mark944turbo

Is ignition input. Is this the same as RPM signal? Where does it come from, the DME or directly from the distributor or something? What does the signal look like on a scope?

I cant find these answers in the manuals, maybe you guys can help.

TT

The tach signal is the same signal that feeds the ignition driver in the 944 but goes to the KLR in the 951. The tach signal comes from pin 21 of the DME. Check out the 944/951 electrical schematics in the workshop manuals. The signal is a square wave with the frequency and duty cycle directly related to timing and dwell.

mark944turbo

Thanks for the reply TT.

What I should have asked is does the 951 have a signal that says where the distributor is, or is the coil told when to fire by the crank sensor? I am a little lost about how the computer knows when to fire the spark plug, basically. I thought that klr 9 was a distributor position input, but I guess it comes from the DME. Could it be DME #1 (ignition impulse at ignition coil) #1?

TT

The DME only knows the cylinder pairs (1&4 or 2&3) to fire ignition, the distributor handles which single cylinder gets the firing pulse. Position info is derived from the speed sensor and reference sensor. Using these signals the DME knows where TDC for each cylinder pair is. The DME only knows that there are two ignition events per rev for a four cylinder engine.

The 968 has a distributor/cam sensor, but I believe that is only used in knock control since it has two knock sensors and needs to know which individual cylinder is firing. Other Motronics use the same type cam sensor to fire individual coil packs, but that signal is used like a distributor to select each coil pack. They all use a toothed wheel on the crank for timing.

mark944turbo

Yes, I remember this about the 968. Once while diagnosing a problem on the dyno, I turned a 968 over by hand with the ignition on. I was stopped by large amounts of smoke coming from the ignition unit! It was trying to make spark, but the problems we were having never let the coil do its thing.

Thanks again for the help tt, now I have to go back to the drawing board on my "top secret" project.

Tomas L

A somewhat related question.
Does anyone know where the reference sensor signal is in degrees BTDC?

mark944turbo

58.64 iirc. Someone correct me please if this isnt accurate, its from my notes not a manual or anything.

Lorenfb

"I am a little lost about how the computer knows when to fire the spark plug, basically. I thought that klr 9 was a distributor position input, but I guess it comes from the DME. Could it be DME #1 (ignition impulse at ignition coil) #1?" - mark944turbo -

The DME controls all the "basic" ignition timing via its' ignition maps. The KLR just
basically retards the DME timing based on a number of variables, e.g. boost, knock,
TPS, etc. The DME outputs a spark signal on pin 21 to the KLR pin 9. The KLR
"decides" the retard amount and outputs the signal on pin 16 to the DME pin 32.
The 944 can run without the KLR by just jumping pins 21 & 32 of the DME
or pins 9 & 16 of the KLR.

Additionally, the 944 DME never "knows" which cylinder is on combustion TDC because
it lacks cam info which the 968 obtains from the Hall sensor. On the 968, because
of the Hall sensor, the DME can selectively retard the pinging cylinder. The 944
type knock system requires all cylinders to be retarded. Also, in the case of the 944,
the DME can lose the reference sensor and still run because it counts the ring teeth.
The 968 knock system functions basically as does the 928 knock system,
i.e. 3 degree retard increments to a max of 9.

Tomas L

AFAIK this is not correct.
The DME only uses knock to decide if the ignition should be retarded. No other variables are used.
The 951 can retard only one cylinder.

Tomas

mark944turbo

I would be interested if anyone knows anything about the knock control algorithms and is willing to share.

According to a Porsche Club North America publication the 951 can retard individual cylinders.

mark944turbo

Somewhere in the filtering process for the original knock signal, rpm, load, or boost must be used. There is obviously a lot more valvetrain/etc noise at higher rpm, so the filter should be different based on these conditions, if it is any good, which Vitesse has said it is.

I think I read something about how the ecu in another car filtered the knock signal. It took the inverse of a normal signal for that rpm/load and added it to the actual signal, and anything left over would be analyzed for knock. It could be something like this.

Tomas L

I don't know if the KLR uses rpm or other variables to filter the knock signal, however I don't think the actual amount of retard is controlled by anything else than knock signal and that was how I read Lorenfb:s post.

Lorenfb

"The DME only uses knock to decide if the ignition should be retarded. No other variables are used." - Toma L -

Some more guessing again! Do some reading, like the 928/964 manuals which provide
info about the Bosch knock system. They all function basically the same.

"Somewhere in the filtering process for the original knock signal, rpm, load, or boost must be used." - mark944turbo -

Some can get it right!

"According to a Porsche Club North America publication the 951 can retard individual cylinders."

This requires info from a cam position sensor (Hall sensor).

TT

The 928/964 systems are completely different from the 951. Now who's guessing. According to Porsche's own 944 Turbo Service manual, the KLR only uses the knock sensor to determine if timing needs to be retarded. It retards timing by 3degrees upon detection of knock. It retards it again 3degrees if knock continues. If knock still continues it reduces boost pressure in increments of 30-50mbar.

And you get it wrong. Just the knock sensor input and KLR pin 24 to know when to listen for knock.

Funny, I don't see a PN in PET for the cam position sensor in my 951 and I don't remember seeing it when changing out the cam.

special tool

I told you that you are not in Kansas anymore, Toto.