Front Mounted CAS

[DJpowerHaus]

I figured I'd finally give moving the sensor to the front a shot. I was inspired by someone else who had done it, but with a 4 tooth trigger and no home to worry about syncing. I'm setting up for my car with a megasquirt (DIYPNP) which allows me to log the trigger signals and make other adjustments. Ideally, this research will result in a trigger disk that will work with the OEM ECU. It would make a pretty simple kit:

custom laser cut trigger disk
mounting bracket
coupling bolt for cam gear.

Making a bracket and coupling seems easy enough.(sorry for big sideways pictures)






Now I really want to figure out this trigger disc.

Using the composite logger I have logged the OEM sensor spinning the stock direction which recognized the correct rpm as expected with no sync errors:


Tuner Studio Log File
http://projectzerog.com/images/members/ ... estOEM.csv

I've also logged the sensor spinning backwards with the disc flipped and the MS recognized the correct RPM with no sync errors.


Tuner Studio Log File:
http://projectzerog.com/images/members/ ... Mount).csv

Comparing the two however reveals that the signals do not match, one is offset, most likely do the the offset of LEDs in the Mitsubishi CAS.

Can I correct this offset with any settings in Tunerstudio?


Ideally, I'd like to solve it by having new trigger discs laser cut so that it will work with the OEM ECU. I scanned the trigger disc into the computer at the highest resolution possible and traced it out in Sketchup (I <3 Sketchup).

Scanned CAS trigger (8mb):Click to Download
Sketchup Drawing: Click to Download

I can export it as a .dxf to laser cut at the local hacker space (Baltimore Node). First I need to figure out what I need to do to modify it. I can rotate the 4 outer trigger slots relative to the center key as well as the inner trigger slots to .1 degree in Sketchup. How can I figure out how many degrees and which direction?

Any help would be great.

[DJpowerHaus]

With a little measuring from my drawing in Sketchup and then rounding* I come up with this as my 2 triggers:

60, 30, 60, 30, 60, 30, 60, 30
30, 150, 70, 110

*I can't imagine the engineers who designed this would use arbitrary angles, so I'm not too scared rounding them to the nearest 5th degree since at most its 1 degree from my measurements which can only be so accurate considering manufacturing tolerances, my scan's accuracy, my tracing of that scan.



These 2 patterns locked, I can rotate them independently pretty easily in Sketchup. Seems as though they are about 5 degrees off from where they would line up in 2 spots. I'm eyeballing my datalogs and guessing that my OEM sensor pattern and flipped/reverse pattern are off around 25 degrees (less than 30, more than 15). Not sure which direction. Will see if I can get some prototypes laser cut so I can benchtest and log.. think about.. dial in again.