for the efi specialists - maf question

[upinthehills]

No manufacturer that I am aware of directly detects coil failure. They just whine that there is a misfire and it is up to you to figure out why.

I would expect they all do. I'm using a Freescale ignition and coil driver IC. Like what Joe 90 was pointing out with the LS ignition, there is a very small resistance in the circuit for the coil primary. This is used to monitor the current flow in the coil primary. When the primary has energy put in to it and it is then shut off, the energy is expected to do work by lighting a spark on the plug and going to ground in the engine block. If the spark does not occur ( and the coil wire isn't shorted to ground ) the current in the coil primary doesn't decay as quickly. All that energy has to go somewhere and it is being absorbed by the transistor that drives the coil.

This extra energy being absorbed by the coil driver can eventually lead to failure. Well, it will lead to failure because someone can save a few cents by using a smaller transistor here if the driver circuit is smart enough to turn it off or reduce it's power when there is trouble. So that is what you get.

[EthylCat]

Really cool stuff, Do you know what vehicles are using these? I'd like to dump the code and disassemble [/quote]


I have not personally come across this generation of technology on a production vehicle yet. I do not know if it is currently being utilized by anybody or not.

[mk e]

I read that some guys from the msextra team tried something with ionsensing but it has been abandonned apparently.

I've been told that's what H-D is doing. They use is for both knock detection and instead of a cam position sensor.

[peejay]

Here is a combustion sensing ignition system:

Pretty cool stuff

Looks like GM disseminated the SAAB technology to Delphi when they bought SAAB That's how they were doing it.

I would expect they all do. I'm using a Freescale ignition and coil driver IC. Like what Joe 90 was pointing out with the LS ignition, there is a very small resistance in the circuit for the coil primary. This is used to monitor the current flow in the coil primary. When the primary has energy put in to it and it is then shut off, the energy is expected to do work by lighting a spark on the plug and going to ground in the engine block. If the spark does not occur ( and the coil wire isn't shorted to ground ) the current in the coil primary doesn't decay as quickly. All that energy has to go somewhere and it is being absorbed by the transistor that drives the coil.

That is all well and good, but out of hundreds of failed coils that I have seen on OBD-II vehicles, exactly none were self-diagnosed as a coil fault. The most that the computer would be able to do would be detect if the low level circuit was open due to being left unplugged (circuit open fault), everything downstream of that is built into the coil's electronics.

The failure mode is not one that would be easily self diagnosed... the secondary windings find a new ground through the case, or sometimes the primary windings. On non-"smart" coils like Fords, this can result in all sorts of electronic gremlins as the voltage spikes feed back to the computer. But it doesn't get self-diagnosed as a "coil fault", just a misfire... assuming that the computer is CHECKING for misfires, which it may not be depending on conditions. Just happily run on nine or seven cylinders, which is almost normal for a Ford OHC anyway

[Belgian1979]

Well the sync loss with the crank trigger sensor is still there. Seems that it picks up a 59th tooth where there should be none.

[mk e]

Well the sync loss with the crank trigger sensor is still there. Seems that it picks up a 59th tooth where there should be none.

Detecting extra teeth normally is noise on the line but it's going to be hard to figure out and solve without a scope trace.

You didn't by any chance run a data log during the run that blew out your muffler did you? I would love to see what that MAF signal looks like across the RPM range.

[Belgian1979]

Well the sync loss with the crank trigger sensor is still there. Seems that it picks up a 59th tooth where there should be none.

Detecting extra teeth normally is noise on the line but it's going to be hard to figure out and solve without a scope trace.

You didn't by any chance run a data log during the run that blew out your muffler did you? I would love to see what that MAF signal looks like across the RPM range.

No I had all additional signals put off to prevent interference. So no maf.

I think I know what happened. I had decel fuel cut set to 80%. I let go of the gas a little because I was already at pretty high speed, the PW dropped and a backfired occured.

However it gave me a chance to see what was going on with AE and map. The map goes to almost atmospheric pressure with any TP over 25%. So in effect it doesn't determine fueling anymore, only rpm does.
I also had a lot of problems with AE not coming on soon enough. I had mapdot ae set to 50% and it worked better as at low rpm, the main factor is map, at anything over 25% tp, I need to use TP AE.

I'm going to try itb load as soon as a get back from holliday.

[mk e]

Well the sync loss with the crank trigger sensor is still there. Seems that it picks up a 59th tooth where there should be none.

Detecting extra teeth normally is noise on the line but it's going to be hard to figure out and solve without a scope trace.

You didn't by any chance run a data log during the run that blew out your muffler did you? I would love to see what that MAF signal looks like across the RPM range.

No I had all additional signals put off to prevent interference. So no maf.

I think I know what happened. I had decel fuel cut set to 80%. I let go of the gas a little because I was already at pretty high speed, the PW dropped and a backfired occured.

However it gave me a chance to see what was going on with AE and map. The map goes to almost atmospheric pressure with any TP over 25%. So in effect it doesn't determine fueling anymore, only rpm does.
I also had a lot of problems with AE not coming on soon enough. I had mapdot ae set to 50% and it worked better as at low rpm, the main factor is map, at anything over 25% tp, I need to use TP AE.

I'm going to try itb load as soon as a get back from holliday.

Do you have logs of baro and MAP? The fact that the throttle is affecting output tells you that you are restricting MAP in a significant way....but the question is whether or not its a clean enough signal to differentiate that change from baro in a repeatable way.

But as you say, TPS will give you an order of magnitude more resolution so I guess why keep messing with MAP.

[Belgian1979]

I don't see any map signal interference. Maybe the 5 mm holes between the runners and vacuum chamber. However i think it's just map being not useable anymore at a certain tp % .

Cannot post the log as i'm in the south of France and haven't got my laptop with me