12-08-2011, 05:02 PM
Joined: Apr 2003
Location: Long Beach, CA
A Primer about ECU Throttle Control
Originally Posted by opposedcyljunkie
I currently use a BoosterPlug which supposedly reduces the temp read by the AIT by 20c. It seems to work but there is still that snatchiness when in the off-idle and lower rpm throttle positions. I'm guessing that it's a sign of a lean condition which could be further improved.
Thanks for any advise.
|Lurch-followed-by-lunge. You all know what I'm talking about. You crack the throttle and there's a moment's hesitation, followed by a burst of power. At times the hesitation is severe enough that it feels like the motor may stall.|
The 'lurch-followed-by-lunge' or 'snatchy throttle' condition that many bikes have is primarily due to the way that Transient Fueling (aka: throttle tip-in) is managed. Not that throttle tip-in is managed poorly per se, but that in order to produce smooth, uninterrupted power, a good shot of fuel is needed at throttle tip-in.
That's fine and all, but the thing about that is the additional fueling needed during throttle tip-in is also the primary source of motor emissions. And because of emissions, there isn't much choice but to operate the motor very lean during throttle tip-in movements, and drivability be damned. The IICE Cool is designed to correct the lean mixture at throttle tip-in.
Just how lean is throttle tip-in? Well, a general rule of thumb for developing a transient throttle control algorithm for the ECU is, "Just make sure the motor doesn't stall." This spartan objective is due primarily to the need to pass emissions within the constraints of both the EPA and the Accounting department, than any abdication of duty from Design or Engineering.
The designer or engineer works within practical constraints. The emissions rules are what they are, and the motor will have the equipment that the manufacturer gives it. And that means, the designer must figure a way to pass emissions using only the ECU and motor control equipment slated for that motor, and do it within the scheduled timeframe and allotted budget for development and testing. Sometimes it's a wonder anything works well at all.
For most BMW bikes, the designer's job to produce a refined drivability from a motorcycle motor is made far more difficult due to the lack of drive-by-wire throttle control. Electronic throttle control is very relevant to drivability because giving control of the throttle to the ECU means that the throttle never does anything unpredictable.
When the ECU controls the throttle butterfly, the ECU can predict the future. What that means is the ECU always knows what the throttle is going to do next, because the ECU is following a script. Here's an example. In one moment the rider twists the throttle, quickly, slowly, any way they happen to do it. And in the next imperceptible moment the ECU executes a slightly different throttle butterfly motion from a pre-programmed routine or script. These throttle-opening routines are crafted to produce good drivability while passing emissions regulations.
Riders open and close the throttle any random way they like. This randomness is an issue because without throttle-by-wire those unpredictable rider-controlled throttle movements are nearly impossible for the ECU to fuel smoothly while also controlling emissions. But with a throttle-by-wire system the random throttle movements never make it directly to the throttle butterfly. Instead the ECU is in control. It opens the throttle, and controls fueling, by following one of its collection of programmed throttle opening curves.
Using a drive-by-wire throttle, the small number of throttle curves pre-programmed into the ECU is a cinch to fuel correctly. The designer can easily make the throttle powerfully responsive and smooth, while at the same time, not cause an emissions spike at throttle tip-in. With the ECU always knowing what the throttle is about to do, drivability does not suffer for the sake of emissions, no matter what the rider does with the throttle.
The reason that throttle-by-wire works so well is the designer isn't forced to pass emissions and have good drivability for all possible random sequences of large and small throttle movements. Instead, the ECU watches the twist-grip throttle as a user-input, and evaluates what the rider wants. Then it opens the throttle by following only pre-configured throttle-opening ramps and curves that were programmed into the ECU. In that way the random, and difficult-to-fuel throttle inputs from the rider are reduced to a small, finite group of pre-determined throttle opening curves.
This is all very wonderful, except for the fact that for the most part BMW bikes do not have throttle-by-wire. And because of that, drivability suffers. Especially at throttle tip-in.
The IICE Cool is designed to compensate for the lean mixture at throttle tip-in.
Poolside screwed with this post 12-08-2011 at 08:35 PM