Cox Abele

I could use some experienced insight on the following issue. My car is throwing a code for P0133. Now obviously it would be nice to simply replace the oxygen sensor and be finished, but nothing is ever that easy.

Here's some back story and all the low down on the car. I have a 98 DAP Civic CX Hatch sporting the bone stock(ish) D16y7. The car has 120k on the ticker and has had little to no problems in the time I've owned it. It's originally from Ohio, so there was more rust than I wanted to deal with, although it's still pretty clean. My little DD needed some much overdue maintenance so I had a shop due the following;

New exhaust from the j-pipe to the muffler (CX has the manifold/cat combo)
2 new axles
Clutch and related components
Timing belt and water pump

There's some very bad air between myself and the shop as they took darn near 2 months to get the work done. I won't take the car back and have left the bickering go.

~5-10 minutes after leaving the garage the CEL came on for the front o2 sensor. I was leaving town the following morning, so I threw a new sensor in figuring if it clears the issue, I've gotten lucky. Of course, it didn't so I start in picking and poking for anything that may be causing the problem.

I monitored the sensors from the ECM via OBDII scanner, and I can see that the front o2 sensor is providing signal and swinging it's voltage. Same goes for the rear. This eliminates a disconnected or broken wire.

I've checked for vacuum leaks countless times. I've come up empty handed each time. I've checked over the exhaust manifold for cracks that could be potentially leaking exhaust gas or introducing air into the system and have found none. If I disconnect the vac line from the FPR, no fuel spills out and I get a little rise in idle due to the newly introduced vacuum leak. The engine will bog a little if I cap the line so I'm assuming the FPR raising the fuel pressure as it should when at atmospheric pressure.

None of the other show sensors any issue that I can tell with my scanner. No other codes are showing. Since I had a cap and wires laying around, I tossed them in and checked on the plugs and rotor. Everything ignition wise is looking good. Pulling each wire individually shows I'm hitting on all 4 cylinders. No excessive oil or A/F ratio issues in the NGK plugs.

The car feels underpowered, but it may also be a mental thing as I was driving my 98 GS-T Spyder while my daily driver was down. On very very rare occasions, it feels like there is a slight miss in open loop, but the car is warm, there is zero hiccup from the engine. This slight miss is on cold mornings, and extremely inconsistent. It never did this before I took the car in or throwing the CEL.

I checked over the exhaust manifold and have not found and cracks or hear any leaks.

I'm really **** about this light. Gas mileage hasn't fallen off drastically, and I've visited a few different gas stations, so it's not fuel quality related. For the life of me, I just can't wrap my head around it.

The only "performance" mods on the car are a short ram intake with a massive filter. This thing has been on the car since I purchased it 40-60k ago and never created any issues. I've eliminated the factory ground straps and upgraded them with some beefy old amplifier wiring I had laying around. I checked the grounds over today and everything is as it should be.


Any advice on where to look? I've been tooling with cars for way too long for this to be such a problem, but I really am at a loss for what I should do next.

lazlong

Stupid question, have you checked for a loose/corroded wire or connector around the sensor? Also sometimes the sensor gets slow response if it's facing the wrong way, did you use the correct torque on the sensor?

Cox Abele

I have replaced the sensor so I am sure that the connection is secure. I've also been able to monitor the voltage it is sending to the ECM via the OBDII port, so I am definitely getting signal, ruling out loose or corroded wires.

Torque spec would be difficult to believe. Given that the threads aren't always cut identically from sensor-to-sensor or manifold-to-manifold, I could find positioning of the sensor to be of minimal possibility of fault. Additionally, before replacing the sensor, the OE unit was throwing the code which leaves me continuing to believe it's not an issue with the sensor.

lazlong

P0133 is for slow response. How fast is it switching? If you have a corroded wire/connection you could still have a signal but, not the right one. Maybe trace the harness back from the sensor and look for bad wire/connection. I say this because the sensor was doing the same thing before and after replacement which leads me to believe it's in the harness. Sorry, that's all I can think of right now.

Newfiedan

If they welded the exhaust and there is an exhaust leak then it can cause the code, sometimes hairline cracks in the exhaust manifold can cause it but so can improper seal between the block and the manifold which may not be easy to see unless you see that telltale carbon spot on the block. I would start with checking the torque on the exhaust studs possibly change the gasket as well to just eliminate that as a possibility if you have the spare money, my advice is to learn about this stuff and start doing the work yourself it not only saves money but decreases the chances of getting ripped off at the garage. Unfortunately it only takes one bad mechanic to ruin ppls opinion of shops and there are to many mechanics out there that are clueless or are **** poor representatives of the trade.

Cox Abele

I've got nearly 15 years of wrenching non-professionaly under my belt. I could have very well done this work myself as I have done every step of every job that I paid to have done on a 93 I sold not long ago. The catch was that I no longer have a facility to work in safely, and throwing a car on jack stands in a parking lot full of moving vehicles isn't something want to do. You see, I relocated across a few states from where I made most of my automotive contacts, so something as simple as finding a good mig welder is difficult. I took the advise to use this shop as the owner is a co-worker (a few rungs up the ladder). Had I known they would let the car sit for as long as it was, I'd of knocked it out in a week's time of after work wrenching, but I didn't have the patience, air tools or desire to fight the rust every step of the way.

The exhaust pieces were entirely bolt in. There was zero welding involved. Aside from the exhaust manifold leaking due to a crack or blown out gasket, a leak in the system wouldn't mean much as it would be post-sensor. I've checked it all over and haven't found any signs of cracking or leaking.

Cox Abele

I don't have the tools to measure dwell, but can say that voltage fluctuates on a scale that is very similar to the S2B1 sensor. Based only on voltage along, I'm seeing a full swing from rich to lean in the same amount of time as my eclipse's wide-band.

Newfiedan

hmm well then perhaps its possible that the cat is not doing its job if there are no noticeable exhaust leaks, and the sensors are showing normal fluctuations any corrosion on the contacts or in the wire would show as low voltage at the ecm. About the only means of finding an exhaust leak that I can think of would be to add some seafoam to the intake, it will clean any carbon buildup out and would surely show any minute exhaust leaks as it kicks out enough smoke to make it look like the engine took a royal crap on you. I found an exhaust leak that was pesky to track with this stuff and it does exactly as advertised it cleans any and all carbon out of that engine, did a kickass job on my mx3 and made her much more responsive. One thing I ought to note is that I did run into an issue in the past where the alternator output was causing a misread of the o2 sensor tripping a code on occasion and wrapping the plug in tinfoil took care of it.

Cox Abele

I ran some sea foam through it Saturday and was a-ok. The alternator is tucked in the rear, front o2 doesn't really run near it. This isn't my engine, but the back and forth about exhaust leaks may be better understood here. It'll give you an idea that the cat isn't really where most of you are probably suspecting.

Newfiedan

I meant the distributor sorry,it gives off an electromagnetic signal that can interfere with sensor signals wrapping the plugs in tinfoil tends to sheild the sensor signals from any leakage due to that interference so yeah anyhow long story short if seafoam did not show any leaks then I would be leaning towards the cat. I am dealing with a lovely p0420 code although I know chances are that I have a cracked manifold.

Cox Abele

p0133 S1B1 is the pre-cat sensor, not post. What would the cat have to do with it? We've discussed that I have not found any exhaust leaks, and if there were a problem with the cat, it would be s2b1 throwing the code.

I can see the distributor creating a signal issue, so I'll consider that and look into it, but that seems pretty far fetched due to the manner of how this all occurred.

Newfiedan

sorry I was on a different train of thought about the cat, was tired and thinking of my own code, however upon further consideration of your problem I would follow the 02 wiring from the main harness plug and look for any damage short of that you would need to check the frequency and amplitude of the oxygen sensor itself which if you do not have the correct diagnostic tools can be hard to do. Maybe check the bolts on the intake mani see if they are loose allowing unmonitered air into the engine perhaps, although I am pretty sure that a vac leak would show other symptoms. I talked to my buddy at a honda dealership and he said to check the secondary 02 sensor make sure to unplug and clean it with some electronics cleaner then reseat it, he said that he had this happen on a 98 civic dx and that got it to stay away once he cleaned and reseated the plug (fyi he put a little dielectric grease in there as well) I know its not indicated by the error code but he said that it worked on this particular car.

Cox Abele

Good news to report. There aren't many solutions available for the problem- most threads with a similar problem end without the started giving their solution.

My situation is a bit unique as I do most of my own work but had a fair amount of routine maintenance done due to rust. I spent a long time going over the vacuum system, checking for exhaust leaks, checking wiring, monitoring voltage... nothing seemed to provide motivation to one direction or another.

I was getting voltage to the ECM, and it seemed to be swinging appropriately. I had no detectable exhaust leaks and no vacuum leaks. There isn't really anything on the car that is so aftermarket that it would throw a code. Simply, I delivered my vehicle to a garage, they had it for a number of months then returned it to me after performing maintenance. At some point they did something to trigger the CEL and I'd even installed a new O2 sensor.

Low and behold, the connectors for the front and rear oxygen sensor are identical. The wiring is all bundled up on itself too. I was pulling electrical connectors for cleaning and die-electric grease when the light bulb went off. I had a wiring diagram close by and was able to figure it out almost instantly! A quit revision to which sensor was plugged into what bank of the ECM and the car ran better. I cleared the code, drove home and pulled the battery to re-set the fuel trims.

It's been a week and I haven't had the light come back on for the oxygen sensor. I did show a lean condition, but that has since cleared itself up. I was also experiencing some power loss and what felt like a dead cylinder at ~65 mph while cruising, however that too appears to have cleared itself up entirely.

So, if you end up with a P0133 and it's not obvious what may be causing the CEL, check to make sure you didn't (or your garage didn't) swap the sensor wiring by mistake (an easy one at that.)

lazlong

^^ Good news. Thanks for posting the solution.