AFR "curve" - flat or trending lean after tq peak?

[BradH]

I think there are generally two schools of thought re what the Air-Fuel Ratio should be across the working RPM range: target a consistent "flat" reading, or richer at torque peak and trending somewhat leaner as the RPMs increase. I'm interested if anyone has comments or data to go along w/ this topic. Thx

[NormS]

It depends on the combustion efficiency of the engine. For those not so efficient engines, a "flat" A/F ratio curve tends to yield the best overall power curve. For more efficient engines, a leaning trend as the rpm's increase, is usually what they like. Do not assume that BSFC numbers are a direct indicator of A/F ratio. BSFC numbers are affected by many more factors, than just A/F ratio. Tune by accurately measured A/F ratios , whenever possible.
The best scenario is to have a wide band O2 sensor in each header primary tube, so that you can see what each cylinder is getting, and hopefully be able to compensate for cylinders that are excessively lean or excessively rich. Having O2 sensors in just the collectors can make you think things are good, when they are not. Those collector O2 sensors are just getting an average of the cylinders feeding the collector. You may find, as we have, that intake manifolds you thought were good, have horrible fuel/air distribution.

[BILL-C]

Many times with dual plane intakes i have seen the leanest cylinder at the beginning of the dyno run turn into the richest by the end of the pull. I'm always amazed by the fact that so few people know about this . The reality is that fuel distribution and curves are absolute train wrecks on most street and mid range race engines. The vast majority of shops don't have the equipment to see it happen so they don't know that it exists. Different story on higher end stuff though.Single plane intakes are far better in this respect. My experience is that for max power N/A engines want to be slightly rich at peak torque and trend slightly leaner as rpm increase.

[NormS]

Yes, 'train wreck' is an appropriate term for the distribution in many intakes, particularly 2 plane intakes. The late Harold Bettes, told me more than once, that "intakes manifolds are some of the worst devises ever invented by mankind". Once we got to using a data acquisition system with 10 channels for wide band O2 data, we found he was right on with that description.

[RW TECH]

It depends on the engine really. There is no for-sure recipe for AFR to produce best power, that covers all engines will all induction systems, port and chamber configurations, piston crowns, exhaust side capacities, at all operating temperatures, etc. etc.

As an example, I've done extensive work with LS7-type engines and have consistently seen where about .83-ish lambda made best torque, then .86-ish made best HP on popular gasolines like Sunoco GTX or VP C12.

Others commented on dual plane intake distribution and in short i agree. Individual cylinder distribution is horrific. For SBC, the Weiand Steet Warrior is about as good as it gets for dual plane distribution vs the other most popular "air gap" type intakes, with same or maybe a tiny bit better power.

Even with a very-very good open plenum 4-bbl worked over by top-level people will often have cylinder to cylinder distributions that are a full AFR ratio apart from richest to leanest. Example: 12:1 to 13:1 from hole to hole.

[racear2865]

Ive been thru this so many times on my dyno, that I dont wont to count. Every freakin engine is different. If you want, you can throw them in wide groups. I end up reading plugs all the way thru especially with a dual plane. Almost invariably, you can make a dual plane run better with a carb spacer. Most of the the time, that it doesnt improve the pull is when the intake is tOOOO large. It will respond similarly if carb is too large but a little different.
reed

[maxracesoftware]

I think there are generally two schools of thought re what the Air-Fuel Ratio should be across the working RPM range: target a consistent "flat" reading, or richer at torque peak and trending somewhat leaner as the RPMs increase. I'm interested if anyone has comments or data to go along w/ this topic. Thx

you cannot make it a "Flat A/F Ratio at all RPM" ... because VE% drops as RPM increase past Peak TQ RPM
as the Engine starves for more Air past Peak TQ RPM , it will respond more to Air increase , than it will to Fuel increase
ex: Dirt Track Engine or any Engine w/Restrictor Plates

a "Generalized Gasoline example" :
the peak Volumetric Efficiency % usually occurs anywhere from Peak TQ RPM point -to- a few hundred RPM above the Peak TQ RPM point
so that if the best Air/Fuel Ratio at or near Peak TQ RPM was 12.5:1
then it would progressively need to be Leaner ( ex: 13.2:1 ~ ) as VE% decreases/drops as it heads towards Peak HP RPM point and beyond !

[agertz1]

Ok, so if intake is TOO big, will an anti-reversion carb spacer help ? Do they sometimes help with cyl/cly AFR' s ?
Thanks, Art.

[HQM383]

I think there are generally two schools of thought re what the Air-Fuel Ratio should be across the working RPM range: target a consistent "flat" reading, or richer at torque peak and trending somewhat leaner as the RPMs increase. I'm interested if anyone has comments or data to go along w/ this topic. Thx

I am also interested in how much does it matter power wise to have either AFR curve described above dialed in to the perfect numbers? Example; the power difference between say 12.7:1 through the entire range or AFR fluctuating 12.5:1 to say 13.4:1 through a power run?

What are people’s findings versus the below videos both on the dyno and drag strip? Should we fret too much over lean or rich spikes that are still within safe ranges?

https://youtu.be/aDSZhy551bo

https://youtu.be/JLRaY3oAKmg

https://youtu.be/JzbLrn-2jyw

https://youtu.be/HXX4zcPr9IE

[rebelrouser]

Just to throw it out there, has anybody played with using a 5 gas analyzer to tune an engine?

[englertracing]

Just to throw it out there, has anybody played with using a 5 gas analyzer to tune an engine?

Marc salvisberg @ factory pro is a big proponent of tuning with a 5 gas analyzer.

[HQM383]

Just to throw it out there, has anybody played with using a 5 gas analyzer to tune an engine?

Never had anything to do with a 5 gas, only AFR gauge.I have experienced on the same engine in same car the same carburetor tuned to two different emulsion and air bleed packages behave different with the same AFR reading. That was at a specific lean spike around 1.05 +/- .03 lambda. One would lean surge, the other drive and respond like it was .92 - .93 lambda. Certainly taught me if you want to use an AFR gauge to tune a carb the butt meter is the final say and AFR numbers are reference points. A novice spark plug reader like myself couldn’t see a difference but no doubt a 5 gas analyzer would tell a story of how the burn was happening. I assume the AFR gauge that (correct me if I’m wrong here) just reports how much oxygen has or hasn’t been consumed in the combustion process won’t give an indication of how close to an ideal state for combustion the mix was before combustion. I refer to atomization and homogenization etc.

It wasn’t just that particular lambda number either. Between the two emulsion/air bleed packages there was noticeable performance differences while driving that had AFR readings in a very similar range. Again, something a 5 gas would pin point as to why. It would be very interesting to know how much the different the two carby set-ups would be on the dyno, chassis or engine, relative to the way the car felt to drive.

[englertracing]

https://youtu.be/yjvPmG123nI