Car question

[glatt]

Pictures are always fun.

This is what it looks like when I start to replace the fuel filter. You can't even see the damn thing, but it's down there.


When I lean in a little closer, you can see the banjo bolt at the top of the guy under there. There is no way I can get my hand in there, even from below, to remove the thing. Well, maybe from below, but the last thing I want is a face full of dripping gasoline when I open the fuel lines.

So here's the filter, and what I need to remove to get to it.

[glatt]

Here it is, with the entire air assembly removed, and half a dozen hoses and cables detached to be able to get to it. Notice I put blue masking tape on everything and numbered them so I could keep track of where it all goes when I put it back.


And look at all the room I have to work with now! I actually had to halfway remove the bracket holding the thing so that I could get the lower (rigid) fuel line to align with the new filter.



It was all pretty straightforward, but took about two hours, including getting all the tools out and putting everything away again.

[Lamplighter]

We're waiting for the denouement.. did you fix it ?

[glatt]

Oh, yeah. There was more resistance blowing through the old filter, but it wasn't dramatic.

[tw]

Originally Posted by glatt

Oh, yeah. There was more resistance blowing through the old filter, but it wasn't dramatic.

So a new fuel filter did not solve anything. An original and still 'wet with fuel' filter might explain a slight restriction. It would not explain fuel restrictions during idle.

Imagine what happens when the car was actually demanding much fuel. Idle demands little. Heavy acceleration requires plenty of fuel. The engine would obviously stumble and almost stall on acceleration if that filter had any relevant obstruction (was dirty).

An intermittent problem can be created by many things that exist constantly. For example, if a fuel pump pressure was slightly low, then the car would work normally almost all the time; only have intermittent stumbling.

The EGR valve would not move smoothly. But the car would work normally almost every time - just not in the rare time when the EGR valve intermittently stuck. Move the valve by hand to feel the restriction that only rarely gets stuck. Unfortunately intermittents not reported by the computer can be this difficult to locate.

Distributor vacuum advance and retard motion can also be intermittent. In one case, I discovered that intermittent was created by grease on those moving parts that had become sticky with age. It always existed. And was located by physically moving the part by hand. The sticky grease always existed. But the resulting bad performance only occurred rarely and without any one common action making it happen. Intermittents are that nasty to find when computer diagnostics do not report them.

As you recall, shop manual provides a list of suspects that are numerous.

BTW, the engine is almost two completely different engines when in idle and when moving the car. Many parts do completely different operations in those two modes. Another way to eliminate suspects from that list. That was the case in cloud's IAV failure. Only the 'idle' engine was failing. The 'moving the car' engine worked just fine.

[glatt]

Two months without a problem. Convinced the fuel filter fixed it. Then it was doing the same sputtering today that it was doing before. No time to trouble shoot or look at it. Got to my destination. Shut it off for about 5 minutes, and it's been working fine since then. Such a weird intermittent problem. But it's fine now, so *shrug*.

I guess I know now why the previous owner parted with this used car.

[Gravdigr]

Try writing down all the conditions/details you can come up with each time it happens.

-how long you have been driving it
-distance
-time of day
-amount of fuel
-engine temp
-ambient temp

Maybe you'll see a pattern that hasn't emerged just yet.

[Griff]

^tw's sock puppet

[glatt]

Yeah. I need to ponder this.

[tw]

Originally Posted by glatt

Yeah. I need to ponder this.

Demonstrated was why shotgunning does so poorly at eliminating problems. So what has changed? As Gravedigr suggests, collect facts. List every part that was touched (not just moved- touched) when you replaced a filter. What has been touched recently? Not when the intermittent occurred. But even a week previously.

The number of possibilities is literally approaching 1000. No exaggeration. Good diagnostic procedure means reducing that suspect list to more like 100. Which is why replacing the filter only on speculation was so unlikely to be the defect. Especially if blowing in the old filter did not expose a restriction. And since a restricted filter only causes problems on acceleration - not at idle.

Imagine a wire of 32 strands. All are broken. Move that cable and some remain in contact. Move the cable and too many are disconnected causing confusion to the computer or insufficient power to a solenoid. Replace a solenoid. That new one might work with two less strands connected. But the problem still exists. Just fails less often. Only symptoms cured.

Above demonstrates what you are dealing with. Previously listed were how to eliminate some more common suspects (ie EGR valve, vacuum advance and retard system, intermittent manifold leak or defective idle control valve, partially broken cable, dirty connector that gets cleaned by making and breaking and then fails months later because the reason for that corrosion was not identified and eliminated, etc).

Well, one symptom that eliminates some suspects is temperature. Apparently (does?) temperature is unrelated to good and bad operation. Has fuel been eliminated as a suspect - were consecutive tanks from the same station or brand? When the intermittent exists, what exactly do you do to make the problem repeatable or make it worse? Do lights change intensity? Do tires make more noise? Irrelevant is a belief that tires are irrelevant. Solving intermittents means collecting all facts irregardless of whehter you believe it is irrelevant. BTW, that is a repeated concept so accurately expressed in a TV show called House.

I routinely get called when others cannot solve strange problems. Many never learn that this is critically important. Collecting facts and symptoms is completely unrelated to identifying the problem. Which is also irrelevant to what comes later - eliminating the defect.

As I said without exaggeration. Your list of suspects is somewhere approaching 1000. Getting that number down means collecting facts as Gravedigr noted. And then identify a defect long before replacing any parts. Yes, even strange noise from tires in rare cases can be the kicker that identifies an intermittent engine problem.

[sexobon]

Originally Posted by tw

... Solving intermittents means collecting all facts irregardless of whehter you believe it is irrelevant. ...

[bold mine]

Eloquently said.

[glatt]

The weather was finally nice enough yesterday for me to work on the car. Except of course, Murphy's law said that I was booked solid for most of the day with other stuff. I only had about half an hour in the late afternoon to look at it.

But I found a problem.

I wanted to start by pulling the spark plugs and looking at them so see what shape they were in. Then I wanted to get a spark tester to check that there was a good spark at each plug.

The car is 12 years old, and I was concerned that the plugs might be hard to remove. I had read about some guy on the internet who tried to check his plugs, and one broke off during removal, leaving threads locked in place in the cylinder head. That would suck, and I wanted to avoid that. So I planned ahead by switching the cars in the driveway so I could use the other one all day running errands, and the Camry engine could get nice and cold. Cold metal shrinks. I also went to the parts store to get the proper sized spark plug socket and an extension, and I got an ignition tester. I squirted a little PB Blaster down into the spark plug tubes about 5 minutes before I tried to loosen the plugs.

So it was time. I pulled the wires off the first plug, put the socket on the extension, attached it to the ratchet, and fed it down into the tube and onto the plug. I tentatively started to push on the wrench, and the plug started unscrewing with almost no effort at all. It was about as hard as unscrewing a light bulb.

I pulled the plug out. It was wet from the penetrating oil I had sprayed in there. But it looked like this:


There's a little carbon build up, but I don't know how much is normal. Each plug looks like this.

The thing that took too long to jump out at me though, is that it's the WRONG spark plug. This car is supposed to have special plugs with two grounding electrodes. One on each side. This 5S-FE engine takes the NGK BKR6EKPB11 spark plug. It's a special metal with two prongs, not your standard single prong. Plus, these plugs were so easy to get out, I think the previous owner had changed the plugs not too long ago and put the wrong ones in.

So then I checked the spark with the tester. This is basically a fake spark plug that you stick into the end of a plug wire and ground against any convenient grounding point. I'm not sure how you're supposed to perform this test. With the engine running, or starting it up each time after making the connection. Anyway, I did it with the engine running, and even without completing any circuits with my body, I was getting zapped while trying to plug it in to the spark plug wire with the engine running. I put on an extra pair of rubber gloves over the nitrile gloves I was wearing, but was still getting a shock through those gloves. It was kind of comical. But anyway I saw consistent sparks in the tester at each wire and they looked equally strong to me. But what do I know? It's the first time I've seen sparks, and I'm not sure how they are supposed to look.

So now two questions:

1. I was running the engine for a while with various spark plugs disconnected. The engine obviously ran poorly during this time, but it was nothing like the way it behaves when it's about to stall. The weird thing though is that the check engine light never came on. You'd think the computer would be pissed off about no spark in various cylinders over a 5 minute time frame, but that light never came on, and no codes were reported. It that a problem?

2. Would the wrong spark plugs with only a single grounding electrode instead of the dual grounding electrodes cause the intermittent stalling problem? The manual said using the wrong plug would result in poor engine performance, but I figured it would be consistently poor, not intermittently poor. I didn't put the correct plugs in yet. Too late in the day to go out and track them down and I was losing light. But I'm definitely going to get the correct plugs and put them in.

[glatt]

Oh, and the voltage was maybe a little high?

12.75 at the battery with the engine off, and 14.75 with the engine on and alternator running. Later after the engine warmed up, it was 14.55 with the engine on and alternator running.

[fargon]

That plug looks good, it is burning properly. I would replace the wires and plugs just because, and drive that car until till the wheels fall off. I would check the cold start injector for function. Go to the library and get the Chiltons Manuel for that car it will tell you the proper diagnoses procedures for the fuel injection and other systems. Do not use Haynes books they don't tell you anything about trouble shooting.

[glatt]

Originally Posted by fargon

That plug looks good, it is burning properly.

Thanks. It's nice to have a second set of eyes looking at this.