Chrysler Repair: Dodge Van Transmission or Engine Surge, dodge grand caravan, heli coil inserts


Question
QUESTION: I posted this to the other expert on this site and received a “too many variables to diagnose” response so thought I would solicit your thoughts.  I have included the original post and also some follow up observations after more days of driving.

Vehicle:  2002 Dodge Grand Caravan, 3.8L motor, 4-spd transmission, about 165K miles.  First some brief history.  A couple of months ago the transmission went into limp mode while my wife was driving it home.  It threw OBDII code 700 (transmission control system malfunction) and code 750 (Shift solenoid A malfunction).  I did some research but due to uncertainty I decided to have a local transmission shop do a driving diagnosis.  They concurred with the faulty solenoid pack so I had them install a new one along with a fluid/filter change.  Problem solved for about a month.  Then last week (again while my wife was driving!) the van began to shudder (slip?) while driving at highway speeds.  After additional extensive research I came to the conclusion that the wrong fluid was likely the cause.  A call to the shop confirmed my suspicions as they admitted to using a generic all purpose fluid and not specific ATF+4.  Half a Saturday and about $120 later, I installed a new filter, several heli-coil inserts where the shop had stripped out pan bolts, and I did a system flush with 11 quarts through the transmission cooler lines (system holds 9.x quarts).  However, the problem does not seem entirely fixed.  After a good 10-15 minute warm up, while driving at 40+ MPH, tach readings between 1200-1800rpm the vehicle will start to shudder (slip) with visible tach fluctuations as the engine appears to attempt to compensate for the problem in the transmission.  I think the condition primarily occurs during the torque converter lockup stages of operation (4th gear?).  The problem appears most prevalent while on level or uphill sections of road – under more load I suspect.  I pulled the battery cable for a few hours to attempt a computer reset, but am uncertain if the computer will go into learn/retraining mode doing this on 2002 vehicles.  After driving the vehicle roughly 50+ miles today (some city, some highway) things seemed to improve, but only during the last 20 or so highway miles.  I have had my scanner connected all day and no codes have been thrown.  My questions:  In your opinion, can incorrect fluid cause a slip or shudder condition?  Can I reset the computer, or does it require a special tool/scanner?  How long does it take to “retrain” itself after a reset?  What are the chances the input or output speed sensors are now fouled?  Thank you in advance for any light you can shed on the subject.

Follow up:
Some follow up and additional Q’s after several more days of driving the van, some checks and fixes, and additional research on this and other sites.  Checked and gapped all plugs and checked plug wires.  Do not have specs for coil pack expected impedance for each of the six outputs, anyone have these numbers?  Pulled throttle body and thoroughly cleaned.  Checked IAC, MAP, PCV valve and TPS while I was at it.  Tried to run down an external EGR valve but apparently it does not exist on early model year 02’ vans.  Cover plates are present where the system should bolt on, but it is not there.  Replaced fuel filter – engineers should be shot for the catastrophe of a location.  The surge/slip/shudder condition still infrequently occurs and a notable bump up/down appears prevalent during torque converter lock/unlock.  Also, another condition now manifests while going downhill - motor (tach) has a notable 600+RMP drop and then almost instant recovery.  Happens both in and out of cruise.  My uncertainty still revolves around rather this is a computer/motor issue or the torque converter ready to shell itself.  When the RPM’s start to surge it almost appears that there are conflicting sensor inputs being sent to the PCM and it develops a sine wave cycle attempting to react to the conflicting inputs.  This is why I continue to have reservations about rather it is the transmission or all the computer hooey that runs the vehicle.  I have read several similar posts with the primary two outcomes as either torque converter or computer issues/upgrade/replacement.  Now the follow on Q’s.  How long (and how many fluid changes) should it take for the computer to re-learn itself with the correct ATF+4 fluid?  With no DTC’s showing up can a shop emphatically diagnose a bad torque converter?  Anything else I am missing and should check?  Thanks again for any thoughts/advice.

Thank you in advance for any insight you might have.

ANSWER: Hi David,
I don't know specifically the length of time/miles for the tcm to relearn. The lack of any codes, and that you have a history of being able to read codes, makes me wonder if this a mechanical failure of the torque converter clutch but which is not able to be recognized under an existing fault code. I don't know of anything that you are missing out on.
One idea I can think of is to unplug the torque converter solenoid plug, which would take the converter out of the picture and see how the vehicle drives under those circumstances. You would find the plug  on the side of the trans that faces the front of the vehicle about mid-level, and it has a three-pin black plug that is located forward of the shift rod where you might expect the converter to be positioned. It has only two wires however: pink/gray and dark blue/white. Let me know whether the behavior remains the same, or disappears. That may help you to conclude whether the issue is the engine or the torque converter.
Roland

---------- FOLLOW-UP ----------

QUESTION: Thanks for the advice.  I think the connector you reference is on earlier or later model transmissions.  I have the complete solenoid pack that I speculate controls the torque converter also (large connector, many wires), an input turbine speed sensor plug below the transmission cooler lines, an output speed sensor to the right of this connector, a neutral start/backup light switch connector below, and I think a PRNDL switch connector.  Any other thoughts on how to test the torque converter?  Again, thanks for your time and advice.

Answer
Hi David,
My error...I overlooked that the tcc solenoid was shown on the 3-speed automatic page. How about removing the transmission control relay which will put you in 2nd gear only to see at least if there are any symptoms normally present are gone? The torque converter is locked out in that situation. Unfortunately it won't let you go above 40-50 mph at a high rpm but the motor's involvement in the situation would at least be observable. It would also cause a fault code to be set so you can see if the tcm notices.
Roland
PS Please 'rate' my answer, and where you see the question about "volunteer of the month" consider a 'yes' response. Thank you