WaryDriver

Car is 1998 Civic w/ D16Y7 engine.
P0170: Fuel Trim Malfunction Bank 1
P0172: Bank 1 Too Rich/Fuel Trim

Things I've done since the CEL, codes, and problems:

• Checked all 16 valve clearances, adjusted a few exhaust and intake. None were way off.
• Replaced fuel filter with brand new.
• Replaced alternator with remanufactured new.
• No vacuum leaks. I hooked up a gauge and reads steady pressure. Sprayed carb cleaner all around intake manifold connections while car running to see if idle noise changes and doesnt.
• Disconnected battery like 10 times and the CEL always returns in a few days of driving. Usually 2-4 days of driving the CEL returns.
• New engine oil and filter.
• New transmission oil.
• New air filter.
• New NGK spark plugs

At red lights, I'll coast in Neutral up to the light. Then it'll feel like its going to stall out in Neutral when Im at a stop (car shakes like its going to stall out). Then all of a sudden, the computer compensates to keep from stalling if thats how it works. It has stalled in parking lots in Neutral just rolling.

If you had the money to replace one part, what would it be? (What would you put your money on?) Or, what would you do troubleshooting-wise? I dont want to rebuild the car guessing what part. I want to test until I can determine what part is faulty.

EDIT: I'm thinking about buying a new upstream O2 sensor, given the codes are saying "Bank 1". Does it sound like it Could be the problem? Any reason why it wouldn't be considered for fixing the problems Im having (It never has in the liftetime of the car and has been 200k miles). Its only $35.

Poopies

iTrader: (4)

Did you try removing and cleaning the Idle Air Control Valve (IACV)?

ezone

That's refreshing to hear. You may need time and equipment to do that.


Unplug the primary (upstream) O2 sensor.
Disconnect the battery or pull ECM fuse for 10 minutes to clear adaptives and memory.

Reconnect battery and drive it with the sensor unplugged, see what happens.
See if your symptoms go away.

Bank 1 does not mean upstream or downstream.
"Sensor 1" or "sensor 2" are what indicate upstream or downstream sensors.

WaryDriver

Removed the IACV and this was what it looked like, 200k miles since 1998..

Flushed it over and over again with solution until solution was clear coming out of it one way then another way then another, did this for like 20 minutes soaking and violently flushing until it looked like this..


Before disassembly, I disconnected battery surely resetting ECU in the time it took me to do everything. Then connected battery once reassembled. Then did the ECU relearning procedures (crank and let heat to operating temp five or 10 minutes, uncrank). The CEL stayed off for one day of driving 40 miles. Then tonight the CEL returned barely accelerating early in fourth gear downhill 40mph, engine was not quite operating temperature (was parked only a mile away), never have A/C on or anything turned on when CEL comes on, nothing turned on except headlights.

As for the feeling of stalling when idle, I think it has gotten better (not wanting to stall out as bad so maybe cleaning this wasnt a waste of time?)

Question: Did it look dirty enough to hurt my car's performance/efficiency?
Question/concern: When I tightened the four tiny bolts that hold the IACV in place, they just kept turning (ALL 4 BOLTS). I was barely tightening them because I read only do it 3-4 N*m. But I think i over torqued and destroyed every one of them. Seemed to hold snug though every turn (weird) but they all just kept turning forever. I should have removed the fuel rail for more room and feel but I didnt.

Next, I will do what ezone has suggested. Plus I ordered a new upstream NTK O2 sensor last weekend for $22. More testing to come...

Poopies

iTrader: (4)

I don't think it looked very dirty, but then again I've never had to do mine. Did the gasket look ok? If it improved your idle a bit, then the free fix was worth it, right?

Weird... if you're worried about it, you can get a new set of 4 for like $3.60 ($.90 apiece) on Honda's Estore. Free shipping if you pick them up at a dealership. Part 15 here:
http://estore.honda.com/honda/parts/...&dl=206925#015

As for your O2 sensor, you can pull it out to see if it looks fouled or not. An 02 sensor socket would be helpful for that!

Wankenstein

OP: if the bolts for the IACV are spinning freely then the IACV is loose and most likely creating a vacuum leak. There is also a paper gasket (IIRC) that is associated with the IACV and you need to make sure that is in good shape as well. As Poopies (him and I are not related though we are both online excrement) said pull the bolts back out and check to see if the threads are stripped.. and hope that it's just the bolt threads that are stripped and not the mating surface's threads or you'll have some heli-coil(ing) to do.
Either way the IACV is not causing the codes but regardless needs to be mounted correctly.

A few free or cheap things you can do that may or may not resolve the codes but, will help your engine run smoother regardless and you should be able to find youtubes or threads on:
1) Reset the ECU (google or youtube)
2) Remove and clean the throttle body
3) Lift and clean all chassis and engine grounds
4) Test MAP sensor: http://troubleshootmyvehicle.com/hon...e-map-sensor-1 (replace sensor if faulty)
5) Clean map sensor: http://www.ehow.com/how_5480692_clea...ap-sensor.html
6) Test o2 sensor (linked video). Alternatively, you might be able to pull one from a local junkyard at a low price and replace the current one to see if that stops the CEL's.
** Some people clean the oxygen sensor by heating the tip with an acetylene torch, thus removing the carbon buildup. i have never tried it and therefor cannot suggest it. maybe others will chime in.

My bet is as your first instinct.. faulty oxygen sensor. However, OE (Denso or NTK) are at least $100.

WaryDriver

Was doing going to do what ezone said and then try replacing the O2 sensor since I have a spare new one when I noticed the exhaust headers/manifold has a crack a few inches before and leading right up to the first O2 sensor. I've never removed the heat shield on this car before tonight. The crack is on top, very visible right after removing heat shield I noticed immediately.


I think this crack is the likely cause for the P0172. Can someone confirm? I did not replace the O2 sensor with my new one after finding this crack. Because I dont want to install it then weld or replace then have to reinstall. What do I do about this crack? Weld or replace headers? If replace, I would like to know what headers are best for my D16Y7 engine that won't crack like this. I have a DX Civic which I've read has different exhaust config than an EX. Is an EX config superior and should I go ahead and convert to EX headers + cat? I have a stock exhaust system right now.

ALSO- another problem I'm having is a STRONG BURNING RUBBER SMELL UNDER HOOD after driving a few miles or even a road trip. Smells like rubber burning around car hood after parking getting out of car and just walking near it its very strong. It's NOT the tires or brakes. It's inside the hood. The tires are not hot or any sign or smell in the wheel well. What does exhaust gas smell like right when it leaves the engine? I'm trying to know whether I'm smelling exhaust gas or not coming out of this crack. There is no smoke when I open the hood even after road trips. It's just the STRONG RUBBER BURNING SMELL under hood. IT'S NOT ANY V BELT EITHER. I have good knowledge on these belts and have inspected them for wear. V belts have perfect tension and I have a new alternator after suspecting it of freezing up causing the belt to burn (that wasnt happening though, because the burning smell remained after alternator replaced and all accessory belts).

Poopies and turd_ferguson, I didnt mean that the bolt heads where stripped. The heads are all good but I overtorqued them all apparently. All four of them turn forever so I must have over tightened each one. And I was being gentle using a short ratchet cant believe i went too hard with my fingers. I stripped the hole threads thats what I meant to say. Anyways its holding good so im not worrying about that. Especially since i did all that after getting the codes. Idling problem remains, car shakes like its going to stall out when idling at red lights. The IACV cleaning did nothing noticeable to fix idling stall out feeling.
turd_ferguson, I've done your steps 1-5 before finding the exhaust crack but engine light kept returning.

Summary of car problems that remain:
1. Burning/burnt rubber smell under hood. Dont know if related to check engine light and codes.
2. Feels like its going to stall out after coming to stops car shakes and sometimes stalls out.
3. Check engine light and codes P0172 and P0170.

ezone

I just said to unplug the sensor and drive it to see if the bog symptom went away.
EDIT: Bog may be the wrong word, I was referring to this statement: (car shakes like its going to stall out). Then all of a sudden, the computer compensates to keep from stalling if thats how it works.

IF the symptom persists, then it might not be the cause.
If the symptom goes away, then stick a new sensor in there and see how it drives.

When it comes time to fix the crack in the manifold, it's only gonna take about 14 extra seconds to remove the new sensor so don't let that stop you from changing it now.
I doubt it is the CAUSE of the code.

You can smell it when you open the hood with it running, it's leaking out that cracked manifold.
(Old car or small engine/lawnmower exhaust smell, if you aren't in California.)

IMO burning rubber (think tire smoke) smells VERY different from raw engine exhaust---exhaust that hasn't passed through a catalytic converter.
The IAC valve has a coolant passage, so if that IAC valve and gasket does not seal up well it could allow antifreeze to get into the intake air stream and cause running issues. Keep it in mind.

WaryDriver

O2 sensor only senses o2 molecules, right? The crack allows exhaust gas to escape before reaching sensor. That means less o2 molecules in the gas mixture hitting sensor. That why I thought the crack is likely the cause of p0172 (rich) reading from the o2 sensor. Because of this rich reading, the car is actually running LEAN because the computer gets info that exhaust is rich then injects less fuel. Which could be why my car stalls out idling I'm thinking. Because of lean mixture theres not enough fuel thanks to misinformed computer. Misinformed by the o2 sensor because of the crack.

Am I misunderstanding? Please say whether my logic is wrong and correct me. A question I have is would this crack introduce/suck in any engine bay air to the manifold or just let gas escape?

ezone

You see hear and smell that a cracked exhaust manifold leaks exhaust out, but you don't see that it also allows fresh air to enter through the same crack under certain conditions (reversion, pulsation), thereby causing the sensor to read (falsely) extra oxygen in the exhaust....at times. Maybe all the time, IDK.

That alone does not cause the sensor to fail nor become inaccurate.
It DOES skew the accuracy of what the sensor is reporting to the computer though.

The computer responds to the O2 sensor input as programmed.

The sensor reports as though the engine is lean (high O2 = lean), the computer adds a bit of fuel injector duration until the O2 sensor gives its desired output. (basic feedback circuit)

IF the computer reaches the limit of its acceptable values to add to the basic calculations for injector duration (seen in short term fuel trim and long term fuel trim values) it will code.
The cracked manifold generally causes the system to go leaner than stoich.
The computer adds fuel until it THINKS it's in stoich- but it's false. If it's so bad it reaches its threshold fuel trim value (+19% or 1.19 depending on the system/scanner used) for enrichment, it would code P0171 lean.


Problem is, P0172 is a rich code.
Long term fuel trim had to exceed -16% (0.84) to set P0172.

Those engines had issues with the primary sensor going "numb" for periods of time, long enough for the computer to go nuts with fuel delivery but sometimes not long enough to code for a sensor problem.
If the primary O2 gets stuck at a high voltage (rich value), the computer pulls trims down trying to get the desired response from the sensor, but since the sensor is stuck it can't respond. The computer has the ability to add or subtract enough fuel to bog the engine.
Then when the sensor starts working again or the computer defaults to open loop again, the bad running suddenly stops.
THAT'S why I said to unplug the primary sensor and drive it a while first, back in post 3.

Cam timing, tight valves, inaccurate MAP reading, wrong injectors, and more could cause a rich code too.
Got freeze data?



OMG I rambled too much again. My brain is fried.

WaryDriver

Replaced exhaust manifold and cat assembly with new assembly that had oem specs even though aftermarket. Replaced upstream O2 sensor with a brand new NGK 24542. Reused downstream O2 sensor in downstream position once I finally got it loose from old cat. Reconnected battery after doing everything and did the ECU relearning procedures. And the CEL returned after a few days of driving with the same codes like it always does.

So, ezone was right. The crack before the sensor wasn't causing the P0172. Brand new exhaust manifold, cat, and upstream O2 sensor. This was an expensive way to find out for sure and was my fault.

Valve clearances I've checked meticulously and made slight adjustments. I got this rich code days after changing my timing belt but don't know if it's related. The timing I checked in multiple ways. Although, that was my first time changing a timing belt.

No freeze data. This time the CEL came on when I was off the throttle and in Neutral rolling at like 35-40mph. Engine just reached normal operating temperature on the temp gauge (had been driving 4-5 minutes). But it has come on when I'm in 2nd gear giving it some gas going 20mph.

ezone

Nah, it doesn't have to be.

Rediag for the code you have now, readjust your thinking capAdd leaking EGR valve to that list, if equipped

Wankenstein

Nerd alert...LOL

Wankenstein

Ezone's been wrenching Honda's since the time you had a 1" pecker and sucking on your mama's *****.

I had a smilar crack on my first 6th gen Civic manifold and it never threw a code. Don't gloat over the repairs though. The crack needed to be fixed (replacement) as you don't want to smell rotten egg farts everytime you drive you car, as well as eventually damage the CC. Never a bad thing to replace an o2 sensor on a high mileage car. The o2 sensor combined with CC replacement should increase gas mileage a bit.

Wankenstein

What model do you have?: EX,LX,DX,HX,VX ? Needs Denso (OEM) and model specific.
Use this link to plug in your car's particulars:

I did mention earlier if the IACV's block mounting holes are stripped..you will have some heli-coiling to do. If you want to cheap out or lazy out you could try to go up in nearest size standard bolt in the current hole... just make sure to use correct pitch, length and highest heat tolerant metal possible.

ezone

OP's profile says he's 59 years of age.
I've only been with a Honda dealer since 2003.

Wankenstein

I didn't pay attention to the OP's profile but, after looking at it, it states 59 post.

For some reason I picture you (Ezone) as a being in your early 50's, grizzled, bearded mechanic that's seen it and done it all in the Honda world.

I just wanted to emphasize to the OP that you're usually right in your statements.

ezone

You might not be all that far off there, but I'm a little younger than that LOL


Oh no, I've only been in this dealer a short while....but I've been around the block a time or three.

ezone

Friday....... a parts department employee brought in an unruly 3 year old grandson...and was kinda babysitting while the dad was buying a car.

The kid was makin a mess, being noisy, gettin on my last nerve....so I duct taped him to a chair.


Just 2 wraps, not hard, not enough to trap the kid, just enough to surprise him and get him to STFU for a while.

Wankenstein

Duct Tape..man's second best friend (after his dog).