The New Life of the 3800 "Formula"

[TopNotch]

Just hope that the bend doesn't crimp down on the wire so you can't get it out.

I got it out, and it works. I will go on a longer test drive tomorrow to see how it holds up.

[TopNotch]

Update on my knock sensor problems. I was still getting an occasional knock sensor error, so it was time to hit the book. A while back, I had bought a 1995 Buick Park Avenue service manual (which also covers other GM 3800-powered cars of that year). The manual said that if a scanner didn't show knock retard when the error occurred, it had to be something else, and suggested that if "secondary ignition cables" (translation: spark plug wires) were near the knock sensor wire, that could cause the false error.
Now, I didn't get the errors until after installing the low mount alternator, and that caused the spark plug wires to take a slightly different path to the rear plugs. So I tried a simple re-routing of the wires, as shown in the picture below.

All I did was move the wires to the other side of the oil filler cap as shown. Then I went on the same test drive that had produced a knock sensor code. No code this time. Could it be that simple? The bundle that the spark plug wires pass over contains, among other things, the knock sensor wire.
On this engine, unlike on newer engines, there is only one wire to both sensors, and only one code set. Newer cars can tell you which sensor set the error.

One other interesting thing. The manual shows both the series 1 and series 2 3800's as possible engines for the 95 cars. That means that the change-over was done mid-year in 1995. And that explains why it is sometimes hard to get the right part for this engine. For example, when I tried to get a new throttle position sensor, the first one I got at the parts store was wrong. I had asked for the 1995 part. I then asked for the 1996 part, and it was correct.

[Fierofool]

So can we assume you have a Series II engine?  Is there any difference internally or just the externals? 

You're so good at troubleshooting, I need to let you foster my two for a while. 

Edit:  Or all 3. 

[TopNotch]

So can we assume you have a Series II engine?

It's definitely a series II. The intake manifold is series II, and all parts that changed between series I and II are series II.
I went on a longer test drive, and then to the store. No knock sensor error. That's one error that always lights the SES light when it happens. Some errors don't, and the manual lists which do and which don't.

[Fierofool]

Even with the mid-year changeover, they still used the OBD 1.5 ECM?  Is it possible it could be outfitted with the more common OBD II ECM?

[TopNotch]

I'm pretty sure OBD II wasn't introduced until '96, regardless of engine. And my OBD ii scanner doesn't work on it.

[Raydar]

...
So I tried a simple re-routing of the wires, as shown in the picture below.

All I did was move the wires to the other side of the oil filler cap as shown.

In their previous location, they would seem to have run parallel to that wire harness for some distance.
In their new location, they cross at nearly a right angle. A right angle crossing is just about always preferred, in the case where one signal can affect another.

[TopNotch]

Is it possible it could be outfitted with the more common OBD II ECM?

The one that's in there works just fine, and, as they say, "If it ain't broke, don't fix it".

[TopNotch]

It turned out that the knock sensor problem wasn't spark wire interference, and moving the spark plug wires to make it better was just coincidence. It was a momentary short, and it decided to be a more permanent short today. I fired up the car, and the SES light came on immediately, and the scanner said it was a knock sensor error. So I measured the resistance at the pin going to the PCM, and it was much lower than when I measured it previously.
There was a frayed section of the wire going to the back knock sensor, and it was touching some grounded metal. So I had to make a new one and had to fish it through to where it connects to the front sensor wire. All better now (i hope).

[Raydar]

Sounds like you got it. Good save!

[Wreck It Ralph]

Good deal.

Does anyone make new harnesses? For stock motors that is.

[TopNotch]

I thought I heard an exhaust leak, so I investigated. Turns out that I have a cracked manifold. In the picture below, I have peeled back some of the heat shielding where the crack is. Since than, I have removed it all.

After removing the heat shielding, I found that the crack doesn't go all the way around. I'll probably have to weld it, but for now, I decided to try something I found at AutoZone. It's called VersaChem Exhaust Seal Joint and Crack Sealer. It's supposed to dry in 4 hours, but I won't try running the car until tomorrow. We shall see.

[Raydar]

...I decided to try something I found at AutoZone. It's called VersaChem Exhaust Seal Joint and Crack Sealer. It's supposed to dry in 4 hours, but I won't try running the car until tomorrow. We shall see.

 

Good luck with the fix. Please let us know. I tried something similar - a JB Weld product, I think. It lasted for about an hour. But that was a long time ago. I hope the technology/chemistry has progressed some, in 15 years.

[scottb]

I cant say for certain but you might have made it impossible to get a good weld. In past experience any kind of silicone based sealer cant be completely removed and contaminates the weld. Im sure the wizardry behind it has advanced but I would still be leery

[Fierofool]

Provided it can be repaired, would brazing work?  It might be accomplished without removing the manifold.  If it has to be removed, it should be bolted down to another head to keep it from warping during repair.