Air fuel ratio,what is proper?

[Warp Speed]

Most all performance engines make peak power between .89-.93 of stoich for the fuel (gas/race gas) used..............
Example 13.9 stoich x .91=12.65 afr. Engine design/efficiency, operating temp, sensor placement and system used can effect this, but it is a very good starting point, and most all will end up in this range if sampled properly.

[Truckedup]

We see many situations which show engines giving more power with what seems to be less fuel. If the head work is giving better fuel atomization, and wasting less, then the engine can make more power on leaner jetting because of that.

Yes, these engines have 1930's hemi heads with what we consider now consider to be poorly shaped large ports. After raising intake port floors without enlarging the ports it made a nice improvement with less fuel

[MadBill]

... Something else to consider about bikes, a broken crank or thrown rod at speed that seizes the engine will likely lock the rear wheel before the ride can pull in the clutch...And it all falls down...

A pessimist* might be inclined to incorporate a sprag clutch in the driveline, JIC... (*Oxymoronic I realize, as how could such a personality possibly justify LSR racing? )

[PackardV8]

When I discuss the tuning of my race bike on the chassis dyno guys often ask what is the air fuel ratio....You know, if it's too lean the pistons will fry, too rich and power is down... Too be truthful I have no idea..

What Mike Jones said on another thread about overscavenging is very appropriate here also.

Over scavenging plays tricks on the O2 sensor. It can't tell the difference between the unburnt oxygen left in the cylinder from combustion, and the oxygen that's flowing from the intake straight out the exhaust during overlap. the (O2 sensor) thinks the mixture is lean when it is not . . . I just tried to explain this to a customer, but he didn't believe me. He's tuning with the O2 sensor, but the cam I sold him, had more duration then anything he had run. When he tuned it to the same O2 numbers he was used to, it didn't run, and showed no heat in the exhaust. As he went leaner, the power picked up, and the exhaust temps started to come up to a normal range. He got it to the point, it was making good power, but he said he couldn't run it with that tune, because he said it was too lean. I tried to explain that it really wasn't lean, it was just the O2 sensor picking up the oxygen from the added overlap, but he didn't get it. If you measure the Air CFM in, and the Fuel lbs/hr, you can see how the O2 sensor can fool you.

Anyone who's spent some time with highly tuned race bikes can appreciate how the A/F ratio will be all wonky down low, because it has too much overlap, then when the cam, intake and exhaust tuning come on sync, it gets righteous and pulls like a freight train, but the O2 meter still won't be in the usual 12.5 range because as Mike says it's suckin' hard and moving more in/out than is being captured and burned and thus the misleading dyno O2 readings.

[David Redszus]

Suppose we have two fuels with the same stoich value but with difference specific gravity values?

Or we have two fuels with the same stoich values, same SpG, but with different distillation curves
and different Heats of Combustion and Heats of Evaporation?

What now brown cow?

A lambda sensor only reads out in milivolts as a function of oxygen partial pressure and temperature.
Every other number has been calculated by some predetermined equation.

It is often assumed that regular gasoline (whatever that is) has a stoich value of 14.7.
Complely false. It can and will vary seasonally, geographically and by brand and blend.

In any real running engine, we never see a single a/f number. We see a dancing variable that changes
with throttle, rpm, temp and time.

Properly, we should refer to a mixture range, not a single value.

And lastly, are we fuel tuning for best performance or to prevent engine destruction?

[MadBill]

..but the O2 meter still won't be in the usual 12.5 range because as Mike says it's suckin' hard and moving more in/out than is being captured and burned and thus the misleading dyno O2 readings.

But at high load and in the design powerband, doesn't any 'over-scavenged' mixture also burn, if only in the header? In which case as long as the sensor is far enough downstream the readings should be relatively unaffected.

[user-23911]

Over scavenging used to be a problem with modified 2 strokes before modern expansion chambers were perfected.
The clue there was that the fuel consumption would go sky high relative to the power produced.
You'd also expect the EGT to go very high at the same time. That's because the wasted fuel would be burning in the pipe.

In real life you won't get over scavenging, not on a 4 stroke unless the cam is really silly long duration.......or you've got a turbo and you're running "antilag".

As far as your AFR goes, it's about the ratio of carbon to hydrogen to oxygen in the fuel. Different fuels require a different lambda for max power.
Stochiometric ratio should ensure clean burning. Theoretically all the fuel and oxygen are consumed.
Max power is always richer because there's 3 main chemical reactions which release different amounts of energy.
The reaction of CO to CO2 makes the least energy.
The reaction of C to CO makes the most energy.
The 3rd reaction is H to H20.
There's a bit more to it than that but it's a good simplification.

If you want to go a bit further then look up the KJs of energy released per mole for each reaction on google.
By running a bit richer you get more of the reactions C to CO and less of the other 2 reactions. That's given a fixed amount of oxygen to use.

[David Redszus]

Two stroke racing engines still suffer from "over scavenging" due to charge purity, pipe length/rpm/temp mis-match.
Sometimes the effect is called "short-stopping" which is not cured by pipe design.

Whether it be two stroke or four stroke, we must also consider misfires which dump unburned air and fuel into the exhaust header where it may or may not burn. ALL engines are subject to mis-fires and malfires.

All sensors lie. Trust your sensors.
An oxygen sensor can correctly indicate what it sees. But it only sees burned mixture in the exhaust pipe and not in the combustion chamber. It will tell us the truth but not the truth we would like to believe.

[turbobaldur]

Two stroke racing engines still suffer from "over scavenging" due to charge purity, pipe length/rpm/temp mis-match.
Sometimes the effect is called "short-stopping" which is not cured by pipe design.

Whether it be two stroke or four stroke, we must also consider misfires which dump unburned air and fuel into the exhaust header where it may or may not burn. ALL engines are subject to mis-fires and malfires.

All sensors lie. Trust your sensors.
An oxygen sensor can correctly indicate what it sees. But it only sees burned mixture in the exhaust pipe and not in the combustion chamber. It will tell us the truth but not the truth we would like to believe.

On a 2 stroke the short circuiting is also very much down to the angles and locations of the ports. People have spend lots of time re-aiming the ports and testing to find the angles that get the best separation between exhaust and fresh charge while still getting good air flow. On a 2 stroke it's also not unusual to see your airflow drop and power pick up at the same time when modifying the tuned pipe.
Most modern 2 strokes inject their fuel into the transfer ports or directly into the cylinder and time the injection event to stratify the charge, delaying the introduction of fuel into the air stream so what fresh charge goes out the exhaust will be mostly air. What has been found in dyno testing these engines is that both direct air:fuel ratio measurements (using air and fuel flow instrumentation) and lambda sensors agree that they're running a pretty lean 0.9 lambda at full throttle. The charge trapped in the cylinder at the closing of the exhaust port is surely far richer. Keep in mind the same 2 stroke on carbs or less sophisticated EFI tuned for the low octane crap peddled as fuel at the pump will most likely be in the 0.7s as far as lambda goes, with the best power ~0.85-0.9 range reserved for racing fuel.

[Belgian1979]

From my experience : every engine is different. Mine doesn't like to be rich and doesn't like to be run lean either. At full power closer to 13/1 and when cruising at 13.8/1

[mk e]

From my experience : every engine is different. Mine doesn't like to be rich and doesn't like to be run lean either. At full power closer to 13/1 and when cruising at 13.8/1

A narrow "happy" band tends to indicate AFR variation is already present in the system.....the window should be much wider once you finish the upgrades you're working on I'd think.

[MadBill]

I would think it would be the other way around. With a lot of AFR variation, any global change would be helping some cylinders, hurting others and going from a tad lean to a tad rich (or vice-versa) on a few. For example, insensitivity to cam index changes is a reliable indicator that the cam specs are sub-optimal. If all the events are ideal, modest advance or retard makes them all significantly worse and you lose 30 HP instead of 3.

[user-23911]

From my experience : every engine is different. Mine doesn't like to be rich and doesn't like to be run lean either. At full power closer to 13/1 and when cruising at 13.8/1

Every sensor is different too.

[mk e]

I would think it would be the other way around. With a lot of AFR variation, any global change would be helping some cylinders, hurting others and going from a tad lean to a tad rich (or vice-versa) on a few. For example, insensitivity to cam index changes is a reliable indicator that the cam specs are sub-optimal. If all the events are ideal, modest advance or retard makes them all significantly worse and you lose 30 HP instead of 3.

Yes helping some but it's the hurting others. With AFR any engine should run fine from what... about 11 to about 15. If it only runs fine at 13-13,8 that generally means the TRUE variation is still 11-15 but at 13 you have some cylinders pushing 11 and having rich issues and when you go the other way at 13.8 others are pushing 15 and having lean issues. I fought for months with an engine like this until giving up and swpaping in an ecu with cylinder tuning....and it was back to a wide, forgiving AFR range.

[RL]

You're doing it right, give an engine what it wants not what you think it should have.

Port surface texture changes the optimum A/F for an engine, the smoother surface the richer you need, the rougher the leaner you want.

Even when you jet and engine on a dyno, you always test different jets at the track - maybe a richer jet fattens up the mid, and speeds up the 60ft -.