Fig3  View shows back probing the TPS signal wire (OR/DB) at the TPS connector. 
 
 
 
 
 
 
 
 

 
Fig7  View of datastream on the Snap-On Scanner while TPS is displaying 4.98 Volts. 
 
 
 
 
 
 

 
Fig4  View of the Snap-On Vantage in graphing mode while back probing the TPS signal wire (OR/DB). Notice the 60 Volt spikes. 
 
 
 
 
 
 
 
 
 
 

 
Fig6  View of the Snap-On Vantage displaying battery voltage and 60 volt spikes on min/max while back probing the TPS signal wire. 
 
 

 
Fig5 View of Snap-On Vantage displaying injector wave form while probing the exposed area of the Injector #1 control wire (WT/DB). 
 
 
 
 

 
Fig. 1  View of TPS signal wire (OR/DB) and Injector #1 control wire (WT/DB) both chafed in the same area. These wires were intermittently touching one another in the wiring harness. 

SYMPTOMS 
    Recently a regular customer dropped off  her 92 Lebaron ragtop with an intermittent problem. The customer reported that sometimes, while driving, she would notice that the engine would seem to stumble and run poorly and sometimes the MIL light would illuminate. The customer also stated that engine temperature did not seem to matter and the problem would come and go, and last for varying time intervals. Our initial road test revealed a well running car with no problems. "Typical darn intermittent" we thought. 

 VEHICLE OVERVIEW 
   Our subject vehicle is equipped with the popular drivetrain package of 3.0 Liter, OHC engine and the Ultradrive A-604 transaxle. Engine functions are controlled by the Powertrain Control module or PCM. An optical ignition pickup provides two reference signals, one for cylinder ID and one for crankshaft position, for fuel and spark delivery. A manifold absolute pressure (MAP) sensor is used to calculate engine load instead of any type of air flow meter. The PCM also controls the auto shutdown and fuel pump relays to power the fuel pump, ignition coil, injectors, and oxygen sensor heater. Besides controlling engine functions the PCM monitors and controls charging system voltage. To aid in diagnoses the PCM also has a reset counter that keeps track of how many ignition cycles have occurred since the last fault code was set or battery disconnect. 
 
   Transmission functions are controlled by the Electronic Automatic Transaxle Controller or EATX. The four speed transaxle is a fully adaptive, electronically controlled transmission that uses feedback sensors to adjust functions and monitor performance. With a good scan tool the EATX can supply a wealth of diagnostic information on transaxle condition, history and current sensor values. These computers and others are connected together by a serial data network called the Chrysler Collision Detection System or CCD bus. Each computer module uses this network to communicate and exchange data with other modules in the vehicle. 

THE DIAGNOSTIC 
    Studying the Technical Service Bulletin titles took about ten minutes but did not reveal any similar symptoms. We called the customer to obtain authorization for a "Level 1 Diagnostic". In short order, a Level 1 Diagnostic at our shop is a package consisting of labor and various tool fees,  normally consisting of: 1) Visual Underhood Inspection (VUI). 2) two hours labor. 3) Scan Tool connect and code interpretation fee. 4) Datastream Analysis. 5) Information Access Fee (TSB & procedures search) and, if used, 6) software fee for the Snap-On Vantage Graphing Multimeter. After obtaining authorization we began our visual underhood inspection. A visual inspection of all underhood equipment did not reveal anything out of the ordinary. It was time to connect the scan tool. We found two codes present in memory. Code 21- O2 Sensor not switching and Code 24-Throttle Position Voltage too high. With the car running hot, the O2 sensor seemed to switch normally from 200 to 800 millivolts. The TPS at idle was. .96V and ok. Studying the rest of the datastream values did not reveal any abnormalities. We decided to clear codes and road test again. Once again the car ran very well. So well that on the test drive I wondered how much time technicians all over the country spend test driving cars in hopes that the car will become symptomatic. 

RESULT TIME 
   In lieu of the fact that the car was running so well and that no codes had returned, we decided to park the car for the day. The next day we started the car again. This time the engine seemed to be stumbling at first, then it would clear up and go to a higher than normal idle. Often Chrysler vehicles will be symptomatic only for a moment when started if  there is a hard fault, because the PCM quickly supplies a substitute value to replace the failed sensors value. A view of the datastream now revealed some interesting information. Code 24 for high TPS was set again and the TPS value in the data stream showed 5.00 V (fig. 7). While the car was symptomatic, we checked the 3 TPS wires with the Vantage in the voltmeter mode. A check of the TPS feed wire (VT/WT) revealed a normal 5.00 V reference. Testing the TPS ground wire (BK/LB) revealed a good ground reading at 0.02 V. However, when testing the TPS signal wire (OR/DB) we found 13.5 V (fig. 3). Now this did not agree with the current 5 V reading on the scan tool. We believe that the scan tool software or the vehicles computer can not display any voltage over 5 V to the scan tool. This is probably limited since there should never be more than 5 V on this circuit normally. 

TWO THEORIES 
    Anyway, the voltmeter showed 13.5 V so we knew this was the true reading. But how can 13.5 V be present on this wire? Even if the TPS is shorted internally the maximum voltage available on this circuit is only 5.00 V. Vantage was also capturing min/max voltage values on this circuit. We noticed voltage was maxing at 60.00 V (fig. 6 & 4). How could this be? We turned the engine off and left the key on. With KOEO we noticed that the problem was gone and TPS came back to the normal .96 V. After some pondering and discussion, we had developed two theories that we wanted to test. Theory 1) The TPS signal wire is touching a wire that carries battery voltage with the engine running. (Although this would not explain the 60.00 V max readings.) Theory 2) The vehicles onboard computer has been compromised internally in which the computer, in error, is sending voltage out the TPS signal wire. This would have to be some internal short or catastrophic failure since the computer should only be sampling voltage at this pin and not supplying voltage there. 

    To test the first theory we decided to let the car run again while tugging on the various  underhood wiring looms. While tugging on the main computer harness near the battery, we noticed the TPS reading would drop to normal and the car would idle down (fig 10 & 11). More tugging created the high idle and high voltage readings again. We were somewhat confident that the TPS signal wire was touching another wire somewhere in the cars intestines of wiring. Now came the moment of truth. Should we run a new TPS signal wire between the TPS and the computer and be done with it, or should we just go a little further? Well, to quote Mitch Schneider, "The deeper you go the more you know". 

WIRING INSPECTION 
    Since the wiring was a fairly short run from the TPS to the computer in the L/F fender, we decided to open the wiring loom. Beginning at the TPS connector we pulled out the TPS signal wire (OR/DB) about twenty inches from the connector. Not far from a junction where two wiring harnesses meet, we found our gremlin. With the wiring loom open and the TPS signal wire out for careful feeling and visual inspection we found a small section of wire exposed (fig. 1, 2,8). The wire's insulation appeared to have been crushed at one time. Only one other wire exhibited the same crushed-exposed symptoms. That wire's color was  (WT/DB). With the car running, this wire showed the 13.5 Volts and the 60 Volt spikes. When testing this wire with the Vantage in the waveform mode, look at what we saw (fig. 5). Look familiar? After studying the wiring schematic we found this (WT/DB) wire to be the Injector # 1 control wire. Since this wire was touching the TPS signal wire, this explained the 60 volt spikes we were seeing. 

    After repairing both wires and taping up the wiring harness, we cleared the TPS code and road tested the car one final time. 

    As to the O2 sensor code, it never reset and the O2 sensor activity appeared normal so we advised the customer that there was no evidence a repair was necessary at this time for the O2 sensor. Most likely the temporary loss of fuel mixture control caused this code to set. 
 

    One of the nice things about this business is that it is never dull. You may see the same type of vehicle or symptom, but that doesn't mean it will be the same old problem. Who would of thought we would see an injector waveform riding on a TPS circuit.