News

Air Fuel Ratio For Optimal Performance

Discussion in 'Technical Q&A' started by pma1010, Nov 8, 2003.

  1. pma1010

    pma1010 F1 Rookie

    Jul 21, 2002
    2,558
    Chicago
    Full Name:
    Philip
    I am installing a wide band O2 in the exhaust system of my carb'd 308. The system (made by Tech Edge in Australia) will provide A/F readout from around 8:1 to 25:1 or leaner. It will be great to have some data to guide tuning, particularly on the track.

    Next step is derivation of some targets. The Weber manual I have has an illustration of AF ratio against engine BHP production showing optimal power is around 12.5 or 12.8:1 (and optimal fuel consumption around stoich at 15:1). Infact the power drop off from being too rich (say 11:1) is modest, while the penalty from being lean is significant. (This ignores detonation/impact of ignition timing, both of which would tend to be helped by a richer mixture).

    Does anyone have any data to support how you'd want the AF ratio varying with engine speed (assuming WOT, as on a dyno) on a 308? While all engines are different, as one reference, Guy Croft's tuning book shows a 1600cc race prep'd FIAT 4 banger where the AF is at 12.8:1 at peak torque, progressively leaning out to 14.2:1 at peak rpm.

    [I have searched the archives on the old site but don't find the information]. Any input appreciated. Data particularly sought.

    Philip
     
  2. Peter

    Peter F1 Veteran
    Rossa Subscribed Owner

    Dec 21, 2000
    6,387
    B.C., Canada
    You need to get in touch with some of the Northwest owners on this site, who regularly dynotest their cars. In fact, "navygakman" (aka Stewart Chung) has/had the .pdf copy of his test which listed the A/F ratio at each point of power development. His carb car is modified, done by Matt "Kermit" Morgan. The numbers you list are in the ballpark of his.

    (I have this .pdf copy, but don't want to dispense it out without his prior permission...)
     
  3. jmillard308

    jmillard308 F1 Veteran
    Owner

    May 29, 2003
    5,756
    Perth West Oz
    Full Name:
    John Millard
    Philip
    I recently dynoed my 308 to re-jet to suit the new exhaust - A/F at max power was 14:1.
    Don't have data for any other revs as it was an old dyno that doesn't produce data printouts :(
    Hope this helps a little
    BTW, Dyno time was the best investment I made
    John
     
  4. mk e

    mk e F1 World Champ

    Oct 31, 2003
    11,453
    The twilight zone
    Full Name:
    The Butcher
    On a dynojet dyno on 3rd gear, my QV likes to be around 13.0-13.2. 1st gear liked about 12.5, 5 gear about 14. The "right" A/F ratio depends on the rate the rpm is changing and how long you want the engine to run. I've seen drag racers jet up and go faster until they got to about 10, then more didn't help. I've also seen road racers on a long fast track (Daytona) go jet lean and pick up mph at the end on the front straight until they saw specks of aluminum on the plugs. The private guy then go back up one jet size, the factories leave it as is and replace pistons at the end on the race. I run my street cars, including the QV at about 13 because it's a good balance, it doesn't foul plugs on melt pistons. It's basically impossible to keep a carb car at one A/F ratio across the whole rpm band, carbs just can't control the mixture that accurately, I'd try to keep it between 12-14 in warm weather, if you're tuning on a cold day, stay to the rich end, maybe 11-13, it changes a lot with temp and air pressure.
     
  5. rexrcr

    rexrcr Formula 3

    Nov 27, 2002
    1,572
    Kalamazoo, MI
    Full Name:
    Rob Schermerhorn
    Hi Philip,

    Air/fuel ratios are considered in either steady state, or transient conditions. Transient also includes warm up, in addition to acceleration.

    At idle, about 12.25:1

    Steady state: 13.2:1 (you will have peak BMEP here)

    Start and warmup: very rich to begin with 11.0:1 to 10.3:1, more fuel required as ambient temperature drops

    Acceleration:
    This is dependent on throttle velocity (change in opening position rate). Faster opening requires richer mixture. At normal operating temperatures, no richer than 12.7:1

    Deceleration:
    Lean to control backfires, though no leaner than 17.1:1

    (Credit Joe Pettitt's book on engine building for most of the above information).


    You can tune on a dyno, though a Dynojet is an inertia dynamometer, and cannot hold steady state. It's fine for full-power pulls. Tuning is best done on an eddy current dyno.

    Philip, is your wide band advertised as a true UEGO? Bosch makes a nice compromise 'almost wide band' Lambda sensor for about $150, and reading voltages directly gives the same information.


    Best regards,

    Rob Schermerhorn
     
  6. pma1010

    pma1010 F1 Rookie

    Jul 21, 2002
    2,558
    Chicago
    Full Name:
    Philip
    Thanks for the information.

    Yes, Rob, a UEGO. Retail on the sensor (alone) is over $300 but they can be had for close to $200 from a shop in Canada.

    The instrumentation is made in Australia (see Tech Edge, www.techedge.com.au). Reads at 19.200 baud (i.e., sampling at 15/sec). Others in another tuning community have had great results with the units and they are very good VFM.

    I'll look for the Pettitt book.

    Peter, I'd like to shoot Stewart a message. Have an email address for him?

    In general I'd have thought that you'd want to run as close to optimal BMEP as possible across the rev range. And, second, that BMEP will be driven by a combination of timing (advance) and AFR. Fattening up the mixture (where possible) around peak torque will enable more advance to be run. All of this, of course, within the limits of adjustment of the carbs!
     
  7. mk e

    mk e F1 World Champ

    Oct 31, 2003
    11,453
    The twilight zone
    Full Name:
    The Butcher
    Here's another thought.

    If you are not octane limited, the 13.2 nuber is probably right, it matches what I got on the dyno.

    If you are trying to run pump gas with high compression or a lot of boost, it seems to work out better to run rich. The rich mixture lets you run more timing without detonation and pulling the timing back robs more hp than the rich mixture does. I never got a chance to play with this on the ferrari because I broke a fuel injector the day I was at the dyno, alls I got was the mixture data. My friend the the drag car running 10 A/F has a 12.5 CR chevy and runs it on 93 octane.

    I think if your engine is stock, the factory timing specs are probably right and you shouldn't have to worry about detonation, so shoot 13.2.
     
  8. pma1010

    pma1010 F1 Rookie

    Jul 21, 2002
    2,558
    Chicago
    Full Name:
    Philip
    CR is stock although cams and carbs are modified.

    Anyone have any sense for the impact of a 1/2 step change in mains (e.g., 155 to 150) or a full step change in ACs (e.g., 175 to 185) has on AFR?
     
  9. pma1010

    pma1010 F1 Rookie

    Jul 21, 2002
    2,558
    Chicago
    Full Name:
    Philip
    I talked to Norwood Performance yesterday about another matter. In the course of discussion we talked about the black "308" (carbon fibre body, v12, t-turbo etc - sounds a beast) and the optimal AF ratio. They echo the thoughts here - 13.2 is the optimal but that you can also only be approximate given the (variables you can't control and bluntness of the) control variables in the carbs.

    Thanks for everyone's help. I'll post results when I have it hooked up. Also looking into some dyno time.
     
  10. Aircon

    Aircon Seven Time F1 World Champ

    Jun 23, 2003
    78,444
    Melbourne, Australia
    Full Name:
    Peter
    I used lamda readings when setting up my motec. .87 for full throttle and 1 for cruise. How does that relate to a/f ratio? I note that I'm not getting as good fuel consumption as I should, so I'm not convinced that lamda 1 at cruise is correct.
     
  11. Steve Magnusson

    Steve Magnusson F1 World Champ
    Lifetime Rossa

    Jan 11, 2001
    18,481
    Texas
    Full Name:
    Steve Magnusson
    This site has a lambda-to-AF ratio calculator (scroll down the page a bit):
    http://www.rbracing-rsr.com/rsrgauge.htm
    For benzina, I think it's as simple as:
    lambda value*14.7 = air/fuel ratio
    (so for lambda =1, the AF ratio is 14.7:1, and for lambda = 0.87, AF = 12.8:1)
     
  12. lndshrk

    lndshrk Formula Junior

    Nov 7, 2003
    753
    SLC, Utah
    Full Name:
    Jim Conforti
    First, optimal AFR can only be determined experimentally

    On a normally aspirated vehicle, start with the fuel you are going
    to run and then tune for 12.5:1 afr. This can be done on the dyno
    or on the street using some form of UEGO sensor.

    From there, to the dyno you go and slowly lean the vehicle out
    until you have both max HP and max Torque. Which, I can assure
    you will occur at both different AFRs and different Spark Timings!

    (With EFI, this is a matter of keystrokes.. you carb guys better have
    a full jet kit and a few days handy :D )

    The key is to either hit peak numbers (not "knock-limited) and leave it,
    or to hit knock (knock-limited) back it off a bit (make it richer) and then
    leave it.

    If you hit 13.5:1 afr you are pushing the envelope extremely.

    The actual AFR depends upon engine design, compression ratio,
    carb/EFI atomization efficiency, and about 100 other factors which
    all relate to BMEP.

    With a forced induction vehicle, depending on the fuel and boost you
    can start as rich as 10.0:1 and lean from there.

    Also, be very wary of "bargain UEGO" sensors. Some work well, and
    others don't. A bad UEGO is worse than NO UEGO.

    Jim Conforti
     

Share This Page