gt37vgt

wel duh???
i know that you will already be looking up aht i'm interested in.
so were did you get it and how much?
it is sure to be around on pirate disc somwhere

Geneqco

Amazon... wasn't too much, forget exactly.

gt37vgt

latest on said research i just ran my theory past a porsche trained tech i met on line he said it should work . i'm now kind of leaning toward running my programmable ecu as a piggyback . what do you understand about the crank trigger .
132 teeth speed ?
or is it 132-1 ?
5volt square wave out put ??
its often referred to as hall then sometimes magnetic ,
although hall is magnetic it should be called hall as it gets confused with old in the distributor sighn wave magnetic. then the reference signal there is one or 2??

Rogue_Ant

Best to get a scope, and find out. There are people that know, but don't share...



Rogue

Bri Bro

On the 944, you have 132 teeth that the KLR is using to adjust the timing. The speed/ref are a Variable Reluctance (VR) speed sensor. The inside of the sensor has a large amount of thin wire (coil) wrapped around a magnet. As the teeth of the flywheel passes close to the tip of the VR, the magnetic field is attracted to the tooth changing the magnetic flux. Anytime you change a magnetic field across a coil, you will get a voltage (Lenz's Law). The amount of voltage is related to the gap between the VR and the tooth and speed. The VR requires no external power source to work. So the output isn't a square wave, more like a half wave sine signal (signal occurs when the sensor has a tooth or pin move over it) that will increase in voltage in direct relationship with the RPMs. The voltage starts out around 1 volt peak at idle and climbs in voltage with RPMs.

fast951

On the 951 there are 2 crank sensors. The "speed" sensor, which reads the teeth off of the starter ring (132 teeth). The second sensor is the "Reference" sensor, which reads 1 tooth off of the flywheel. The speed sensor is used to calculate the engine speed and the ref sensor to indicate TDC. Both signals are sine wave.

The KLR does not read the Speed and Ref signals, just the DME.

edit: The KLR does not advance the timing, it retards it when it detects knock.

gt37vgt

awsome thanks guys i've got a scope waiting for mre at home . be home in a couple of weeks .

Geneqco

Hey John,

You obviously know a bit about these systems... would it be more accurate to say that the KLR neither retards nor advances timing but provides signals to the DME and it is the DME that controls timing?

The Workshop Manual seems to indicate that in the absence of any signal from the KLR, timing will be retarded 6 degrees as a default/safe mode which would tend to suggest that the DME advances timing when it receives an appropriate signal from the KLR (the 6 degree retard being the defalt position when no signal is received from the KLR).

Do you know if this is in fact how it works in practice?

The terminals of interest are:
KLR T16 – output to DME T32: TD* final stage of input from K/CP control;
KLR T24 – input from DME T31: trigger signal for K/CP control.

It appears the KLR gets an ignition input signal to T9 from T21 of the DME (which is the same signal used for the tacho). It seems that the KLR then manipulates that signal before returning it to DME T32.

The KLR also uses a trigger signal from DME T31.

Do you know the nature of the trigger signal from DME T31 to KLR T24 and how the KLR uses this?

Do you know the nature of the output signal from KLR T16 to DME T32, how it differs to DME T21 and how it is used by the DME?

If you are able to shed some light on these two questions, it would be a tremendous help.

Thanks for your interest.

gt37vgt

also just found that the later twinn cam has a nice hall out put of the back of the cam gear giving a nic 2 5 volt square waves per revolution and if disconnected yeh 6 degree retard . if a knock sensor fails 6 degree retard from the dme so seems that dme has a -6 safe mode .

Transaxle

The KLR is a pretty simple system...

It gets the the ignition signal from the DME and sends it normally directly back to the final stage in the DME. If necessary (e.g. knocking), it retards the signal in two steps of ~3° before returning it. The DME does not touch the returned signal anymore - it is directly feeded into the final stage.

Without the KLR the DME/Engine losses power because there is no WOT signal. Nothing else happens.

The critical path using the KLR with another ECU is the trigger signal. This is send some time before the ignition signal and resets the KLR to prepare it for the next ignition.

Geneqco

Thanks Transaxle, that's great information. Can you tell us a bit more about the trigger signal to the KLR... it's form and nature and the best way to get an aftermarket ECU to produce the appropriate signal?

Thanks again.

Transaxle

A short peak signal (~15us). 3 teeths in advance of the ignition signal.

gt37vgt

so lets sum this up the reduced performance in the absence of the KLR is simply due to the dme never seeing a wot signal any more .
you said the ignigton signal returned to the dme is unchanged ...... so is it then then simply fed striat out pin 1?????? the time diferntial you speak of does this mean that the signal to the KLR is one or to rpm behind the ignightion signal . guess it doesn't matter if it is a full cycle out of phase.
the more i lern the more i feel i should run my ecu in a piggy back configuration .
but if i can the timing signal through the klr untimed at 0 to -6 degrees i can then use my aftermarket on to set the advance and stil get antilag wich needs heps of retard and you can't go past static.

perhaps lost in translation i dont understand Transaxles last post describing the signal .
I will go study the Transaxle page now looks great..

also Geneqco a after market ECU is more flexible with the signal in to than it is with the signal it puts out the a ignighter this is why i lean toward the klr modifiing the signal pre ECU.

Transaxle

Sorry for my bad English...

The ML3.1 includes a internal final stage - firing the ignition coil via pin #1. In the 951 version the line from the CPU to the final stage is splitted and leads through the KLR. So the KLR can retard the ignition timing.

To prepare the KLR for a the next ignition signal, it needs to be resetted. That is done with the trigger signal. The timing is controlled by the DME. The trigger signal is sent to the KLR 3 flywheel theeths before the ignition signal is sent. The duration of the trigger signal must be at least 15us.

gt37vgt

your English is better than my German will ever be

the 2 things i dont understand are "ML3.1" and "15us" is that 15 milliseconds? millivolts? and is the signal from the klr go stait out pin 1?