In Front of the Flywheel - June - 2018

Pressure and Vacuum Diagnosis, Part III: Running Compression

In the last two issues of GEARS Magazine, we covered engine mechanical testing using high current probes for relative compression testing in conjunction with transducers for cranking vacuum testing. We also touched on in-cylinder cranking compression testing using a pressure transducer.

This month we’ll expand our in-cylinder testing by using a pressure transducer to analyze running compression.

The Test

Much like our previous compression test, you’ll need to install the transducer in the spark plug hole to observe cylinder pressures. But in the case of running compression, you’ll need to enable fuel and ignition to allow the engine to run.

Ignition needs to function for the cylinder you’re testing because you’ll want to see when the spark for that cylinder occurs. If possible, you’ll want to disable fuel for only the cylinder you’re testing. Once the engine is running, there’s a plethora of data you can observe.

The Anatomy

Before you can analyze an oscilloscope capture using this technique, you need to understand the anatomy of a known good waveform. In figure 1, the green trace is a pressure waveform from a cylinder while the engine is idling. I’ve used cursors, from top-dead-center compression to top-dead-center compression, to mark the 720º of crankshaft rotation that occurs in the four-stroke cycle.

It’s easy to identify TDC compression because cylinder pressure is the highest at this point. Notice that the running compression pressure is only 74 PSI. It’s normal for the running compression at idle to be lower than the compression pressure of a cranking test.

The red trace on the capture is the ignition firing event for the cylinder where the pressure transducer is installed.

After identifying the 720º cycle, I divided the capture into four equal, 180º sections using the scope’s rulers. Each of these four sections corresponds to one of the four strokes, labeled in blue.

During the power stroke, the pressure decreases steadily as the piston moves down. Because there was no combustion, the pressure continues to move into the vacuum range. Shortly before the piston reaches bottom-dead-center, the exhaust valve opens. This is the point where the pressure stops moving down and starts moving back up again.

The next stroke is the exhaust stroke. During this stroke, the exhaust valve should be open and pressure in the cylinder, although fluctuating slightly, should remain near 0 PSI. Near the end of the exhaust stroke, the intake valve opens and valve overlap occurs. The timing of this valve event is very hard to determine because of the valve overlap occurring at the same time.

The intake stroke is the next area to focus on. The exhaust valve closes shortly after the beginning of this stroke. Again, due to the valve overlap, this event is very hard to pick out. As the intake stroke continues, you can see the vacuum generated.

The final 180º section is the compression stroke. At the beginning of the stroke, the piston is on its way up in the cylinder but the intake valve is still open. The pressure in the cylinder should start to rise right around the time the intake valve closes. You can see final compression pressures when the piston reaches TDC.

Now that you’ve seen a good-running compression waveform, let’s take some time to look at some faults you can identify using this technique.

Where’s the Distributor?

Nowadays, an often-overlooked cause of a low power-related issue could be retarded ignition timing. Most of today’s vehicles don’t have that octopus-looking thing we used to call a distributor and there are no provisions for ignition timing adjustment.

That doesn’t mean that ignition timing can’t be off. Usually it occurs when something is physically broken and, with no timing marks, there’s no way to check the setting.

For example, let’s look at this Ford Econoline with a 4.2-liter engine. The vehicle was at a shop; the complaint was low power on acceleration. The technician had already replaced all the usual suspects including the MAF sensor, spark plugs, ignition wires, ignition coil, CMP sensor, fuel pump, fuel filter, catalytic converter, and the entire exhaust system. Even after all of these repairs, the van still had low power.

We performed an in-cylinder compression test (figure 2), which confirmed that the ignition timing was retarded.

In this case, the ignition-firing event (red) occurs after top-dead-center: To the right is retarded. With the engine running, we’d expect the firing event to occur well to the left of top-dead-center, or advanced.

To calculate exactly what the ignition timing is requires some simple measurement and math. First, measure the time it takes for the 720º, 4-cycle event to occur. For discussion, let’s say that measurement is 136 milliseconds. Next, divide 720 by the time measured to figure out how long 1 degree of crankshaft rotation takes. In this case, 720 ÷ 136 = 5.3 milliseconds.

If we now measure how far away the ignition event is from top-dead-center, say 45 milliseconds, then 45 ÷ 5.3 = 8.5. If the measured ignition event was before top-dead-center (or to the left), then the ignition timing would be about 8º-9º advanced. If it were after top-dead-center (or to the right), then the timing would be 8º-9º retarded.

Ignition timing on almost all modern vehicles is based on the crankshaft position sensor’s input to the PCM. If the timing is off, examine the CKP reluctor closely. In the case of the Econoline, the crankshaft pulley keyway had worn (figure 3) and caused the reluctor to shift on the crankshaft.

How About the Valves?

The next value of this technique is to distinguish between different valve train faults. When you understand the known good anatomy of a compression waveform, you can begin to analyze the waveforms by looking for issues, or anomalies, that don’t fit the norm.

This next vehicle was a Dodge with a 4.7-liter engine that had a misfire. It had sufficient spark and fuel. The technician also opted to swap ignition coils and fuel injectors to no avail; the misfire didn’t move with either attempt. Once he performed a running compression test (figure 4), the mechanical fault became obvious.

In the waveform, you can see a compression stroke and a power (or expansion) stroke. But notice the pressure buildup when we should see an exhaust stroke. It almost appears to be an extra compression stroke. Actually, it’s an unwanted compression stroke because the exhaust valve isn’t opening.

If an exhaust valve sticks closed, the exhaust has nowhere to escape and pressure builds again. Notice the rapid pressure release the moment the intake valve opens to prepare for the upcoming intake stroke. The Dodge in question had a cam follower that had fallen off (figure 5) and was no longer opening the exhaust valve.

Every compression related issue will appear a little different on the scope. They all have different “signatures” if you will. The next vehicle didn’t have a dead misfire but did exhibit some rough running. A relative compression test revealed that one of the cylinders was a bit lower that the rest. When performing an in-cylinder running compression test (figure 6) we obtained a very interesting waveform.

What this capture shows is an unwanted rise in pressure near the end of the exhaust stroke. Obviously, compared to our last example, the exhaust valve isn’t stuck closed. But could it be closing early?

That’s exactly what’s occurring. This could be caused by advanced exhaust camshaft timing, but in this case, only one of the cylinders is affected. Given that piece of information, we can take camshaft timing off of the table. But could the valve timing still be off?

Notice the maroon bracket labeled Exhaust Valve Open: The pressure rise where the power stroke ends and the exhaust stroke starts occurs much later than expected. This indicates that the exhaust valve is opening late and closing early.

We definitely have a valve timing issue, but only on one of the valves. The trick is to confirm what’s causing it. It could be a collapsed hydraulic lifter, but in this case, the fault was a worn camshaft lobe.


This technique can be valuable in two major ways: The first would be a quick, clean, and accurate method for diagnosing mechanical or ignition timing issues. The second would be as a learning tool.

How do we learn from this technique? If you obtain a capture, you can’t make sense of, save it for future analysis. After you’ve disassembled, diagnosed, and repaired the engine, revisit the capture knowing the fault and figure out why the pressures behaved the way they did.

Then you’ll be more apt to recognize the “signature” of that specific fault.

Engine or electrical diagnostic issues you’d like to see addressed? Let Scott know. Send him an email at and you just may have your question covered in a future issue of GEARS Magazine.