P0320 - Car Completely Immobile


Veteran Poster
Sep 22, 2023
Richmond, VA
Country flag
Alright...this happened to a lesser degree several months ago. Back then I jumped onto @Rodeo Joe 's thread about the camshaft sensor because the two issues seemed related.

I decided to start a new thread at this point because A. I don't want to confuse us all by mixing two separate topics and B. this is now a major problem.

Looking back:

A couple of months ago, I had P0320. Symptoms included tach needle erratically going to zero while engine was running, occasional very rough idle, and on a few occasions engine dying at a stop while in gear.
Back then I replaced the crankshaft sensor and camshaft sensor without any result. I then put my original sensors back in. I cleaned connectors to the best of my ability. I checked continuity. And I regapped all plugs.
Ultimately the issue disappeared.

For the last five weeks or so:

I've been noticing occasional rough idle immediately after starting the engine, which usually disappeared after a few minutes. No codes.

A few days ago:

Code P0320 was back. Symptoms got increasingly worse. The erratic tach needle is back, and the engine died a few times when coming to a stop.


Engine extremely difficult to start. Once it starts, it idles rough. Rpm fluctuating. When shifting into gear, it usually dies immediately. In Neutral, it may stay running or die.
I'm halfway down the driveway and at this point unable to move the car even a few yards.

So...please HELP!!!!!!!!

I'm totally lost honestly.

To reiterate, coil (Motorcraft), wires, and plugs are all less than two years old. There's no visible rodent activity, and the only other significant problem I know of is a small coolant leak which I believe to be a faulty lower intake manifold gasket.
Check the condition of the harmonic balancer. The tone ring for the crank sensor is mounted to the balancer. I've seen the outside half of the balancer separate from the inside half of it and cause an imbalance. That will cause shaking for sure.
Didn't the 3.8 still have a distributor in 1997? What camshaft sensor did you replace? Do you have a scanner? Usually if the PIP sensor in the distributor fails you won't have any spark and it just plain won't start. Have you ever replaced the TFI (ignition module)?
Well, my housemate and I pushed the car up the driveway. I started the engine twice to turn the steering, which worked in spite of the rough idle. Putting it into gear still caused the engine to die though.

Now, since I didn't have anything else to do, I took out the crankshaft sensor, put it back in, and...the car started and idled normally.


Maybe the sensor was loose?

Or could this just be because I had disconnected the battery, and the car forgot about the problem?

Looking at the sensor, there was a lot of whiteish/pinkish dirt of it. Possibly road salt?


My best guess at this point is some issue inside the wiring loom leading down there, because removing the sensor includes some wiggling, which may have inadvertently (but only temporarily) fixed my issue.
That or it was a bad ground. Clean the mating surfaces with a wire brush and re-install. See if that keeps it working.
I'd hook it up to a laptop and see what it says on a datalog. The fact it worked after you took it out and back in points to a wire that's corroded or broken. I'd see if I coud find a harness for it next time you're in the jy. Keep the tools with you; next time it gives you crap, do the rame replace/reinstall and see if it fixes it again. That is the sign of a bad wire getting moved.
Last edited:
I'd hook it up to a laptop and see what it says on a datalog.

I don't have the hardware and/or software to do that. All I have is the OBD Fusion app on my phone. Not saying I couldn't get it, but this is a budget operation...

The fact it worked after you took it out and back in points to a wire that's corroded or broken. I'd see if I coud find a harness for it next time you're in the jy. Keep the tools with you; next time it gives you crap, do the rame replace/reinstall and see if it fixes it again. That is the sign of a bad wire getting moved.

I do agree. And based on that theory, I can pretty much pinpoint the problem area of the harness because I only touched the portion around the crankshaft sensor, i.e., the harness portion below the camshaft sensor. This part of the harness only carries the crankshaft sensor wires and the wires for one O2 sensor (plus presumably some grounded shielding though that's not visible near the connector).

And basic tools live in my trunk. Not that I'd like to do this on the side of the road.
Counterpoint to what I just said: I let the engine idle, and then I wiggled the wire while it was idling. No effect at all.

I mean ultimately I'll have to take the harness apart I suppose. It's a guessing game.
The thing with wire rot is that it opens when it sits, and when you dink with it, it makes back up. I've had this issue with a starter wire. The big one, lol. When you reached down and wiggled it, then hit the switch, sometimes it blew sparks out one end, but it would work for a while.
Mating surfaces? Meaning the pins in the connector?
Mating surfaces as in where the sensor sits on the block and where it bolts into the block. I'm not sure about the 3.8, but some sensors ground where they mate to the block.
Side note:

What is this single wire suspiciously loose connector aft/inboard of the power steering pump?

That's the oil pressure switch.

Take the wire loom off that section of the harness and cut off/replace as much of the wires up to the connector as you can. Replace the connector and pigtail if you can find a good donor. It's entirely possible there's a short in the harness that vibrates and/or heat-creeps itself into and out of continuity. There's no good way to test the harness without a load on the wires - a basic continuity test won't show if there's a short that fails with current on the wire.
Are you running Android? Check out Torque app. Even the paid version is cheap and gives alot of data logging info.
I went in there last night...but sort of lost patience and put it back together. No obvious damage. I didn't take off the shielding because I have no new to replace it with.


The shielding has an uninsulated ground wire running through it, which looks messy but doesn't seem like it could cause any harm.

This morning the car runs normally. I went for a short drive around the neighborhood.

Should I just forget about it until the next episode?
that wiring looks a hot mess. what is it actually connected to? the white wire with purple around it? and the metal sheathing?
It's grey and blue for the crankshaft sensor. They go directly to the large connector on the firewall passenger side. The shielding ground wire, I'm not sure where it actually connects to ground.

And the general messiness of it is by design. Just looked at a harness on a Mustang, and it looks exactly the same.
What exactly am I looking for? It spins fine (I mean it doesn't wobble.)
Well when a balancer breaks the outer half can move back and forth. Or it can spin slower than the crankshaft. Get a paint marker or nail polish or something similar. Simply paint a straight line on it that crosses the inner and outer ring. If it moves when it runs the paint line will separate.
Does that make sense to you?
So to bring this thread to a close, at least for now, the car has been running impeccably this week. And I'm really not interested in addressing a "phantom problem". Cars have bad days, too.
The code incl. SES light came back today. Symptoms:
  • Starts normal
  • Drive short distance
  • Car starts shuddering under acceleration
  • Erratic tach
  • Pull over
  • Car idles, then dies
I was able to make it home.

So then I did this:
I made an all new crankshaft sensor harness with an extra connector I had. Routed the harness to the main PCM connector on the firewall (directly below the right side hood hinge), cut the original wires there, and connected my new harness.

Initial test drive went fine. But since this problem seems to come and go, I won't be confident I fixed it until I can go without the code for a decent while.

Interesting side observation: after the repair I started the car three times. Once just to check, then twice during my test drive. Each time the engine started instantly. Not that it usually takes long...but it takes distinctly longer than what I'm now seeing. Again, final verdict when I had a decent while to monitor.

Similar threads