|Vehicle details:||Ford Fiesta 1.3 16 V|
The topic of this case study is not so much "How we managed to diagnose the fault in question" but more about the multiple options at our disposal when it comes to diagnosing component failure.
With scan tools, scopes, multimeters, vacuum/pressure gauges, gas analysers, jump wires, neons, power probes, clamps and smoke machines (to name just a few), we are often faced with a dilemma of what tool to select for the right test, to bring the right conclusion at the right time (to quote James Dillon).
Here we have a 2001 1.3 16 V Ford Fiesta, with a complaint of misfire under all engine run conditions. The vehicle arrived at local business Ives Garage (owned and run by Kevin Ives) where we are often invited to experiment with vehicles demonstrating a variety of faults in order to gather real life data.
To get straight to the point a scan of the car showed diagnostic trouble code (DTC): P0301 - Cylinder 1 Misfire Detected. This was a permanent fault code and a permanent Cyl 1 Misfire that proved to be fuel injector Cyl 1. However, I hope the mention of the cause does not prevent you reading any further as with this case study, we knew from the outset why the vehicle misfired, what proved of additional value here was:
"Could we apply alternate techniques in which to conclusively test and condemn the injector?"
Imagine a vehicle arrives in your workshop with identical symptoms, how would you most likely proceed? I guess exactly as Kevin did with a scan of the vehicle, followed by a visual inspection, 4-Gas test, and scope test (voltage and current) of the injector. All lead to a conclusive diagnosis of the injector and a happy customer.
It is at this point I would like to step back and add more tests to the diagnostic process. I hope that these tests will highlight other options we have at our disposal while also mentioning some pitfalls of what on the surface is a straightforward diagnosis.
The scan tool is the first port of call allowing the technician to record the fault code and evaluate live data for parameter errors such as fuel trim, lambda, O2 sensor activity, misfire counts etc.
|Oil temperature||69 °C|
Very quickly we can establish the misfiring cylinder and a lean engine run condition, leading onto confirmation using a 4 gas analyzer test producing the following results.
A word to the wise here regarding lean misfire detection:
The PCM has the ability to switch off the fuel injected into an underperforming cylinder very soon after a misfire is detected (Catalyst protection-Emission regulations).
Therefore if we are presented with a misfire as a result of a "no spark" condition, the PCM will shut down the fuel into the relevant cylinder presenting the technician with a lean misfire 4 Gas test results!
So how do we know our 4 gas analyzer test results are symptom and not cause?
Relatively easy by monitoring Cyl 1 fuel injection signal using PicoScope after the fault code has been cleared, while starting and running the engine.
The PCM will need to gather data from numerous sensors in order to make an informed decision on which cylinder to cut (misfire assessment). During these precious seconds, conventional fuelling will be resumed and our 4 gas analyzer test results will display true emission values.
High CO% eventually changing to low CO% with high O2% (once injector is cut) indicates a misfire relating to a rich condition (fuelling therefore must have been present before injection cut).
Low CO% with high O2 continually from engine start indicates a misfire relating to lean condition outside the control of the PCM (Injection signal initially present during lean 4 gas analyzer test).
In the case of our Fiesta, our CO% remained consistently low with high O2 levels while Cyl 1 injector signal remained stable throughout (no injector cut).
We therefore have a misfire as a result of a lean condition outside the control of the PCM.
The waveform below indicates Cyl 1 and 2 injector signals overlaid for comparison.
Using the Reference Waveform feature of PicoScope we have captured Cyl 1 and 2 injector voltage and current signals and aligned them for direct comparison. Immediately we are drawn to the Pintle Hump sector of the waveform.
Cyl 1 Injector voltage signal is clearly missing the characteristic pintle hump (blue trace). The injector pintle hump is formed when the injector pintle passes through the solenoid winding core on closure, disrupting the magnet field on route and revealing this event in the induced voltage decay. The absence of the pintle hump confirms no injector close event suggesting there was no injector open event, hence a lean misfire.
Note: The injector open event (pintle lift point) can often be seen in the injector current ramp, but on this occasion there was no difference between Injector 1 or 2 current. Therefore do not rely on current waveform alone during injector evaluation, use both voltage and current.
With all the above information we have enough evidence to warrant an injector replacement. However, below we have some additional tests you may wish to consider that confirm beyond all doubt an injector is required.
How about listening to the injector (a very simple and non-intrusive test)! Accompanying injector operation is the characteristic ticking sound that will increase in frequency with engine speed and stop during over-run fuel cut off. The faulty injector on this vehicle produced a normal ticking sound! If we could visualize the ticking sound using our scope would it be possible to identify an injector failure?
Using the NVH kit we simply applied the accelerometer (incorporated within the kit) to listen to the injector operation and more importantly, present a visual display.
Channel D (yellow trace) has revealed the inactivity of Cyl 1 injector even though an audible ticking noise was present. Each injection event is clearly detected by the accelerometer attached to Cyl 1 injector only. What is intriguing here is the amplitude of Cyl 1 injector was by far the lowest (should have been the highest), and the amplitude of the remaining injectors reduced relative to the position of the accelerometer. Number 4 injector has the lowest amplitude (after number 1 injector) as this injector is furthest away from the accelerometer. Number 2 injector therefore demonstrated the highest amplitude (adjacent to number 1 injector).
What is clear from the accelerometer method of testing the injector is the fact that activity during the injection event can be monitored and displayed to produce a direct comparison between functioning/non-functioning injectors.
Moving onto fuel pressure we know we have an injector that is not contributing to Cyl 1, therefore the temptation was too great not to monitor both the high pressure delivery and fuel return lines for activity under engine run conditions.
Please observe all the relevant health and safety procedures before and during any fuel related measurements.
With the WPS500X placed in the return line:
Channel A indicates Cyl 1 injector voltage while Channel D reveals the activity from the remaining injectors using the accelerometer connected to Cyl 1 injector.
What is clear from the captured signal is the sudden plunge in the fuel return line pressure after each injection event for all cylinders, except Cyl 1 where the pressure actually rises. We can also identify a large fall in pressure for Cyl 2 injector, in comparison to Cyl 3 and 4. This could indicate a greater contribution of fuel from Cyl 2 injector or a measurement error! Air contained inside the WPS500X pressure transducer could present sporadic test results when measuring fluctuations in fluid pressures but I can confirm no air remained after extensive bleeding of the transducer.
A useful tip when bleeding the pressure transducer for this style of measurement is to turn the pressure transducer upside down momentarily. This purges air trapped inside the measurement chamber allowing it to vent through the bleed port when the bleed button is pressed.
The next port of call was inevitably the activity present in the fuel delivery line.
First impression of the above waveform using WPS500 pressure transducer placed into the fuel delivery line (monitoring fluctuations only) is not a favourable one! This is no reflection of the pressure transducer, more my expectation believing a failed injector would be clearly visible in the fuel delivery rail.
While there appears to be reduced fluctuations in the fuel delivery line during Cyl 1 injection event the results are far from conclusive and on this occasion (with this style of test) an injector diagnosis could not made.
Given our vehicle incorporates a fuel return, the fuel pressure regulator can often reveal the activity present in the fuel return line as a result of movement of the vacuum diaphragm. Attaching the WPS500X or First Look Sensor to the fuel pressure regulator is another opportunity not to miss, and one that presents a far less intrusive insight into fuel return line activity.
The above waveform indicates the current passing through each injector via fuse 9 of the engine bay fusebox (green trace), revealing all injection events. Both the red trace (First Look Sensor) and the black trace WPS500X (reference waveform) confirm pressure transition after each injection event except Cyl 1.
I think at this stage it was safe to say Cyl 1 injector is most certainly at fault but could there be another option? The answer most certainly is YES and the process below allows the technician to calculate injector contribution and balance between each injector (in situ) in conjunction with PicoScope, the WPS500X pressure transducer and a third party injector driver.
Here the WPS500X pressure transducer is installed into the fuel delivery line, air is purged from the measurement chamber and all fuel injectors disconnected. The ignition is cycled twice (Ignition on-ignition off twice) to prime the fuel line to the relevant line pressure. A third party injector driver is then connected to Cyl 1 injector while PicoScope is used to monitor driver activity and residual fuel line pressure (see image below).
All things being equal, fuel line pressure should fall in direct proportion to injector operation (controlled by the 3rd party injector driver). The decay in residual fuel line pressure should be equal across all injectors (driven individually in turn) if contributing evenly.
The above capture takes further advantage of the Reference Waveform feature of the PicoScope software enabling all injectors to be displayed simultaneously and matched against the operational time of the third party injector driver (six waveforms from our 4-Channel scope).
The blue trace indicates supply voltage into the injector while the dark green trace indicates the ground trigger signal (both provided by the third party injector driver).
There can be no doubt looking at Cyl 1 injector trace, (red) that during injector operation there is no drop in residual fuel line pressure, hence no contribution into the chamber from Cyl 1 injector
Cyl injectors 2, 3 and 4 all demonstrate similar contribution levels based upon their respective line pressures.
Be aware that injector contribution testing (in situ) will introduce fuel into the combustion chamber, crankcase and ultimately the catalyst.
As a result please limit the number of contribution tests to one per cylinder or remove the fuel rail (complete with injectors) from the vehicle. Secure injectors into the fuel rail and utilise measurement tubes to capture any fuel dispensed from the injector nozzle during contribution testing.
To digress slightly from the fuel injector diagnosis, the effects of a lean mixture on ignition firing voltages (voltage initially required to jump the spark plug gap) can be clearly measured to provide further evidence of a lean condition.
Firing voltage is not only dependent on the integrity of the ignition circuit, but the conditions inside the combustion chamber. A lean mixture presents an increased resistance between the spark plug electrodes and so an increased firing voltage is required to initiate the ignition event. The waveform above highlights how the peak firing voltage reduced by approximately 35 % under full load conditions (WOT) with the new injector installed (black trace). The point I am trying to make here is that while we knew we had lean mixture condition, when no gas analyzer is available further evidence is always at hand.
Think of the combustion process as a triangle where each side represents the 3 key factors required for combustion. Compression-Mixture-Ignition.
With all three factors in place, in the relevant quantity at correct time we can be assured of complete combustion. However should we remove or adjust any these factors not only do we spoil the combustion, we experience the effects in the remaining factors and here is where we search for evidence.
Our lean mixture resulted in no combustion with a direct effect on ignition (firing voltage).
The purpose of this case study is to highlight the advantages of spending time with vehicles presenting stable faults while sharing some of the pitfalls as the evidence gained is invaluable.
Below is a summary of the techniques used above that will allow you to make an informed decision regarding which test best suits your application as it will vary from vehicle to vehicle..
Remember, above we have diagnosed a fuel system with a return line and no delivery from Cyl 1 injector. What if we are presented with a partially performing injector (poor spray pattern) with no return line? Not an easy question to answer but some of the tests above remain valid and would yield results. We must adapt to the vehicle under test with an array of test equipment and knowledge at our disposal of which PicoScope and Pico accessories remain essential.