Race Engines: Why no fuel cut during over run

[adrag]

I first really noticed this when I went to see my first grand prix and since then I've noticed that many fuel injected race cars pop and backfire like crazy on the over run. I'm referring to naturally aspirated cars, not turbo charged cars which are often equipped with anti lag systems that causes popping on the over run. So what is the benefit to injecting fuel during the over run? During the traction control era in F1, I figured that maybe the ECU fired certain cylinders during braking in order to prevent rear wheel lock up, but that should be illegal now and yet I still hear them on TV popping during braking.

 

[malbeare]

it helps to keep the exhaust valves cool as the fuel fryes on them and evaporates.

A tidy mind not intelligent as it ignors the random opportunities of total chaos. Thats my excuse anyway
Malbeare

http://www.sixstroke.com

 

[BrianPetersen]

I'm speaking from a motorcycle roadracing background but with bikes having gone EFI across the board on high performance models, it's the same issue. We have to deal with OEM decel fuel cut systems and their consequences all the time.

In order to reduce emissions, the decel fuel cut threshold has to be assuredly above the engine "load" (or negative load as the case may be) that results in misfiring. This results in a noticeable abruptness in the throttle response at slightly-cracked throttle. You go into a corner on the brakes (fuel is cut off), try to smoothly apply throttle on the exit, at first the engine does nothing then "whack" the fuel comes on and the engine takes up the load. This is not much fun when you are on a bike cornering at the limit of traction, particularly when the bike weighs 400 lbs and has 120 - 160 horsepower - driveability is extremely important. Some models are worse than others for this; I rode a stock 2003 Yamaha R6 and did three laps then brought it in ... horrible driveability. My stock ZX10R has decel fuel cut, but it's below the misfire threshold and is scarcely a bother. I suppose they just let the catalytic converter deal with it.

If you want to fix this, you have to keep injecting fuel down to a smaller "load" threshold. This means there will be some misfiring, but it's better to have that, than to have an abrupt "on/off" behaviour. The misfiring means there will be some popping out the exhaust.

In the bike applications, the piston and exhaust valve temperature isn't a big enough issue to warrant injecting gross amounts of fuel on deceleration, but that doesn't mean it isn't an issue in other applications.

Fuel consumption is an important issue in MotoGP because of the restricted fuel tank size, but those engines are using drive-by-wire throttles and are deliberately opening throttles on deceleration to cut back on engine braking. I don't know if they are cutting off fuel simultaneous with opening up the throttles, but it's quite possible with the way engine management controls are going these days.

 

[SomptingGuy]

In the carb days, throttle control was all. Rapid movements had bad effects on the engine. Today's ECUs protect the driver, be (s)he a commuter or a racer.

- Steve

 

[mattsooty]

"I'm speaking from a motorcycle roadracing background but with bikes having gone EFI across the board on high performance models, it's the same issue. We have to deal with OEM decel fuel cut systems and their consequences all the time."

This is not a problem of EFI per se but more a symptom motorcycle EMS being about a zillion years behind that of cars - although they are rapidly catching up!

MS

 

[kneedown]

I'm trying to determine how my motorcycle's ECU decides when to implement fuel cutoff and when to restart fuel injection. It seems as if cutoff is implemented by fully closing the throttle at "high" rpm and restart is either as soon as the throttle starts to reopen or else when a "low" rpm is reached.

Anyone know the strategies used by manufacturers in their ECUs.

 

[BrianPetersen]

kneedown, you've pretty much got it. Inside the stock ECU is the map in which RPM and throttle position are looked up to establish the duration that the injector is supposed to fire. It has zeroes in the almost-shut-throttle / higher RPM regime.

You can trick some of them by telling it that the throttle is a bit more open than it actually is, but that has other side-effects.

 

[PQatPIT]

Also ignition timing plays a role here. When on high rpm's and low loads the ignition advance can be as high as 40 degrees. This makes incoming charge to burn inside combustion chamber and popping sounds and possible flames and so on are not so much an issue.

When deceleration fuel cut is on and throttle is opened engine should catch easily and not with abrupt power. If it does then the ecu tuning is not good.

 

[TurboXS]

We were told by an Australian V8 Supercar team that the organiser wanted flames on the overrun for the spectacle.

 

[izzmus]

I am involved with a regional race series where the #1 rule seems to be "give the spectators a spectacle". The grandstand tickets are paying the bills so the course and race format are designed to appeal to the masses: lots of noise, sliding, and squealing tires. (The winner of the race, for example, is the first to come to a stop inside a roughly 10'x20' box. First across the line would be easier to judge but this way is more exciting to watch)

As such, I've gone as far as installing an ignition cut on shifting for a nice 2 foot white flame (and LOUD report) from the tailpipe on every shift. Races are by invitation and I want to make sure that I'm invited!

 

[turbocohen]

The problem that even the best ECU cannot overcome is Inadequate injector linearity. To accomodate high rpm pulse periods along with low map decel the injectors linear range needs to be much higher than what is on the market.
A guy who retired from Siemens, now continental, figured out how to develop a solenoid injector with over 22:1 turndown using conventional injector driver electronics to accomodate the trend towards oem's desire to maintain closed loop during decel for optimum catalyst storage. I ran a cng fueled vehicle and he ran a gasoline vehicle supplied by chrysler just for this activity. The injector with its own electronics molded internally.. Management killed it. Now BMW uses scaled injector pressure to bridge the gap but high linearity injectors are the real answer and if the manufacturers had some gonads they would promote what is now an 11 year old field tested idea..

http://www.google.com/patents?id=y0IFAAAAEBAJ&printsec=abstract&zoom=4&dq=siemens+danny+wright

http://www.google.com/patents?id=ro...stract&zoom=4&dq=siemens+danny+wright#PPA5,M1

 

[dgallup]

I seriously doubt it is more cost effective to wind dual coils & put the electronics in the injector than simply use commonly available peak & hold drivers in the ECU. I like the idea in the second patent of winding half of the secondary coil clockwise and half counter clockwise to negate the inductive component & make it purely restive. Many European & Japanese cars used similar external "ballast resistors" with low impedance injectors 20 years ago. Again, I think it would cost less to implement the switching in the ECU and buy low cost resistors than add a circuit board and all that expensive copper wire to the fuel injector.

 

[turbocohen]

It is not more cost effective to use ph drivers and 62v clamp zeners either when high linesarity is the objective. The described inejector can open as fast as the same solenoid with an optimized 4x1 driver and low impedance coil cept for one problem.. it can't close fast. The onboard circuit applies optimum opening moxie then retreats to a minimal hold current. With a wire harness and peak hold drivers there is a bit of variation as the wires and injector coils heat up.. The warmer the injector the longer it takes to peak which is typically after open. The "dekatronic" with onboard circuitry compensates for voltage and temperature. On the test bench the dekatronic opened and closed faster at any voltage than the peak hold driven injector that was matched for Qd and Qs. The total cost taking drivers and a wire harness that is suitable for 4 amps vs 1 is also a factor.

Several clients ran into calibration snafu's when cylinder to cylinder a/f strayed too far.. Saw over 10% var after hot shut down restarts and some other events.

T