Not all bugs are the same

You know how it goes. One car comes in with a problem, and then several follow. We had a 2006 Nissan 350Z come in the lab with a simple, so we thought, cylinder 5 misfire Diagnostic Trouble Code (P0305 DTC). The car showed all the normal symptoms of a bad spark plug. The 3.5L V6 ran rough, idled bad and smelled worse. But it became a head scratcher before we found the real root cause.

Just after we won this battle, a 2007 Nissan Maxima with 3.5L V6 came in with basically the same codes and symptoms. The students said, “We got this,” but it turned out to be a completely different problem. Two Nissans, same DTCs, same engines, but the root causes turned out to be totally opposite of each other.

The 350Z had been running rough and repeatedly turned on the Malfunction Indicator Lamp (MIL) for the same DTC. The normal parts had been thrown at it. New spark plugs were installed and the No. 5 cylinder spark plug was changed three times. Then a new coil was installed on No. 5 cylinder after the coils from other cylinders had been switched around with it, but no change was noticed. When I walked up to the car, the idle was noticeably rough and a rich running engine smell (that noticeable “rotten egg” odor) really caught my attention. These are all signs of a run-of-the-mill misfire, right? The only thing that seemed abnormal was the MIL not blinking. A blinking MIL usually means excessive emissions and possible catalytic converter damage, but the owner only complained of poor gas mileage and drivability.

The Maxima came in a few weeks later with the same symptoms, except it had a cylinder 4 misfire (P0304) DTC and the occasional random misfire (P0300) DTC. It ran rough, idled poor and smelled rich also. The students checked for a misfire by doing a cylinder balance test, but reported they found nothing conclusive. They pulled the plugs, found them worn and very dark in color, and replaced them with new ones. No change was found when the engine was started. They then decided it must be the same problem as the 350Z. Does that type of statement sound familiar? I suggested they run some tests before replacing any more parts.

350Z’s Diagnostic Process
The conventional thinking process always centers on the DTC. The first thing to remember is an Engine Control Module’s (ECM) logic only points to problems it cannot correct for. If an ECM could figure out all problems then life would be simple, but they can’t. I like to do the simple tests first then proceed to the harder tests until the problem is found and the root cause is corrected.

This amp meter reading shows the bad AF sensor heater’s reading. Notice the .3 amp as compared to a normal of a little over an amp after the engine had been running for a while.

We had a scope handy, so we performed a cranking compression test using an amp clamp around the negative battery cable. The idea here is that as each cylinder on the compression stroke approaches top dead center the amperage draw increases to a peak. All the peaks were equal, meaning the compression for each cylinder was about the same. So we started testing for vacuum leaks. Using both propane and sonic ears, we painfully tested for leaks, but found none. Next, we moved the amp clamp to the wire feeding power to the cylinder 5 coil. This way we can see if the circuit is flowing proper current and look for connection issues.

The oscilloscope image showed a consistent amp ramp that was the same as the others. We tested the spark output voltage of that coil and two others, with all three reading about 8 Kv (kilovolts). Finding nothing, we looked at the service information and technical service bulletins for any hints. The usual procedures ranging from spark plugs to vacuum leaks were suggested with no one statement caching my eye. The next to last possible problem in the service information listed the bank 1 Air Fuel Sensor (AFS) as a possible cause. I was unsure how that could cause a miss, but wondered why they only listed bank one and not both, especially since they listed all misfire codes from random (P0300) to each specific cylinder (P0301-6) under the same procedures. We dismissed this idea for now and continued with the normal issues related to misfires.

We originally tested the total amperage draw for both heater circuits on the 350Z at the fuse panel behind the left kick panel. Our total amp draw for both AF’s heaters read only about 1.4 amps. With the new sensor in place our meter is showing 2.2 amps a few minutes after the engine was started.

Now it was time to look over all the tests we had performed and the scan tool data. I went back to the Freeze Frame data looking for when the DTCs were setting and under what conditions. I noticed that about five minutes after starting the engine cold and before the engine was up to normal operating temperature, the code or codes would set. The next step was to test the fuel injectors to see if one was sticking or leaking causing a rich condition that only happened on engine warm-up.

I noticed bank 2’s old spark plugs were almost black in color while bank 1’s plugs were of normal color. I went back to the scan tool looking at the AF sensors to see if bank 2 was richer than bank 1. That is when we noticed the lack of response from the bank 2 sensor. We forced the system rich and then lean looking for a response. Its readings changed, but not as much as bank 1. The service information at step 14 and 15 of the OEM flowchart has you disconnect the AF sensor 1 and ECM connectors to check for continuity and then check for grounded circuits.

At step 16, it mentions testing the AF’s heater but not how. The students wanted to replace the AF sensor and I was about to agree when I decided to tell them to test the heater circuit first. They asked why the AF heater monitor passed if there was something wrong. I responded with, “The AF monitored passed, but you think it’s bad.”

They decided to test the heater circuit. We looked at the wiring diagram and decided to test total amp draw at the fuse. They jumped the heater fuse with a test lead and used their meter to test total amp draw. It seemed a little low so I had them use their jumper kits to test each AF sensor. Both AF to harness connectors were located over each valve cover making it somewhat easy to do. Bank 1’s amperage was about 1 amp while bank 2 was down around 0.3 amp. I prefer amperage rather than an ohm meter. Testing a circuit while it is flowing amps usually yields better results when looking for connection issues, plus the heater element changes resistance with temperature and may reveal an open circuit when hot.

The duty cycle of the heater showed the same on the scan tool, so we checked for voltage drop at each AF sensor connector. They both read the same voltage so we decided the AF sensor was, indeed, bad. We ordered a new one and installed it the next day. We ran several total amperage draw test with each test reading around 2.1 amps. This solved the problem, but remember the service info only mentioned bank 1 as a possible cause, not bank 2. Don’t follow service information steps blindly, always think and question why you’re performing the tests outlined in the flowchart. The AF sensors would warm up enough to pass the monitors, but triggered the engine to run rich enough to run and idle rough under certain conditions. When driven, the heat from the engine would heat up the sensor enough to allow it to work normally.

At a 70°F engine start up, both AF sensors on the 350Z pulled almost 3 amps, but by the time I took this picture the amperage flow had dropped down to 2.64. 

On To The Maxima
A few weeks after the 350Z left the lab, another Nissan appeared with the same symptoms. It had a cyllnder 4 misfire (P0304 DTC) instead of the P0305 the 350Z suffered from, and added a random misfire DTC (P0300). The 2007 Maxima is equipped with the 3.5L V6 that has basically the same engine and ECM programming as the 350Z. Naturally everyone, including me, is thinking we know what the problem is.

The students grabbed the scan tool and hooked it up to verify the DTC. The P0300 random misfire code was an added issue, and I still like to stress to students the need to test before they replace any parts. We looked at the AF sensor data first and were surprised to find both values similar in both generic and enhanced modes. The voltage values were interpreted differently, 14.4V in enhanced mode and .38V in generic, but were equal and varying although they were mostly staying on the rich side. This changes the game so now we are back to starting at the beginning.

Would this reading have caught your eye? The Maxima’s MAF varied from 4 to 7.1 g/s. The engine was idling as low as 500 RPMs while reading 4 g/s. But the MAF reading went to 7.1 g/s at only 750 RPMs. With the car in park we noticed the MAF reading increased faster than the RPMs increased as we revved the engine.

The diagnostic flowchart is the same as the 350Z’s which groups the random code with the cylinder specific codes. The way the engine was running and the heavy rich smell made me think more on the fuel management side, but could it still be an AF sensor heater? What if they were not heating up the AF sensors properly? This would make the ECM add fuel.

We took the car for a short test drive to fully warm up the engine and the exhaust system. The engine ran better while driving it but we had a little hesitation on acceleration and some chugging as we backed off the throttle. As soon as we let the car idle, the roughness returned along with the rotten egg smell. If the car idled for around five to 10 minutes, the engine would start missing and you could see a little black smoke. The P0300 or a cylinder specific code would return if we let it run long enough or upon restart. We actually had a P0305 DTC set once, which is spooky.  

We checked the normal fuel-related issues like fuel pressure, cylinder balance and ran a compression test. The spark plugs looked fine, except they were all blackish in color. We hooked up a scope with an amp clamp to the injectors to look at injector ON times and for a circuit issue that may cause the injectors to spray too much, but everything looked normal.  The fuel trims were a -25 percent short term and -10 percent long term, showing the engine was running rich and knew it.  We looked over the service information and at step 16 it displayed a chart with what Nissan considers as normal Mass Air Flow (MAF) sensor readings.

What are the odds? Just goes to show it pays to do a visual inspection!

The problem with their suggested readings is the range is too large. They suggest a range of 2.0 to 6.0 grams per second (gm/s) at idle and 7.0 to 20.0 grams per second at 2,500 rpm as a basis for diagnosing a contaminated MAF. That is a big range when you are looking for an issue like this. The old rule I have used is 1 gm/s for each engine liter of displacement at 600 RPMs of no-load idle. While it’s not foolproof, it is far more accurate. The engine had a rough, fluctuating idling from 500 to 700 rpm. Doing some quick math I divided 700 by 600 then times 3.5 gm/s equals about 4.1, but the readings were always a little high.

Now there are several ways to use a scope to test a MAF. You can use an amp clamp on the battery wire supplying power to the senor. This method allows you to see the amp draw increase as you accelerate. If the MAF outputs a voltage this test works the same but you need a good scope program that can convert frequency. I also hook a channel of the scope to the Throttle Position Sensor (TPS) for a response comparison. As you blip the throttle the volts or amps increase similarly to the TPS voltage. We just decided to do a visual first so we removed the two screws holding it and lifted it out of the tube. This time the visual inspection really paid off! A bug was touching the element in the MAF causing it to read off. Once we blew off the bug and reinstalled the MAF the car ran great. 

Subscribe to Motor Age and receive articles like this every month…absolutely free. Click here