Mi pregunta:
Hi Al,
I have a question:
Regarding to ECM/02 reading While the bike was idling, it was sitting at .50 and while the engine warm up then it slowing started to drop little by little to 47, 45, 42, 39 and when the engine was fully warmed up got down to 23. I stop the bike a few minutes and I start the engine again and While the bike was idling was the ECM/02 reading 50/48. is this normal?, is good work condition?
I KNOW YOU ARE BUSY,
THANKS FOR THE GREAT SUPPORT AL!.
Respuesta de Al:
The ECM has two modes of operation: Open Loop and Closed loop. Closed loop operation is when the injector duration is determined from the A/F ratio obtained from the rear cylinder O2 sensor. Note that this means that the injector duration of the current combustion cycle is determined from the measurement of the exhaust gas output of the previous combustion cycle. Buells use a low cost narrowband type O2 sensor that is highly non-linear with an output response as shown in Figure 1. As such, they are not particularly useful for holding the A/F ratio at anything other than the 14.7:1 stoichiometric value, ON AVERAGE. The ECM holds the A/F ratio by maximizing the .45V crossings of the sensor. It commands the mixture alternately slightly rich, then slightly lean, to keep the O2 sensor voltage alternating back and forth across the .45V value. So while the engine runs ON AVERAGE at 14.7:1 in closed loop, it is rarely running at EXACTLY 14.7:1 at all.
For normal operation in cruise mode, 14.7:1 is the right A/F value for maximizing fuel economy and minimizing emissions. The ECM operates in closed loop any time the engine is operated steady state at normal operating temperature, from 1500 RPM to 4000 RPM (Tuber frame Buells), or Idle to 5000 RPM (XB9) or 4000 RPM (XB12), up to 30% of maximum throttle input, with “light” engine load. These are not exact figures but are close enough for the purposes of this document.
Figure 1. Narrowband O2 Sensor Response
The engine runs in open loop any time the ECM isn’t operating in closed loop operation. Open Loop operation is when the injector duration is determined completely from the internal fuel lookup table map for the current operating conditions, as determined from the sensors other than the O2 sensor. There is no feedback loop from the O2 sensor in open loop operation, and the previous combustion cycle has no direct impact on the current combustion cycle fuel input. Open loop operation is used at idle for Tube frame Buells, at high speed/load conditions (i.e., when accelerating) at throttle settings above 30%, whenever the throttle isn’t relatively steady state, and whenever the engine isn’t running at normal operating temperature for all Buell DDFI engines.
The way the Buell ECM works when running in closed loop (and most likely any other closed loop EFI system that uses this narrowband sensor) is by maximizing the .45V crossings. So while it is true that the Buell ECM runs ON AVERAGE at 14.7:1 in closed loop, it actually is rarely at exactly 14.7:1. The ECM bounces the AFR slightly richer, than slightly leaner, all the time. If you monitor O2 sensor voltage with a scanalyzer or VDSTS, you can watch this occur. Most of the time, the O2 voltage when running steady state on a fully warmed up motorcycle bounces between .2 and .7 Volts or so, with .15 and .75 VDC not all that uncommon. But I've monitored it while driving down the road and seen lower than .10 VDC and higher than .80 VDC on occasion. That means that the AFR can hit 16:1 or even higher briefly. It is my belief that a subtle lean miss occurs when the ECM overshoots a little on the lean side for a few combustion cycles, and that the uneven-ness of the engine operation when this occurs is the source of some folks complaints with the way this engine operates. But it doesn't hurt anything, as it never stays there. And as soon as you twist the throttle, it is yanked out of closed loop and NOT at 14.7 or higher any more.
The AFV is set during a subset of closed loop operation. If you try to map the bike at anything other than 14.7:1 across AFV learn mode, you'll only screw things up by causing the AFV to learn improperly. So even the open loop map across the closed loop operational domain is mapped to 14.7:1.
There is nothing wrong with 14.7:1 during cruise. You don't need to generate more power from richer AFRs during cruise, as it only wastes gas and pollutes more. If you need more power, you get it by twisting the throttle, not by dumping more fuel. The only evilness of the 14.7:1 AFR is the oscillation algorithm that causes it to occasionally be leaner than that. But that is the reason these bikes make 55-60MPG.
There are folks complaining of surging issues that MAY be more severe than what I am discussing above. Davo suspects timing issues are responsible, I'm not so sure. But mine doesn't do it, and I haven't personally experienced any surging other than the mild one generated by the subtle lean miss discussed above. I don't know what to make of it. I have no complaints with the way my bike is operating on our custom map with the micron, with the possible exception of a little bit of decel popping that I haven't been able to tune out, and an off-idle acceleration lag that seems to be a characteristic of DDFI, since no fueling or accel table changes seem to reduce that lag.
Al
Espero que esto os sea util para entender el funcionamiento del 02 sensor, aunque desde luego hay que entender algo de english.