Hey Bill, just cheerleading on here, I bet it's crazy simple.

Have you put the VR right on the trigger wheel? I pushed mine until it touched and just wiggled a little bit back and all seems to be ok.

Just curious if you experimented with different gaps.

Question; do you have BOTH ends of the VR sensor grounded???

I was told here that ONLY ONE end should be grounded for a proper "Faraday" cage.

Sorry if this has been addressed......

Tippy,

I haven't played with the sensor gap (yet). I have checked that it is set at what the manual says ... 1mm. I've had suggestions to both increase and decrease the gap, but it seems more logical to try and close it up. I'll try that at my next session.

As far as shield ground is concerned, it is grounded at one end only.

Can you just pop the pins out of the connector and switch position 1 and 2? I am not sure where you tie into the system.

I would have to undo some of the heatshrink behind either connector to do this, so I'll make sure I need to first. Bought some M and F connector terminals and I'll solder up a short three-wire jumper to plug into the existing connectors.

Tried a few things tonight.

First, I switched the FW sensor polarity. This made no difference to the log.
Then I switched ignition capture to rising edge. Finally I have found something that has an effect ... this log showed NO signals at all!
Next change was to put sensor polarity back to what it was originally, still with rising edge ... still no signals.
Switched back to falling edge and log returned to normal ... sorry, abnormal!

Final adjustment was reducing FW sensor gap from 1mm to 0.2mm. This also had no effect and log still looks like this.
The red signals are at 58 spacing and the other peaks are 38-40.

http://forums.pelicanparts.com/uploa...1403618583.jpg

Just as a reminder, the log below is what we have with no compression. The large pulses are at 58 spacing.

http://forums.pelicanparts.com/uploa...1403619083.jpg

Before doing that, do a continuity check between the sensor connector pins 1, 2, 3 to the DB37 pins 24, 1, 2, respectively. Also check that sensor connector pin 1 is NOT shorted to anything ELSE )I doubt it is, or you would have OTHER symptoms.)

Also disconnect the DB37 and check to see if connector pin 3 is grounded at the engine. It should NOT be grounded to both the MS3 and the engine -- ground at one end only.

While you have the DB37 disconnected, test if any OTHER sensors are grounded to the engine. They should NOT be. I found my CLT sensor was inadvertently grounded this way --fixed the connection and cleared up some noise problems.

If this does NOT check OK, then you might have to attack the harness. If OK, then go with Jamie's oscilloscope thoughts.

Brian

Here is a sample from the MS3 forum with annotations showing reverse polarity of the sensor. The "phantom missing tooth" is evident, the same as it appears on yours.

http://forums.pelicanparts.com/uploa...1403623623.jpg

Bill,

Since you have the pins for the crank sensor and you have already opened the ECU case side of the db37 I would suggest that you build a single, long connector from the VR sensor to the DB37. Use a coax shielded wire and make a single run outside car away from the harness. Disconnect the current wiring and connect your bypass. This will eliminate any chance of a crossed connection and noise pickup through the harness.

Right now we need to isolate the problem even if it means taking apart some of the wiring or just bypassing it.

VR polarity is wrong for the selected detection (rising or falling edge) giving you a double missing tooth.

Bill,

The spikes at 58 counts are your "missing" tooth area on the flywheel. See the information below from the MS Extra manuals:

"TunerStudio Screenshot of real 60-2 signal during cranking with compression effect. "

http://forums.pelicanparts.com/uploa...1403628421.jpg

"In the above 60-2 screenshot you can see 57 normal teeth and the single extra long "missing" tooth. The cyclic nature of the teeth is due to the engine speeding up and slowing down during cranking. The OEM has place the missing tooth in a good place to avoid confusion with the compression effect. With poor choice it can be difficult for the code to reliably detect the tooth."

The consensus certainly seems to be pointing to setup of the sensor - polarity, input capture or noise in the harness.

When I used rising edge input capture, there were no pulses in the log regardless of polarity, so it seems like falling edge is best option.

Reversing polarity didn't make any difference to the logs. But looking at the sample logs in the MS manual it seems like the obvious answer.
Just thinking about that - I only ran yesterday's logs WITH compression, never thought to try with plugs out. Might try that tonight and reverse polarity again to see the effect with no load.

I have been contemplating running a separate shielded cable from the sensor to the ECU just to discount any interference from within the harness.

What is the effect of R56 and R52 in all this? My understanding is that they can be used to tune out this type of interference. I have incrementally adjusted both with no obvious effect. I got R56 up to 4 turns and R52 up to two turns.
Should I have carried on further with the adjustment?

The old geezer from next door came over again to see what I was up to. He knows zip about this stuff but is quite inquisitive so I tried to explain what was happening. His response was "if the spark is all f'd up when you crank it, why not just push start it?" That might be worth a try too!

Get the old geezer back. My mom used to come in while I was stuck on a homework problem and ask me to explain it to her. Grumpily, I did ... and in the process figured out the answer. She had left long ago, not really wanting to hear the answer ... only wanting to get me to think in fundamentals.

I like jpnovak's idea of a separate wire to feed the VR to the ECU, bypassing the harness.

I'd turn both pots back to full CCW to eliminate another variable.

I believe this may have been part of my early suggestions in this thread. I isolated my trigger noise issue in such a manner years ago.

I turned both fully CCW some time ago after they didn't appear to make any difference. I still wonder whether I should try going more than 4 turns CW.

I've only dealt with my install, but I believe full CCW is the base setting.

I bet if you had an oscope, you'd see big changes as I did with little movement of the pots. That's why I said it'd probably be a good idea to return them to their base settings.

Sorry Bill to throw out another question that has possibly been covered already, but are using the correct trigger method, ie VR, optical, Hall?

I assume you have since you are getting readings but thought I'd ask.

Tippy, I checked most TS setup screens but I can't find where to make that setting.

For reference, this is what current Ignition Options looks like.

http://forums.pelicanparts.com/uploa...1403663820.jpg

What is "Spark hardware in use"?