Tech Tips

Use this comprehensive information library whenever you need helpful hints from car repair professionals. Whether you are a professional or a DIY mechanic, you’re sure to find the tips you need to insure a job well done. Click here to go back to the previous page any time during your search.

Subaru

1995 Subaru Legacy 2.2L Ignition Diagnosis & Timing Belt Service

edited from an article by Gary E. Goms, ImportCar, September 2001

It’s an unfortunate fact that many import vehicle owners tend to ignore scheduled maintenance intervals. Owners usually ignore maintenance, at least until the malfunction indicator light (MIL) illuminates and/or the vehicle begins to demonstrate one or more symptoms related to a driveability issue. Only then does the deferred maintenance come to light.

Worse still, the owner attempting to diagnose and repair the vehicle himself further complicates the maintenance issues. Too often, a set of new spark plugs, fuel, and air filter merely mask the root of the driveability complaint. The driveability technician must carefully go over the vehicle to make sure the owner has installed the new parts correctly.

To illustrate, let’s look at a 1995 Subaru Legacy equipped with a manual transmission and 2.2-liter engine. With 74,000 miles on the odometer, the owner complained of an intermittent sharp engine miss or bucking sensation at low speeds. A flashing MIL would occasionally accompany the misfire condition. In an attempt to remedy the problem, the owner installed new spark plugs, but the engine miss persisted.

MISFIRE MONITORS
Fortunately, the 1995 2.2-liter Legacy is one of the first OBD II-compliant vehicles produced by Subaru. Although the 1995 on-board diagnostics aren’t as comprehensive as later versions, the basic OBD II diagnostic process is the same.

On the 1995 Legacy, OBD II misfire monitor can detect a misfire much easier than a tech using a conventional ignition scope. If the misfire condition is severe enough to damage the catalytic converter, it becomes what is termed a "Type A," catalyst-damaging misfire. This will result in a flashing MIL and will occur in one OBD II "trip" cycle. A Type "B" misfire, on the other hand, is a misfire condition that will increase tailpipe exhaust emissions 1.5 times the Federal Test Procedure Standards. This becomes a two-trip monitor that must detect a misfire condition during two consecutive trips and the MIL will illuminate instead of flash.

When I retrieved the diagnostic trouble codes (DTCs), I discovered a P0304, which is a misfire on cylinder #4. No other codes were retrieved, so now it’s time to determine exactly what is causing the P0304 DTC to set in the diagnostic memory.

ROAD TESTING
Depending upon the abilities of the scan tool, freeze frame data may describe the conditions under which the misfire took place. According to one set of data, the enabling criteria for running the misfire monitor on 1995-’96 models is the engine running at idle or cruise speeds and the Vehicle Speed Sensor (VSS) input over 3 mph. When the engine is test-started, the PCM detects a misfire condition in the 200 to 1000 rpm range.

But let’s keep in mind that the P0304 DTC doesn’t tell us if the spark plug, wire, or ignition coil is causing the misfire. When we apply simple logic, we find that, if one of the ignition coils were at fault on the 2.2’s distributorless ignition system, a misfire DTC would be probably be set for the companion cylinder, since both fire on the same coil cycle.

A misfire code can also be set by a mechanical failure such as a broken valve spring, low cylinder compression, or an inactive port fuel injector. But, if any of these mechanical failures were present, the engine should demonstrate a repeatable rough idle condition. In this case, our focus Subaru idled smoothly enough that I began to suspect that the misfire was caused by the ignition system itself. With those thoughts in mind, a road test would be the quickest way to get a hands-on feel of when the misfire would take place.

When road testing, the most important objective is to expose the engine to a variety of operating conditions. The Subaru, for example, would idle smoothly and accelerate at full throttle without a miss or hesitation. But a miss could be felt when the engine was lightly lugged in fourth gear at about 30 mph. The ignition misses occurred as one or two sharp jerking sensations just as the throttle was being applied. As the throttle opened up, the miss disappeared.

ENGINE LOAD
The reasons for the misfire occurring at light throttle applications and light loads aren’t immediately obvious. First of all, most of us were taught that maximum spark voltage occurs at full throttle, when cylinder pressures are at their maximum. But this theory doesn’t explain why we would have a misfire at a light throttle setting.

In the case of our Subaru, when the engine is at full throttle, the air/fuel mixture becomes richer and therefore offers less resistance to the spark jumping the electrode gap. The demand for high spark voltage is also reduced as the spark timing advances to compensate for increased engine speed.

During tip-in, however, the opposite normally takes place. The spark is retarded, the air/fuel mixture is lean, and cylinder pressures are beginning to build. In some cases, the spark demand actually becomes higher than at full throttle. This would explain why I was getting a P0304 DTC that matched the enabling criteria for our Subaru.

DIAGNOSIS OF SPARK LEAKAGE
Neglect and sloppy service procedures can cause a variety of problems with modern distributorless ignition systems. First of all, if the spark plug electrode has eroded to the point that firing kilovolts (KV) is excessive, the spark will seek a path to ground that has the least resistance. This path can be through a coil tower, coil case, ignition cable, or down the side of the spark plug insulator.

If the owner has attempted a repair by replacing the spark plugs and/or wires, then he’s eliminated much of the evidence we need to diagnose the misfire. If we had an open spark plug wire, then it stands to reason that we might have a perforated ignition coil. But, if the spark plug wire has been replaced, then we still might have a perforation in the coil, which will set another misfire code, especially under high-humidity and voltage conditions.

To date, I’ve found that most misfire DTCs are caused by spark leakage. If the spark plugs have been replaced, it’s a safe bet that the spark is arcing down the side of the spark plug insulator or through a perforated wire or spark plug boot. If the spark has arced down the side of the spark plug, I recommend replacing both the plug and the wire, since a companion carbon track forms inside the boot itself. As for spark leaks through wires, they’re very difficult to detect, so I would recommend replacing all the wires to save time and money.

For diagnostic purposes, however, it’s important to detect spark leakage before removing the wires or spark plug boots. As for wetting agents, soapy water simply won’t soak through the wire well enough to cause a spark leak. Instead, I recommend a commercial ignition wetting agent or a homemade mixture concocted from a 50-50 mix of windshield washer fluid and household ammonia.

With the engine running, our Subaru Legacy proved to be an easy diagnosis since number-four cylinder immediately shorted out when sprayed with a wetting agent. The misfire was immediately confirmed by the MIL flashing on the instrument panel.

What caused the misfire? Actually, the spark leak wasn’t visible at first, but upon closer inspection, I found a crack forming where the spark plug boot seals against the camshaft cover (Photo #1). Upon removal, it showed that the other wires had also started to crack when the owner initially replaced the spark plugs.

After replacing the spark plug wires, I cleaned the throttle plate assembly with an intake cleaning solvent, and inspected the air and fuel filters, which appeared to have recently been replaced. Last, I asked the owner if the timing belts had been replaced at the recommended 60,000-mile interval. They had not been replaced, so I proceeded with replacing the timing and accessory drive belts.

TIMING BELT REPLACEMENT TIPS
Like earlier 1.8-liter engines, the 2.2-liter engine is a non-interference engine in which the valves will not be damaged if the timing belt breaks. The 2.5-liter Legacy engine, on the other hand, is an interference engine that will bend valves if the camshafts slip out of time.

Before proceeding with a timing belt replacement, always consult an up-to-date service manual for detailed information. To ensure accurate valve timing, the 2.2 engine’s timing belt incorporates timing marks on the belt that align with timing marks on the camshafts and crankshaft. The timing belt is also directional in that, on the passenger’s side, the belt has 44 teeth between the cam and crankshaft timing marks, while on the driver’s side it has 40.5 teeth between the two timing marks.

After the timing covers are removed, the best procedure is to turn the engine over until the timing marks on the camshaft sprockets and crankshaft reluctor (not the sprocket) align with the timing marks on the rear camshaft sprocket covers and engine block, respectively.

TDC MARKS
At this point, it’s easy to mistake a mark on the crankshaft as the TDC timing mark. The two crankshaft marks are 90 degrees apart, so if the wrong mark is used, the crankshaft will be 90 degrees out of time (Photo #2) or the camshafts will be 45 degrees out of time (Photo #3).

In reality, the mark is on the reluctor, not the sprocket. In order to establish a point of reference and, if the timing marks are worn off the belt, it’s best to highlight the alignment of the sprocket, belt, and rear cover timing marks with white typing correction fluid. If you’re re-using the old belt, mark the direction of rotation for good measure.

TENSIONER ADJUSTMENT
The timing belt tension adjuster is a unique hydraulic design that bears against a pivot on the tension adjuster pulley mount (Photo #4). The tension adjuster incorporates two slotted mounting holes. The tension adjuster pulley itself pivots on its mounting bolt. A slight trace of oil around the tension adjuster rod isn’t reason to condemn the tension adjuster. With tension released, the tension adjuster rod shouldn’t extend more than .606-.646" from the cylinder.

The tension adjuster should also require about 33 to 110 pounds of pressure to gradually (taking up to 3 minutes) compress the adjuster rod into the cylinder (Photo #5). The rod shouldn’t compress in less than 8.5 seconds, nor should it compress with less than 33 pounds of force applied.

A small Allen wrench can be used for a stopper pin to secure the piston in the adjuster cylinder (Photo #6). The adjuster can then be installed on the engine block (Photo #7 on page 44) by hand-tightening the mounting bolts to the block.

ACCESSORY INSPECTION
As always, check the tension pulley and water pump for smooth rotation before installing the timing belt. In addition, check the water pump for leaks. Many shops recommend replacing the water pump with each belt change. Others go so far as to remove the oil pump for inspection before the belts are replaced. If the crankshaft seal is leaking, it should be replaced. Some aftermarket data mentions that replacement seals may be missing the garter spring, so check for the presence of the spring and fill the seal cavity with grease to prevent it from popping off as the seal is driven into the engine block.

Last, the harmonic balancer should fit snugly on the crankshaft. Always check the woodruff key and keyway for wear and always tighten the balancer retainer bolt to the specified 69-76 ft.-lbs., using Subaru Tool 49997000 or an alternate method to hold the crankshaft in position. In this case, I placed the manual transmission in fourth gear and locked the wheels with the parking brake to hold the crankshaft while I torqued the harmonic balancer bolt to spec.

Subaru recommends removing the camshaft sprocket pulleys before installing the timing belt. If you remove the sprockets, keep in mind that the left- and right-hand sprockets are different. Removing the sprockets requires the use of an ST 499207100 sprocket wrench or equivalent to hold the camshaft in position while the retaining bolt is removed. The belt can be installed without removing the sprockets, but it can be difficult to do without damaging the belt by twisting or prying.

After the belt is installed and the cam sprockets torqued to 54-61 foot pounds, loosen the tension adjuster mounting bolts and remove slack in the belt by sliding the adjuster toward the passenger side of the engine and torquing the mounting bolts to specification. With the timing belt in position, release the tension adjuster by removing the stopper pin. With the slack removed from the timing belt, the timing marks printed on the new belt should be in the exact position described in the repair manual.

Go back