yep, lean cruise i shoot for 16.2:1, anything beyond that and lean misfires can happen due to the variability of the fuel around here. running that high, i managed 37.6MPG on a ~450 mile trip this summer. not bad for a 21 year old midsize car that has since been retired.
PE: 11-12:1 for boost, 12-12.5:1 N/A, depending on the exact engine. there are always exceptions to the rule though.
and NEVER a PE delay.
1995 Chevrolet Monte Carlo LS 3100 + 4T60E
Well, you need to get your head clear on this. 14.7 is a ratio that makes the best balance of what comes out. It is not the optimal ratio for making the engine run. As engineers learn more about combustion they're figuring out how to make the engine run better on leaner mixtures, but many engines run best at mixtures richer than 14.7. Old stuff like the 304 and the IH 345's like a mixture around 12:1. The old GMC 60 degree V6 truck family had huge bores and liked mixtures as low as 10:1. Look through forum after forum and you'll find questions like "My engine runs great in open loop but when I enable closed loop and try to run at 14.7 it's got problems." It takes a little more work to make an old engine run at a leaner mix.If 14.7 is the optimal raito, then why would you want to go to something diffrent?
Not all fuel is burned in any cylinder and a richer mixture is used to get more fuel and more oxygen together so they're likely to burn. Unused fuel is wasted but the total amount of fuel burned (and power made) is higher than at 14.7:1. Larger bore engines with tall piston tops and large amounts of cast iron tend to keep fuel from burning at the outer areas of the piston and cylinder. Lean mixtures can burn well but the temperatures produced tend to make NOx which is harder to clean up than HC and CO.
I've got to run, but if you look around for "rich best power" , "lean best power" and "BSFC" you might get some good reading.
Maybe another way of saying the same thing is that 14.7 is the optimal ratio in chemistry, not necessary the optimal ratio for the engine.
Yeah, really! PE delay, How Rude!Originally Posted by RobertISaar
and NEVER a PE delay.
If I put my foot in it it's becaue I want power. Why would I want PE delay? That goes back to OEM priorities, not mine...
1990 Chevy Suburban 5.7L Auto ECM 1227747 $42!
1998 Chevy Silverado 5.7L Vortec 0411 Swap to RoadRunner!
-= =-
My Camaro idles real smooth and cruises nice at around 15.5:1.
1978 Camaro Type LT, 383, Dual TBI, '7427, 4L80E
1981 Camaro Z-28 Clone, T-Tops, 350/TH350
1981 Camaro Berlinetta, V-6, 3spd
1974 Chevy/GMC Truck, '90 TBI 350, '7427, TH350, NP203, 6" lift, 35s
In my Datsun, I have it tuned (no highway lean cruise mode), to run at between 15 and 16:1 while at a steady state cruise.
The man who says something is impossible, is usually interrupted by the man doing it.
How do you do that? So it must be open loop? Is that a Code $59 thing?
1990 Chevy Suburban 5.7L Auto ECM 1227747 $42!
1998 Chevy Silverado 5.7L Vortec 0411 Swap to RoadRunner!
-= =-
Yes, I'm running open loop, no, not just a $59 thing. I seem to get better milage and drivability in open loop, though, I only tried closed loop briefly, I decided that I wanted to get the VE tables and such in better shape before enabling CL again.
The man who says something is impossible, is usually interrupted by the man doing it.
Hopefully not to get too far off topic, but I have always wondered exacty why does a lean mixture make a engine run hot?
79 Jeep Cherokee, AMC 401, T-18 manual trans, hydroboost, 16197427 MPFI system---the toy
93 Jeep YJ Wrangler, 4.0L, 5 speed, 8.8 rear, homebrew hub conversion and big brakes, hydroboost, 2.5in OME lift, 31x10.50's---the daily driver
99 Jeep WJ Grand Cherokee limited, 4.0L, auto, 2wd, leather and power everything, 99% stock---the long distance highway ride.
You know you ask questions that drive me nuts sometimes because they are so simple I can't answer them! It has also caused some of the best learning threads we have had! Just so you don't misinterpret that, it is a good thing.
Lean in itself does not make an engine hot, certain leaner than stoich mixtures can actually cool an engine because of there lean condition leave a thin barrier against cylinder walls and when ignited the extra heat from more oxygen and less fuel never hit the walls. Combustion happens in mid air if you will. More air (oxygen) then fuel is what causes leaner mixtures to be hotter. This can be measured in Exhaust Gas Temperatures (EGT). So exhaust valves will be the first to go if to lean. People think the pistons will melt a hole. That is not true, that is caused by detonation when a mixture is to lean for to long (or other reasons) and fuel is ignited before the spark plug fires, like ignited from a hot exhaust valve or glowing tip of spark plug, then you have a two way flame front in combustion chamber reaching full combustion way before piston has compressed, like knock because of to much timing on steroids.
Richer, more fuel has a cooling effect from more fuel evaporating or actually wetting the walls of cylinder which is to rich because you are wetting rings and things with fuel taking away the oil protection. Ever have a carb engine and the oil smells like gas? Ever notice proper tuned EFI engine oil stays cleaner longer and does not smell like gas?
1990 Chevy Suburban 5.7L Auto ECM 1227747 $42!
1998 Chevy Silverado 5.7L Vortec 0411 Swap to RoadRunner!
-= =-
I have been accused of asking lots of questions before...lol
I just like knowing as much as I can, even if my memory only lets me remember so much of it.
79 Jeep Cherokee, AMC 401, T-18 manual trans, hydroboost, 16197427 MPFI system---the toy
93 Jeep YJ Wrangler, 4.0L, 5 speed, 8.8 rear, homebrew hub conversion and big brakes, hydroboost, 2.5in OME lift, 31x10.50's---the daily driver
99 Jeep WJ Grand Cherokee limited, 4.0L, auto, 2wd, leather and power everything, 99% stock---the long distance highway ride.
Emissions are affected greatly by how close to stoich you can keep. I can't overemphasize how greatly. I generally calibrate using lambda, so lambda = 1 is stoich, below 1 is rich. Anyways.... catalyst efficiency is greatly affected by the averaged lambda going in. On the inlet of the catalyst, assuming that the fuel in question wants a lambda of 1.0 for best conversion efficiency, I like to see lambda staying from between 0.96 and 1.04, but with an average of 1.0000. If you run an average lean shift of 1.002 what will happen is that the conversion first looks good, then after a period of between 20 and 200 seconds the NOx goes to hell - it'll rise from say 5ppm exhaust out to 250ppm. The 250ppm might meet 1995 emissions standards but not 2011. So the trick is to put an oxygen sensor after the cat and use it to learn an offset for the front sensor. That works in steady state, but not in transients. The postcat sensors are often used for fuel control, especially for US emissions levels but also on EU vehicles as well.
For transients, it is most common these days to run an adaptive oxygen storage model for catalyst oxygen management. When you go through a transient the ECM estimates the oxygen imbalance from the engine and calculates a revised step-and-ramp for the front sensor (or in extreme case, an open loop AFR step) to restore conversion efficiency. But what you first try to do is eliminate as many transient air-fuel excursions (open-loop) as possible through careful calibration. You basically need to map out not only steady-state volumetric efficiency (needed even on an ECM with a MAF), but also thermal models that include a lot of stuff. Normally three airflow estimates are used - MAF, speed-density, and speed-alpha (throttle angle), and used in different operation modes. Filtered MAF is normally used for steady-state. Speed-density and speed-alpha are all weighted in with the MAF reading in different transient modes, depending on which weighting tracks the real airflow better. Read the Delphi paper on 'Thermal and Pneumatic State Estimators' and it explains a lot of this. Then you do the air transient estimation calibration.
Then if you're dealing with a liquid film, then you need to dial in fuel puddling. These days that's normally done with a physically based compensator (commonly an X-alpha compensator).
I guess normally my order for calibration is roughly this. I do it always on an AC motoring dyno. It takes at best a few weeks and at worst a year or two to do.
1. Enter in the base engine and sensor information
2. Calibrate steady-state VE, first pass, plus MBT and knock limited timing, and engine torque at MBT, first pass.
3. Calibrate transient air temperature model
4. First-pass calibration of fuel injection timing
5. Recalibrate steady-state VE, reconfirm engine torque.
6. Calibrate transient air flow model
7. Calibrate transient fuel flow model, engine warm
7.5 - Calibrate engine drag, frictional, and parasitic load (alternator/AC/power steering) torque, Refine engine torque map.
8. Calibrate idle controller for reasonable idle return; this usually needs to be revisited on-vehicle. Also calibrate engine trag torque.
9. Calibrate accelerator pedal feel with a 'reasonable' calibration. This needs to be revisited on-vehicle.
10. Calibration of catalyst temperature model, steady-state.
11. Calibration of front lambda, step, ramp for emissions. If goal cannot be met, consider using timing retard for emissions if EGR is not available.
12. Calibration of postcat setpoints and gains for emissions.
13. Repeat calibration gathering for different coolant temperatures to refine temperature models, air/fuel shifts, and ignition timing shifts.
14. Run engine dyno emissions test to see where you are. If this is an engine test cycle (ETC, WHTC, HDDT) then you have a good idea where you are.
15. Calibrate easy DTC's - rich/lean/sensor
16. Calibrate hard DTC's - sensor drift, emissions
The more complicated the engine is and tighter the emissions laws are, sometimes you end up 'looping' trying to get something tuned up right so you can go on to the next step. Another challenge is that if you find a software bug, you may need to go back and recalibrate a bunch of stuff once the software guys or gals fix it. When you add VVT or VGT that adds another level of complexity because there is now not just one optimum solution - there are several.
That is the new best first post I have seen!
and I know know why there are post cat O2 sensors!So the trick is to put an oxygen sensor after the cat and use it to learn an offset for the front sensor. That works in steady state, but not in transients. The postcat sensors are often used for fuel control, especially for US emissions levels but also on EU vehicles as well.
1990 Chevy Suburban 5.7L Auto ECM 1227747 $42!
1998 Chevy Silverado 5.7L Vortec 0411 Swap to RoadRunner!
-= =-
Rumor is Honda will release a 60:1 next year in the states and currently using it overseas. I will ask more questions from the source.
Anyone know what the overall lean ratio of the CVCC engine was? How about the 49 state 91 - 92 Civic with WBO2?
Posting Permissions
Reply With Quote
Bookmarks