PDA

View Full Version : How are engine misfires detected?



Stephen Tashiro
02-19-2022, 10:37 AM
How are misfires in automotive engines detected?

The computers in modern cars can report misfires in individual engine cylinders. How are these misfires detected? Are they deduced from the motion of the crankshaft? - or are there sensors inside each cylinder?

Paul F Franklin
02-19-2022, 10:45 AM
Misfires are detected by monitoring the acceleration of the crankshaft after each cylinder fires (or is supposed to fire) compared with the average acceleration of the crankshaft over a complete revolution. The acceleration is calculated from the crankshaft position signal, so the main sensor involved is the crankshaft position sensor.

Erik Loza
02-19-2022, 10:53 AM
My understanding is that the ECU detects a fault in the wiring circuit and reports it as a misfire. There is no sensor for this fault, per-se. A squirrel chewed through the fuel injector harness on my 4Runner once. I started the engine, heard/felt the misfire, and of course got the CEL. So, I think if the ECU detects a fault in either the fuel injector or ignition coil circuits (the only two things that could really cause a misfire in a modern vehicle), it is programmed to report that fault as a misfire. Now, there is an actual sensor that detects combustion cylinder detonation or pre-ignition (aka "pinging" or "knocking") called the Knock Sensor. This sensor detects harmonics and is actually threaded through the block, into the cylinder water jackets, since liquid (engine coolant in this case) is so efficient at transmitting soundwaves. Not sure if any of this answers your question but this is how I understand it.

Erik

John M Wilson
02-19-2022, 11:24 AM
How are misfires in automotive engines detected?

The computers in modern cars can report misfires in individual engine cylinders. How are these misfires detected? Are they deduced from the motion of the crankshaft?

Way, way, back in the old days ('70s and '80s) I was an engineer for Cadillac, responsible for the assembly line machines that interfaced with the on-board car computer & electronics. Very primitive stuff compared to what they have now. One of the car sensors is located in the exhaust manifold, and measures the amount of oxygen remaining in the exhaust gases. This is used as part of a closed loop system with the fuel injectors to make sure that the engine is not running too rich or lean. Even with the technology we had then, we could detect each cylinder as a "pulse" and could diagnose a malfunction (like an inoperative cylinder). At the time, we couldn't pinpoint it to a specific cylinder, but the advances in the technology have been astronomical since then, and my semi-illiterate guess is that the on-board computer has access to the crank position signal and the oxygen sensors (many cars now have multiples of these) and can pick out an individual pulse that is way rich (spark malfunction) or way lean (fuel malfunction) and tie it back to a particular cylinder.


- or are there sensors inside each cylinder?

All of the sensors that I am aware of are external to the actual cylinder, located in strategic places like the engine block (knock sensor), intake air system (mass air flow sensor), exhaust system (oxygen sensors), etc. The combustion chamber located between the piston and the head is a very, very, highly engineered space, and is also quite hostile to electronics. I would be surprised if consumer-oriented engines had any electronics in there, but I wouldn't put it past those Formula 1 guys, who have limitless budgets.

Rick Potter
02-19-2022, 12:30 PM
I just had a mechanic figure out a very slight intermittent miss on a 2005 T-Bird. His portable scanner told us immediately which cylinder it was. There are coil packs on each sparkplug, and we switched the indicated coil pack to another cylinder, and it immediately showed that one as bad. It was using the ignition control module info, as I recall.

Impressive, and all done in my garage.

Michael Drew
02-19-2022, 12:46 PM
It's complicated, but Paul Franklin summed it up well.

Modern engines with a computer monitoring and controlling how it behaves use a lot of math in their functions. Misfire detection is a hot mess to try and understand. And each manufacture has their own strategy too. And - and, each manufacture changes strategy with each model of controller that gets introduced. I calibrate my personal vehicles that I modify, but I could never do it professionally. There's simply too much to learn, and too much to keep up on..... Good "tuners" are friggin geniuses. "Tooners", however, are clowns that should be avoided.

Here's a simple description from the HPTuners misfire detection help menu.

Misfire Detection Temps


Max Temp: The maximum engine coolant temperature to which Misfire Detection tests will run
Min Temp: The minimum engine coolant temperature to which Misfire Detection tests will run


Misfire Max Time Between Events:


Idle: Increase these values to max misfire tests less sensitive.
Low RPM: Increase these values to max misfire tests less sensitive.
Med RPM: Increase these values to max misfire tests less sensitive.
High RPM: Increase these values to max misfire tests less sensitive.


Misfire Max Time Between Events By Cylinder vs. VE% vs. RPM


Idle (Not Moving): Increase these values to max misfire tests less sensitive.
Idle (Moving): Increase these values to max misfire tests less sensitive.
Med RPM: Increase these values to max misfire tests less sensitive.
High RPM: Increase these values to max misfire tests less sensitive.


This is the where adjustments can be made to the misfire detection logic for one of my cars. This one has a pretty rowdy camshaft, so I essentially disabled misfire detection at idle and low rpm by maxing out the cells for that rpm range <1600.

https://i.postimg.cc/J4zYjNrT/missfire-DTC.png (https://postimg.cc/bd4Rh22b)