I'm still learning this beastie. I've got a 4423 Automotive scope that I'm using to death. I've got a car, a 2006 Dodge Status 2.4l AT that is giving me some grief. I'm getting a P0335 crankshaft position sensor circuit failure. A new Standard Brand (not OEM) crank sensor has already been installed. This vehicle rarely exhibits it's problem. Most of the time when cold. By definition this code states that the ecm hasn't seen a crank sensor signal while 8 cam sensor signals have come and gone.
Currently I've set my scope up with three channels. One for the signal line, one for the 5 volt reference line, and one for the ground line. I've set the time division to 20ms, and the sample rate to 200MS. With this I get 32 screens of data. I believe the problem is bad enough when it is happening to get caught with that sample rate. I also used pulse width based triggering off the signal line. The signal has a repetitive long pulse width that allowed me to lock this pattern in really good.
I need to get in this car and crank it to see the problem.
How best to catch this? How would you set this up to catch it? I am thinking the first thing I'm going to need to do is change my trigger. It worked good for looking at a good signal, but will probably be prohibitive in looking at what I hope is a bad pattern.
Would you start the scope running. Wait for the 32 screens to be filled, and then crank away? When would you freeze the recording? Hopefully it won't start right away. So the moment it starts running while cranking hit stop?
I would set the trigger up so that a window trigger was used. Whereby if there was no voltage between the thresholds for x ms, it would trigger the capture.
Use 'Window Pulse Width' and set voltage for say 1v lower and 10v upper - you may have to play about with ranges to account for noise.
You can work out how long to make x, by first looking at the time length the signal spends in each voltage state under normal conditions for a known rpm, and then multiply for number of teeth and crankshaft revolutions required for code to be set.
Try 50% for the trigger offset.
You don't have a bad connection somewhere, do you?
It looks to me to be a bad sensor. I've changed it, and it is running without fault. I'm going to let it sit the rest of the day and restart it again. Honestly I'm not really looking for help at figuring out what's wrong with the car. I'm looking for help as to the best way to set the scope up to catch it. It does seem that setting the sample rate much lower helped me catch this much easier. I have to get over wanting to set the sample rate as high as I can. It did allow me to catch the problem without any help from anyone else. I set the scope up...walked over and cranked the car. When it finally started I ran over and stopped the capture. With 32 screens I was able to go back and see it. I didn't think I was going to be able to go back and catch it but I was. It was about 15 screens back, so perhaps I could have walked instead of run.
Image File
It was apparent that I was not getting a signal from the Sensor. At least that is my take on it. The ground and power lines remained constant. My signal line went to 5 volts and sat there until the vehicle finally decided to start. Maybe 5 seconds. But seems like forever. I would think if the computer was disrupting the signal it would be being pulled to ground. Seeing the 5 volt told me that the computer was waiting for input from the sensor, which would latch to ground.
I have to say I'm really liking all these different trigger methods. No scope I've had before would have been able to lock onto the missing tooth pattern that was being generated as my signal. I think next time I look at one of these....since the computer is stating within the code it is getting the cam signal all during the loss of crank signal...that I will use that signal as my trigger signal. I will have to try that.
One big question for me right now is - Is there a way to keep all the screens? I noticed after saving it that I am left but with what I was looking at on the screen at that moment. Is it possible to save it all? And if not can you scroll halfway through a screen and halfway into the next to better center the moment you were really interested in? Is there any other way to get a long recording - I'm thinking of when we have a vehicle where the modules won't go to sleep. I'd like to record for hours at a low sample rate.
This vehicle rarely exhibits it's problem. Most of the time when cold
Does that statement mean it mainly faults from a cold start ? Yes Absolutely
Post a capture of one complete revolution of the crank to see the complete signal .
When open circuit does the signal voltage stay high or low ?
Stays High.
I apologize for taking your question wrong. I am here to learn, and get better at this....so please ask away and I'll do my best to listen and learn.
I really want to figure how best to use the buffer. I was using an old tektronix DSO before this, and my biggest gripe was always trying to catch the problem. I'd have to sit there and pray that I could hit the button fast enough to get it while it was on the screen. I dream of being able to leave the scope hooked up to a car that occasionally stalls, walk away, notice it has stalled, and look back through the buffer to see what happened.