sneaky98gt

Hi folks, the name's Will. I'm new to the forums, but certainly not new to cars or my Civic. I've got an 01 LX, 2-door, 5-speed, 269k miles. Bought it around a year ago. So far has been a decent car, and you guy's write-ups have been a great deal of help for me (so far, I've replaced the head gasket, intake manifold gasket, valve seals, camshaft seal, oil pan gasket, interior speakers, along with a half-dozen other things I'm forgetting), so thank's a lot.

I've recently come across a new issue. I've searched around a good bit, but haven't found the answers I'm looking for. Be warned, though, I'm a little long-winded sometimes. I'd rather be thorough than not. But, if you want to cut through to my real questions, they're bullet-ed at the bottom of my post.

Ok, so I'll start from the beginning. For the first 6 months or so, I got fantastic fuel mileage. On the highway trips I'd take to and from work, I'd easily get 37-38 mpg; on highway trips back home to see the folks, I knocked down over 41 a couple of times; and then for in-city driving to and from school, I'd see a worst of around 34 mpg.

Well, around Christmas, I started to notice a decline in fuel mileage. First, it was just 1 or 2 on average, then it was 3-4, and so on. Maybe February or so, I noticed the infamous exhaust manifold crack, and shortly thereafter, the ensuing P0420 code. Naturally, I attributed this as the cause for the gas mileage decrease. Because of school, work, and life in general, I ran it like this for a couple months and just sucked it up.

Well, I finally got around to fixing the crack a little over a month ago. Reset the CEL, and went on my merry way, fully expecting my mileage to pick back up. It didn't. Not long after I reset the code, it came back. I ignored the MPG over the first tank, thinking the ECU might need some time to adjust. Checked the next tank, which was all city. BAD. 27 mpg bad. Checked the next tank, 100% highway, that I've done a few dozen times before. Also BAD. 32 mpg. Next tank: mixed driving. 29 mpg. So on the same trips and driving habits I've had for the past year, it's losing a solid 20-25% of the fuel mileage it used to have.

And for anyone asking, all of the general maintenance has been done since I bought the car 15,000 miles ago. Spark plugs, tires, alignment, cleaned air filter, etc.

Long story short: I fixed the exhaust manifold leak and the fuel mileage is at an all-time low, while all the general maintenance is at 100%, and still have a P0420 code.

This leads me to believe that either the cat is bad, or the O2 sensor(s) is bad. Here's my questions:

--Which is more likely (O2 sensor, or cat) to be defective?

--Is there any way to check whether or not an O2 sensor is working?

--Is there any way to check whether the cat is good or not?

--Anything wrong with the $250 exhaust manifold/cats on Ebay?

--Anything wrong with the $50 NGK oxygen sensors on Ebay? (I understand these are the OEM brand).

--Any other advice/help you may have to give concerning my issue?

Thanks!

-Will

ezone

Gas mileage....Look at the tires first, and tire pressure. Next killer is cold weather., idle time, and winter blend gasoline. Ethanol gas will drop your gas mileage. Next would be overall vehicle condition, dragging brakes, etc. Over a quarter million miles--cat. Are there any codes related to the O2 sensors? No. The only other way is by reading a Scantool Datalist, and most average people still wouldn't be able to do much with that info.

O2 sensor(s) could have issues, and the likelihood of a sensor problem that the PCM didn't detect is pretty small....but not impossible by any means. Trust the code. If you fixed the exhaust leakage, then you can pretty much trust the code. Yes. Probably.... What is your OEM brand? HINT: Look for a name stamped into the side of the original sensors. DENSO and NTK were both OEM, but use whichever brand came out. DENSO and NTK have their own websites, you can look up your own part numbers and you can probably cross-reference the OEM part numbers right out of an online catalog.. A dealer would throw a cat at it.

sneaky98gt

Tires are good. Air pressure is at 36 in the front, 32 in the rear, just like I've always ran (including when it was getting good mileage). Weather has been getting warmer; it never idles, and we've gone to the summer blend. With my other car of 8 years, I've always, without exception, gotten better mileage in the summer. Vehicle is in great condition. I know the front brakes are fine. Haven't specifically checked the rears, but there's certainly nothing noticeable (noises, not coasting well, etc.). I'm 99% confident none of those things are the issue.

I'm not doubting you at all. But my other car had O2 sensors go bad to the point of driveability issues, and never once threw a code.

I always hear people say a bad cat gives off a foul odor, and that you usually notice power loss with it. I haven't noticed either of those. I also know that O2 sensors don't often last 250k miles, and have no idea if/when the previous owner changed them.

Bottom line is that I don't mind spending money to fix a problem, but don't want to needless throw money at parts simply guessing what might be wrong.

Please explain. Because there is NO WAY I'm spending the kind of money that Honda wants for one of these parts. For the amount a new cat costs for this 4-cylinder Honda, I could do a complete exhaust on my other car. Long tube headers, hi-flow catted mid-pipe, and full cat-back. Stupid kind of money for this dumb thing.

I've had them out before, but never noticed what brand they were. I'll check that out soon. I know I've read in various places that it was NGK/NTK, but could be Denso also. Doesn't really matter to me.


Thanks for the help!

ezone

I've had to diagnose those too, that's why I didn't completely exclude it. 2 different issues, even though both can be cat related... Neither have anything to do with what the computer is monitoring for a P0420 code for decreased efficiency. The computer is judging the ability of the cat to store and release oxygen. True. I'd replace the Cat first, with OEM. Then monitor data, then sensor(s) if necessary. Not much to explain here. Cheap is cheap for a reason. You get exactly what you pay for.

If you get a supercheap cat from ebay, just how long do you think it will last? Who will warranty it when it sets another P0420 code in 6 months or less? One year? Who will pay to install another one?

I believe if you get a California certified aftermarket cat, it is warranted 5 years. A reputable supplier and installer may handle warranty issues if they arise.

The OE cat lasted a quarter million miles. I guess it's entirely up to you, it's your baby. How much longer does the car need to last? Brand matters a lot. You wouldn't believe how many people try to put the cheapo Bosch sensors in from DangerZone and get burned by it.

got2run5

Just an FYI, a week ago NTK 02 sensors were on sale at Advance Auto for $60 + a $10 rebate so $50. Not sure if they're still on sale.

There are ways to test your cat. There are 2 ways a cat gos bad-clogging and not operating at efficiency.

P0420 means that latter but poor fuel economy means the former.

To test for a clogged cat you need to measure the pressure buildup with a gauge before the cat. This is usually done by taking out the upstream 02 and measuring the pressure there. If there is a large buildup-it's clogged and exhaust can't get through well.

To test for a cat that no longer "cleans" the exhaust there are a couple ways you can use in combo with each other.
-Buy an infrared thermometer ($30 ish at walmart last time I looked) and check the temp of the cat. Google this since idk how much of a variance to look for.
-If you have a nice scanner tool (or a bluetooth adapter with an app like Torque), graph both 02 sensors. The upstream one should vary up and down, but the downstream one should remain somewhat flat. If the downstream mimics the upstream, your cat isn't "cleaning" enough.

I've never actually done the first couple, but I extensively researched it when I had P0420.

ezone

When a cat gets restricted or clogs, most people notice the loss of performance either long before or right along with noticing a loss of gas mileage. Poor fuel economy can be the result of the PCM repeatedly attempting to run the self diagnostic test for the cat monitor, or attempting to attain a desired output from the rear O2 sensor while running and driving. Yes, some of these cars can and do use the rear sensor for some amount of fuel control at certain times. It all depends on the operating strategies programmed into the PCM, and I can't say that all of them do it but I know that some do. This is actually not a valid test as far as trying to 'outjudge' the computer. Yes, measuring the inlet and outlet temp can tell you if the cat is present and has lit off, but you cannot judge its efficiency in any way shape or form more than working or dead. The computer makes that decision.

Besides, just what do you think you would be able to do if you COULD outjudge the computer....Slap it and tell it the cat is still there and working?? No. Computer still wins.

Computer can fail its own cat long before a state emissions test would find a problem with it too. Computer still wins.

Same deal for watching/graphing the O2 sensors. I've been doing this for an awful long time, and I won't even try to judge any more than "it's there" or "it's hollow".

sneaky98gt

I'll come back to this, as it looks like you sorta answered my upcoming question in a later post.

I don't know how long it'll last, but a lot of folks seem to get away with them just fine. The ones I'm looking at have a 5-year warranty. I do all my own work, so labor isn't an issue. I understand the piece of mind you can get from buying OEM and paying a stealership to install it, but that is just too much money. Simply outrageous price for what it is.

Haha, not in the slightest. Simply my good-gas-mileage-beater. My 'baby' has twice the cylinders, half an atmosphere extra induction, and about 5 times the amount of power. LoL!

Honestly, not much actually. I graduate in December. Assuming I have a job when I graduate, I'll probably be selling both of my cars and going for a single car somewhere between their two levels of performance and fuel consumption. Something along the lines of a Mazdaspeed 3.

I know that even with el-cheapo parts, they'll never pay off in fuel mileage before I get rid of it. BUT, I know it's a lot harder to sell a car with issues (that's how I got this car: SUPER cheap). So if I'm going to fix it anyway, I figure go ahead and fix it now and save the fuel mileage as long as I can.

I worded that incorrectly. I meant that it doesn't matter to me which I have to buy, because as far as I can tell, they're both about the same price.

That's what I started asking above. You're saying (I think) that the cat has failed in the manner of simply not breaking down the NOx emissions, and the downstream O2 sensor is detecting that correctly, and therefore replace the cat. You are NOT saying that the cat has become 'plugged' or whatever since I'm not smelling anything or noticing a big power decrease. (Correct me if I'm wrong)

OK, I got that. And my next logical question is: why would that affect the fuel mileage as badly as it is? You're saying that the rear O2 sensor is affecting the fuel trims. That could be true on a Honda, but on my other car, it is not. Ever. Matter of fact, they have been turned off for around 5 years now.

If what you're saying is true, wouldn't guys who put a catless pipe on their car have a huge decrease in fuel mileage? I honestly don't know the answer to that concerning Hondas, but my other car actually picked up mileage when I took off the cats.

What might seem more reasonable to me is that the upstream O2 sensor is going bad, while the downstream is OK. It might be sensing that the exhaust gas has a lot of oxygen in it (when it doesn't), therefore richening the mixture (bad gas mileage), and then the downstream is also measuring the extra oxygen, not sensing a huge difference, and thinking the cat is bad. IDK, hypothetical situation.

I guess this is why I started a thread. I had all of these situations running through my head, and wasn't sure what was most logical and/or probable.


Oh, one more thing. Sometimes, the CEL actually goes away for a while. It did that today on the way home from work. It'll stay gone for a couple hundred miles, then come back randomly for another 500, or whatever. Seems a little weird, but nothing seems different (gas mileage or performance wise) other than the light being on or off.

Thanks again for all the help.

sneaky98gt

Thanks for the info, I'll check that out.

ezone

Get whichever brand is OE for that particular application. I tried to explain that already....The theory as I understand it anyway.

Do you have any idea how much gas mileage drops on a Prius with a cat code? I read a case study on one, it was amazing. It can on many of these, yes. It all depends on what the engineers wanted. Too bad you can't read the case study I mentioned.

Think of a gas engine as nothing more than a life support system for the catalytic converter. Because clean emissions is of utmost importance in the eyes of the EPA (and therefore the design engineers), that's the main purpose of the engine....Motivating the car down the road is entirely secondary. An O2 sensor cannot detect NOx. An O2 sensor, by design, can only detect O2.

The PCM in this case, can only judge the the cat by its ability/capacity to store and release O2. It judges this by manipulating fuel control in order to exact a reaction from the rear sensor...under certain conditions. It compares the number of sensor transitions between front and rear (ratio) and compares to a threshold. I'm not an engineer by any means, and I'm not a teacher either. I'm sure general cat monitor info is available elsewhere if you really need to know more.

Handy chart found via GIS....sorry if it doesn't open here, my HTML privileges seem to have been taken away. So your PCM is judging the cat, and the cat is right at the borderline between good and bad. I get this too, many times if I get a good running vehicle with a first time cat code I erase it and let them go. It could take 6 months before the code comes back. Or it might take a week. It's up to the customer to decide what they want to do (we don't have emissions testing in my locale). The rotten egg/sulfur smell (hydrogen sulfide, H2S) originates in a reaction of high sulfur content in SOME fuels, a lack of oxygen during the burn process (rich), and the catalytic reaction of some compositions within the cat itself. Newer Hondas aren't known for the horrible sulfur smell that some other manufacturers were well known for (older GMs come to mind right away) under normal running conditions, but I have smelled it immediately following a merciless flogging of the gas pedal. Sometimes I might describe it as a gunpowder smell too. Some engineers are far smarter than others. Some have entirely different strategies and goals. Honda did not hire Ford engineers for a damn good reason, I'm sure. Your ***** must be HUGE! (insert know your meme pic here)

LOL had to do it. Refer to the above explanation of life support for a cat. It takes fuel to keep the cat lit and hot. You no longer care about keeping the cat alive, you can focus on other goals. Sure, and it is reasonable at the kind of miles you have. (I'm not going to analyze the rest of the paragraph you wrote.)

If I were to attempt this route, knowing what the computer is looking for as far as sensor activity or inactivity......I'd change the front O2 sensor first....then see how it goes as far as mileage and keeping the light off.

/can't proofread

sneaky98gt

So let me sum up what I'm thinking what you're saying. The PCM actually uses the rear O2 sensor to adjust the fuel trims, so as to get a certain exhaust gas mixture to properly heat up the cat, and thus reduce emissions. That correct (in a nutshell)?

Therefore, if the cat isn't working properly, the rear O2 is directing the PCM to dump in extra fuel in an effort to increase the cat temperature and decrease emissions.

Seems complicated, but does make sense in theory.

Quite the irony here, considering my 500 horsepower, 3600 pound Ford is currently getting the same fuel mileage my 100 horsepower, 2500 pound Civic is. :P


All said and done, I'm almost definitely going to change both the cat and O2 sensors. Considering the shape all the rest of the car's general maintenance items were in when I bought it, I'm fairly certain the O2 sensors are well past their recommended life cycle. So replace them, and if the problem isn't fixed, new cat will be on the way.

Thanks again for the help. I'm a big fan of learning new things, especially car related (mechanical engineer, here). And I've certainly learned some new things about Hondas today.

ezone

ONLY at certain times and conditions (however, we don't actually know what all of the possible conditions could be). It uses the front sensor under most other normal conditions, like you would expect from any other manufacturer. The computer may attempt to get a certain output from the rear sensor by manipulating fuel control. It could ignore the front sensor entirely during these times. It might not ignore the front one too.

The computer can only see as O2 content in the exhaust goes higher or lower than the threshold designed into the O2 sensor. The O2 sensor can only switch between a high and a low voltage as Oxygen content goes lower or higher than the designed 14.7:1 threshold.

The computer can control the O2 that enters the cat by watching the front O2 sensor and controlling the fuel. You see this as the sensor cycling up and down rapidly, but the PCM is watching the time span of each high and low transition.....

It watches for the desired change in the state of the rear O2 sensor, based on what amount of O2 it thinks is entering the cat. If it doesn't see what it wants, then it can manipulate fuel as it reruns its tests (possibly repeatedly, possibly once or twice...IDK without serious research).

Most vehicles run the cat monitor during steady state driving, so this is the condition where one might notice the mileage drop as the tests are run......IF the drop is solely attributed to a cat code. Some of these cars have a calculated cat temperature parameter in the datalist, but it is based on a model. So the computer should know when the cat is supposed to be at operating temperature.......but there is no way it can actually monitor temperature. Overheating the cat does not necessarily reduce emissions. Not really. See above.

The computer might only need to run a cat monitor test one time in a drive cycle as long as it passes that test. It might be free to repeat the test every time it fails the test, attempting to get a passing result. You see this as "the light went out on the way home".

I'm tired, and I'm not sure this is worded right. It sounded good in my mind though. Not proofread much either.