The most famous last words in automotive diagnostics are, “We’ve always done it this way.” In many cases, “always done it this way” consists of diagnosing a fuel pump by squirting some type of liquid or gaseous hydrocarbon into the air intake. Case in point, a 2011 Nissan Rogue had suddenly stalled on a busy summertime highway. See Photo 1.
According to service information, the Rogue’s fuel pump is a single-line system with the fuel pressure regulator mounted inside the fuel pump module.
The technician diagnosed the 2011 Rogue as a bad fuel pump by squirting propane into the air intake. With propane, the engine runs, without propane, it stalls. But, after the fuel pump was replaced, the Rogue exhibited the same problem, which was a very slow, unsteady idle followed by an engine stall. What we have here isn’t a bad fuel pump, but rather a no-fuel condition, which is an entirely different situation.
To begin, all high-pressure fuel injection systems incorporate safety protocols that prevent the fuel pump from emptying the fuel tank after a serious accident.
Most safety protocols follow this general strategy: Immediately after the ignition is turned on, the engine control module (ECM) activates the fuel pump relay for several seconds to prime the fuel injectors. In late-model vehicles, this same function is performed by the identification fob as the driver approaches the vehicle. During cranking, the ECM will activate the fuel pump relay only after it receives a signal from the crankshaft position (CKP) sensor. After the engine starts and reaches idle speed, the ECM keeps the fuel pump relay activated via the CKP signal.
Since some imports are re-badged domestic models, remember that the fuel pump circuit might include an inertia switch that disables the fuel pump circuit during a sudden impact. Other fuel pump safety devices include oil pressure switches that disable the fuel pump circuit when oil pressure drops to zero during a rollover. Some older imports with mechanical vane-type airflow sensors include a safety switch that disables the fuel pump if the engine stalls during an accident.
The words “fuel system,” denote a group of vital parts that, in conjunction with each other, activate the fuel system. For example, the CKP sensor signal is vital because it activates the fuel pump and injectors via the ECM. The camshaft position (CMP) sensor, which times the fuel injectors via the ECM, can be considered a vital sensor because some engines won’t run in a default mode with the CMP disconnected. The mass airflow (MAF) sensor reports intake air as frequency or voltage to the ECM. The MAF can also be considered a vital sensor because some engines won’t run in a default mode with the MAF disconnected. See Photo 2.
FUEL INJECTION FACTORS
Let’s not forget that fuel injectors can cause a no-fuel condition. For example, a closed throttle position on a cable-driven throttle and position (TP) sensor should report less than 1.0 volt on a 5.0 reference voltage. Wide-open throttle (WOT) should report a maximum of about 4.5 volts. Above 4.5 volts, the ECM normally enters a “clear-flood” mode that inhibits fuel injector activity. If a TP voltage signal shorts to 5.0 reference volts, the ECM might automatically enter a clear-flood mode, causing a no-fuel condition.
The engine coolant temperature (ECT) sensor can similarly inhibit fuel injector activity if the signal is shorted to reference voltage. A shorted ECT circuit can indicate about 300° F coolant temperature, which can reduce injector pulse width to nearly zero. Remember, too, that fuel injectors and some fuel pumps can also become disabled when the vehicle security system can’t identify the ignition key or driver identification fob.
DATA STREAM SOLUTIONS
Going back to our Nissan Rogue, we couldn’t keep the engine running long enough to set codes or get a good read on the serial data stream. But our data stream will indicate how many grams per second of air is flowing into the engine. As a rule of thumb, the airflow in grams per second through an engine at hot idle should equal the cylinder displacement in liters. Hence, the airflow through the Rogue’s 2.5 L engine should roughly equal 2.5 grams per second (gps) at idle speed.
In this case, the airflow at hot idle was showing a much lower 1.8 gps which reduces fuel injector pulse width to the point where the engine will barely idle. We tested the MAF by adding enough propane to increase engine speed. Oddly enough, the grams per second didn’t increase with engine speed, which indicated a badly contaminated MAF sensor or loose ducting between the MAF sensor and throttle body.
With those issues verified, case histories revealed that 2011 Nissan Rogues have a history of unusually catastrophic MAF sensor failures. Replacing the MAF sensor solved the no-fuel problem.
Last summer, I was called to diagnose a no-fuel problem on a 2003 Mitsubishi Eclipse. The Mitsubishi’s DIY owner had previously diagnosed a cranking, no-start condition as a bad fuel pump by squirting starting fluid into the air intake. See Photo 3.
Needless to say, replacing the fuel pump didn’t cure the problem. After the pump assembly was reinstalled, I used my scan tool to command the Mitsubishi’s fuel pump “on.” Since the fuel pump is accessed from under the rear seat, electrical testing was easy. Going through the bi-directional fuel pump test, I couldn’t hear fuel pump noise nor could I measure any voltage at the fuel pump connector.
Using the fuse and relay locator in service information, I verified circuit continuity at the fuel pump fuse. According to the wiring diagram, this Mitsubishi model uses two relays to activate the fuel pump. Since the fuel pump relay is powered by a second relay, two relays must engage before the fuel pump activates. I could feel the first relay click, as I activated the fuel pump circuit with my scan tool, but this wasn’t the case with the second relay.
A quick look at the interior fuse box schematic revealed that the DIY mechanic had installed the fuel pump relay in the wrong bay. Reinstalling the fuel pump relay restored fuel pump operation. Now we have fuel pump operation, but still have a no-start, no-fuel condition.
GIVING IT THE GAS
OK, so now the fuel pump motor is drawing about 3.5 amps, which indicates that it’s producing fuel pressure. The engine still won’t start on its own, but it responds to starting fluid sprayed into the intake. A lab scope test verified that the fuel injectors were opening for about 5 milliseconds, which is more than adequate for room-temperature starting.
So, let’s go back to measuring fuel pressure. Unfortunately, the Mitsubishi’s 2.4 L engine doesn’t include a Schrader test port on its fuel injector rail. Upon inspection, I also discovered that it would be very difficult to install a fuel pressure gauge in series with the fuel inlet line.
But, since the Mitsubishi has a conventional two-line fuel system, the fuel pressure regulator was connected to the fuel tank return line by a rubber hose held in place with a spring-loaded hose clamp. By removing the return hose and replacing it with a longer hose so that the return fuel could be easily directed into a secure container, we cranked the engine and discovered that we had an excellent volume of liquid flowing through the regulator by-pass. See Photo 4.
Assuming that the fuel pressure regulator was correctly calibrated, we could also assume that we had enough fuel pressure at the fuel rail to start the engine. But, let’s take a closer look. The “liquid” exiting the fuel pressure regulator had a suspiciously milky look to it, like water mixed with gasoline. This would explain why the fuel pump was drawing about 3.5 amps and why we now had fuel pressure, but had no combustion inside the engine. The fuel tank had, at some point, been contaminated with water. Cleaning the fuel tank and adding fresh fuel solved another “no-fuel” problem that was caused by something other than the fuel pump.