'Is It Compression, Spark, Fuel?'Posted 8/13/2004
By Brian Manley
The contest began when the teams ran to their cars, flipped over the work orders under the wiper blades, read the listed customer concerns and started diagnosing their vehicles. We instructors watched in puzzlement from the sidelines as some of the Mustangs ran for a while, then died, while others were complete no-starts. "What could cause the cars to act differently?" I asked another instructor. "They are supposed to have the same bugs."
Time passed, and the students attacked other concerns: Trouble codes led to defective sensors, light harnesses were faulty and fuses and bulbs were broken. The bad parts were swapped with the judge for good parts and then reinstalled in the car. Many vehicle systems were brought back to life until finally, one of my teams faced the inevitable: Why won't the car start?
I saw them looking at the engine and talking with each other. "What could be the cause?" I was thinking with them from a distance. Then I saw them poke around the engine compartment and remove the air intake ductwork, where they discovered a tennis ball wedged halfway into the air intake, just before the throttle body. They threw the ducting back together, fired up the Mustang and went to final judging. Their diagnostic abilities earned them a second-place trophy, some new hand tools and scholarships to automotive colleges.
When we're faced with a vehicle that has one, two or all of its cylinders not contributing horsepower, how do we isolate the root cause for a wimpy cylinder or a full-blown no-start? No-starts seem to be the easiest to isolate - a slipped timing belt, a defective fuel pump, an ignition coil that's arcing to ground, a module that has failed - there is always a straightforward series of tests that lead to the root cause.
If the engine cranks, but fails to start, I'll usually check the most obvious things first. I'll hook up a couple of spark testers to plug wires and a noid light to an injector connector. If I'm missing one or both of these signals, I'll funnel diagnosis to a specific component or connection. Often, there is a spark and injector trigger, but no fuel pressure when checked with a gauge. Still other times, the engine sounds as though it has low compression, which is confirmed with a compression test. But which methods do you use to isolate a single cylinder misfire? What is the most effective method to find the root cause for one cylinder that does not contribute any power?
Older engines with distributors are easy enough to power balance; I still have a power balance tester that hooks to the No. 1 plug wire and to the negative coil terminal that allows me to manually kill cylinders. Distributorless systems need a different method to kill cylinders. I use a set of metal spacers that allow me to use a test light to ground the spark to that one cylinder, or I insert inch-long pieces of vacuum hose. Most vacuum hose has enough carbon to be used for this purpose. This one is great to show students!
Many OBD-II systems, and some OBD-I systems, allow you to power balance the engine by shutting down injectors or by killing coils to see if the rpm changes when this is done.
'98 Olds Cutlass - 3800 V-6
We recently had a customer bring this vehicle to us with a "shaking at idle" concern. Our Tech 2 scanner was able to perform power balance and injector balance tests. The power balance test revealed that the No. 2 cylinder wasn't contributing (no rpm drop when killed). A spark test on that wire verified good fire, and a compression test showed 125 psi. As I worked through the possible scenarios with my students, several listed the injector as a suspect, so I asked them how we would go about testing it. The ohms test answer came up, so we tested No. 2 and compared it with the other five. They ranged from 11 to 13 ohms; all within specification.
Next, I grabbed our hand-held injector balance tester that can be used on any port fuel injected system to energize the injectors. This unit will let you simulate idle, cruise and wide-open throttle, or manually control the injector by holding down the button. The other three modes have a calibrated time built in, such as 10 milliseconds for the idle selection. This unit, along with a fuel pressure gauge, will elicit the same results as the OBD-II/scan tool method.
If, after power balancing, you discover that a cylinder or two has no fire to its plug, then simply backtrack until you find the defective part. On a conventional distributor ignition, if - for example - one cylinder out of four has no spark, then the problem would be traced to a plug wire or a plug; some component after the spark entered the cap. I still see burned-through sections in distributor caps that divert one cylinder's spark to ground, or an adjacent cylinder.
On a distributorless ignition system (DIS), the faulty part would be the wire, coil or module. If the engine has two spark plugs that are without spark and they share a common coil, swap that coil with an adjacent one and see if the no-spark condition follows the coil. If it does, then the coil is at fault. But if the problem remains, the module could be the culprit. This same method can be used to diagnose coils on a coil-over-plug (COP) system.
We Have Fuel, We Have Spark
If you have a lab scope that will perform relative compression testing, often this test can lead you to the cylinder with low compression in the fastest time. Usually, however, we must screw in a mechanical compression gauge.
Once a low compression cylinder is identified, a cylinder leak-down test must be performed to isolate the root cause. Is the intake or exhaust valve burned? Has the sealing ability of the rings finally given up? Has a camshaft lobe started to go flat?
Introducing air into a cylinder that is at top-dead-center on the compression stroke will reveal where the compression is escaping. A burned valve will allow air into the intake or exhaust systems, and a poor cylinder-to-piston seal will fill the crankcase with air. A flat camshaft lobe won't show up on a leak-down test; this will most likely show up during running compression test.
I still recall Chevrolet 350s that would have a "dead" No. 8 cylinder when the vacuum hose for the transmission vacuum modulator would leak vacuum. Vacuum leaks can kill a cylinder as much as a plugged injector. If the fuel mixture is too lean, it won't burn. When searching for the reason for a dead hole when spark, fuel and compression are normal, never forget the integrity of the air intake system. A good method to locate intake leaks is with a smoke machine. This is less scary, less dangerous, and often more accurate than using carb cleaner or propane around a running engine.
Oh, and don't forget those pesky field mice and tennis balls!
AutoInc. Web Site |
ASA Web Site |
U.S. Supreme Court Rules on HMO Suit Limitations |
Collision Repair Consumer Protections Still Remain Elusive in Many States |
Making the Right Scan Tool Purchase |
Estimating Systems |
Steering to 'First Position in the Mind' |
Guest Editorial |
Tech to Tech |
Tech Tips |
Around ASA |
Shop Profile |
Net Worth |
Stat Corner |
Copyright (c) 1996-2011. Automotive Service Association®. All rights reserved.