1991 Buick Park Ave. intermittent crank no start

1991 Buick Park Ave 3.8L vin L

code 42 for EST system fault is present
vehicle starts and runs probably 70-75% of the time
when the problem starts to occur the starter will buck as though the spark and fuel have ignited before piston is at TDC
when the problem is in full effect the vehicle will crank over but no RPM signal from the PCM and intermittently the tach will show cranking RPM but has also shown no RPM with the same crank no start result
I scoped both the 3x and 18x signals at the ignition module and found they were present when cranking also verified power and ground to the module
crank sensor, wiring pigtail, and ignition module have been replaced prior to me getting the vehicle (crank sensor pigtail wiring is suspect but nothing changes when the vehicle is running and the wires are pulled/moved)
I was able to scope the 18x signal yesterday while the vehicle wouldn't crank and noted that the square waves were rounded on the leading edges (I know the module mainly cares if they pass a threshold) but while cranking the square waves weren't uniform in their spacing in that some of them would be wider than others (I don't have a picture of this but it wasn't like changes in rotational speed it was that the wider square wave was in between evenly spaced square waves)
When the vehicle runs the 18x pattern is uniform
The included scope capture is as follows:
blue trace - 3x sync pulse
red trace - 18x ckp signal
green trace - starter current (since sometimes it would run)
yellow trace - 12v power supply to the ignition module
battery was low when cranking but still cranking the motor over well

Since the 18x waveform on the CKP sensor is the only thing I can point to as out of the ordinary (don't have a capture of this) and since doing swaptronics on the ignition module had no effect I'm leaning towards replacing the CKP sensor after inspecting the back of the harmonic balancer for irregularities. Any help or insight would be greatly appreciated.

Can you zoom in on the crank signal in your program and give us another photo.

As soon as I get it back inside, it got moved to the back of the line when the starter quit and the customer wanted to bring his own.

Ok, so after a few days I was able to stay late Saturday and today and I found that the 18x signal wasn't changing widths within the waveform as I had suspected. Looking at the Uscope on the wrong time scale made me think that was the case but looking at more time on the screen with the Pico showed me that the sync pulse (3x signal) was correct in that there were three distinct and non-uniform square waves for each revolution and the 18x (crank signal) was 18 evenly spaced up and down cycles per revolution. I ended up looking at the camshaft sensor even though I didn't have any code for it in the PCM the sensor was dead and when you hold a magnet against it terminal A of the pigtail goes to ground with the new sensor but there is no change with the old sensor. According to the service information the vehicle should have run without the input from the camshaft sensor but today the vehicle was a consistent crank no start until the sensor was installed. Hopefully we have a happy running car in the morning since I've spent way too much time working on this without us being able to bill for that time.

Sorry for the delayed reply i have been busy. With the cam sensor when they go bad like that it can trigger a ckp code. The pcm will use the cmp sensor to determine ignition on start up. When it goes bad the pcm will fire the spark plugs at random points till it fires. With the new sensor in did you recheck ckp signal to see if its still rounded over?