Diagnose spark plug damage (pic included)

ASSISTANCE Stock Engine-Bay Components
41

[QUOTE=welfare;2321934]…never at idle. and closed loop is not initiated near or at wot. this leaves a pretty small window for the single wire o2. and that was the reason a heater circuit was instituted. the element can now reach operating temp within about 30 seconds of key on. [/QUOTE]ok, I believe I have read that similar year Civics only run closed-loop when at 10 - 90% WOT.

If that is true, then the O2 sensor is not in the picture at idle, cold or hot. It would be interesting to find a complete description of the normal idle regulation, instead of the piecemeal one in the Acura manual (!)

Be that as it may, you would agree that an old/dirty O2 sensor would read lean, and therefore over-correct the fuel to make the engine run rich - so if the sensor is old, fuel mileage may suffer? I believe this is true of other cars, except that a CEL comes on in later models with more specific O2 error codes.

Flaky FITV or IACV → rough idle, hunting, etc…
Old O2 sensor → engine runs rich → plugs covered with soot, poor mileage

(?)

42

[QUOTE=PMI;2321945]ok, I believe I have read that similar year Civics only run closed-loop when at 10 - 90% WOT.

[/quote]
not just civics, or hondas. this is true of all vehicles with a single wire o2

If that is true, then the O2 sensor is not in the picture at idle, cold or hot. It would be interesting to find a complete description of the normal idle regulation, instead of the piecemeal one in the Acura manual (!)

idle regulation on these cars is extremely simple. the iacv is a stepper motor. basically just a controlled vacuum leak. the passage bypasses the throttle plate. inside looks like a screw that turns in and out opening or closing the airway to the intake. it operates in increments called “counts”. a typical range is 0-50 counts. at 50 counts it is all the way closed (iirc). the ecm uses inputs to determine how many counts to turn the valve in order to maintain 750rpm. if you could datalog this, you’d see exactly what was going on. if you saw say 45 counts, you’d know that there was another point that air was getting in. seeing as how the iacv is the only passage for air past the plate, aside from your throttle stop adjustment (base idle). or that your base idle was out of adjustment. you could go outside right now, warm up the car, and pull any vacuum line off and the iacv will close and maintain 750rpm. the ecm may be a little confused and may hunt a bit, but you won’t get a high idle. this hunting idle is similar to a maladjusted base idle. this is why you must disconnect the iacv when adjusting idle. take it out of the picture. otherwise the iacv will just counter your idle adjustment to try and maintain 750rpm. you open the throttle plate more, the ecm closes the iacv more. until you’ve got an iacv sitting at 50 counts. around there is where the idle starts to hunt. the ecm is confused. this is a very common occurance and this is why on a hunting idle, i always suggest to go in and check base idle. it’s free, it’s common, and literally takes seconds to perform.
i won’t go into anymore detail on the iacv cuz it would take up an entire page. what i can say is if you have an obd2 vehicle. honda or otherwise, and have access to a device capable of viewing live data, go have a look at the iacv PID. mess around. watch what it does. open and close the throttle. pull a vac line. turn on the ac. viewing data is the best way to really understand it. and pretty much all iacv’s are the same design. stepper motors.

Be that as it may, you would agree that an old/dirty O2 sensor would read lean, and therefore over-correct the fuel to make the engine run rich - so if the sensor is old, fuel mileage may suffer? I believe this is true of other cars, except that a CEL comes on in later models with more specific O2 error codes.

Flaky FITV or IACV –> rough idle, hunting, etc…
Old O2 sensor –> engine runs rich –> plugs covered with soot, poor mileage

(?)

yes. a contaminated o2 sensor will read lean. most common cause is coolant in the exhaust stream. it’ll contaminate an o2 pretty quick. there’s really only a few o2 error codes. heater circuit. and range faults. but ask yourself; how does the ecm determine whether the o2 is reading out of range, or if it’s simply the engine running lean? that will be your homework today hahaha

43

[QUOTE=welfare;2321954]…there’s really only a few o2 error codes. heater circuit. and range faults. but ask yourself; how does the ecm determine whether the o2 is reading out of range, or if it’s simply the engine running lean? that will be your homework today hahaha[/QUOTE]Yeah, thanks, I think :wink:

Let’s see… a conventional O2 sensor output is some kind of alternating waveform, which represents the changes in the air-fuel mix, around the ideal 14.x-to-1. If the O2 sensor reads lean, the ECM enriches the air-fuel mix, and the output of the sensor changes state. So, I would guess that if there is a steady-state output after some period of time, high or low, the O2 sensor is out of range (?)

Now, here is something I did NOT know until I did some reading online last night, typical O2 sensor life:

Unheated 1 or 2 wire wire O2 sensors, 1976 - early 1990s: 30~50K miles (!!)
Heated 3 and 4-wire O2 sensors, mid-1980s - mid-1990s: 60,000 miles
OBD II 1996 & up, ~100,000 miles

Although I understand how the O2 sensors are used on OBDII cars, on this car, not so much. So, do we even get a CEL if the O2 is getting old or dirty? Can a bad coolant sensor prevent the ECM from going into closed-loop on this car?

The OP had a fouled plug, and now his mileage is down even though the car runs and idles ok. So, if the O2 sensor is dirty b/c partly burned fuel went into the exhaust when the plug fouled, his engine may be running open-loop for the simple reason that the sensor is not reading correctly.

(PID loops may be over my head :D)

44

[QUOTE=PMI;2321956]Yeah, thanks, I think :wink:

Let’s see… a conventional O2 sensor output is some kind of alternating waveform, which represents the changes in the air-fuel mix, around the ideal 14.x-to-1. If the O2 sensor reads lean, the ECM enriches the air-fuel mix, and the output of the sensor changes state. So, I would guess that if there is a steady-state output after some period of time, high or low, the O2 sensor is out of range (?)
[/quote]

while that would throw a code, the o2 signal will typcally not reach the specified ranges for a given time, programmed into the ecm. say, above 700mv, and below 200mv. the voltage will still rise and fall tho.

Although I understand how the O2 sensors are used on OBDII cars, on this car, not so much. So, do we even get a CEL if the O2 is getting old or dirty?

yes. the ever popular code 43. fuel supply fault. really just a blanket for everything involving the fuel system lol. this goes back to the question i’d asked. how does the ecm know if the o2 is at fault? or whether we are indeed jst running lean? these early ecm’s don’t know. they just see it as running lean even though injector PW is near max…fuel system fault

Can a bad coolant sensor prevent the ECM from going into closed-loop on this car?

a bad ect sensor won’t prevent closed loop on any car. it is solely the o2 sensor’s ability to create a voltage reading from oxygen content that determines whether or not the ecm can initiate closed loop. on a heated o2, this takes place within about 30 seconds after starting the engine. long before the ect has reached operating temp. in fact, it’s that cold start that really prompted the institution of the heater circuit in an o2 sensor. cold start is when the most CO emissions were being dispelled. number one cause, excess fuel to combat HC dropout. being able to enter closed loop in such a short period, on a cold engine, lowered CO and HC emissions immensely.
ect, tps, map, etc…are inputs, of course. but the o2 alone allows closed loop.

The OP had a fouled plug, and now his mileage is down even though the car runs and idles ok. So, if the O2 sensor is dirty b/c partly burned fuel went into the exhaust when the plug fouled, his engine may be running open-loop for the simple reason that the sensor is not reading correctly.

(PID loops may be over my head :D)

could be, sure. easy enough to test. while a lab scope is the ideal method, because you can check cross counts and hashing, a dvom will suffice to at least range test it

45

Very helpful, thanks.

A lab scope may not do much for the average diy mechanic. I would like to give it a shot sometime, and I have an old Tek 2221A, but getting a good trigger is a challenge in the presence of ignition noise and a floating ground. I don’t see how I can easily do this without some experience with a known good system.

In the 1990-93 Acura manual, there is a single section for a HEATED oxygen sensor (no separate test for the single-wire unheated sensor). The guide for code 43 asks for a “test harness”, which is not very helpful.

Reason why I asked about the coolant temp sensor:

… A bad coolant sensor can also prevent the system from going into closed loop because the computer also considers engine coolant temperature when deciding whether or not to go into closed loop…
(this is from http://www.obdii.com/articles/Understanding_Oxygen_Sensors.html, and may not apply to this car, or it may not even be correct, but I’ve seen similar comments elsewhere)

I replaced my O2 sensor (<$30 at Autozone) when I started driving this car 10~12K miles ago, but my exhaust tip looks like I am running a little rich anyway. I have never replaced either one of the coolant sensors.

I think that next time I have a chance, I will just take my O2 sensor out again and see what the tip looks like. If I were the OP, that is what I would do too. If I thought one or more plugs were fouled, there is a chance the O2 sensor tip is too.

46

i wouldn’t take much regard in the ect comments on that site. what i would do is try it out for myself. use a scan tool capable of viewing data, and see for yourself if the closed loop were affected with it disconnected. it wouldn’t make sense to avoid feedback from the only sensor capable of providing it because one of the inputs was seen as malfunctioning. but hey, till you see for yourself, don’t take either word.

a visual test of an o2 sensor isn’t really very accurate. they are fairly rugged in design. as i’d stated, you can range it with a dvom, but you won’t be able to check reaction time without a scope.
to range, simply probe the voltage signal at operatiing temp. then, fog propane into the intake tract. you should see voltage rise to 9-9.5v. once propane is removed, should drop to .5-1v. if voltage doesn’t reach either ranges, sensor is bad. that’s the easiest way.

47

typically, the sensor won’t reach the higher voltage range (rich), due to contaminants. this causes the ecm to believe that a richer mixture is needed.

48

Thanks. It may take a few days b/f I have a chance to try this, but thanks for your time!

49

I got 2 videos for you guys. One is normal engine start, and I was able to get the rough idle this morning. It didn’t last long and eventually corrected itself. There was no input from me in adjusting the idle on the video (I did not touch the accelerator in either video). In both videos, temperature outside was in the 40s an the engine was not warm, which I think is why the idle is converges to about 1500RPM to get the engine warmed up. (I’m not sure where it should be actually). When it was running rough, I smelled gasoline (could have just been that it was because my door was open).

When the car was running really poorly a few weeks ago, it would be worse that than the video, and really lacked power (i.e. as if running on 3 cylinders).

Rough idle: http://youtu.be/S8af8QuHcOE

Normal idle: http://youtu.be/GihrKmNHgkg

It looks like the outdated phone I had didn’t capture the audio all that great, but the idle is apparent at the tachometer.

In reading the previous posts, and checking my notes, it looks like my oxygen sensor (Denso 234-1017) is way past its life. I last changed it at 244500, and I’m at 328000 now. That’s a cheap part to replace to remove it from the list of possible issues. You agree?

50

on a cold start, your single wire o2 is not going to have any effect on idle. realistically, a single wire o2 is not going to have any effect on idle at any time, as closed loop wouldn’t be initiated.

if that’s a cold engine, that is almost certainly a fitv issue. service it, or replace it. i am 95% sure it will solve the issue

51

Right now it seems to only occur on cold starts, but it HAS happened on warm starts. I’m about to check the FITV now.

52

if you understand the air passages and how high idle is achieved, it is quite obvious.
the fact that you have a strong fuel odour from the exhaust tells you right there that the ect is likely working, dumping fuel for cold temp, but not enough air in the mixture. only 2 passages for air. iac, and fitv. if the fitv were opening as it should, it wouldn’t really matter what the iacv was doing. because the fitv passage should allow plenty of air to pass on a cold engine…if it were working

53

Wait, I already checked the fast idle valve last week. It works fine – I checked the air flow in the port by the throttle body. I’m getting confused with the terminology you’re using and what’s found in the Helm’s manual. What’s the IAC? I also just came across something in the Helms manual which is the fast idle control solenoid valve which is only on automatics. I’m going to test that one out, but get back to me on IAC.

Does EACV = IAC?

I still don’t have a check engine light / code on the ECU.

54

fitv issues are actually very easy to diagnose. you just need to isolate. knowing that there are only 2 air entry ports (fitv and iac), you just tape off the fitv port in the throttle body. it is the lower of the 2. if the driveability issue ceases, you know you’ve found the culprit.
however, this is when the concern is much more frequent than yours. it’s always easy to find a problem if it’s occurring. not as easy when it’s intermittent

55

[QUOTE=N FUL FX;2322095]Wait, I already checked the fast idle valve last week. It works fine – I checked the air flow in the port by the throttle body. I’m getting confused with the terminology you’re using and what’s found in the Helm’s manual. What’s the IAC? I also just came across something in the Helms manual which is the fast idle control solenoid valve which is only on automatics. I’m going to test that one out, but get back to me on IAC.

Does EACV = IAC?

I still don’t have a check engine light / code on the ECU.[/QUOTE]

if the problem wasn’t occurring when you checked the fitv, then of course it will check out fine, at the time.

iac=idle air control. same thing as eacv. electronic air control valve

56

Maybe this will help, it is a page from the Acura manual. Btw, I did not mean to suggest the O2 sensor is a solution for the idle issue, that was in reference to the possibly low mileage, and fouled plugs.

I think that the first line on that page explains what makes the idle go to close to 2K briefly, before it settles around 1500 rpm. After that, it should stay at ~1500 until the coolant temp starts to rise, and then the idle drops gradually, as the needle on the coolant temp gauge rises, until they both settle to a normal “warm” state.

57

The “rough engine start” video you posted does not show the initial rise to 2000 rpm, for whatever reason, so your cold start is not consistent with statement 1 in the manual.

After that, your videos (both of them) DO follow statement 2, as far as one can tell. The graph illustrates statement 2, it does NOT show the initial bump to 2K rpm, which is a bit confusing. I think that the transition from the 1500 rpm to the warm idle, is where the FITV gradually closes down. It is controlled mechanically, by the coolant temperature rising an heating up a slug of wax, so that makes sense.

The coolant stays hot, and therefore the FITV stays closed, for a few minutes after ignition off. However, you (should) still get a brief bump to 2000 rpm on a warm start, which is a function of the IAC valve, controlled by the ECM.

58

I couldn’t reproduce the rough idle while I was testing things just now. :auto: :hmm:

If I were to tape up the fast idle port in the throttle body, it seems to me that the idle would be really low until the engine warmed up, so I wouldn’t be able to recognize the rough idle. :shrug:

59

Well, if you tested the FIT valve, then you know if it is open when cold, and closed when warm, is that correct?

60

Another suggestion, how about removing the IAC valve and cleaning it a bit?

The one I took off my parts car looked pretty sad. The little metal screen was almost plugged with grime, so full air flow was close to impossible. I have do doubt that it would still regulate somewhat, but I would say that the open area was reduced by over 50%. If the amount of dirt is proportional to the life of the car, yours might be in similar shape.