Engine mechanical testing: Good, better and best

June 1, 2018

In this first of two articles on engine mechanical testing, we will introduce two fundamental engine mechanical tests that will be performed with electronic tools. This first test is the relative compression test performed with a scope and high amp current probe. The second is the cranking vacuum test performed with a scope and pressure transducer. In the next article, we will look into the engine with in-cylinder pressure transducer testing. 

Let’s look into the fundamental tests that have traditionally been performed to determine engine mechanical condition to see how times are changing. For decades, technicians have used vacuum gauges to measure intake manifold vacuum to help determine engine-sealing issues. If there was a cylinder power contribution problem suspected, the compression gauge and cylinder leakage gauge were used to determine the root cause of failure. The shortcomings of analog vacuum gauges are their inability to pinpoint a cylinder-specific problem. A vacuum gauge measures the average vacuum inside the intake manifold and some gauges are so damped they may mask a valve sealing problem. A scope and pressure transducer will display individual cylinder vacuum “pulls” or vacuum events so more detailed analysis is possible. As engines have evolved, more elaborate intake manifold designs have been incorporated on V-style engines, which can make access to spark plugs difficult and thereby render gauge-style cylinder compression testing a last resort. There must be an easier way to determine base engine mechanical health. The good news is there is a simple yet powerful mechanical test — the relative compression test. The relative compression test is not new; big-box engine analyzers performed cranking and running tests and used input from a current clamp around the battery cable to ferret out base engine mechanical problems. These large analyzers are all but gone in most shops, and many younger technicians have never seen one used at all. Many shops today have high-quality digital storage oscilloscopes available and by adding a few inexpensive probes, the scope will do a great job at displaying relative compression tests.

Inside relative compression

Let’s start with some basic theory and build on how this test will expand your diagnostic capabilities. A relative compression test relies on the fact that as a starter motor turns over a normal engine, as each cylinder is pushed to top dead center on the compression stroke, the force required to compress the trapped air increases and hence the current required by the starter motor to overcome this force increases as well. An old-style analog battery/starting/charging system tester would show the ammeter needle wavering during a cranking test so we know the current is varying in the circuit. A scope and good high-amp current probe is so fast that this changing current can be displayed on the scope. I will mention here that if the current to the starter motor is oscillating, the voltage in the circuit changes as well. It is possible to do a simple voltage test at a battery-positive voltage point and ground with a scope and see this voltage oscillation during cranking, but the current probe will give a clearer picture and will be used as the test tool of choice for this article. When you perform this test, it may be done either synced or un-synced, meaning you can add a second scope channel to an ignition trigger signal so that individual cylinders can be identified. If the engine has spark plug wires, you may need a sync probe for your scope to trigger from a specific cylinder. Sync probes are inexpensive and readily available. Let’s look at a normal relative compression test and point out the basics of how to use the test for engine diagnostics.

Figure 1 - Relative compression test from a 2012 Chevrolet Suburban, 5.3 V8 engine. The engine was cranked over for 9 seconds.

The first waveform (Figure 1) shows a good relative compression waveform captured from a Chevy Suburban with the raw scope settings I used during the test. The scope is set to a slow time-base and the entire test is captured. A Pico scope has very good zoom capabilities so the capture will be manipulated to show much greater detail for the analysis. If you use a scope such as a Snap-on product, you will want to perform this test at a faster time-base and then use the zoom out function to look at the entire test. The Pico scope was set to 2 seconds per division, but a Snap-on scope setting for this test may be 50 milliseconds per division and then let the scope buffer fill up. This capture is a synced relative compression test because the ignition-firing event was captured. If you only use a current probe or do a battery voltage test without the ignition trigger, you can see a problem, but you would not be able to identify a particular cylinder so you may as well just get used to connecting two leads and do synced tests. The next waveform is the same capture but with heavy manipulation of the stored patterns for better viewing.

Figure 2 - Known good synced relative compression waveform. The current waveform in red shows a peak to valley amplitude of 31 amps, which is okay.

In figure 2, the original capture has been zoomed in, filtered and annotated to show the engine firing order. There is a lot of data on the screen to be analyzed. Most techs are generally looking for a low current event and not much else, but there are many items to discuss about these captures. The first item to discuss is the current waveform in red. The saw tooth pattern is what we should see and each peak represents a piston reaching TDC as the starter current increases to push the piston to TDC and past, then the current drops quickly until the next cylinder in the firing order comes up on TDC. This is a “relative” compression test so it only displays each cylinder relative to the other cylinders in the engine — it is not an actual compression value. If all the cylinders have too much or too little compression, the overall test may look normal. If the engine had a valve timing problem causing high or low compression, it would affect all the cylinders unless the engine were an OHC V-style and the valve timing was off on only one side. In that case, you could have three or four high peaks and three or four lower peaks (as seen in figure 3) depending on the number of cylinders in the engine, and the test would point you in the direction of valve timing right away. Normal relative compression waveforms typically show a peak to valley current between about 30 to 70 amps. If the current is too low, suspect an engine with low overall compression and do a cylinder compression test to verify. If a chain jumped on one side of a Ford 4.6 or 5.4 V8, think of how quickly this test would lead you to verify camshaft timing.

Figure 3 - This un-synced relative compression waveform was captured on a 1998 Ford Explorer with a 4.0 VIN E engine. The timing chain had jumped on bank 2, causing low compression on that side of the engine.

The next important item in the known-good synced relative compression waveform is the location of the ignition pulse in comparison to the current waveform. A typical engine cranks with ignition timing very close to TDC, sometimes slightly ahead or possibly slightly retarded if the engine uses a catalyst heating strategy. Most technicians do not have a go-to test to verify proper ignition timing if a vehicle is a cranking no-start, but a synced relative compression test will help verify good ignition timing on late-model vehicles with no timing marks. As seen in the known-good synced waveform (Figure 1), the ignition trigger pulse lines up with a current peak, which we will consider cylinder #1, indicating good spark timing. Compare that to Figure 4, where the timing is obviously incorrect. The only caveat here is if the engine has a distributor and the distributor is installed incorrectly, this ignition pulse could occur in sync to a current pulse, but the pulse is not cylinder #1. An in-cylinder compression waveform would uncover this problem and will be covered in the next article. Keep in mind that if cylinder #1 is not accessible due to intake manifold design you can always use the sister cylinder to #1 as your ignition sync input, which will be on the other bank of the engine. Also consider that when performing this test, you will need to prevent the engine from starting. Do not disable the ignition system if you are triggering off a spark event. Remove power to the fuel pump or injectors when performing a synced relative compression test.

Figure 4 - This synced relative compression test from a 2002 Ford Escape 3.0 shows very retarded spark timing causing a no-start condition. The engine cranks in DIS mode, so two firing events are seen in 720 degrees. The crankshaft sensor trigger wheel has two keyway slots, and the wheel was installed in the wrong slot.

When it’s wrong 
When viewing a relative compression test on an engine with low compression in one cylinder, the problem will be fairly obvious; there will be a low or missing peak depending on how low the compression is. Typically, the cylinder following a low compression cylinder will have a slightly higher peak due to the starter speeding up when the low compression cylinder pushes past TDC and then slows down when the next cylinder with good compression comes up to TDC. The next two waveforms (Figures 5 and 6) were captured from a 2005 GMC Yukon with a rough-running, low-power complaint that was diagnosed as a bad catalytic converter at a GM dealer. As the relative compression test displays, the first 4-stroke cycle showed good compression, but then the #3 cylinder lost all compression as the engine continued to rotate. The first cycle proves the cylinder can seal and a gauge-style compression test was done at the dealer and the tech saw good compression on the gauge, but the gauge captures pressure due to the Schrader valve in a compression test hose. As the engine rotates, the cylinder seal is lost. This can only be a mechanical problem, most likely caused by a valve sealing issue due to the complete lack of compression pressure seen on the relative compression waveform. A broken valve spring was suspected and confirmed when the valve cover was removed, the exhaust valve spring was broken (Figure 7).

Figure 5 - Relative compression waveform from a 2005 GMC, 6.0 V8, showing a normal pattern at the beginning, but a problem occurring as the test continues. The arrow shows a loss of compression that repeats for the rest of the test.
Figure 6 - The zoom in of the 2005 GMC capture shows cylinder #3 changing from good compression to no compression as the engine turned over.
Figure 7 - Broken exhaust valve spring on 2005 GMC Yukon. This was found in less than five minutes of testing with a relative compression test after several days of testing and parts swapping by other technicians.

Sometimes it may not be a low or missing current event that is seen while performing a relative compression test; sometimes it may be the opposite. The next waveform capture (Figure 8) is from a 2008 Pontiac G8 with a 6.0 V8 with GM’s AFM (Active Fuel Management) system, which can cancel four cylinders during light-load operation. The engine is logging a cylinder #4 misfire code and this is an AFM cylinder. A synced relative compression test was performed. The waveform shows almost twice the level of current needed to push cylinder #7 to TDC on the compression stroke. Can one cylinder have higher compression than the rest without a high domed piston or longer connecting rod? Not possible or likely, unless an engine builder is playing games. This is not a carbon issue either as the current is double the other cylinders.

Figure 8 - Relative compression waveform from a 2009 Pontiac G8. Note the high current seen on cylinder #7. This indicates there is increased effort needed by the starter to push cylinder #7 to TDC on its compression stroke.

The problem is easily explained: there are two cylinders on compression at the same time, #7 and its sister cylinder #4. The exhaust lifter has collapsed on cylinder #4, so when it should be on its exhaust stroke with the valve open, the valve is closed and both cylinders are compressing air, hence the misfire code for cylinder #4 and the high current event on the sister cylinder #7. It should be clear by now the importance and capabilities of this simple yet effective mechanical test. 

“Seeing” cranking vacuum 
The second test mentioned was cranking vacuum. To perform this test, a pressure transducer will be required and there are two types. Absolute pressure transducers like the Pico WPS500 (shown in Figure 9) can measure both actual vacuum levels and display what are called vacuum events or pulls created by each cylinder in the engine. There are also differential pressure transducers like the Sen-X Technologies First Look sensor. This transducer displays the change in vacuum or pressure seen by the sensor and not the actual level, but it is very sensitive and well suited to this particular test. Figure 10 shows both sensors connected to the same port and you can make your own decision if you want one or both. There are other companies making high-quality vacuum transducers; I am only mentioning units I have personal experience with. Whatever brand and style transducer you use is not important, only that you actually own them and perform the test.

Figure 9 - Connecting pressure/vacuum transducers to a GM V8 engine to perform a cranking vacuum test. Connecting to a centrally located vacuum port and keeping your hoses short will produce the best test results.

As with the relative compression test, the cranking vacuum test should produce a clean, oscillating pattern with good uniformity. Again, it is important to identify which signal event is created by which cylinder in the engine, so I will illustrate the procedure in a captured waveform. There will be a vacuum event every 180 degrees in a four-cylinder engine, every 120 degrees in a six, and every 90 degrees in an eight-cylinder engine. A vacuum stroke occurs 360 degrees after the cylinder fires on the power stroke, so you need to capture an ignition event when performing a cranking vacuum test to properly identify the events on the screen. In the annotated waveform shown in Figure 11, I have used rotation rulers found in the Pico scope software to break up the waveform into eight evenly spaced intake events. The numbering order on top is the firing order, but it is overlaid on the vacuum events, not the ignition event. The vacuum pull happens 360 degrees after the ignition coil firing signal.

Figure 10 - Cranking vacuum waveform with both a Pico transducer and a First Look transducer connected to the same test port. Note the Pico waveform pulls down from zero volts due to it being an absolute pressure transducer while the First Look transducer’s signal oscillates above and below the zero voltage reference. Both transducers display vacuum events or pulls.
Figure 11 - This cranking vacuum waveform from a 2005 GMC Yukon, 6.0 V8 is considered good. The individual cylinder vacuum events are labeled. There should be the same number of vacuum pulls as the number of cylinders, and they should be relatively even.

Once you capture a cranking vacuum waveform and ID the events, you can begin to make conclusions on what you see relative to the problem on the vehicle. Remember we are looking at mechanical engine sealing with electronic tools, so keep in mind what could go wrong in an engine when analyzing an abnormal pattern. Valve leakage, valve timing, ring sealing and intake or exhaust path restrictions can all have an effect on these patterns. The 2005 Yukon with the broken exhaust valve spring was tested with cranking vacuum. The pattern in Figure 12 shows the effect of the sticking open exhaust valve on cranking vacuum.

Figure 12 - Bad cranking vacuum waveform from broken exhaust valve spring on cylinder #3.

When this pattern is looked at by itself, it is difficult to say the problem is a broken exhaust valve spring, but it is easy to see there is a problem with the engine and the problem is mechanical! When the intake valve opens on cylinder #3’s intake stroke, there is a loss of vacuum in the engine; the top Pico trace shows this because it is an absolute pressure transducer. Both transducers are momentarily connected to the exhaust manifold through the open intake valve in cylinder #3 and the First Look sensor clearly shows a pressure rise as exhaust pressure connects to the intake manifold when the intake valve opens. This is only the beginning of what can be seen when pressure transducers are connected to the engine. We will continue the discussion in the next article with in-cylinder pressure testing. Once you add these tests together on the same scope screen, the operation of the internal combustion engine becomes much clearer. I am sure you will not be disappointed if you acquire the necessary tools and begin using these tests in your diagnostic routine. Practice is important and testing good engines first should be a priority. Remember, an engine can only operate properly if its base mechanical condition is good! The quicker you can determine this, the better. Best of luck!

Voice Your Opinion!

To join the conversation, and become an exclusive member of Vehicle Service Pros, create an account today!