Modern Day Testing Strategies for Mechanical Integrity Issues
Tips for Diagnosing High-Mileage Engines
You likely have noticed that a significant number of vehicles that have come into your shop lately have six-digit odometer readings. What this could mean is that a mechanical integrity issue may be the cause of a misfire code or a misfire symptom. The average age of today’s vehicles is now over 11 years old. In addition, a common problem seen on GDI engines is leaking intake valves due to carbon build-up on the back side of the intake valve and valve seat and face causing compression leakage and misfire codes. The traditional cylinder leak-down tester usually will not find this problem. We will cover this problem in detail later in this article. We have seen valve timing issues, leaking intake or exhaust valves and wiped cam lobes well before the odometer reading turns to a six-digit reading. In the past, mechanical integrity issues were often difficult to pinpoint with the old diagnostic strategies. The use of a DSO coupled with a pressure transducer now gives us a clear dynamic picture inside the combustion chamber to identify these problem areas. One of the major issues in this industry is what and how much do we charge for diagnostics.
In this article, we are going to cover some specific tests that can be done inside of one hour. I have had many tech calls to other shops whereas a misfire symptom and code existed when the technician had replaced the spark plugs, the injectors and the coils to no avail. A simple relative compression test would have alerted the technician as to a low compression problem. An electronic compression waveform would greatly benefit technicians in pin-pointing cylinder mechanical issues. We will cover this test in detail later on in this article.
I am dating myself here, but I bet some of you can recall a procedure we used to use on distributor-equipped engines when we suspected a loose and worn timing chain that may be the cause of loss of power. The procedure pretty much went like this: With the distributor cap removed we simply used our breaker bar and a socket on the vibration dampener bolt and rotated the engine in a counterclockwise direction until the TDC marks lined up. Now as we rotate the crank in a clockwise rotation we simply note how many degrees the crank turned before the rotor moved. Typically any value over 14 degrees pointed to a loose or worn timing chain. Some of us actually put a degree wheel on the dampener to do this test.
On modern-day systems with engines equipped with variable valve timing, the rules have changed and in some ways are more accurate and easier to do with the use of scan tool data. There are some specific rules when looking at scan data if we have a variable valve timing issue. We will look at this data a little later on in this article. Also in this article, we will cover valve timing issues, cylinder compression testing and cylinder leak down testing.
We will conduct most of these tests electronically using pressure transducers dynamically. We will also use a high inductive amp probe to conduct a relative cranking compression test to monitor the amperage value from the starter as each cylinder comes up on its compression stroke. When a cylinder is low on compression the amperage draw on the starter for that cylinder on its compression stroke will be lower. We will cover this test later on in this article.
Compression testing has been around since the advent of the internal combustion engine. The conventional compression gauge was only designed to do a wide-open throttle cranking compression test. When using a conventional compression gauge keep in mind that there is a Schraeder valve inside the hose that acts like a one-way check valve. What this means is that for example, during a 3-second wide-open throttle cranking compression test, each compression stroke bumps the compression values up. What I have always taught is that the second compression reading is the one that should be noted. The reason we don’t focus on the first is that we may not have gotten a full intake stroke. The second compression value will assure us that a full intake stroke would have occurred. Again the conventional compression gauges were designed to do wide open throttle cranking compression tests only. In reference to a retarded valve timing issue you can actually have good cranking compression values.
The tests we are going to conduct involve using a DSO and electronic pressure transducer coupled to our conventional compression gauge hose. If we leave the Schraeder valve in the hose you can see the compression values in Fig. 1. This test is on a good 4 valve Honda engine. Notice the second compression stroke indicates a good value of 170 PSI. During WOT cranking conditions most engines are in the clear flood mode by shutting off the injectors. If you are doing this test on an engine that doesn’t have clear flood mode, simply remove the fuel pump relay or fuse and start the engine until it runs out of fuel. Then you can proceed to do a WOT cranking compression test. The electronic pressure transducer will allow us to view compression values during cranking, at idle, off idle and during a wide-open throttle snap condition with the Schraeder valve removed. The difference in compression values on good engines during these 4 tests are indicated on the chart in Fig. 2. These 4 tests can help us isolate a low compression problem, a retarded valve timing problem, a wiped cam lobe or a restricted catalytic converter problem as you will see in our case studies. Notice in Fig. 2 that off idle at about 1500 RPM the compression values are at only about 40% that are shown during side open throttle cranking conditions. The reason being is that as the RPM increases and the throttle plates are barely off idle, the engine simply is not pulling in a lot of air under these conditions. In addition, note that a wide-open throttle snap test will equal that of a wide-open throttle cranking compression test.
Now let’s look at the anatomy of a compression waveform with the Schraeder valve removed and the engine at idle. See Fig. 3. We can now view all four strokes occurring dynamically. We are now calling the power stroke the expansion stroke since we are not firing a spark plug. We can hook a spark tester to the plug wire or coil and sync off the firing event to verify good spark timing. We will demonstrate that later on. As you will see later on we can pinpoint the exhaust valve opening point (EVO) and the intake valve opening point (IVO) in the event we suspect a retarded valve timing problem or a wiped cam lobe.
The cylinder leak down tester was improved in the early 90s whereas the amount of pressure applied to a cylinder at TDC has been reduced to about 40 PSI. The leak-down gauge has been designed to be more sensitive. See Fig. 4. This tester may not be sensitive enough to detect leaking intake valves on GDI engines caused by carbon build-up on the valve face and seat. We are going to demonstrate a new strategy later on in this article.
Now we are going to diagnose cylinder pressures on a high mileage 3.8L Chrysler push rod engine during cranking, idle, cruise and WOT snap conditions.
Fig. 5 indicates a compression value during cranking at 125 PSI. This is a high mileage engine with an oil consumption problem. If you recall the wet test, a couple of ounces of oil was inserted into the cylinder and we repeat the compression test. An increase of 20% compression values pretty much-indicated piston ring blow-by. Carbon buildup on the compression rings may be the problem. Please stay tuned for more info on that area.
Now let’s go ahead and look at the idle compression values in Fig. 6 indicating 45 PSI. Note that the off idle values in Fig. 7 indicate a value of 35 PSI. The reason for this is that the throttle plates are barely open and the RPM has increased and the engine is simply not pulling in a lot of air. Keep in mind that this is a high mileage 2-valve engine that indicates low compression values across the board. Fig. 8 indicates a WOT snap compression value that matches that which we captured during WOT cranking conditions of 125 PSI. Compression values are measured from the exhaust pocket up to the TDC point.
A new test involves performing a relative compression test using a DSO and a high inductive amp probe. The amperage waveform values don’t represent cylinder pressures but the relative starter current flow values as each cylinder reach TDC on its compression stroke. We simply clamp the amp probe around either the positive or negative battery cable. Make certain that the arrow on the amp probe points in the direction of the current, towards the starter on the positive battery cable or towards the battery on the negative battery cable. We can use a synch probe around #1 secondary lead to sort out the firing order. If the vehicle is equipped with a Chrysler COP ignition or a Ford COP system we can synch off of #1 primary. On COP systems where the igniter is integrated into the coil, we can synch off #1 igniter trigger signal. Obviously, we would need to have the firing order in hand. A cylinder low on compression will cause less starter draw current. We will show some case studies later on in this article.
Let’s assume that you suspected a retarded valve timing issue or a wiped cam lobe. We can detect that by viewing the compression waveform during idle no-load conditions by focusing on the exhaust valve opening point (EVO) or the point of intake valve opening (IVO). Note Fig. 9. We have captured a single-cylinder compression event with two TDC events. The distance between the two TDC events represents 720 degrees crank rotation which is needed to complete all 4 strokes. Note that the distance between the two cursors measures 176 milliseconds. We simply divide this value by 4 and can determine that a separate stroke will occur every 44 milliseconds. To find the 180 degrees BDC point we simply drag the second cursor to the 44-millisecond mark. This identifies the 180-degree mark and the bottom dead center. Notice in Fig. 10 that the EVO point occurred before the 180-degree BDC cursor. The number of minor divisions between the TDC cursor and the 180-degree cursor indicates 20. !80 divided by 20 minor divisions tells us that the crank turned 9 degrees per minor division. Note that there are 5 minor divisions between the EVO point and the 180-degree cursor. 5 times 9 tells us that the exhaust valve opening occurred 45 degrees before the bottom dead center. To locate the 360-degree mark we now move the cursor over to the 88-millisecond mark. This identifies the 360-degree TDC mark.
Look at Fig. 10A. Notice that the intake valve opening occurred 5 degrees after TDC. Notice how distinct and uniform the pressure changes that occurred at the EVO and IVO points were. On an engine with a wiped cam lobe, these points would be reduced. In addition, on engines with a wiped cam lobe, the valve would have opened later and closed earlier. The valve timing specifications on this 3.8L Chrysler engine are indicated in Fig. 11. Notice that the OEM spec says that the exhaust valve opening should occur 46 degrees before BDC and the intake valve opening should occur at 1 degree after TDC. On engines with fixed valve timing in this example, the 180-degree cursor should pretty much split the EVO ramp in the center. See Fig. 12.
The good news is that on engines with variable valve timing systems we have robust scan data to alert us to a retarded valve timing issue. This data will not help us in diagnosing a wiped cam lobe, but it will help us pinpoint a retarded cam timing problem. When looking at this data, a very important cardinal rule is that the RPM must be stable. If you are in and out of the throttle you will see these values unstable on your scan tool. In addition, on most GM engines equipped with variable valve timing, the PCM will not vary the valve timing in your bay when you increase the throttle. The PCM must see a VSS input as in a test drive. You have the ability to command the variable valve timing control solenoid in your bay with your scan tool to say for example a 25% command and then to track the camshaft position. GM refers to the camshaft out of its desired position as cam variance. See Fig. 13. In Fig.14 we see GM cam variance data during a test drive. The Ford systems use the term cam error. Chrysler and the Asian systems may use the terms actual and desired. See Fig. 15. Not all scan tool companies will use this same nomenclature.
On domestic engines equipped with variable valve timing, the exhaust cam timing is advanced at idle with 0% duty cycle command to the variable cam timing control solenoid. Off idle, the PCM will increase the duty cycle control to the solenoid and retard the exhaust cam timing. The intake cam timing at idle is retarded with a 0% duty cycle command and off idle the PCM increases the duty cycle command to advance the intake cam timing. The net result is valve overlap which eliminates the EGR system and improves the engine’s volumetric efficiency off idle. The Asian engines equipped with variable valve timing will do the opposite of the domestic engines. The exhaust cam timing is retarded at idle and advances off idle while the intake cam is advanced at idle and retarded off idle. The net result is the same in that it creates valve overlap during off idle conditions.
The GDI engines have been plagued with a carbon build-up problem on the back side of the intake valves. We no longer have the benefit of the solvent effect of fuel being sprayed on the backside of the intake valves as in the PFI systems, although some Lexus V-8 engines incorporate both GDI and PFI systems. This carbon build-up is caused by vaporized oil from the PCV system. This carbon build-up can occur on the valve which impedes the engine’s ability to breathe and can cause compression leakage past the intake valve face and seat. See Fig.16. Many chemical makers supply an aerosol spray to introduce into the intake system during off idle conditions. Two examples are shown in Fig. 17. This procedure is becoming very popular for GDI engines. In my opinion, this procedure should fall under a maintenance requirement. In cases where the carbon build-up is extreme, some shops have reverted back to the walnut shell blasting process to remove the carbon. This would obviously require the intake manifold to be removed. The folks at BG Products have come up with a kit that includes a chemical known as 44K, along with large wooden toothpicks along with a wire brush and some zip ties strapped together. The procedure involves removing the intake. On cylinders with the intake valves closed, you would add a couple of ounces of 44K. After a 15 minute cold soak, you would now use the wooden toothpicks to break apart the molecular bonding of the carbon. The wire brush is used to scrub the backside of the intake valves. A bunch of zip ties are supplied and tied together. The straight end of the zip ties are inserted into the end of a drill. This is designed to scrub the intake runners. Rotating the engine over until the other intake valves are closed, we simply repeat the process on the remaining cylinders. The Lexus dealers are well aware of this problem on their GDI engines. Some Lexus dealers use an additive during an oil change interval known as MOA (motor oil additive) also available from BG Products. The additive helps prevent vaporized oil from forming and building up on the intake valves.
Compression losses from carbon build-up on the intake valve seat and valve face normally cause an intermittent misfire code, usually flagged during a cold startup. The misfire usually goes away after the engine warms up. The cylinder leak-down gauge usually will not be sensitive enough to find this slight compression leakage thru the intake valves. Let’s say that the PCM has flagged a #3 cylinder misfire and the freeze frame data indicates it was flagged during engine cold conditions. Since this is a GDI engine, you may suspect this common intake valve leakage problem. Do you all have a smoke machine to help pinpoint EVAP leaks?
With a barb vacuum fitting, we can apply smoke pressure to the #3 cylinder at TDC thru a conventional compression gauge hose. The flow gauge will alert you to a compression leakage from a carbon build-up on an intake valve and seat. With the air duct removed you should see smoke waffling out of the throttle body. Keep in mind that valves rotate when the engine is running. If the flow gauge is bottomed out indicating no leaks, start the engine briefly and repeat the smoke test. This is best done on a cold engine. In addition, carbon buildup on the compression ring lands is known to have been a problem on all engines and even on GDI engines. This carbon prevents the compression rings from expanding, causing compression blow-by. See Fig.18. The folks at BG Products have a special chemical known as 109 to help address this carbon build-up problem on the piston ring lands. I have had success using this chemical on low compression problems caused by carbon build on the compression ring lands on PFI engines as well GDI engines. In addition, I have done many engine rebuilds where I have literally had to chisel out the compression rings due to carbon build-up on the pistons' ring lands. Carbon build-up on the tip of the injector can also cause carbon deposits which can easily cause a lean condition and a lean density misfire with double-digit positive fuel trim values. In addition, carbon build-up on the GDI injector pintle can cause the injector to leak fuel, causing a rich condition and double-digit negative fuel trim values. Chemically cleaning the injectors thru the injector rail may or may not solve these problems. Technicians would need to communicate this with the car owner with regard to trying this procedure. It’s the lowest cost potential fix compared to replacing the injectors.
Now let’s cover the diagnostic value of the relative compression test. Using a high inductive amp probe, we clamp it around a battery cable. With the throttle pedal to the floor and the engine in the clear flood, mode let’s take a look at what the good cranking amps waveform is telling us in Fig. 19. Notice the uniform starter current draw as each cylinder reaches TDC on its compression stroke on this good engine.
Now let’s share some case studies with you. The first example is from a tech call to another shop. The high mileage 3.4L engine had a dead miss on cylinder #4 with a MIL. The car owner had replaced his spark plugs and secondary leads. I decided to do a compression test on #4 cylinder. See Fig. 20. Do you notice that the compression values are low? But look closer... Did you notice that the firing event occurred during the exhaust stroke from a crossed secondary lead? The cylinder’s compression was low because the injector was still pulsing fuel into the cylinder, washing down the cylinder walls, causing low compression readings. I reversed the #1 and the #4 plug wires to their proper spark plugs.
Another case study involves a misfire on cylinder #3. A cranking amps test was performed. Note the waveform in Fig. 21 which indicates a low starter current draw for the #3 cylinder on its compression stroke.
The next example is from a V-6 engine with misfire codes from all cylinders on bank 1. Notice in Fig. 22 that the starter current drawn from the cylinders on bank 1 are considerably lower than those on bank 2. The firing order on this engine is 1 2 3 4 5 6. This could be caused by a retarded cam timing problem on bank 1 or a restricted converter on bank 1. We can verify an exhaust restriction by looking at a compression wave form. Look at a compression wave form captured at idle in Fig. 23. Notice the rising pressure during the exhaust stroke. Now let’s do a WOT snap compression test on cylinder #3. In Fig. 24 note the abrupt rise in pressure right after the exhaust valve opening caused by a restricted converter.
Thank you for your commitment to this industry.
This industry is better because of your commitment.