EcoModder.com

EcoModder.com (https://ecomodder.com/forum/)
-   Off-Topic Tech (https://ecomodder.com/forum/off-topic-tech.html)
-   -   Puzzle: vacuum leak, injector crud, or O2 sensor fail? (https://ecomodder.com/forum/showthread.php/puzzle-vacuum-leak-injector-crud-o2-sensor-fail-31871.html)

California98Civic 05-03-2015 10:25 AM

Puzzle: vacuum leak, injector crud, or O2 sensor fail?
 
After a month of normal and flawless operation, I am back to the very same symptoms that caused me to replace the fuel pressure regulator. No CEL, no pending codes. But the ECU thinks the AFR is off and is therefore cutting fuel trim, sometimes at much as 35% and often 10-25%. when I hit the throttle, trim will be slightly increased, which seems normal for this Honda. Then trim steadily decreases fuel as I hold steady throttle, at either high or low load. It will even do this at idle. Though it has bucked a little once or twice, it has never stalled (yet).

My power is reduced. Fuel economy is suffering badly.

Last month I tested fuel pump pressure (normal) and fuel pressure regulator (normal). Replaced fuel filter (it was past time). Compression test too was good.

Yesterday I removed each spark plug, all black around base and gray around insulator.

No smell of gasoline in any cylinder.

Vaccum leak? MAP sensor? fuel injectors? O2 sensor?

I replaced the O2 with a good quality unit nearly 3 years ago. It has maybe 40,000 miles on it.

At this point I decided to drive it until it fails or I get a CEL. I have AAA+ and therefore free towing up to 100 miles.

I figure I'll patiently test items. Learn. Fix. I need my DD.

Any advice appreciated.

james

California98Civic 05-03-2015 02:30 PM

Cold start this morning, it idled raggedly in open loop, and then steadied at idle in a few seconds once the UG indicated closed loop operation... at the same time the trim started getting lean. It stabilized in the negative teens (-14 to -18%).

Once warmed up it stalled once while coasting with the engine on in neutral.

There was one episode where the primary O2 reading on my UG showed voltage 0.995 for something like 5-10 seconds straight with no flutuation.

Warm, the engine generally idles comfortably in the 600 rpm range, with a fuel trim -24% or so. Sometimes it idles almost at normal 0.00 fuel trim.

California98Civic 05-03-2015 03:51 PM

more reading, more thinking outloud
 
Maybe I can rule out the vacuum leak idea, because if the ECU is reducing fuel the must either be more fuel getting into system than ECU expects or less air. A vacuum leak would mean more air getting into the system, which the ECU would adjust to by adding fuel, not subtracting it.

Maybe I can also eliminate the O2, since the problem occured during open loop in cold start this morning. In closed loop the idle stabilized as the system detected a rich condition and took away fuel. Maybe eliminate the MAP sensor too, for the same open/closed loop reason.

This leaves fuel injection or a fuel leak from somewhere.

BabyDiesel 05-03-2015 03:59 PM

I have had a similar problem in my Escort. What I was told by one of the respected members at TeamZX2 was it is a vacuum leak. A bad MAF/MAP would cause + fuel trims. A vacuum leak causes - fuel trims. I think it is related to the EVAP on mine, or PCV lines. I know you have a Honda, but these are worth a shot considering our issues are similar.

California98Civic 05-03-2015 04:06 PM

Quote:

Originally Posted by BabyDiesel (Post 477866)
I have had a similar problem in my Escort. What I was told by one of the respected members at TeamZX2 was it is a vacuum leak. A bad MAF/MAP would cause + fuel trims. A vacuum leak causes - fuel trims. I think it is related to the EVAP on mine, or PCV lines. I know you have a Honda, but these are worth a shot considering our issues are similar.

Thanks. But a vacuum leak would mean more air getting in, which would mean the ECU would add fuel, no? That would mean positive fuel trim, and I am seeing negative fuel trim.

user removed 05-03-2015 05:12 PM

Try some techron first. Uneven injector delivery volume or poor spray pattern could be an issue. It could be MAP related, trim is probably through the 02 feedback. Reading almost 1 volt is very rich if my memory still serves me.

regards
mech

BabyDiesel 05-03-2015 05:51 PM

Quote:

Originally Posted by California98Civic (Post 477867)
Thanks. But a vacuum leak would mean more air getting in, which would mean the ECU would add fuel, no? That would mean positive fuel trim, and I am seeing negative fuel trim.

You are correct, good sir! I got them switched in my brain before I typed them :o

- FTs is bad MAF/MAP
+ FTs is vacuum leak.

Sorry for the confusion!

user removed 05-03-2015 08:25 PM

Look at it like this, how many ways is it possible to get too much fuel for the proper amount of air.

I've seen the injectors in a Z car that were continuous when they were supposed to be momentary, ecu fried by a reverse polarity jump start, another time fried by leaking windshield. The ecu regulates fuel delivery through the ground. Obviously a continuous ground would be a big problem.

Listen to the injectors with a stethoscope compare each ones sound to the other. If one is silent, clean the connection and try it again. I've seen engines not miss with a dead injector, the fuel trim was increased by the remaining injectors until it provided enough fuel for the car to actually run better than you would think.

Injector issue
ruptured diaphragm on the fuel pressure regulator
Open circuit on the ambient air temp sensor (would massively flood out a z car in two seconds).

Now here is the real problem. Even if you replaced the fuel pressure regulator yesterday, it could still be defective, even if you replaced it three times in the last week.

Had a customer replace 4 starter motors in his Z before he came back to me and apologized for calling me a thief, after I tried to explain the difference between the $39.95 direct drive cheap rebuilt for Advance, versus the $129 Nissan factory rebuilt gear reduction starter motor.

This is one of the reasons why I retired

regards
mech

California98Civic 05-03-2015 11:09 PM

Amazingly, I appear to have solved the problem from a "surprise" direction. I was wishing for a mechanical vacuum gauge because it seems odd to always rely on the mediation of the UltraGauge or my OBD2 diagnotic reader thing. Then I remembered someone saying cleaning old grounds can help the car run. Then I remembered cleaning the grounds has been on my list of long-procrastinated maintenance for a year or two. Then I took 30 mins to clean engine ground, transmission ground, battery ground, and one other grounding cable. And.... PROBLEMS ARE GONE! Just pulled nearly 70mpg in traffic on a 30 mile test run. Mostly normal fuel trim for a car doing a lo of P&G (slightly on the + side by 2 or 3%). Wow. They had not been cleaned since 1998. Hahaha!

BTW, I also got fuel system cleaner, Mech. What the hell, it can't hurt. I bought a big bottle to use in each tank for a few months. Thanks!

Chrysler kid 05-04-2015 08:31 AM

Use marvel mystery oil as a fuel additive it works great in Hondas

user removed 05-04-2015 08:39 AM

Quote:

Originally Posted by California98Civic (Post 477899)
Amazingly, I appear to have solved the problem from a "surprise" direction. I was wishing for a mechanical vacuum gauge because it seems odd to always rely on the mediation of the UltraGauge or my OBD2 diagnotic reader thing. Then I remembered someone saying cleaning old grounds can help the car run. Then I remembered cleaning the grounds has been on my list of long-procrastinated maintenance for a year or two. Then I took 30 mins to clean engine ground, transmission ground, battery ground, and one other grounding cable. And.... PROBLEMS ARE GONE! Just pulled nearly 70mpg in traffic on a 30 mile test run. Mostly normal fuel trim for a car doing a lo of P&G (slightly on the + side by 2 or 3%). Wow. They had not been cleaned since 1998. Hahaha!

BTW, I also got fuel system cleaner, Mech. What the hell, it can't hurt. I bought a big bottle to use in each tank for a few months. Thanks!

Fantastic. I like to polish the connection areas and dab a little grease on them. Never had to redo one with that procedure. I got 6 bottles of techron at Costco for $20, supposedly the ticket for the wife's direct injected Sorento (Kia relabels the bottles). Saved me pulling the rail on the Sentra which requires pulling the upper manifold.

regards
mech

California98Civic 05-04-2015 01:47 PM

Quote:

Originally Posted by Chrysler kid (Post 477923)
Use marvel mystery oil as a fuel additive it works great in Hondas

Thanks, both of you. I actually have MMO already and use it sometimes. I ought some specific injector cleaner on Mech's advice too... what the hell. The BlueDevil oil leak stuff worked and I never would have believed it. When the claims are modest, i am willing to try it, so as to prevent or repair something minor. My injectors are old. Perhaps they're a bit gunky. I think I'll treat them with techron and MMO for a good string of tanks, a few months worth. Can't hurt. Might save me a hassle.

Thanks for the advice re: coating the electrical grounds like that, Mech.

Sincerely,
James

California98Civic 05-13-2015 03:05 PM

Interestingly, once again, after maybe 200 miles "fixed" the exact same problem has returned: wildly leaning-out fuel trims, associated with a decrease in power, and decrease in fuel economy. My tank average is dropping fast. This is the third return after the initial repair. Each time the problem goes away for a few hundred miles, or even nearly 1000 miles, and then returns. First time I made it go away with a fuel pressure regulator, then replaced the regulator again when the problem came back, then fixed it by cleaning ground wires when it returned a second time. Now it is back again, for a third time.

Why does it run normally for so long and then return to this fail state again?

Puzzle!

WD40 05-13-2015 05:00 PM

Do you reset the ecm?
If you do it might take while for the cars issues to show up.

California98Civic 05-13-2015 05:28 PM

Quote:

Originally Posted by WD40 (Post 479191)
Do you reset the ecm?
If you do it might take while for the cars issues to show up.

It's a good question. On the second repair with the pressure regulator, I did not reset the ECU (I left the battery connected) and the problem still went away. Then, another time, all I did was reset the ECU by disconnecting the battery for an hour or two. But that did not get rid of the problem for even a moment. When I cleaned the grounds, the ECU was of course reset.

Weird, huh?

BabyDiesel 05-13-2015 07:20 PM

I'm going to put money on it being the EVAP system. I have a P0446 code that took over 1000 miles to pop back up, and I had horrible idle and acceleration when it did. It was undrivable.

Might be an electrical connection for an EVAP component? That would explain it going out only every so often.

P.S. I am not a mechanic, so take my words with salt :) I can get my way around most vehicles though!

California98Civic 05-13-2015 11:50 PM

Thanks. MAP, Injectors, EVAP... there is my testing list. Maybe an intermittently open wire? But I can't see how changing a fuel pressure regulator or grounds cleaning would change the function of any of these components. Maybe there has always been three or so problem sorta layered on one another? Ugh.

But there has never been a CEL and the car does run, though not great.

Chrysler kid 05-15-2015 09:41 PM

Ok let's start over again

How many miles do you have on:
Distributor cap
Ignition rotor
Spark plugs
Spark plug wires

Have you tried to use carb cleaner around the intake manifold and hoses to check to see if the rpms change when sprayed on a certain part?

Have you cleaned the idle air control valve lately?
Is the coolant still fresh ?

Do you have any oil leaks around the valve cover

When is the last time you changed the air filter and inspected the throttle body for any dirt or build up


There was a Toyota echo 5 speed idling super low that just had tons of gunk built up on the throttle body the other day.

California98Civic 05-16-2015 01:03 AM

Quote:

Originally Posted by Chrysler kid (Post 479513)
Ok let's start over again

How many miles do you have on:
Distributor cap
Ignition rotor
Spark plugs
Spark plug wires

Have you tried to use carb cleaner around the intake manifold and hoses to check to see if the rpms change when sprayed on a certain part?

Have you cleaned the idle air control valve lately?
Is the coolant still fresh ?

Do you have any oil leaks around the valve cover

When is the last time you changed the air filter and inspected the throttle body for any dirt or build up


There was a Toyota echo 5 speed idling super low that just had tons of gunk built up on the throttle body the other day.

Here's another crazy deet: for the last two days the car has run quite normally, with longterm fuel trim moving to normal and then slightly rich. It has been raining these last two days.

Looking at your list, the cap, rotor, igniter, and spark plug wires have about 30,000, maybe. Maybe less. The plugs have been fuel fouled and cleaned. I have three sets that rotate in. No oil leaks. Compression is good. Filters are clean. I cleaned the throttle mechanism probably 20,000 miles ago, maybe a bit less. The carb cleaner would be for testing vacuum leaks, yes? I have not done that yet. I also have not cleaned the Idle Air Control Valve... both are good ideas.

Chrysler kid 05-16-2015 09:28 AM

Then the next question is have you checked your exhaust manifold for cracks? The car could be getting funny readings from the primary oxygen sensor if there is a Crack

ksa8907 05-16-2015 10:04 AM

Im curious of your fuel trims. Negative numbers mean the ecu is trying to lean the mixture and positive mean it is trying to enrich the mixture.

So you have a problem with negative fuel trims indicating a rich condition. The ecu is detecting too much fuel or too little air as compared to the amount of fuel it expected to need.

I would suspect bad injectors or leaking injector o-rings.

Map sensors are cheap and personally, i would replace that asap to rule it out.

Chrysler kid 05-16-2015 06:43 PM

It could also be a dirty fuel strainer or weakened fuel pump

ksa8907 05-16-2015 10:11 PM

Quote:

Originally Posted by Chrysler kid (Post 479600)
It could also be a dirty fuel strainer or weakened fuel pump

Remember, negative trims mean too much fuel.

California98Civic 05-17-2015 08:06 PM

The fuel pump tested normal.

Would a dirty strainer fail intermittently? Would injectors or their o-rings leak intermittently? After two or three rainy days in which normal function resumed, the problem has suddenly emerged agajn and with a vengeance. Return of the sun, warmth, and dry conditilns has meant a return of my fuel trim wackiness.

Part of the difficulty is knowing I cannot necessarily count on the sensors, meaning that the ECU might be cutting fuel based on false data from the MAP Sensor and/or the O2 sensor, right?

Bah! I gotta start testing stuff. Thanks all.

user removed 05-17-2015 08:39 PM

Have you checked every ground. Grounds usually fail when it gets dry. Fuel pressure good?
All testing must be when it is in failure mode, obviously.

Have access to another CPU?

I had a customer with a 75 Z car. " I drove all the way to Jacksonville Fla, where the car died. It restarted after a while, ran rough, until I revved it up a few times, blew some black smoke, then ran fine ever since."

I had the 5 most likely suspects in inventory, installed them all, then removed one at a time over the next 6-8 months and it never acted up again until I reinstalled the original CPU.

Maybe next time it messes up spray some water on the engine compartment and see if it clears up with the added humidity.

Run an additional ground wire from the cpu to the negative batt terminal.

regards
mech

user removed 05-17-2015 08:41 PM

Also spray some water on and around the plug wires at night,see if you see any sparks, indicating bad plug wires.

regards
mech

Chrysler kid 05-17-2015 11:18 PM

I would check your exhaust manifold. Is it still the stock one?

pgfpro 05-18-2015 12:09 AM

Quote:

Originally Posted by California98Civic (Post 479726)
The fuel pump tested normal.

Would a dirty strainer fail intermittently? Would injectors or their o-rings leak intermittently? After two or three rainy days in which normal function resumed, the problem has suddenly emerged agajn and with a vengeance. Return of the sun, warmth, and dry conditilns has meant a return of my fuel trim wackiness.

Part of the difficulty is knowing I cannot necessarily count on the sensors, meaning that the ECU might be cutting fuel based on false data from the MAP Sensor and/or the O2 sensor, right?

Bah! I gotta start testing stuff. Thanks all.

This sounds silly but what is your ground wire like (condition) on the intake???

2000mc 05-18-2015 02:58 AM

In the process of replacing the fuel pressure regulator, is there a wire harness or connector that you move, or even touch? If so try to figure out if / which wiring that could effect fuel trim is there. Then with the engine running and monitoring fuel trims and anything specific to what might be in the harness there, move the harness around / push/ wiggle connectors.

California98Civic 05-18-2015 11:01 PM

Car ran so poorly by the time I got home today it would shudder as I accelerated, fuel trims got as lean as -29%, and in the driveway I was able to idle it until it stalled. The computer took the car out of closed loop and ran it in open loop. So I got out the mitivac vacuum gauge and my scan tool, figuring I would measure vacuum mechanically and via the MAP sensor in inches Hg simultaneously. But first I tried three suggestions from this list: I hunted unsuccessfully for a ground wire on the intake; I sprayed water onto the hot engine to mimic rainy day humidity (no improvement); and I wiggled/pushed every wire and connector I could remember moving or brushing in previous efforts at repair. I also wiggled a couple others in the EVAP canister (I think).

That wiggling suggestion worked! The car runs perfectly normally again.

user removed 05-19-2015 07:57 AM

Injector connection? Make sure you mentally note the precise thing you did that "fixed it", cause it ain't fixed until you clean that connection thoroughly. If you can't find a ground from a valve cover bolt to the radiator support, MAKE ONE. You can't have too many grounds. My 84 CRX had a bad ground when it was built. Right radio speaker open circuit, factory new wiring harness. I just added a ground.

Also no need to garden hose the engine, a spray bottle like windex is fine. You could damage the engine if done completely wrong.

regards
mech

2000mc 05-19-2015 02:02 PM

Quote:

Originally Posted by Old Mechanic (Post 480012)
Injector connection? Make sure you mentally note the precise thing you did that "fixed it", cause it ain't fixed until...

If it was mine, I wouldn't be disapointed to end up eventually replacing the connector w/ pigtail, and possibly the component it's hooked up to. Considering how much trouble you've gone through. Of course if you find a broken or chaffed wire, maybe just a simple wire repair. Either way I would continue pursuing this issue until it's resolved better than just getting it to go away for now.

California98Civic 05-19-2015 09:44 PM

update
 
After 20 or 25 miles of normal operation, the problem returned as I began the return commute. I stopped repeatedly, checking one connection at a time until the problem went away again. It turned out to be the connector into the MAP sensor. Back in March or April, when I repaired the fuel pressure regulator the second time, I relocated my extra brake booster and it seems the vacuum line began to conact and stress the wires into the connector for the MAP. Since it was behind and under the air cleaner box, against the firewall, I did not notice. As soon as I repositioned the vacuum line and reattached the MAP sensor connector, the problem went away.

Perhaps the problem is solved, but only driving will "prove" it.

Thanks!

James

phh 07-30-2015 06:42 PM

Update?
 
It's been a while since your last post. Is it fixed? Anything new to report?

My wife's car is experiencing the same problems. Since it's an 1989 model, the on-board diagnostics give limited info. I've taken many of the same steps as you but so far no success. I'm checking the wiring harness next.

Thanks!
Peter

California98Civic 07-30-2015 08:31 PM

Quote:

Originally Posted by phh (Post 488494)
It's been a while since your last post. Is it fixed? Anything new to report?

My wife's car is experiencing the same problems. Since it's an 1989 model, the on-board diagnostics give limited info. I've taken many of the same steps as you but so far no success. I'm checking the wiring harness next.

Thanks!
Peter

The problem has not returned. Maybe take a look at this other thread and the responses... if you are not familiar with this technique for pulling codes on OBD1 and OBD0, I bet this will help you get more info: http://ecomodder.com/forum/showthrea...tml#post487287

some_other_dave 08-04-2015 01:36 AM

89 would be non-OBD. The only way to check the codes is to pull up the front of the passenger's carpet and look at the ECU when the key is on. There is a window in it, with an LED that will blink out the codes. No long/short blinks, just count the number of blinks until it pauses.

Generally the codes are only blinked out if the check-engine light (CEL) is on.

-soD

phh 08-05-2015 03:44 PM

Quote:

Originally Posted by some_other_dave (Post 488856)
89 would be non-OBD. The only way to check the codes is to pull up the front of the passenger's carpet and look at the ECU when the key is on. There is a window in it, with an LED that will blink out the codes. No long/short blinks, just count the number of blinks until it pauses.

Generally the codes are only blinked out if the check-engine light (CEL) is on.

-soD

The car is actually a Mitsubishi Mirage rebadged as an Eagle Summit, not a Honda. I was able to retrieve the trouble-codes with a voltmeter. Unfortunately the codes aren't concise enough and I still need to diagnose by hand (ugh!)

I used California98Civic's post as a rough guide and found it helpful.

some_other_dave 08-07-2015 08:29 PM

Hahaha! Whoops, I assumed you were talking about an 89 Civic, since the thread started on Civics. :)

-soD


All times are GMT -4. The time now is 10:15 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com