Chrysler Repair: 96 Town & Country Dies, Stumbles, gas vapors, fuel pump relay


Question
Hi,
I have a 1996 Town and Country with 208,000 miles and live at 5,000-foot elevation. It intermittently dies, and intermittently "stumbles" where revs drop and loses power for a few seconds. When it dies, the engine simply stops smoothly without warning or bucking. The engine has stopped after slowing to a stop. The engine has stopped at various speeds. The engine always immediately and easily restarts in neutral or park.

Recently the car failed to start when at an automotive shop, failed to produce spark and the solution was to replace the crank sensor, the cam sensor and the PCM.

Before and after the above service I was/am able to read a code of 31 using the manual method. It initiates the 31 code as soon as the engine is turned on, always. The Check Engine Light does not illuminate with this code. The engine mostly stalls or stumbles without issuing a Check Engine Light, ODB-II code--but always has a 31 code. Sometimes 31 31

Now, a friend's ODB II code reader does not show a code when 31 can be read by the manual method. I replaced the EVAP solenoid and it did not fix the problem. The wires from the EVAL solenoid to and from the PCM, test in good shape with correct voltages and correct ohm readings. The EVAP solenoid works immediately upon turning on the engine with it's pulsating vacuum.

The vacuum lines appear appear good and check out as unblocked and uncracked. Lines to the manifold/plenum appear to be in good shape. The canister appears in good shape with all gas vapors contained.

The car appears to have about 50 psi of fuel pressure when tested (parked.)

Occasionally after a stall, I get a code of 61 and 0106 (ODB) indicating MAP sensor problem. The wires look good, voltages to the MAP sensor are within spec. MAP sensor appears to function when tested. The engine continues to run when the MAP sensor is removed at idle.

I swapped the ASD relay with the fuel pump relay without change.

I tested the fuel pump relay and found during a stall event, the engine began its stall...then the fuel pump shut off...as it died. It appeared that the fuel pump was receiving a signal during the stall event.

Any tips, suggestions or next steps? Your insight would really be appreciated.

Thank you,

Joel  

Answer
Hi Joel,
I appreciate the detailed history of the problem. I have a couple of thoughts: the egr valve may be sticky in its action or the MAP sensor may be inaccurate. Both of these possibilies would impact the mixture and thus explain the die off. I assume you have the 3.3 or 3.8L engine. Interestingly I don't find either a 61 or 0106 code in the powertrain troubleshoot manual for the that year, but the 0107 and 0108 do pertain to the MAP. And the egr valve can misbehave and yet not set a fault code.
The egr valve is located near the throttle body air intake at the end of the engine, mounted in pipe that recirculates exhaust gas from the rear cylider bank exhaust pipe back around to the intake manifold. The exhaust gases have some fumes that can plate out a crud on the valve stem and thus keep it from closing tight when you are at idle. That makes for a too lean mixture so the engine stalls. The valve proper is mounted horizontally with the stem visible in a space between the body of the valve mounted on the pipe and the round top of the valve which is flanged and so if you look carefully you will see a metal rod (stem of the valve) with a slot around its circumference. You can take the tip of flat blade screwdriver and insert it in the slot and then lever the valve back and forth to check if it is moving freely (against spring action in one direction) or not. If it doesn't seem to close easily with the help of the built-in spring, then I would spray the base of the stem with solvent from a pressure can (such as WD-40 or carb cleaner) while moving the stem back and forth.  Then see if that solves the issue.
I have experienced problems with idle/start when a MAP sensor started to go inaccurate without actually failing to the degree it would set a code. So given that hint of a code, and the absence of anything else in your thorough efforts so far, it might be a good choice for a part investment at this point. And it is not surprising that it will run when the MAP is disconnected as the computer will substitute a 'guess' when that happens, based upon other engine sensor/operating parameters at the time.
On the LDP issue, you might try rechecking the connections while shaking the harness in case of an intermittent. There are 8 pages of fault/event tree approach to that code that I can xerox copy and mail to you if you would like them.  
Other than that, I can't give you any other ideas. Please let me know how this works out for you.
Roland