Car question

[xoxoxoBruce]

Originally Posted by BigV

Just my

Not to worry, I'm aware. I've made my suggestion and don't intend to pursue it, as I'm sure tw will have a solution.

[glatt]

I watched the engine for a while today. Can't find anything wrong. Motor mounts are fine. I used a flashlight and small mirror to look at all 4 of them very closely. Can't find any vacuum leaks. Don't see anything touching that would transmit vibrations. I listened to the injectors with a plastic pipe, and can hear them all clicking at the same speed and strength.

It sounds a little louder near the belts, and it's a throaty exhaust sound, but I can't find anything wrong with the exhaust other than a pinhole leak in the muffler that drips water. The hangers are fine.

Does this O2 sensor look right? This is a highway trip from about a month ago. I've got 4 charts from the trip.

Speed:



RPMs:



I think this is the O2 sensor up by the engine:



And I think this is the O2 sensor after the catalyst:



Shouldn't this second O2 sensor be pretty much flat? It's fluctuating almost as much as the first one.

There are no codes, and the check engine light is off.

[tw]

Originally Posted by glatt

Shouldn't this second O2 sensor be pretty much flat? It's fluctuating almost as much as the first one.

An engine is constantly adjusting (ie short term fuel trim, timing advance) for changes you may not even realiize exist. View how stable your foot is by monitoring the throttle position with Carchip. Also see how much an engine changes with throttle and load variations (ie monitor manifold vacuum, timing advance). Simply approaching another vehicle means those parameters must vary significantly with less air flow resistance. An O2 sensor must detect changes so tiny that you do not even realize they exist. Those same changes would also appear after the catalytic converter.

O2 sensor is important during normal operation. Install a defective O2 sensor and see little to no idle mode change. Your gas mileage implies everything necessary for proper normal mode operaation (engine timing, spark advance, manifold air pressure, EGR valve operation, O2 sensor) is ideal. Concern is something different during idle mode.

For example, an EGR valve must open and adjust position during normal mode. Same EGR valve must be completely closed during idle. Same EGR valve acts completely different in the two operation modes. EGR valve could be leaking (defect during idle mode). But performs as a completely different device (100% functional during normal mode). IOW the same part has two completely different operation modes.

Your EGR tests said the EGR valve in both operation modes was OK.

O2 sensor says little during idle mode operation. O2 sensor is mostly about detecting minor changes in normal mode operation to cause minor adjustments to fuel injector fuel trim. Your mileage suggests fuel trim is just fine because an O2 sensor is accurately detecting minor changes such as your foot on the accelerator, minor change in road grade, air resistance due to the vehicle in front, etc.

How much must parameters change when you think everything is constant? Use CarChip to monitor thottle position, manifold vacuum, ignition timing advance, and short term fuel trim. Notice how much your engine changes when you think everything is constant. The O2 sensor, as Carchip shows, also reflects those variations.

Sensor readings and control adjustments vary much during normal mode operation. Those same sensors are typically irrelvant when in idle mode operation. During idle mode, a computer monitor a crank or cam sensor to measure engine speed. Then makes adjustments to the Idle Control Valve to maintain RPMs. Other sensors and control functions change little or are ignored. When entering idle mode, a part that does not operate or change during normal mode operation (Idle Control Valve) goes from doing nothing to doing the most.

Again, exhaust noise suggests no missing cylinders. A vacuum gauge would probably confirm that. 700 RPMs, although low for most cars, may or may not be too low for your engine.

We have avoided the Idle Control Valve. Nothing at this point suggests a problem exists there. However manually testing it for smooth operation (with fingers) might discover a minor problem. I seriously doubt it. But really, we are running out of suspect. And the 700 RPM number is slowly sounding relevant.

A most complicated operations occur during normal mode operation. What we have seen and what gas mileage confirms implies all those functions are OK. Idle roughness only at 700 RPMs involves many fewer sensors and parameters. O2 sensor typically is not on a list of idle mode suspects. An O2 sensor defect would be more apparent in normal mode operation. But everything we have seen implies the O2 sensor is just fine.

Don't remember if an answer was provided. Are the voltage limits for that O2 within specs defined by the shop manual? At one point I wondered if your O2 voltage variation was too much. If remembering correctly, you said it was OK.

[glatt]

Originally Posted by tw

For example, an EGR valve must open and adjust position during normal mode. Same EGR valve must be completely closed during idle. Same EGR valve acts completely different in the two operation modes. EGR valve could be leaking (defect during idle mode). But performs as a completely different device (100% functional during normal mode). IOW the same part has two completely different operation modes.

Your EGR tests said the EGR valve in both operation modes was OK.

When I checked the EGR valve, I could open and close it by sucking on the hose that goes to the diaphragm, but I couldn't see inside the valve to see if it was sealing properly when it slapped shut. Maybe there's a little grit keeping it from closing. I sprayed A LOT of carb cleaner in there so I think it's clean, but I can't know for sure. I suppose I could have tried somehow blowing through the exhaust channel to see if it was blocked. Even after cleaning, there's no way I'd want to press my lips against that surface.

[infinite monkey]

I our favorite so ing him. Go elsewhere or and kick back with a nice cold

I'll buy the first round.

[BigV]

glatt, when the car's all warmed up or whatever conditions are needed for this rough idle symptom to appear, would the roughness disappear if you gave it a tiny amount of gas? Like just enough to raise the idle, as if the idle were, say 100 rpm greater?

[glatt]

Originally Posted by BigV

glatt, when the car's all warmed up or whatever conditions are needed for this rough idle symptom to appear, would the roughness disappear if you gave it a tiny amount of gas? Like just enough to raise the idle, as if the idle were, say 100 rpm greater?

Not sure if only 100 extra rpm makes it go away, but a couple hundred more do make it go away. If it goes from 700 to around 1000, the vibrations go away.

[tw]

Originally Posted by glatt

but a couple hundred more do make it go away. If it goes from 700 to around 1000, the vibrations go away.

That is about the point where an engine changes from its idle mode operation to normal mode operation. More accurately, point centers closer to 1200 RPM. Its not a sharp transistion. But at 1000 RPM, idle mode parameters are changing to normal operation mode.

What exactly is that point? Decelerate while not touching the accelerator. It begins entering idel mode when fuel injectors go from no operation (no fuel into the engine) to providing fuel for idle mode.

Idle for most cars is 800. But again, what determines that idle? Its computer times the engine by monitoring only the crankshaft or camshaft sensor. Computer only reads that sensor to adjust fuel for 700 (or 800 on most cars) RPM once the engine is warm.

Asking if 700 is too low is a valid question. Sometimes that spec number is on a label glued under the hood or attached to the door pillar (that label covered when the door closes). I have never seen a shop manual not provide that number even when no adjustment is possible. Since a wrong number could be symptoms of a specific problem.

Anything you might do to change idle will be compensated for by the computer. So if idle should be 800, well that would imply a master clock problem in the computer. And therefore might explain roughness only at idle only because the computer wants an idle that is too low.

Well, touch the accelerator to raise RPMs up to 800 may or may not report something useful. Says little by itself. But in combination with other facts, may imply a suspect.

[BigV]

im, you crack me up too. not like Ron White or anything, but your own special smart funny.

[infinite monkey]

Puh. I got pieces of Ron White in my stool.

[BigV]

hm.

1000 rpm is a high idle. Does the roughness diminish in a linear fashion from 700 to 1000? What's the lowest rpm that is acceptable? Maybe you have a tach, maybe not.

What I'm thinking is how low can you go to get the idle to be smooth? I don't really care about the number, I guess. If it was my car, I'd do the stuff you're doing. But if I could not find anything, I'd rig some way to get the throttle to stay in a position that let it idle smoothly. If that meant using the adjustment screw on the throttle linkage, ok. If that meant bending a bracket to accomplish the same thing, ok. If that meant putting some kind of washer/spacer in the linkage, ok. Pressing eversolightly on the gas pedal while stopped would not be ok. Something like that.

[BigV]

it is still about fuel/air/fire, as long as we're talking about internal combustion engines.

...

however...

when the fire is sent, when the fuel is sent, how much fuel is sent, the control of the air (via variable valve timing), those kinds of things are MUCH MORE strictly controlled, and affected by MANY, MANY MORE factors that didn't used to be considered. Those factors are evaluated by VASTLY MORE sensors, and that means lots and lots and lots more points of failure.

[Gravdigr]

Yeah there's about twelve dozen electronic fuck-ups that can leave ya hoofin.

[glatt]

What's the resistance value for "continuity" in a relay? The manual says to check for continuity, but the meter reads 0.9 ohms when I test the meter by touching its leads, and reads 118 ohms when testing the relay's prongs. When I apply battery power to those same prongs, I hear the relay click, and the other set of relay prongs go from no continuity to 0.9 ohms. Seems like it's working, but I'd expect "continuity" to have a far lower value.

By the way, I checked the battery cable, and it felt secure, but I was able to tighten it up considerable. There's a little rubber gasket between the two sides of the clamp, and it barely squeezed before. I tightened it up until the gasket was bulging a bit.

Tried to get the check engine light to come on by driving hard, including up a hill. But it's still off.

The only thing I've found wrong is the fuel pump doesn't come on for two seconds when key turned to "on."

[xoxoxoBruce]

Less than .9 ohms. A fingerprint will give you .9 ohms, that's nothing. Sounds like the relay is cool.

Homeland security saw you were speeding and punished you by sending a signal on skynet for your car to hiccup, you know, just to scare you a little.

Sounds like at this point, without throwing a ton of money at it, or making it a possibly fruitless quest, you'll have to wait for a solid failure, or at least less intermittent symptoms.

As far as the pump not giving you that two seconds, it doesn't matter as long as it's starting right up when asked.