by Richard McCuistian 

The concern was that the Jeep (the same one from the oil pressure story) was misfiring on a couple of cylinders, but the problem was intermittent.  Jeff (the tire store tech) and I didn’t discuss it as extensively as we could have, but in the end, Jeff swapped the crank sensor and the spark plugs and sent the Apache pilot on his way. 

The tire store guys I know are really good at brakes, suspension, vibrations, and front end work, and they do a lot of it, but they really don’t have the time or inclination to chase an intermittent driveability concern, so when the Jeep came wheeling back in, their suggestion was that he call me to see if I was willing to tackle the problem. 

 It sounded interesting, and at the very least I wanted to have a quick look under the hood, so I told the serviceman to bring the Cherokee by so I could try to duplicate and experience the problem for myself.  Being the proud owner of a solidly dependable 2001 Jeep Cherokee  myself, I was more than a little interested in this concern. I did explain to him that I might not have time to do more than just give it a cursory look.  

Dropping Companions 

This look-over I gave the Jeep could have been more in depth, but it wasn’t for a couple of reasons.  To begin with, my primary scan tool was at the repair center for the second time in 6 months to have the software reloaded. Thus the only thing I had on hand that would communicate with the Jeep wasn’t my tool of choice and had only limited functionality at best on that platform. 

Secondly, this is a Brakes, Steering & Suspension, and Transmission semester, and I wasn’t sure I wanted any of my students to take a lot of time fooling around with something that would put them behind on their NATEF training schedule of tasks.The pilot said the Jeep had run just fine during the 30 mile drive to the college and that he wasn’t sure we would be able to duplicate the concern when he got there. 

When he arrived, I took a quick look under the hood just to see what I could see.  One thing I pay particular attention to on 4.0L Jeeps is the cam sensor synchronizer.  If anybody has moved it, the previous marks of the anchor area are apparent, but sometimes the shaft will try to gall and turn the cam sensor synchronizer as well (In most cases, the shaft locks completely and destroys the driven gear and the engine won’t even start), and when that happens, you can generally see a short scratch on the anchor area that was made when the synchronizer movement occurred. 

 I saw none of that, so I closed the hood and we drove the Jeep a short distance before duplicating the concern.  At first the Jeep began bucking like it had a spark leaking past a plug boot, but the further we went the worse it ran..“All you have to do in order to correct this is switch the Jeep off and restart it,” he told me.I didn’t want to do that, because I was anxious to get a little more personal with what was going on out there, and so I carefully limped back to the shop and popped the hood.  The 4.0L had dropped 4 cylinders. 

The fact that it would even run on two cylinders was astounding.  I began disconnecting injectors to determine which cylinders were silent, and I found that the only ones popping were 3 and 4.  In the midst of my testing, the engine stalled, and when it was restarted, the problem had evaporated and another test drive failed to reproduce the anomaly, although I did manage to use a cheap generic scan tool to squeeze a coil primary circuit code (P0353) and a (P1391) ‘loss of cam or crank signal’ code out of it. 

This seemed to be heat-related, and that conjures up a host of possibilities.  If the cam sensor is dead or disconnected, the engine won’t start, but it will keep running if the cam signal is lost.My students were trickling in from their academic classes by this time and so I told the red Cherokee owner that if I was going to make a wild guess, I’d say the rail coil assembly was going sour, based on the fact that the dead cylinders were all companions (1/6, 2/5).   

I know you may be asking,

“What about the fact that the engine skip disappeared on restart?” 

The Jeep PCM (like so many others since the mid-‘90’s), kills injectors on misfiring cylinders in an attempt to preserve the catalyst and leaves them offline until the engine is restarted. Thus a malfunctioning coil rail could cause the injectors to drop out in companion pairs the way this 4.0L was doing.  

New Rail Coil, No Change     

I was tired of not having a serviceable scan tool, so I requisitioned and bought two OTC Nemisys scan tools that week.  It was a good thing I did.  The Nemisys is faster and more sensible than its big brother Genisys, although it basically uses the same software. There are no color graphics on the screen, but who needs them anyway?

 It comes with a neat software package for storing and playing back files saved in the tool’s memory.I bumped into the Apache pilot in the grocery store parking lot and got his report on the Jeep:“I drove it around quite a bit with the new coil rail on it and thought it was fixed, but it started sputtering again.”  I was becoming more and more interested in knowing what the problem was to the point that I decided that an auto mechanics instructor needs to tangle with one of these once in awhile to stay sharp and remain professionally developed. 

We filled out the proper paperwork at the college the following Wednesday and he left the Jeep with me.   I studied the scan tool cam and crank counters for awhile on the Nemisys screen to get a feel for what I was looking at, but saw no concern.  On Friday, I let it idle all morning to get it good and hot, then drove it to lunch; it has frequently seemed to me that a vehicle will be more prone to exhibit an intermittent concern when you aren’t driving the vehicle just for the purpose of finding it. I knew this guy wouldn’t mind where or when I drove his Jeep (within reason) so I drove it like it was my own.  Well, it ran great all the way to Hardees that day. 

I allowed the Jeep to hot soak while I ate my burger.  The Jeep ran great almost all the way back to the college before it started misfiring and stalled.  Knowing I had hit paydirt and that I had some free time (the students leave just before noon on Fridays), I spent a couple hours fiddling with it.  With the OTC Nemisys connected and peering into the OEM datastream, I noticed the cam sensor counter began to appear erratic and eventually flattened out completely while I was watching it.  Remembering the loss of cam sensor code, I wasn’t surprised. 

I had another cam sensor on hand, but installing the new sensor did nothing to produce anything less flat on the datastream. 

Connecting my Interro PDA handheld scope to the cam sensor signal wire (I had to split the tape at the PCM connector), I found a good square wave cam signal all the way to the PCM.  The signal was making it to the PCM but the PCM wasn’t sending counter information out to the scan tool. It was time for me to leave for the week, so I plugged the problem Jeep’s PCM into the harness on my personal vehicle along with the Nemisys (leaving my PCM mounted for easy reconnection), and about ten miles into my journey home, the Nemisys showed a flat cam sensor signal! 

This seemed to indicate a PCM concern, but for the entire weekend, that was the only time I saw a flat cam signal and my Jeep never ran bad or stalled.  I wasn’t at all satisfied. A new PCM for this Cherokee tips the cash scale at a whopping $610 (list), and I didn’t want to throw a part that expensive at the problem before I was sure.  Too many PCMs have been changed for no good reason (I’ve done it myself, in years past), and I didn’t want to go there.  

Digging Deeper 

By Monday afternoon, the Jeep had deteriorated to the point that the problem was almost always present, even with the engine cold. The engine would idle just fine but as soon as the throttle was opened, the PCM would drop injector pulse to first one pair of cylinders, then another, and then it would stall.  What was interesting is that it didn’t always throw the Cam/Crank code, but it virtually always threw ignition coil primary circuit codes, and I cleared them each time. 

What was I supposed to think? Aren’t those codes there to give us some direction?  If that wasn’t enough, I kept getting a transmission fault code, a development that muddied the water even more. 

Having known of transmission circuits being shorted (the same 9 volt reference voltage wire that feeds the cam and crank sensors goes to the VSS), I disconnected the transmission wire harness but to no avail. The problem was still there.  I quickly clipped the coil primary wires at the PCM and ran a wiring overlay to the rail coil and accomplished nothing. I inspected the crank sensor wiring but found nothing pinched, chafed, or lying against a hot pipe. 

Having used the Interro PDA (this is a $2000 handheld oscilloscope, not a Palm Pilot) to check the current ramping pattern on the coils (all three circuits can be passed through the probe for a nice parade pattern on the scope), I found that each coil was intermittently giving two current ramp (see photo below).  Was the Jeep actually supposed to have repetitive spark at idle?  Unknown, but I surmised that My thinking was that Crown Victorias and Escorts do!  It makes the engine idle smoother and cleaner, but the repetitive spark evaporates as soon as the throttle is cracked. 

 This was new information.  But the fact that the current ramps remained strong even as the engine was dropping cylinders indicated that the primary circuits were still intact; an open circuit provides no current ramp, although a shorted circuit might provide one, but this harness was short, easy to inspect, and pristine, so I discarded the primary circuit short idea for the time being.I decided to install a different type of ignition coil pack (I removed the coil rail, put a different coil primary connector on my overlay, and wired a 1998 Ford Explorer coil pack so the Jeep could start and run on the Explorer coil. I connected a set of secondary leads to the plugs.My students were astonished that I could make this happen, but it wasn't really that hard to do.

Now, with the Explorer coil pack installed I could access the secondary with my ignition o-scope.  The vehicle ran the same way, which eliminated the new coil rail he had purchased as a possible source of the problem, but now I was able to connect the ignition o-scope and watch the secondary pattern while the stall-stumble was under way.  The o-scope agreed with the Interro PDA; there were indeed two firing spikes per spark plug, but the second spike seemed to come and go, and when it did, the sound of the engine would change.  This was getting downright weird.  Did the Jeep have repetitive spark or didn’t it?   The book didn’t say so, but as far as I’ve been able to tell, the Ford shop manuals don’t mention it on those cars either. 

Breakthrough     

I buy equipment from Young’s Equipment Systems in Houston, Texas, and the phone rang while I was fighting with the scopes and scan tools.  It was Glen Young telling me that he was shipping some equipment I had ordered.  Glen also has a shop of his own, teaches classes on emissions, electronics, o-scope reading, whatnot, so he’s a lot more than just an equipment salesman.

     “Glen,” I broached, “let me run something by you.”  I described my trouble with the Jeep and his answer was simple.

     “When I reach the end of my thought processes on something like that, I check the cam sensor synchronization with the scope.”  Well, that was fine, but short of splitting the wire harness on my own Jeep and making a comparison (which I could do if necessary), I had no idea what to look for.

     “I haven’t seen anything on that on paper anywhere… What is it supposed to look like?”

Glen described what I should be looking for (see illustration at left), and when I went back to the Jeep and adjusted the cam synchronizer to match the pattern Glen described, every problem that engine had went away, including the bogus repetitive spark pattern.  I felt silly.Bringing the crankshaft around to zero degrees TDC compression, I removed the cam sensor and checked the hole in the vane where it was supposed to line up with the hole in the housing.

I’ve performed that operation dozens of times and never even thought about doing it on this Jeep because the cam synchronizer hadn’t been touched for 47,000 miles (two of my students had put cam bearings in this engine 47k miles ago), except by timing chain wear, which probably allowed the cam to lag just enough to tip the camshaft-driven cam synchronizer over the line and caused this problem.

Incidentally, there is a utility in the DRB III scan tool for synchronizing the sensor, but my experience has been that the toothpick method works best, believe it or not.  Normal timing chain stretch must have allowed the cam sensor to work itself slightly out of line, and the rest is history.  It was a perfect match.  I later told Glen that I had gone through fire and flood all week long on that doggone Jeep and the whole time I could have fixed it with a toothpick.

Never mind the fact that the scan tool readings and my own logic led me all the way around Robin Hood’s barn before I got back to where I should have started!  Never mind that the problem was initially extremely intermittent. All in all, it was a happy ending.                                                                                    R.W.M.