Drivability From The Front Seat

Things like airflow restrictions, changes in fuel pressure, camshaft timing and fuel quality always will show up in the fuel trim data on your scan tool. Only a few PIDs are needed: throttle position, engine rpm, engine load, STFT and LTFT for all cylinder banks, oxygen sensor voltage for all sensors and MAF data if you are working on a MAF engine.

Recap of Fuel Trim
Fuel trim is nothing more than the corrections in fuel injector pulse width that is made by the ECM (Engine Control Module). The fuel trims start with STFT, which is the product of the feedback from the oxygen sensors. As STFT is captured, the fuel trim correction is moved to LTFT, where it is stored in fuel trim cells. Here the LTFT data is stored in thetrim cell lookup tables so the ECM has the correct fuel trim data each time the engine speed or engine load is 

changed.

The oxygen sensors will notice anything that alters the normal air mass or normal fuel mass entering the engine. These sensors will give their feedback to the ECM, the module will make a correction in the fuel injector pulse width and this goes on as long as the engine is running in closed loop. By using your scan tool to graph the fuel trim data, the technician can get a good direction on most any drivability problem.

Sitting in my bay is a nice looking 2006 Chevrolet Avalanche. The odometer shows the vehicle has traveled 135,000 miles, the engine is the popular 5.3 VIN Z flex fuel powerplant with an automatic transmission attached. The only complaint on the vehicle is the Check Engine light is illuminated. Test driving the vehicle uncovered no apparent drivability problems.

With a scan tool hooked up, three DTCs (Diagnostic Trouble Codes) were found in memory: P0101 (MAF circuit out of range), P0172 (system rich bank 1) and P0175 (system rich bank

2). If we take these diagnostic trouble codes at face value, the problem could be something as simple as a misreporting Mass Air Flow (MAF) sensor. Before you shotgun a new MAF on the vehicle, though, let’s fire up the scan tool, create a custom PID list, take a quick test drive and use the captured data to analyze the problem.

The testing process
Data PIDs I’m recording on the test drive are engine RPM, throttle position (TPS), MAF grams per second, engine load, LTFT for both banks, STFT for both banks and oxygen sensor voltage for all four oxygen sensors. These few PIDs will give you all the needed information you need to analyze the problem and determine if the P0101 DTC is caused by a misreporting MAF sensor or not.

The vehicle was taken out and data captured at an idle, at a Wide Opent Throttle (WOT) acceleration up through one transmission shift, then at a short ¼ mile cruise. Because the test parameters are the same for all vehicles I diagnose, I have a preset route that allows for consistent results in a safe manner.

System analysis
Start the problem analysis with Figure 3, point No. 1, MAF data. The MAF data displays 6.67 G/S of airflow. A rule of thumb for MAF is, “with the engine at idle and all loads turned off, the air mass should equal 1 gram/second (G/S) per liter of engine displacement, at 500 rpm.”

The air mass shows 6.67 G/S with an idle speed of 544 rpm. This looks a little high to me, but file this piece of information in your mind to use at a later date. Point No. 2, LTFT, is showing a negative 19 percent pulse width correction. This data is screaming unmetered fuel in the air stream, in other words, there is too much fuel for the amount of air flowing into the engine.

Point No. 4, STFT is showing a negative 4 percent correction, so if we add the long term and short term fuel trim together to get total trim, the scan tool is showing a total negative 23 percent of injector pulse width correction. Point Nos. 4 and 5, oxygen sensor voltage, show the engine is running in closed loop and is in proper fuel control. This platform will stay in closed loop at all load conditions from idle through wide-open throttle.

Figure 4 is the data collected at wide-open throttle acceleration, or the Volumetric Efficiency (VE) portion of the test drive. Data point No. 1 is showing the engine at wide-open throttle. Point No. 2, engine load, is showing a maximum of 88 percent load at wide-open throttle at the 1-2 shift point. This is a little low for this platform. The known good value should be 99 percent engine load. This is another piece of information to store in your mind.

Point No. 3, LTFT and STFT are both showing zero percent correction. This is normal, since the fuel enrichment is taken care of with the fuel enrichment strategies. Point No. 4, front oxygen sensor voltage is showing both banks above 900 millivolts, which is telling us there is plenty of fuel available during the maximum power demand the engine is under.

Let’s move on to Figure 5, taken during the cruise portion of the test drive. Point Nos. 1 and 2, LTFT and STFT are showing a total trim of negative 6 percent. This looks normal, although I would be happier with a total trim of zero percent.Putting the pieces together One thing that stands out from the data from the first three data captures is the higher than expected airflow at idle. I would be happier with 5.5 G/S airflow. The other is the negative 23 perecent total fuel trim, making me wonder where the extra fuel is coming from. Is this excessive fuel problem caused by a MAF sensor that is incorrectly reporting

the airflow into the engine, a dribbling injector, or a fuel quality issue? The recorded scan data speaks for itself. After all, the reason for using scan data is to get the problem to come to you, the technician, sitting in the comfortable front seat.

At wide-open throttle, the engine load is too low. I would expect to see 99 percent engine load, and the scan tool only displays 88 percent. This engine load is a product of the airflow being reported to the ECM.

During cruise, the total trim is a little out of line at negative 6 percent. There is one more test I would like to do, using a volumetric efficiency calculator as a tool, to see if the MAF is doing its job. This will be one more tool to help out with the MAF diagnostic process.

When using a volumetric efficiency calculator, be aware that things like a restricted exhaust or air intake, or retarded camshaft, can and will affect the calculations. The volumetric efficiency calculator must be used along with the fuel trim data to ensure its accuracy. Take a look at Figure 6. The calculator shows the Chevy’s volumetric efficiency is only 65 percent, absolutely too low.By using the captured data, MAF data at idle, engine load PID and VE calculator and understanding the criteria for a P0101 DTC, I have determined the MAF is not reporting correctly. In fact, it is over reporting at idle, and under reporting at high engine loads. I can

now pull the trigger on a new MAF sensor with confidence that it is the cause of the code.  But what about the others that had set?

New MAF installed
With a new MAF sensor installed on the vehicle, the vehicle needs to be taken on its “after the repair” test drive. The same data should always be used to prove out the repair that was used to find the original problem. In this case, the recorded data has changed a little, one problem has been fixed, but there is still a problem that is shown in the fuel trim. Let’s examine the data and see what it tells us.

Looking at Figure 7, we can see the results of the whole test drive but concentrate on

point No. 1, the idle MAF data. This is showing 5.6 G/S, which is inline more with a 5.3 engine at idle. Point 2, though, the fuel trim still is not correct. I don’t think the DTC P0172 and P0175 have been fixed and if the vehicle is shipped like it is, it will be back with the Check Engine light glowing once again. At that point, the diagnostic work to fix the real problem will come out of my pocket. The total fuel trim is showing a negative 14 and negative 12 on bank 1 and 2, respectively.Moving to the WOT VE test seen in Figure 8 and point No. 1, the engine load is now at 99 percent, which is correct. Point No. 2, the total fuel trim is behaving correctly for the wide

open throttle, and point No. 3, the oxygen sensor voltage is showing there is a good supply of fuel to the injectors, and the proper amount of fuel is being injected into the air stream.

The cruise portion of the data found in Figure 9 tells a different story about the fuel/air ratio, with a total trim of negative 18 on bank 1 and a negative 15 on bank 2. One more test is needed,- we need to plug the scan data into the VE calculator and see how the results pan out.

The VE calculator in photo 10 shows the VE calculation of 86 percent, which is very good for this engine. This is evidence the new MAF sensor has fixed one problem, but there is still an over fueling issue lurking in the wings.

If this were a dribbling fuel injector, or fuel injectors, I would expect the fuel trim to be different between the two cylinder banks, and if it were a leaking fuel pressure regulator, I would expect to see the fuel trims very high at idle and lowering with the vehicle at cruise. This scan data is waving the flag of a fuel quality issue. I do not want to open the hood and connect a fuel pressure gauge; it’s cold outside and I want to find the problem in the comfort of the front seat.

Finding the second problem The vehicle we are working on has the VIN Z engine under the hood; this is a flex fuel engine. The scan tool is the place to go to see what the engine management system is doing with the flex fuel calculations. Take a look at Figure 11 and point No. 1. The scanner is displaying a fuel alcohol content of 47 percent alcohol. If this vehicle were in a place where E85 could be purchased, the displayed number might be correct but in my area, there is no E85 for sale, only E10. Before you leave the scan data, note the loop PID at point No. 2. This vehicle will stay in closed loop at all driving conditions, including wide-open throttle.The next step was to test the actual alcohol content of the fuel in the gas tank, and I found

there was only 6 percent alcohol. Now the scan tool has waved the red flag about the negative fuel trim problem. The fuel/alcohol content is not being reported properly. Because alcohol has less BTU (British thermal unit) per pound than gasoline, there is less energy per pound in the fuel being reported than the fuel that is in the tank and the ECM is commanding a longer fuel injection pulse than is required for E10.

This vehicle uses a virtual alcohol sensor where the oxygen sensors are used to make the fuel/alcohol calculations. Is there a possibility the misreporting MAF sensor has something to do with this miscalculation of the fuel/alcohol?

A fuel composition relearn can be commanded with a scan tool, but I opted to do it the manual way to verify the system would work properly on its own. I added four gallons of fuel to the fuel tank and drove the vehicle. This allowed the system to do its relearn on its own. The relearn was done, with no change in the fuel alcohol content.

Searching service information, I found TSB 06-06-04-010E, which outlines this problem very well. An ECM reflash is required to fix the software problem that is causing the miscalculation of the fuel’s alcohol content. The reflash was performed, another four gallons of fuel was added to the fuel tank and the vehicle driven. Figure 12, points Nos. 1 and 2, shows the ECM running the fuel composition calculation. The first time the calculation was run, the system reported 25 percent alcohol.

The next data capture, Figure 13, was taken after two refueling events. The total fuel trims run around negative 4 percent and at point No. 1, the fuel/alcohol content is showing 25 percent alcohol content. I find this consistent with this vehicle platform. The fuel/alcohol content is always reported higher than what is measured in the tank. This causes the total fuel trim to run at a slight negative rate.

Learn to use your scan tool. It makes problems like this come to you and makes them easy to solve.