You generally don't need an oscilloscpe to check a PWM signal, at least on a Cat machine anyway. A DMM that will read duty cycle will do fine. I swear by the Fluke 87, had the same one for over 20 years now, and it's still a current model so it must do something right. Currently seems to be priced at about $300.
IMO unless as a starting point the stabilized supply voltage from the ECM is not checked, and also the ground back to the same source, no PWM signal can be relied upon as being correct. See below for the list of things to check for a typical PWM sensor fault. I'm not sure the self-test even checks the stabilized voiltage (5v, 8v, or 10v) that is being produced by the ECM.
The possible causes of this diagnostic code are listed:
- The ECM + supply circuit is open.
- The ECM - return circuit is open.
- The sensor signal circuit is open.
- The sensor signal circuit is shorted.
- Debris on the sensor.
- The sensor has failed.
- The ECM has failed. ECM failure is unlikely.
Going on a bit further, and maybe this is the answer to your question...........
The duty cycle of a PWM sensor is measured versus time. See the illustration below that shows what the output signal looks like at 10%, 50%, & 90% of the sensor Duty Cycle. Note that the only thing that varies is the "ON" time (grey shaded area) vs the "OFF" time (white shaded area).
The key here is that the "height" (red arrow) of the square wave is proportional to the voltage in the sensor signal wire - not to be confused with the stabilized voltage supplied to the sensor by the ECM.
So in addition to receiving a % Duty Cycle signal the ECM also has a range of upper & lower acceptable voltages (heights) of the square wave signal between which the signal is judged to be correct. Anything outside that range and the voltage in the signal wire is judged to be either "too high" (FMI03) or "too low" (FMI04).
View attachment 230603