This diagnosis begins like many others have in the past. The local guy from the used car lot drags in a "great deal" he picked up at the auction. You know the vehicles — clean, southern, low mileage, and (oh yeah!) it doesn't run! This time it happened to be a 2008 Jeep Liberty 3.7 brought in on the trailer. It would crank and attempt to start, smelt heavily of gas when it did run, accompanied with a heck of a lot of mechanical noise from under the hood. Prior to dropping it off the throttle body was replaced as well as the intake manifold gaskets and spark plugs. However, now that it is at our shop it is time to see how much of a "great deal" this vehicle really was.
 |
| Figure 1 |
Now that the vehicle is in the bay it is time to get to work. Where to begin? I initially scanned the vehicle for codes to see if there was anything that might help give some direction. There were approximately 12 codes stored in the ECM, but at first glance none of them seemed to be of any value for the current situation. (Figure 1) It was obvious when I pulled the vehicle in there was a hard misfire on one or more cylinders, along with a fair amount of popping and mechanical engine noise, so I made an attempt to use scan data to see what cylinder(s) were having a hard time but at this point I was not able to get the vehicle to run long enough to gather any useful data. Seeing that gathering scan data was not going to be an option it was time to move on.
 |
| Figure 2 |
The next step in my diagnostic routine was to set up my Pico scope and run a relative compression test on the engine to gather a "general health check" of the engine. With the heavy amount of mechanical noise, you could hear when the engine did run, I figure this would be the best next step. Often times the relative compression test can give us an insight to the cylinders’ ability to seal and show us how they are in comparison to each other. I am sure if you have used this approach you are familiar with the classic waveform and the benefits it can hold. After disabling the fuel pump and unplugging the ignition coils, I clamped my high amp clamp around the battery cable and set up a trigger on the #1 ignition coil (Figure 2) Now we are ready to crank the engine!
 |
| Figure 3 |
This is where we gather our first bit of usable data and direction. Have a look at the waveform we gathered and tell me what you see. (Figure 3) The blue trace is our starter current and the red trace is our #1 ignition coil trigger. Quickly we can see that the ignition timing seems to be about right occurring just before TDC. Ok we can put that in our memory bank. Let's focus on the starter current now. At first glance, on the starter current waveform, we might think there are two "good" cylinders (the higher humps) and perhaps three other cylinders that are not contributing as much (the lower humps.)
Five FAQs Answered Only Here | |
 | With the answers to these five questions, you will build better estimates for your repair shop. See how in less than 10 minutes. |
 | |
But before we draw any types of conclusions based on this let's have a close look at the amperage scale (Figure 4). We can see that the amperage draw of the higher humps is near 300 amps! Experience will kick in at this point and tell us that "normal" starter current draw is around 150 to 200 amps on these engines. After observing this the big question becomes, how can the starter draw too much current? Can a cylinder have too much compression?
 |
| Figure 4 |
 |
| Figure 5 |
After some pondering and knowing that these 3.7's along with the 4.7 liters are notorious for the cam followers (rocker arms) falling off, can we make a hypothesis based on this data as to what might be going on in this engine? After all we are trying to gather as much data as we can before we open it up. Knowing the firing order 1,6,5,4,3,2 and knowing the companion cylinders 1 and 4, 6 and 3 and 5 and 2, could this help us (Figure 5)?
After a few moments, a cup of coffee and some pondering I came up with the conclusion that if #4 and #6 exhaust valves were not opening then it would appear on the relative compression test that the #1 and #3 were doing extra work on their compression stroke. Make sense? So let's say the #1 cylinder is traveling up on its compression stroke and starting to draw current from the starter. At this same time the #4 cylinder is on its way up on the exhaust stroke. TECHNICALLY this should not affect the starter current because the exhaust valve is open. But what if it wasn't? In that case it would be building compression in #1 AND #4 drawing near twice the current as normal from the starter. Same goes for cylinders 6 and 3.
 |
| Figure 6 |
Now that we have a pretty darn good idea what could be happening and we know what two cylinders are the potential culprits, can we prove this 100% before pulling out the wrench set? The easiest way I figured would be to use the Pico and an in-cylinder pressure test to prove or disprove our theory. It should be pretty easy to identify a nonfunctioning exhaust valve with the pressure test. I started on cylinder #4 and there it was (Figure 6)! 180psi compression and 150psi compression on the exhaust stroke of the engine! It did not take long to move the transducer down one hole and see the same exact waveform on the #6 cylinder. This made the diagnosis 100% complete. The #4 and #6 exhaust valves were not opening. On a side note, what would you have seen if you were using a regular compression gauge? 180psi? Would you perhaps have moved on at that point, assuming the cylinder was good? Always keep in mind that the standard compression test only shows the cylinder’s ability to seal, not breath. This is where a pressure transducer and a scope can really shine over a compression gauge.
 |
| Figure 7 |
Data has been gathered, pondered and proved so now it is time to get dirty. Fortunately for us both 4 and 6 are on the same bank so we only had to pull one valve cover. As soon as the passenger valve cover was removed both exhaust rocker arms were found laying on top of the head (Figure 7). Ahh, the sweet feeling of victory! I discussed with the used car lot as to my thoughts on what causes this and what the corrective action should be and you may know how this ends. The rocker arms were snapped back on, valve cover reinstalled and it was off to the sale! The customer was happy that he really did get a "good deal."
If you take anything away from this, I would hope it would be that sometimes we need to slow down to go fast. Gather as much data as you can before tearing into an engine and take the time to look at the scales on your relative compression test and not just the overall picture because as we proved sometimes and engine can have too much compression!
|
|