflow at idle and underestimates airflow at higher speeds. This means the long term fuel trim (LTFT) will learn lean corrections at idle and rich corrections at higher airflows.

When a vehicle is driven at wide-open throttle (WOT), or high loads, the PCM normally goes into open-loop. This provides a rich A/F mixture for maximum power. If the MAF is contaminated, the fuel system will actually be lean because of the underestimated airflow. During open-loop conditions, the PCM applies LTFT corrections learned during closed-loop operation. This combination of underestimated airflow and lean fuel trim corrections can result in spark knock/detonation and lack of power complaints at WOT and high loads.



[Modified by 5 Liter Eater, 9:48 AM 4/18/2002]

EE_Chris

Rubbing my magic lamp....Poof: PST-500

I have one, so I know it displays LT & ST Fuel Trim.

But this is what I'm getting at: While cruising around, the ECU uses the fuel trim to keep the emmisions clean. Clean emmisions mean a leaner burn. Leaner burn equals more prone to detonation....I know you know this.

So when you say "I'm trying to come as close to it as possible without the ECU having to correct a lot"....your still running lean. Whatever mods you did to increase LTFT, the ECU is going to compensate to still run cleaner = leaner.

I think one thing that is wise about this method...is to be able to tune IDLE conditions.

I also think that using this method to tune part throttle conditions is wise...to a point. I mean, it would get you into the ballpark....however, after finding these values using a stock ECU + hack devices, your still running lean or to the point your producing clean emmisions. Once you step up the setting for more power, I believe that the ECU is going to use fuel trim to put you right back where you came from.

I'd personnally like to meet up with up someone, who has such devices, and take a look at there fuel trim.

Any guinnea (sp?) pigs?

5 Liter Eater

Well, I'm OBD1 so I can't be a guinnea pig. I can't use most any scan tool they're all for OBD2, but I'd still rather have OBD1.

I want to run lean and clean in closed loop while cruising. The ECU is going to make you do that anyway by using the STFT and LTFT. I'm not worried about detonating in closed loop. When in open loop I will tune my wide throttle map for max power with a dyno. So this method is really only for closed loop, idle and cruising fuel tuning. That's what's cool about the VAFC is that you can tune a cruising map and a WOT map. The point of tuning the cruising/closed loop VAFC map is so the corrections the ECU makes in closed loop don't carry over in to open loop.

So, if a newbie installs a high flow fuel pump and his fuel pressure gets jacked up he might not know it because the ECU is constantly taking out fuel to lean out the mixture to stay stoich while in closed loop. The car runs OK in boost because the pressure is then controlled by an FMU but over time the LTFT will trim out fuel bacause it's always having to take out fuel in closed loop.

Xerxes

You've got some good ideas but it will be hard to get the car to drive just like stock without a standalone.

when you set your vafc for -25% fuel, for example, you aren't getting exactly -25% fuel, because the vafc modifies the MAP signal that the ECU sees, as opposed to modifying the injector pulse width, and that's what makes this so hard.

also i don't think that looknig at the a/f gauge right after turning on the the car is the right way to go, because there will be varying amounts of fuel left over from the starting process that won't neccessarily reflect whether or not you're getting positive or negative STFT. IMO, without a hondata or some other tool that can datalog your actual trim values, this is not possible.


actually i have ONE idea; you can rig up your o2 input to see the exact voltage for a 14.7 a/f ratio. that way your ecu will not adjust anything; then you can look at your a/f gauge which is still hooked up to your actual o2 sensor, and then see if you're rich or lean.

it's ghetto but it's the only way i can think of to do what you want to do.

-Xerxes

5 Liter Eater

You're right about taking the reading at idle, that's not a good indicator since there is a lot of correction there that wouldn't be there while cruising (IAC).

I guess I did start thinking I was talking pulsewidth even though I knew better, but I don't think that matters as much as you'd think; the whole principal of the VAFC hack is moving the fuel map down enough to compensate for the boost and the bigger injectors so as long as -40% (or shatever setting) gets the desired result it should carry over throughout the map. But this method I'm proposing has use even when not use in conjunction with a VAFC. For guys still using say 310's and an FMU it will allow them to find the perfect vac/static pressure for off boost driving and tune the rise rate on the FMU for max power. The VAFC hack works really well as is, I'm just trying to get it even closer.

Sending a false 14.7:1 ratio voltage to the ECU is a really good idea IMO! After doing an ECU reset there should be no STFT or LTFT and that way you are seeing the true richness/leanness of whatever fuel pressure and/or VAFC setting you're at. Such a device would not be hard to rig up. Then just rev to say 3000 RPM standing still in neutral and see where you are. Assuming you keep the same 1:1 rise rate the ECU expects, that setting should be good across the board.

Thanks for the idea Xerxes. I think I will look into making one or maybe one of ZIP's O2 simulators could just be used.

5 Liter Eater

I think I have the best way to tune where the correction should be to avoid any LTFT thanks to ION's thread with link to an article about larger injectors and MAF/MAP manipulation:

I am thinking about putting a toggle switch in-line with the O2 signal just before the ECU, after where the A/F taps in. That way you could just flip a switch and be in instant open loop and tune the entire RPM while cruising, leaving WOT tuning to a wideband on a dyno.

I think it will work well!