Tech Feature: Lack of Power Leads to Exhaust Restriction Analysis – UnderhoodService

Tech Feature: Lack of Power Leads to Exhaust Restriction Analysis

In the early days of carburetors and mechanical fuel pumps, most lack-of-power complaints stemmed from vapor locking on the fuel pump or the fuel filter clogging. But that's becoming the exception rather than the rule.

photo 1: i made this exhaust tap by welding a short length of brake line into a section of exhaust pipe. Given today’s high-mileage vehicles and the complexity of their fuel delivery systems, it’s no wonder that we don’t encounter more lack-of-power complaints than we do.

Having lived and worked for nearly 32 years at an 8,000-foot altitude in the central Colorado mountains, I’ve witnessed my quota of lack-of-power driving complaints.

In the early days of carburetors and mechanical fuel pumps, most lack-of-power complaints stemmed from vapor locking on the fuel pump or the fuel filter clogging. Even today, I’ll run across a case in which an in-tank fuel pump vapor locks or begins to lose pumping efficiency as the fuel tank temperature increases. But that’s becoming the exception rather than the rule.

In other, more common cases, the fault lies with exhaust system restriction. Causes for exhaust restriction range from a crushed tail pipe to a mild clogging of the honey-comb matrix in the catalytic converter. Given the cost of replacing a converter assembly, proving that an exhaust restriction is responsible for a lack-of-power complaint can be a tough call, especially if it’s a marginal or intermittent condition.

Years ago, the problem became so bad with the old two-way converters that I made an exhaust tap by welding a short length of brake line into a section of exhaust pipe and attaching it to a vacuum pressure gauge with inexpensive vacuum hose.

See Photo 1.

Drilling a 3/16” hole into the exhaust pipe allows the tap to be inserted into the exhaust stream and held on by a hose clamp. The hole can be sealed with a 3/16” pop rivet. I don’t, however, recommend drilling or cutting any converter assembly covered by the auto manufacturer’s eight-year, 80,000-mile 2: a casual drive in this 2001 caravan would have given it a pass.

Mobile Mechanic
Most recently, I’ve been working part-time providing mobile diagnostic and training services for local shops. I was called on a case involving a 2001 Dodge Caravan with no diagnostic trouble codes (DTCs) and a low-power complaint.

See Photo 2.

A third-party shop had changed the fuel filter and spark plugs in an attempt to address the lack-of-power complaint.
As a preliminary step, I took the Caravan for a climb up a local test hill with a scan tool attached in the movie mode.

Because the van had more than 160,000 miles on the odometer and didn’t feel especially low on power, I was suspecting a fuel delivery problem.

The most notable symptom was that the engine had a slight difficulty reaching the tachometer red-line at wide-open throttle (WOT).

Other than that particular symptom, I would have given the vehicle a “pass” on the lack-of-power complaint. To my surprise, the long and short fuel trims were under 10% and all other data stream information was normal except that the down-stream oxygen sensor didn’t achieve the normal flat-line status after warm-up. photo 3: because the downstream oxygen sensor tends to track the upstream oxygen sensor voltage, this graphing data appears to indicate a catalytic converter failure.

I’ll elaborate more about that later.

Pinpointing the Complaint
Verifying the exact symptoms is extremely important in any lack-of-power diagnostic procedure. I’ve had customers complain about lack of power when, in fact, their automatic transmissions had defaulted into second gear because of a mechanical or electronic failure.

In other cases, I’ve had clogged fuel filters show up as very high fuel positive trim numbers. In even worse cases, I’ve had mass air flow sensors coated with dirt, which again cause positive fuel trim numbers and lack of power. I even had one Volkswagen with a lack-of-power complaint come in with a multiplicity of fuel trim and MAF codes.

Oddly enough, it passed through three specialist shops, none of which bothered to check for an air filter inlet screen clogged with debris. Each symptom generally produces its own diagnostic data, both in trouble codes and in data stream information. But regardless, verifying the actual type of performance complaint is a vital first step in the diagnostic process.

The Second Road Test
After discussing the first road test with the shop owner, we decided to make another test run because he claimed that the vehicle also had a problem with the transmission hesitating on the up-shift. photo 4: the exhaust pressure would exceed 6 psi during a momentary 1,500 rpm torque converter stall test.

With the shop owner driving and me operating the shop’s aftermarket scan tool, the second test was a bit more revealing. One disappointing part of the experience was the scan tool’s limited ability to record movies. I’ve used this type of scan tool to do this on previous jobs with no problems. But, for this reason, I read each data line while the shop owner drove the vehicle.

On this trip, the shop owner could make the engine “hang” in gear with the throttle released. Although it would eventually shift into the next gear, it would take a while. During the “hang” time on the gear changes, I noticed that the scan tool data stream was momentarily showing nearly zero vacuum with the throttle closed. Excess exhaust backpressure seemed to be the only explanation.

This particular vehicle uses a speed density system that includes the throttle position (TP) sensor, manifold absolute pressure (MAP) sensor, and engine speed inputs to determine the weight of air flowing into the engine. Speed density systems are a little more difficult to diagnose than an engine equipped with a mass air flow system simply because the air flow into the engine is a calculated, rather than actual, value.

Because the transmission is more than likely using throttle position and intake manifold vacuum to calculate engine load, a low MAP reading would affect shifting points and firmness of clutch engagements. In this case, the TP sensor was showing closed throttle, but the PCM was apparently giving more authority to the zero-vacuum MAP input.

Back at the Shop
I had two significant pieces of data indicating a catalytic converter failure. The first was the jagged voltage line from the downstream oxygen sensor and the second was the low intake vacuum reading at closed throttle. The odd part was that, although the jagged downstream oxygen sensor line was indicative of a catalyst failure, no catalyst efficiency codes were stored.

See Photo 3.

The enable criteria for a P0420 catalyst efficiency code on this 2001 Caravan is “engine running in closed loop for three minutes, ECT sensor input over 147° F, 1,200-1,700 rpm with throttle open, VSS over 10 mph, MAP at 1.5-2.0 volts, with the PCM detecting that the switch rate of the rear H02S-12 reaching 70% of the front HO2S-11 switch rate.”

With this in mind, the catalyst monitor probably ran just as the vehicle was pulling onto the highway. Because the catalyst monitor is non-continuous, it runs only once and during that single test, the downstream oxygen sensor was likely hovering below the 70% switch rate required to store a DTC. Whatever the scenario, no P0420 DTCs were being stored in the pending or current diagnostic memories.

Indicative Vs. Definitive Data
At this point, all I had was “gray-area” indicative data. The fact that the data stream was showing low manifold vacuum at closed throttle during an up shift could be attributed to a slow update on the scan tool.

The classic method of testing for excessive backpressure is to attach a vacuum gauge to the intake and compare idle readings against a steady-throttle, 2,500 rpm vacuum reading. In theory, the two readings should be very close if no backpressure exists in the exhaust. In reality, I believe this test is no longer valid because it simply isn’t sensitive enough to detect borderline cases of excessive exhaust backpressure. Photo 5: This converter is obviously being clogged with contaminates from the engine.

A more definitive test is the direct backpressure test. By replacing the upstream oxygen sensor with a backpressure tap, we can get a direct measurement of actual exhaust backpressure.

The initial pressure tests were anything but definitive. Although the gauge showed 1 psi at idle, that’s well within the margin of error for the gauge. When the engine was stalled at about 1,500 rpm against the torque converter, the pressure rose to between 6-8 psi.

See Photo 4.

While this reading is high, there’s no standardized test limit for exhaust backpressure. According to my experiences, many engines routinely generate 6 psi under load with no pronounced lack-of-power complaint.

But the most definitive number was 15+ psi at snap throttle. In general, the highest backpressure at snap throttle should be less than 5 psi. Fifteen psi under any operating conditions is excessive. Once removed, the converter showed evidence of partial clogging.

The problem we have in Photo 5 is that a contaminated exhaust stream is clogging the catalyst matrix.

A second phone call to the owner indicated that the engine regularly consumed a small amount of coolant. In this case, the client shop’s customer elected to replace the catalyst with a disclaimer on the catalyst warranty. This repair obviously wasn’t a textbook solution, but more of a practical solution in light of the mileage and market value of the vehicle itself.

Failure Patterns
But there’s more to exhaust restriction testing than taking a backpressure test. Always keep in mind that the catalyst substrate can disintegrate and either clog the muffler or allow a large piece of catalyst to intermittently clog the outlet of the converter. If the converter rattles when tapped with a rubber hammer, the substrate has come loose inside the converter. Similarly, tapping the muffler can reveal mufflers that are contaminated with loose substrate materials or have loose baffles.

Scan Tool Testing
Over the years, I’ve read and heard a lot about diagnosing an exhaust restriction with scan tool data. Although the scan tool is the first piece of equipment I’ll use to diagnose any type of driveability complaint, there are limitations. First, always keep in mind that scan tools don’t work in real time. A slow update rate on the scan tool might alter relevant data and result in a false diagnosis.

Next, the real problem with exhaust restriction is the degree of restriction itself. Every vehicle has some degree of exhaust restriction built into its original design and that degree of restriction tends to increase with mileage. The design of the engine also dictates how much exhaust restriction an engine will tolerate before it begins to lose power. Valve overlap, valve timing, variable camshaft timing, tuned intake systems and turbo charging all change an engine’s tolerance for exhaust restriction.

The primary issue behind restricted exhaust is that it changes the volumetric efficiency or fill rate for the cylinders. The more exhaust restriction, the less the cylinders are filled with air and the less power produced. In extreme instances, an exhaust can become so restricted that the engine no longer pumps air, which results in a cranking, no-start complaint.

So if the question is volumetric efficiency, could we use mass air flow data to diagnose a restricted exhaust? If we knew that the MAF sensor itself was functioning correctly and we knew the actual values for correct volumetric efficiency, we could do that. Myself, I think there are too many variables, but I’m sure there are many techs that can and do use MAF data to diagnose exhaust restriction.

More Case Study Notes
Let’s also keep in mind that scan tool data from different vehicle platforms require different diagnostic strategies. A few years ago, I had a late-’90s Chevrolet pickup come in with a complaint alleging that the transmission wouldn’t shift out of third gear. A quick road test revealed that the customer was correct. In fact, the truck wouldn’t accelerate past 40 mph.

Scan data included misfires for all driver’s side cylinders. The truck also produced 6 psi backpressure at idle on the left-hand catalytic converter. The solution, of course, was to replace the converter assembly and, at 90,000 miles, replace the spark plugs and wires as well just to ensure that a cylinder misfire didn’t cause a repeat failure of the new converter.

I also tested a 2001 Lincoln Navigator for a lack-of-power complaint. The outstanding bit of data stream information on this road test was the left- and right-bank fuel trim numbers were running at opposite extremes of over 25% at highway speeds.

Despite this data, the client shop claimed that dropping the exhaust header pipes down a few inches to relieve exhaust pressure didn’t make any difference in performance. Puzzled, I responded that we had a serious fuel trim problem and that something other than a vacuum leak was causing it. Upon further inspection, the shop discovered that one of the two catalytic converters had collapsed, completely clogging the exhaust.

There are two points to be made here. First, fuel trim numbers, not cylinder misfires, indicated that I had an exhaust restriction problem. Second, I can only speculate that, when the exhaust was dropped, the atmospheric oxygen surrounding the oxygen sensors skewed the fuel trims so much that it decreased engine performance.

Whatever the case, dropping an exhaust pipe to test for exhaust restriction is not, in my opinion, a valid exhaust restriction test on modern OBD II vehicles. Keep in mind that to be valid, any test must produce repeatable results. When a tech drops the exhaust, he simply introduces another set of performance issues. That’s why I recommend using an accurate gauge and pressure tap to confirm suspected exhaust backpressure problems.

You May Also Like

Alternator Testing For No Charge Conditions

Many alternator problems turn out to be nothing more than a bad connection at the alternator or a bad wiring harness.

If only battery voltage is present at the battery on a running engine, does this mean the alternator is “bad?” No, it does not. It only means that the alternator is not charging, but does not reveal why. Therefore, it does not point to a faulty alternator. All too often, the alternator is condemned by technicians due to this test alone. The cause could instead be a module that distributes the power to the vehicle and regulates the alternator.

Tech Tip: No Start in Cold Weather Due to No Weep Hole

This condition can be caused by condensation freezing in the muffler assembly.

Determining Catalytic Converter Efficiency

Almost every part on the engine determines how long a catalytic converter will last.

Talking To Customers About Oxygen Sensors

One of the toughest components to sell as a maintenance item, or even a replacement item, is an oxygen sensor.

PCV System DTC

PCV systems are far more complex now, but they are also much better at controlling crankcase pressure.

Other Posts

Hybrid Vehicle Fluid Maintenance

There are opportunities your shop can’t pass up just because the car or SUV has a hybrid badge on the back.

Using a Scope

Ignition coil and plug diagnostics.

Repairing Wiring Harnesses

A guide to probing and poking.

Diagnostic Procedures For Stop/Start Systems

Stop/start technology is more than just a switch wired to the brake pedal and starter.