How does the PCM detect a misfire?

[PurplePiss]

How does the PCM detect a misfire?

[jimmyblue]

This is a subject I am very interested in. The misfire
tables set a "time between 'events'" value for declaring
a misfire, but nothing about what constitutes an "event".
I gather that it's some amount of computed crank jerk
but setting the threshold for that, is probably much
more useful and to-the-point than deciding how many
of them to tolerate.

There is a rolling misfire counter and so many counts in
a cycle (see cylinder mode index, revolution mode index
rolling, this is said cycle) before you add to misfire history.

I would really like to get ahold of the individual misfire
event threshold, I think it's directly relevant to lightweight
converter and big-cam driven misfire detects elevation.

[Shifted]

I believe it uses the knock sensor. The ECU is expecting a certain amount of noise during a fire event, and if the noise isn't what it expects, its a misfire. I believe this is how it works on the 4 cyls anyway, I'm not a big V8 guy.

[PurplePiss]

I think the knock sensors are tuned to a specific frequency which I think would be different than if a cylinder didn't fire and it just pumped air and fuel out the rear. I dunno.

If it knows the crank position and it sits and watches to figure out how much the crank is rotated for a certain amount of airflow and fuel vs. load, I would think it could average those across 8 cylinders or maybe 2 cylinders before and after. Maybe then it could say, "hey I expected the crank to be at this position, but it isnt, so the previous cylinder must not have put any effort into the crank". How it knows which cylinder is another mystery. It can't tell a misfire from the O2's that's for sure. Maybe the voltage discharge resistance/rate of the coils changes if it doesn't light it up? Just brainstorming here.

[Shifted]

Is the V8 a batch fire or sequential fire system? I know the older V8's I worked on (carbeurated engines) were sequential fire, so its easy to tell which cylinder is supposed to be firing (not that they had an engine managment system). I know V6's and I4's are batch fire (2 cylinders per coil, 2 cylinders firing at any one time, wasted spark system). If its similar, it may be able to tell which piston is supposed to be providing the power at any one given time, and which one is the wasted spark cylinder. By your logic it should then be able to tell which one didn't fire.

[PurplePiss]

I bet a really smart programmer could tell us how it does it.

[Chris@HPTuners]

it uses crankshaft events, not knock sensors. The diagnostic basically expects a delay between crankshaft events, that if outside the calibration signals a misfire has occured. Misfire tests are adapted for all kinds of things that can apply external forces to the crankshaft, such as rough roads, tire slip, transmission shifts, ac engagement, TCC lockup etc.

[Billiumss]

What would you suggest is the proper way to adjust them for converters and bigger cams?

[all-flavors]

If you want to know how Misfire operates Pm Me and Ill Send you a PDF of the control stragey and how to use your scanner to evaluate misfire events


Regards
Tech-link

[jimmyblue]

This is what I'm talking about, how to widen
the "calibration" window to tolerate more
crank ripple before calling an event. Rather
than just allowing more of, something that
pretty much happens constantly.

it uses crankshaft events, not knock sensors. The diagnostic basically expects a delay between crankshaft events, that if outside the calibration signals a misfire has occured. Misfire tests are adapted for all kinds of things that can apply external forces to the crankshaft, such as rough roads, tire slip, transmission shifts, ac engagement, TCC lockup etc.

[Nithros]

I got this info from another web site...

Circuit Description

The Crankshaft Position sensor is mounted through the side of the engine block at the rear of Bank 2 behind the starter assembly. The Crankshaft Position sensor works in conjunction with a 24X reluctor wheel on the crankshaft. The reluctor wheel is inside the engine immediately in front of the rear main bearing. The PCM provides a 12 volt power supply to the CKP sensor as well as a ground and a signal circuit.

A misfire causes a change in crankshaft speed. The PCM times the interval between each pulse and compares each new time interval with the previous one in order to determine when an excessive change in crankshaft speed has occurred. You can expect a certain amount of acceleration/deceleration between each firing stroke, but if the crankshaft speed changes are greater than an expected amount, the PCM interprets this as a misfire.

The PCM uses the Crankshaft Position sensor for both spark and fueling. As the crankshaft rotates, the reluctor wheel teeth interrupt a magnetic field produced by a magnet within the sensor. The sensors internal circuitry detects this and produces a signal which the PCM reads. The PCM uses this 24X signal in combination with the Camshaft Position sensor 1X signal in order to accurately determine crankshaft position. The PCM also calculates a 4X signal from this information. The PCM uses the 4X signal for internal calculations. The 4X signal also provides a tach signal for any device which requires one.

Observe that as long as the Crankshaft Position sensor 24X signal is available, the engine will start. The PCM can determine top dead center for all cylinders by using the Crankshaft Position sensor 24X signal alone. The Camshaft Position sensor 1X signal is used by the PCM to determine if the cylinder at top dead center is on the firing stroke, or the exhaust stroke. The system attempts synchronization and looks for an increase in engine speed indicating the engine started. If the PCM does not detect an increase in engine speed, the PCM incorrectly synchronized to the exhaust stroke and will re-sync to the opposite cam position. A slightly longer cranking time may be a symptom of this condition.


Conditions for Running the DTC

DTCs P0101, P0102, P0103, P0107, P0108, P0117, P0118, P0121, P0122, P0123, P0125, P0335, P0336, P0341, P0342, P0343, P0500, P0502, P0503, P1258 not set.
The engine speed is between 450 RPM and 3,001 RPM.
The ignition voltage is between 11.0 volts and 16 .0 volts.
The engine coolant temperature is between -7°C (19°F) and 130°C (266°F).
Fuel level greater than 10%
The TP sensor angle is steady within 1%.
The ABS and Traction Control systems are not active.
The transmission is not changing gears.
The AIR diagnostic test is not in progress
The A/C clutch is not changing states.
The PCM is not in fuel shut-off or decel fuel cut-off mode.
The ABS signals not exceeding rough road thresholds.

Conditions for Setting the DTC

The PCM determines that an emission type Misfire is present.
The PCM determines that a catalyst damaging Misfire is present.

Action Taken When the DTC Sets

The PCM will illuminate the Malfunction Indicator Lamp (MIL) under the following conditions:

The PCM illuminate the Malfunction Indicator Lamp (MIL) on the second consecutive ignition cycle that the diagnostic runs and fails, if the diagnostic fails under the same conditions (load, RPM, temperature, etc.) as the previous ignition cycle that the test ran and failed.
The first time the diagnostic fails, the PCM records the operating conditions in Failure Records.
The second consecutive ignition cycle that this diagnostic fails [under the same conditions (load, RPM, temperature, etc.) as the previous ignition cycle that the test ran and failed], the PCM stores this information in Freeze Frame. The PCM copies any data previously stored in Freeze Frame and copies the data into the Failure Records. The PCM then overwrites the Freeze Frame. The only exception to this is if a Misfire DTC was already recorded in Freeze Frame. In this case, the Misfire data stays in Freeze Frame and the PCM updates the Fuel Trim data in the Failure Records.
The PCM determines the percent of misfire over a 1,000 revolution period is high enough to cause excessive tail pipe emissions. The PCM illuminates the Malfunction Indicator lamp the next consecutive ignition cycle that the diagnostic runs and fails, if the diagnostic fails under the same conditions (load, RPM, temperature, etc.) as the previous ignition cycle that the test ran and failed.

Conditions for Clearing the MIL/DTC

Important

If the last failure was during a non-typical driving condition, the MIL may remain ON longer than the three ignition cycles. Review the Freeze Frame/Failure Records for the last failure conditions.

The PCM turns the MIL OFF after three consecutive ignition cycles that the diagnostic runs and does not fail within the same conditions that the DTC last failed.
A History DTC clears after forty consecutive warm-up cycles, if this or any other emission related diagnostic does not report any failures.
A last test failed (Current DTC) clears when the diagnostic runs and does not fail.
Use a scan tool in order to clear the MIL/DTC.

[PurplePiss]

I bet a cam drives it crazy.

[EC_Tune]

Definitely. I usually multiply the low end (Idle events) for cams by 2 as a starting point.