[gnoon]

I have a 98 Plymouth Grand Voyager "Expresso". I've been having issues, I replaced the BCM and the radiator quit working. I replaced the relay for that, no good so I direct wired it off the battery to a switch and everything seemed good for awhile. The other day I was idling down the street and it just died, no warning just died. It would turn over but not start. It sat for about 12 hours and started right up, ran for a bit than died again. Sat for another 12 hours and started right up again.

I was wondering about fuel so I put a pressure gauge on the fuel rail and with the engine idling the gauge rapidly oscillates between 50-65 pounds. I let it run for about a half hour, shut it off and it won't start again. I put a plug tester inline in the plug wire and it shows no spark.

Just wondering if anyone has come across this before or has any thoughts on why it will start, run and than lose spark and quit running after a period of time.

 

[TheDevilIKnow]

Unscrew the HVAC module in the center of the dash and have a look at the Connector to see if it is burned or corroded. Occasionally the data bus gets shorted and interferes with the PCM. Are you aware of the instrument cluster connector fault that can manifest any number of symptoms? This is usually an electrical problem when the vehicle stops dead. It could be a fault with the PCM as it warms up. It's a good idea to knock down all the known quirky faults that can be show stoppers. Are there any codes that come up on the On Board Diagnostic. The fact the engine starts then dies is a bit of a hint. Let us know what you find! Good Luck

 

[1994 Sport]

Welcome to the site gnoon.

Cam or crank sensors that short or open circuit when warm can go back to working again when they cool off.

Like "TheDevilIknow" asked; are you setting any codes?

 

[TheDevilIKnow]

I was thinking that you can probably save yourself a lot of grief and get a proper scan diagnostic because this very fault has happened to many. Sometimes it's bad news with the PCM coding a nasty fault and needs replacement other times the Vehicle Security System regulated by the Body Control module has a fault or is receiving no data and won't allow the PCM to operate. Just a suggestion. You can eliminate the HVAC problem with a quick look see in case there is a problem brewing.

 

[gnoon]

Thanks for all your input. I've got a busy week and won't be able to get to the van for awhile but I'll post and let you know what I find.

 

[TheDevilIKnow]

1994 Sport has a good point in his post. The Crank and Camshaft sensor need to be in good shape for the PCM to operate properly as these sensors are integral to the operation of the injector triggers. I've noted that some PCM vendors recommend the replacement of the cam and crank sensors if you're replacing the PCM because there have been many folks replace the PCM only to find a faulty crank or cam sensor. This procedure would help vendors limit their returns as you wouldn't be certain of the fault should the vehicle return to operation.

 

[Mr. Fixit]

Locate the crank sensor. normally on the drivers side in the bell housing on the transmission.
get a transmission funnel the one with the long nose and a jug of water. when this quits running take the funnel and pour cold water on the sensor and see if it will start. if it does you have found the problem, changing that thing is another joy.

Hope this helps.

 

[BillGrissom]

It would help if we know what engine you have.

 

[alloriginal]

It a Grand so its NOT going to be a 4 cylinder. An "Expresso" is going to be a 3.3/3.8L.

Besides the diagnostics is nearly identical for all of the V6's 3.0L included weather it is fuel, spark etc.....

 

[BillGrissom]

It a Grand so its NOT going to be a 4 cylinder. An "Expresso" is going to be a 3.3/3.8L. ....

Good point. I missed the "Grand" and never heard of "Expresso".
I agree w/ the posts on bad crank sensor (or wiring) or similarly a flexplate issue (crank sensor "toner ring" is welded on). A bad crank signal will not normally flag a code because the PCM relies on that to know the engine is running (and thus spark at correct times). I don't think a distributor-less ignition with "wasted spark" (coil pack w/ 3 coils in V-6) like our vans (2.4L, 3.3L, 3.8L) needs the cam signal to spark since it fires on every TDC stroke (needed or not). I have heard mixed stories on that. I think the purpose of the cam sensor is for sequential (timed) fuel injection. However, both share the same 8 V supply for their Hall-effect sensor, so either can draw down the other. When I had a similar problem in my 2002 3.8L, going thru the whole wire harness (pulled it down below car on ramps and opened up) fixed it, but I don't know exactly why. It did appear that the 8 V supply wire also went to a box behind the left inner fender (ABS?) and I wiggled those connectors and harness. Several years ago and hasn't missed a beat since, other than a bad coil pack.

 

[1994 Sport]

Up thur 2003 the 3.3/3.8 engines in the vans need the cam and the crank sensors both operating for the engine to start and run. Lose either one and the engine stalls.
Starting in 2004 with the NGC (PCM) the engine will start and run with just one or the other.

The Chrysler 4 cylinder engines for the most part will run with just the crank sensor, even the early ones.

 

[Mr. Fixit]

The crank sensor is used to tell the ecm the position of #1 cylinder. once the ecm locks on and fires the # 1 cylinder the crank sensor is no longer needed because it starts a sequence of firing order and this is where the cam sensor comes into play. the crank sensor says when to fire and the cam sensor says what to fire. if you were to unplug the crank sensor once the engine is running the ecm wouldn't know it because it doesn't look at it again till you start the engine again. i started working on mopar back in the early 80's and have gone to school for training on these vehicles. i am not here to argue or debate i am here to try and help and repairing something over the computer is a challenge at best. as far as the service engine soon light is concerned if it comes on the ecm has detected a fault. i don't care if people say it runs perfectly there is a problem period. an engine that misfires puts unburned gas in the exhaust and this is bad for the catalytic converter. obd 1 didn't have the ability to detect misfires at highway speeds and as a result raised the level of bad gasses that comes out. obd II on the other hand is one big tattle tail it reports the slightest fault's. I have repaired electronics sense i was 11 years old and have a working knowledge of how things work and i can tell you this electricity will follow the path of least resistance and bad grounds are the number one problem in cars today and with more and more sophisticated electronics grounds are going to get worse.

Hope this helps somebody.

 

[1994 Sport]

Locate the crank sensor. normally on the drivers side in the bell housing on the transmission.
get a transmission funnel the one with the long nose and a jug of water. when this quits running take the funnel and pour cold water on the sensor and see if it will start. if it does you have found the problem, changing that thing is another joy.

Hope this helps.

Didn't you just contradicted yourself?


Unplug the crank or the cam sensor on any 3.3/3.8 Chrysler V6 1990 to 2003 with the engine running and you will see for yourself what happens.

Hands on is the best training. My hands have been on a lot of these vans.

PS: I passed on a job for Chrysler to be a trainer at their learning center. :thumb:

 

[TheDevilIKnow]

I'm hoping Original poster "gnoon" gets back on the forum to let us know of any progress. On a '98 Expresso" I would feel a lot more confident if he had told us that he had resoldered the instrument cluster connector. As these vehicles age I pretty much think that the instrument cluster solder job is a prerequisite for our arm chair diagnostics. The connector ground pin that usually fractures on the instrument cluster circuit board often looks like a cold solder joint. This makes sense because of the size of the circuit board grounding trace and would likely have required a longer heat application than other pin connections to make a proper solder joint connection.
We have a vehicle with random intermittent ignition failure. No codes to look at. (Not that they aren't there) The vehicle cools then starts again. The Crank sensor would be a good start as suggested. Does any one want to comment on the failure modes of Crankshaft sensors?

 

[BillGrissom]

... the crank sensor says when to fire and the cam sensor says what to fire. if you were to unplug the crank sensor once the engine is running the ecm wouldn't know it because it doesn't look at it again till you start the engine again.

I don't quite understand this. I can't totally dispute it, but can ask a simple question. If our V-6 "distributor-less" engines have only 3 coils and 2 spark wires run off of each coil, doesn't the coil have to fire every time that pair of cylinders is at TDC? One is on its compression stroke and needs the spark and the other is on its exhaust stroke and doesn't (hence term "wasted spark"). There are only 4 wires to the coil pack - 3 coils & a common. Thus, why is the cam sensor needed for the spark ignition timing? For those neophytes about engines, the camshaft turns at half the crank speed and can tell you which of the two cylinders is on its compression stroke. I don't dispute that unplugging the cam sensor will stop the spark, but doesn't appear the PCM would need to behave that way. Were the designers mean, or was this a way to purposely disable the engine to protect something from damage?

 

[Mr. Fixit]

I don't quite understand this. I can't totally dispute it, but can ask a simple question. If our V-6 "distributor-less" engines have only 3 coils and 2 spark wires run off of each coil, doesn't the coil have to fire every time that pair of cylinders is at TDC? One is on its compression stroke and needs the spark and the other is on its exhaust stroke and doesn't (hence term "wasted spark"). There are only 4 wires to the coil pack - 3 coils & a common. Thus, why is the cam sensor needed for the spark ignition timing? For those neophytes about engines, the camshaft turns at half the crank speed and can tell you which of the two cylinders is on its compression stroke. I don't dispute that unplugging the cam sensor will stop the spark, but doesn't appear the PCM would need to behave that way. Were the designers mean, or was this a way to purposely disable the engine to protect something from damage?

Years back GM had a problem with their cars just shutting down where ever they were. you could be running down the highway and bang it stops. this is because GM uses the crank sensor all the time and when it went bad the ecm had no clue where #1 cylinder was. ford v8 engines with a distributor for spark distribution only has a damper pulley with 8 fingers on the back of it. 1 of the fingers is shorter then the other 7 and the crank sensor has an 8 thousandth gap looks at the damper and when the short finger passes the c/s it's signal is different form the other 7 and fires #1 cyl. for every pulse of spark you get a pulse of fuel. you can unplug this sensor and ford will continue to run. same with mopar. this is why you will NEVER get a crank sensor code because if the sensor is bad the ecm has no idea the engine is cranking. if a mopar c/s goes bad when it gets hot it simply WON'T START. this is when you put water over it and if it starts you have found the problem. these rinky dink code readers will not tell you much except the trouble codes. now if you had a scan tool i used a snap on scanner you could hook it up and look at the crank rpm and if it is not present you have a bad crank sensor. the cam sensor on the other hand can cause rough running because it says what cyl to fire and as long as it puts out a signal the ecm is happy and will not set a code. should be a square wave signal and if it is a saw tooth signal it can fire the cyl early or late and cause all kinda of trouble here again you need a lab scope the see the signal and what it is doing. the cam/crank sensors consist of 20,000 turns of very fine wire and a magnet. this wire can have shorted turns and produce a bad signal and make for a ratty running engine.

Hope this helps.

 

[1994 Sport]

Stating that you will NEVER set a crank sensor code on a Mopar if the signal is lost is pretty bold statement.:eekkkk:

If the PCM sees a cam signal pattern without seeing a crank signal pattern for 2 seconds, bingo Po320 sets.

I have seen this code. I'm I just lucky? I think not.

http://www.dtcsearch.com/P0320/Chrysler/

Others have seen this code too.

https://www.google.com/?gws_rd=ssl#q=p0320+chrysler+minivan

 

[Mr. Fixit]

Stating that you will NEVER set a crank sensor code on a Mopar if the signal is lost is pretty bold statement.:eekkkk:

If the PCM sees a cam signal pattern without seeing a crank signal pattern for 2 seconds, bingo Po320 sets.

I have seen this code. I'm I just lucky? I think not.

http://www.dtcsearch.com/P0320/Chrysler/

Others have seen this code too.

https://www.google.com/?gws_rd=ssl#q=p0320+chrysler+minivan

The ecm gets the tach signal from the crank sensor. if this part fails it will set a code! It's called a ghost code sort of like the ( 1684 ) code you get saying your battery has been disconnected with in the last 50 key on. you know that's bull because you have had your van longer then that.

Hope this helps.

 

[BillGrissom]

... the cam sensor on the other hand ... says what cyl to fire ...

I agree w/ most of what you explain, except that the cam sensor signal can control spark timing in a "wasted spark" system like ours. Each cylinder must spark on each turn of the crankshaft, because 2 cylinders share a coil. If using just the camshaft for this, you would need 6 "fire" signals per camshaft rev. The camshaft also isn't as accurate because of stretch of the timing chain (or belt in 4 cyl). I think the cam sensor is used to time the fuel injection to the valve openings ("sequential"). I guess one could disconnect the crank sensor while the engine is running and see if it dies, but pretty hard to reach it on a V-6 and easy to burn your hand on the exhaust.

 

[jee34]

I also agree that crank sensor signal is more precise. They also use the cam sensor to detect bad timing (too stretched timing chain) or a bad timing belt.Also,the crank sensor tells computer each tdc,and cam sensor tells computer valve timing,used for injection timing,not ignition timing.