PDA

View Full Version : How PCMs "see" a misfire.



Stevonator
04-18-09, 06:27 PM
Misfires can be caused by literally just about anything....and by NO means just ignition. So before you go out and buy the seemingly obligatory cap, rotor, plugs, and wires to fix a misfire, let's take a look at what a misfire is to you and isn't to a pcm....and vice versa.

The first very important thing to mention is the ignition system has absolutely nothing directly to do with setting a misfire code. For fact, while the PCM commonly tells the spark to happen....it has absolutely no idea whether anything high voltage even made it to the distributor....let alone the cylinder.

To you a misfire can be heard and felt.... Not so for a pcm. So how is it that a pcm can not only tell you that you have a misfire, but tell you what cylinder it is on? Read on.

First, understand that a PCM is a machine. It is not a magic box, but a machine that does whatever it is told by its program. Without its program...it's a PAPERWEIGHT. This program comes from a HUMAN BEING during vehicle development. This person has to figure out how to find a misfire (and a ton of other things). He can feel and hear a misfire just like you can....but that doesn't help the machine that he has to program.

No doubt he has thought about the ignition, but he has to have reliable results....and just checking the ignition isn't going to catch a bad injector, or anything else. So unless I want to hang a couple hundred sensors on the engine solely for detecting a misfire.....he needs to get creative.

Let's take a look.....well, can we paint a solid picture of what a misfire IS. It needs to be very simple and to the point so he can figure out what he has to try to detect and how to actually do it.

A misfire is any event that means a fuel/air mixture does not ignite and expand sufficiently inside of a cylinder......Wow....that involves pretty much the whole engine! How in the hell do you do that?

The crank sensor.

Hmmm, every time a cylinder fires it pushes the piston down. If I know what degrees of rotation this is supposed to happen (and I can figure that out) I should be able to watch the speed of the crankshaft at those degrees of rotation and SEE the crankshaft ACCELERATE.

Crankshaft speed may be measured in RPM, but it DOESN'T turn at a constant speed. I only need one tooth on a reluctor ring to measure RPM but to measure crank acceleration through the firing order takes quite a few more....I can do that. Wow...Slam dunk I've got this thing nailed....and with ONE SENSOR!!

Not so fast.....If two cylinders' degrees of acceleration are close (we adjust firing orders and to make these shaking, quivering 3/4 350s be smooth (3 times on a 4.3...odd fire, semi even fire, and even fire.) That we have to be careful not to FALSELY set a misfire code.

On top of that ALL engines misfire. We simply can't make the magic happen in every cylinder every time. So we have to track misfires over time and decide how many to allow. Who decides this? Not the programmer.....the EPA!. Once the total misfire count causes the engine to fail the threshold test on the dyno....the EPA says the light must come on.

How touchy can this process be? There are TSBs out there in which some vehicles can set misfire codes BY DRIVING DOWN A ROUGH ROAD....no joke. The PCM only knows what the sensor tells it. What happens when a stone flies up and puts a ding in one tooth of the crank sensors reluctor wheel (been there)....you get a misfire code....Is that cylinder misfiring. NOPE.....but you've got a code. Remember, the PCM only knows what that sensor tells it....and in this case it's a big fat stinky LIAR. Hook up a scan tool and all you see is the same thing as the PCM......A LIE.

There's also a flipside. In many cases, you may have a misfire and the PCM doesn't pick up on it. Particularly true when it's intermittent.....remember, there's a threshold to get a light.

If the PCM can't pin down a specific cylinder, but the total threshold is met, You get a P0300. If the threshold is met by any one cylinder, a cylinder specific (P0304) is set.

I get the WTF problems every day..... it's WAY more common than you might think. And no I'm not saying put a crank sensor in it because you have a misfire code !!!!!!!

King_Ice_flash
04-18-09, 06:32 PM
All this and no mention of the CASE relearn?

spideyjected
04-18-09, 06:46 PM
So does this apply to OBD 1 and 1.5 trucks as well? Our codes are different.

King_Ice_flash
04-18-09, 06:53 PM
There is no crank sensor on a 94/95.

spideyjected
04-18-09, 06:54 PM
Or a 93 YO!

12sws27
04-18-09, 07:25 PM
Maybe the title of the thread would be better if it was, "How OBD2 PCMs "see" a misfire"?

Stevonator
04-18-09, 10:03 PM
Come on now guys, you want me to be constructive, so let's not nitpick....If you're smart enough to know you don't have a CKP, you probably can tell the post does not apply to those without them.

I will now add the OBD1/1.5 section:

These trucks do not have crank sensors and therefore cannot set misfire codes.
The end.

spideyjected
04-19-09, 06:57 AM
I was hoping to maybe make this post be of some use to me actually. That's ok, do what you gotta do.

Stevonator
04-19-09, 04:18 PM
dt...I can't help the technology that came in your truck....Your truck doesn't even have a clue that it's got a misfire at all. That only makes my job (or yours) even harder to find it. Believe me I wish the earlier vehicles had it.