Fuel trim analysis always is a very interesting subject, especially when it comes to using the fuel trims to analyze drivability problems. Have you ever stopped to think about the fuel trim and what it is telling the technician?
Several years ago, I remember being in a training class where the trainer stated that “the art of diagnostics is to get the problem to come to you, instead of you chasing after the problem.” This is an interesting concept, since it leads a person to a different thought pattern of diagnostics.
This shows fuel trim data taken at engine idle. Notice the bank 1 trims are -14.8 and the bank 2 trims are +16.4. Short trims look normal, which is an indication this valve timing problem has been going on long enough for the short term trim data to have been all learned in the long term trims.
Consider something as routine as finding a leak in a tire. I know the easiest way is to inflate the tire, then dunk it in a tub of water and see where the air is squirming its way to the atmosphere. Here again, the person looking for the leak is using a diagnostic process to quickly and efficiently make the problem come to them. Using a scan tool and fuel trim numbers for diagnostics is similar — with the tech using the trim numbers to tell him or her when and where to look for the problem.
One simple way to explain fuel trim is that the engine management system changes the fuel pulse width to maintain the air fuel ratio (AFR) at proper feed-gas levels for proper tail pipe emission and proper operation of the catalytic converter(s). Keep in mind, when using the trims, we need to be watching long term fuel trim (LTFT), short term fuel trim (STFT) and rear O² sensor trim numbers to come up with a total fuel trim. (A few OEMs actually list a total fuel trim and/or a rear fuel trim data PID.) It is easy to think of this AFR as a stoichiometric 14.7:1 ratio of air and fuel, but this is not always the case. Many times the engine will not run at the stoich AFR, because computer strategies will allow the engine to run either richer or leaner depending on the power needs of the engine. This is especially true with engines that are using wide band air fuel ratio sensors.
Here is fuel trim data taken at 2,500 rpm. The long term has changed to -10 on bank 1 and +20 on bank 2. Short trims still look normal. Again, this problem has been around long enough, so all the short-term corrections have been moved to long term.
From General to Specific
When starting with a drivability problem analysis, it is most important to be aware of the kind of fuel management system the vehicle is using. First determine whether it is a speed density or a mass airflow (MAF) system. Without knowing this information, your resulting mistakes in the interpretation of the scan data can lead you down some dark, slippery roads.
Once the type of air/fuel management system is known, a correct diagnostic strategy can be designed. Without the proper diagnostic plan, the technician is just wandering around in the dark with no direction.
Here is in-cylinder running compression on cylinder bank 2. This waveform shows the exhaust valve opening at 30ov before bottom dead center. This is normal for this engine.
Fuel trim data is dynamic, which means it is alive and active. If you are trying to use the trim information to root out a drivability problem, you must view the data under a variety of engine load conditions. This is often best accomplished by driving the suspect vehicle and recording the data on your scan tool for later review.
Fuel trim corrections will give the technician a direction for nearly any engine drivability problem. Before you throw your hands up and say that isn’t true, sometimes information telling you what the problem isn’t is very good information. How many times do you start a testing procedure and find information that is correct? This correct information helps the diagnostic tech to narrow the problem down to the pinpoint testing.
Let’s say you are working on a poor power complaint. You test the fuel pressure and find it to meet the manufacturer’s specification. This is good information, and with it you can rule out many things such as a fuel supply problem to the fuel rail.
Here is fuel trim data at idle with the new timing belt. Notice the long-term trims are back to normal at -4.6 on bank 1 and -2.3 on bank 2. Short term is also normal at -1.5 and 0.08.
So it goes with using fuel trim data. The trim data is telling you only the corrections the fuel management system is making to keep the air/fuel ratio where the programming says it should be.
In the case of a poor power complaint, the fuel trim data should be gathered at idle, 2,500 rpm and on a test drive. If you want to test the volumetric efficiency (VE) of the engine, you’ll need to do a full throttle run through at least one shift point, followed by a steady cruise. If all the information is found to be in order, you have gained a lot of good information in a relatively short amount of time. With this information, you can rule out air/fuel ratio calculation and fuel delivery problems.
At this point, it might be a good thing to test the VE of the engine. The fuel trim information has given you a direction for more testing. In the testing business, I find it rare to find the solution to a drivability problem with the first test procedure. Drivability analysis is a process that always starts out wide and narrows the problem down with each different testing procedure.
An Ailing Trooper
Because most automotive technicians are visual learners, I find it easier to explain a concept by using a case study. This makes it easier for me to explain how to apply the theory to a real-world problem. The vehicle for the case study is a 1997 Isuzu Trooper. The Trooper is powered with a 3.2 L V6 engine, and there are 168,000 miles recorded on the odometer. The engine is hooked to an automatic transmission, and the fuel management system is using a MAF sensor to measure the air the engine is inhaling.
Fuel trim at 2,500 rpm with new belt. Long-term banks 1 and 2 are 1.6, and the short term trims are near zero. This data is telling the technician the problem is fixed.
The customer complaint is rough idle and poor power. Scanning the engine computer for DTCs found a P0300 (Random Misfire) stored in memory. As diagnostic technicians grow in experience, we pick up little habits that are used in our craft. One of these habits I use is to walk to the back of a vehicle and put my hand over the exhaust pipe and feel the exhaust pulses.
When I put my hand over the end of the tailpipe on this car, I noticed the exhaust flow was not smooth. I got the feeling there was a slight misfire on some of the cylinders, which called for more investigation.
I started my investigating with a scan tool and fuel trim data. By looking at the fuel trim data, I was able to decide whether I needed to be looking for a fuel issue, an ignition issue or an airflow issue.
Looking at the scan data, I can see the long term trims on banks 1 and 2 are the reverse of one another; bank 1 is a -14 percent and bank 2 is a +16 percent. I’ve seen plenty of cars in the past where one bank was normal while the other was too rich or too lean, as I’m sure you have. But what would cause one side to be too rich (bank 1) and the other to be too lean (bank 2)? This data was telling me there is an airflow restriction on bank 1.
On this engine, the air mass entering the engine was being measured by a MAF sensor. The engine is designed to share this air equally between each bank. The engine also is designed to inject fuel equally into each cylinder bank. When the airflow to one cylinder bank is restricted, the division of air between banks is imbalanced. The restricted side is not able to flow the same amount as the unrestricted side. But the MAF sensor cannot distinguish this difference — and neither, at first, can the engine control module (ECM).
This shows fuel trim on test drive taken at wide-open throttle. Add the long- and short-term trims together to come up with the total trims, and they add up to a nice round zero.
The oxygen sensors will pick up this air/fuel imbalance and do everything possible with fuel injector pulse width to correct for the air imbalance. The cylinder bank with the negative fuel trims is always the cylinder bank with the airflow restriction. So what could cause this imbalance, this restriction?
Continuing to use fuel trim data, I could quickly determine that this restriction to airflow is not a plugged catalytic converter. Restrictions don’t always have to be physical ones, do they? The data is pointing directly to a cam timing problem. I can tell this by bringing the engine speed up to 2,500 rpm and watching the fuel trim change. In this case, they remained separated with bank 1 at -10 percent and bank 2 at +20 percent.
Verify and Repair
I am a firm believer in using more than one test to verify a problem. For cam timing problems, I like to use a pressure transducer and lab scope to back up my scan tool data. In this case, I removed a spark plug from each cylinder bank and screwed a pressure transducer into the spark plug hole, then started the engine and captured a waveform. This allowed me to see when the valves open and close while the engine is running.
This shows running compression with the new belt. This waveform was taken from cylinder bank 1. Both cylinder banks 1 and 2 are the same.
The pressure waveform gives more information more quickly than taking a vacuum reading, compression test and a leak down test on the engine. Another plus for the in-cylinder running compression test: I didn’t have to remove a lot of engine parts while looking for the problems. This makes the problem analysis procedure quicker, more accurate and at a lower cost to the customer. The best thing I like about this diagnostic process is that I get to play with the cool toys — and I love finding problems without getting my hands dirty!
The in-cylinder waveforms showed the left bank 1 camshaft lagged in timing by about 30o of crankshaft rotation. With these two different tests, there is enough evidence to pull the accessories and brackets that are mounted to the front of the engine, then remove the timing belt covers and inspect the valve timing components.
With the timing belt covers removed and the timing marks aligned on the crankshaft and both camshaft sprockets, it was easy to verify that the bank 1 camshaft sprocket had jumped timing by two teeth retarded. By using a logical diagnostic process, finding the problem was easy and accurate.
This shows in-cylinder running compression on cylinder bank 1. Notice the exhaust valve opens at the 180o mark of crankshaft rotation.
A new timing belt, rollers, cam and crank seals were installed and then the after-repair testing was performed. By using the scan tool and fuel trim data, it was easy to see the fuel trims were back to their normal position and the engine idles nice and smooth. The normal engine power returned, and the vehicle owners were able to get back to their vacation plans.
As engines become more complicated and technology keeps marching on, the need for better diagnostic processes follows suit. This Isuzu engine is a rather simple one, but what would you do if a problem like this came in on an engine that is very difficult to gain access to the timing belt? What would you do if there were no timing marks on the camshaft sprockets to line up?
The need for better and more accurate diagnostic processes is not in the future. It is here now.
Subscribe to Motor Age and receive articles like this every month…absolutely free. Click here
