Hot Rod Forum banner

Oxygen sensors, whats better...factory or aftermarket?

8K views 71 replies 12 participants last post by  2-manytoyzs 
#1 ·
You know what I mean, it seems the factory ones last a 100k but aftermarket seem to last 30-80k. Is it because the older the engine gets the more oil contaminates the sensor? Or is it because the factory ones are better made, they do have a slightly different design on the diffuser?

What do you think?
 
#52 ·
It could run AFR's of 18:1. This would allow compliance without running an EGR. At AFR's in that range NOX drops. <hr></blockquote>

bcwrench, great reading but I must ask about this statement. Just wondering how running a leaner mixture such as 18:1 could reduce NOX when almost every car I've repaired for state inspection for high NOX has usually failed because it was running too lean. I can see the benefits of running a system like the Honda system but I can't see how reducing NOX could be one of them unless the PCM/ECM was programmed to drop that AFR when not idling or under load. I do not know the entire fuel strategies for that system.

The abnormal behaviour displayed only takes place during full throttle or high fuel demand conditions and the Pulse Width for bank 2 goes to 10 while bank 1 hovers around 6-7. It takes some time for the sensor to recover hence the low volt readings during recovery. Also the new sensor goes high (1.00+ volts) during this period leading me to believe a low fuel pressure condition is causing the problem with the computer unable to correct the condition without going full rich and then full lean, could be injectors too? <hr></blockquote>
It looks like from what you're saying that the O2 sensor voltages are high during load and accel. This is normal and they don't always match identical from bank to bank. When your accellerating your PCM is programmed to increase pulse width (richening fuel mixture for power) which will cause your O2 sensors to stop cycling and read rich (high). When the PCM sees that the load is no longer there it changes strategy, reduces fuel delivery, and the O2 sensors go back to normal as the system leans back out. If you had a vacuum leak or low fuel pressure you would see the O2 sensor voltages drop under load from a lean condition and increased pulse width as the PCM tried to compensate. It looks to me (if I read correctly) that you are just seeing WOT fuel strategies, the PCM increasing fuel delivery and the O2 sensors verifying this.
 
#53 ·
I also thought the same thing DM but the problem is when this happens the CIL lights and if you watch the sensor readings it takes a good five minutes for sensor two to recover, during this time I can watch the computer try to take control and correct the situation but it takes too long.

Here is the order of events as I have witnessed them;
  • WOT.
  • During recovery O2 sensor 2 goes lean and reads around 0.02 volts.
  • Fuel control at this point ignores the lean sensor and the CIL lights up.
  • Computer attempts to reign-in the rogue sensor and starts adjusting fuel trim to compensate eventually going full rich at +41. CIL stays lit.
  • Control attempt seems to work as sensor swings the other way and goes full rich and produces a 1.12 volt signal, CIL goes out.
  • 1.12 volt O2 signal lingers and the computer again tries to compensate and goes full lean on the fuel trim, my calibrated nose tells me that it indeed is going lean and driveability symptoms appear.
  • After what seems like a period of 5 minutes or so the sensor recovers and begins switching behaviour again, pulse widths even out between banks and all is well once again.

The interesting thing I should note is the old O2 sensor did not switch rich (1.12 volts) like the new one does during this period. All KOEO codes are 111 and KOER also, memory codes tell of the lean sensor but nothing else shows up in the codes.

Other than this unusual behaviour the car runs like a top and light city driving does not illuminate the CIL. These cars were known to be shipped new with either the standard Taurus fuel pump (60 gal/hr) or the proper SHO/Mustang GT (100 gal/hr) pump which can lead to a WOT lean condition even new. I am wondering if this might be the problem since it only happens at WOT and the factory pump now has 100 000 miles on it. It makes no unusual noise other than being kinda quiet compared to my Mom's 89 Mustang 4 cylinder.

I am going to plumb in a fuel pressure gauge (if I can find the proper fitting) and see what the pressure at idle is.

Any other suggestions? I'm stumped? <img src="confused.gif" border="0">
 
#54 ·
well 4Jaw, while I have NO idea what you need to do next, (sounds like your on the right track.) I feel compelled to respond to this statement,(mainly to pass some time and keep the discussion lively)
Originally posted by KULTULZ:
<strong>FORD today is not the FORD of yesterday.</strong><hr></blockquote>
thats for sure..... but you have to go back aways.....
ford knew when it released the 64.5 mustangs, that utilized the top of the gas tank as the floor of the trunk, that in an accident, gas might spray all over the passenger area, thus turning the car into a molotov ****tail. but it was cheaper to pay off a few lawsuits than to redesign and retool the car. but I still wouldn't buy a new gm vehicle.

sorry to have digressed.
Kultulz, would you do a how-to on where to get the best deals on parts, cos I think I been getting my stuff at the wrong place.
 
#55 ·
Originally posted by 4 Jaw Chuck:
<strong>The lag question was one I had considered especially when you consider how slow the control system is to begin with. The pathetic processing power in the typical EEC IV is pretty much maxed out in stock configuration as it is and adding speed to a mule is like supercharging a lawnmower, it won't cut anymore grass without an architecture change. </strong><hr></blockquote>

Believe it or not FORD has one of the fastest processing speeds.

Originally posted by 4 Jaw Chuck:
<strong> I might add to this conversation that the sensor change did not solve this cars problem, </strong><hr></blockquote>

O2 sensors are only relied on and used for steady state feedback. Any transitional or heavy load operation is strictly controlled by lookup tables and feed forward input. Replacing O2 sensors in hope of curing a "power" problem is a waste of ones money.


Originally posted by 4 Jaw Chuck:
<strong> I am now investigating fuel pressure and other possibilities. The abnormal behaviour displayed only takes place during full throttle or high fuel demand conditions and the Pulse Width for bank 2 goes to 10 while bank 1 hovers around 6-7. It takes some time for the sensor to recover hence the low volt readings during recovery. </strong><hr></blockquote>

Are your PW's processed or real-time (scope) readings. Datastream for most domestics is processed and unreliable for a full diagnostic. Chryslers have a real-time display but the refresh rate is too slow.


Originally posted by 4 Jaw Chuck:
<strong> Also the new sensor goes high (1.00+ volts) during this period leading me to believe a low fuel pressure condition is causing the problem with the computer unable to correct the condition without going full rich and then full lean, could be injectors too? </strong><hr></blockquote>

All conventional O2's will go full scale under heavy load and accel conditions. 1 volt is equivalent to approximately 2.5-3% CO output. Full load fuelling is around 5%. This is why one cannot use a conventional O2 sensor for setting load fuel.

Originally posted by 4 Jaw Chuck:
<strong> Fuel trim of course takes some time to react and will go from full lean (-20) to full rich (+41) trying to correct the problem which leads to some driveability problems before it recovers. What are the possibilities that a blocked cat is causing this problem? The car was picked up with a bad thermostat and other problems with the cooling system and consequently ran cold for a long time. This is apparently bad for converters because they load up and overheat and then collapse, cutting them off would not bother me a bit? </strong><hr></blockquote>

STFT will bounce around like a cat on a hot tin roof. Under heavy load the learned fuel trim numbers will be added base calcs but the fuel trim for most part is a direct reflection of the lookup tables and will remain static if the heavy load remains static.

The worst thing for cats is lean engine operation with a misfire, or a slightly rich operation with an air pump dumping continuously. A rich running engine will just put the fire out. I firmly believe in cats. My favourite saying is that if you are not part of the solution you are part of the problem. One would not like a neighbour dumping their household garbage in ones back yard, why should it be any different for the automobile. I have found cats do not affect power until the engine reaches a continuous load of 85% or more for long periods. I have even run accelleration tests on stock GM trucks and found that removal of the cat actually slows the truck down. Where the idea of more power comes from is that the exhaust makes more noise, it is a psychological thing.

In some applications removal of the cat will enable EGR and knock control codes.

I can say that I have yet to see a muffler shop dual exhaust system that has been fabricated in place of the factory system to show lower backpressure than the stock system. I have dyno'd some that were 3 times the OE backpressure but made a heck of a lot of noise.

A piece of info that I would suggest, is that any time a component is replaced, that is an output or input of the fueling strategy, it is best to clear the memory of the ECM by an ECM disconnect. This allows the system to revert to base programming as it was one day one. If the system is 100% intact the engine should respond correctly. This suggestion does not work that well with OBDII applications (96-up). It is best to use a scan tool to perform a reset.
 
#56 ·
Well chuck I must admit I'm a bit confused reading what we've got here. In your steps you say WOT then right to recovery. When you hit WOT are the O2 sensors going rich like they should? Also I don't have access to any manuals at home but doesn't a 111 code mean systems normal (also code 11 in two digit systems). If your Check Engine Light is coming on she must be storing a code, do you have a code definition? Check to make sure you don't have any vacuum leaks around the intake. Check for any problems with the secondary induction system, those SHOs have that weird intake systems (like 4 barrel secondaries). Also if you hook up a fuel pressure gauge (cheap at most auto parts stores) check for correct pressure regulator operation. Most Fords have a small schrader valve in the fuel rail. Regulators are generally in the return side of the system and when vacuum is removed the regulator causes a restriction in the return line and results in a pressure increase. If your fuel pressure is below spec see if you can pinch off your return line (usually a section of hose between the engine and frame, make sure you don't pinch off the supply hose) and monitor pressure. If pressure max's out with the return line pinched it could be a bad regulator. Also disconnect the vacuum line to the regulator and make sure it's dry, if there's gas in it the regulator is shot.
 
#57 ·
Originally posted by dmorris1200:
<strong> quote:
--------------------------------------------------------------------------------
It could run AFR's of 18:1. This would allow compliance without running an EGR. At AFR's in that range NOX drops.
--------------------------------------------------------------------------------

bcwrench, great reading but I must ask about this statement. Just wondering how running a leaner mixture such as 18:1 could reduce NOX when almost every car I've repaired for state inspection for high NOX has usually failed because it was running too lean. I can see the benefits of running a system like the Honda system but I can't see how reducing NOX could be one of them unless the PCM/ECM was programmed to drop that AFR when not idling or under load. I do not know the entire fuel strategies for that system.
</strong><hr></blockquote>

If you look at a 5 gas chart you will see the NOX rises and reaches a peak at around 16.5-17:1 AFR. This is an area where, without computer control, most engines will not run nicely if at all. In fact most engines let you know something is wrong before 16:1 is reached. This the area you are most likely seeing. Even "average" stoichiometric operation can have high NOX because numerous things will increase it. Leaner mixtures are just one.

Hondas main goal was not to just reduce NOX but to boost fuel economy and this required quite an elaborate setup.

After the NOX peak is reached it takes a big dive. Chrysler tried a lean burn concept back in the late 70's that didn't work well because they tried working it with a conventional carb. IT also wouldn't allow operation much less than 15.5, so they still had to use an EGR system.
 
#58 ·
Chrysler tried a lean burn concept back in the late 70's that didn't work well because they tried working it with a conventional carb <hr></blockquote>

We were just discussing one of those not too long ago here. Haven't actually seen one in years.

If you look at a 5 gas chart you will see the NOX rises and reaches a peak at around 16.5-17:1 AFR. <hr></blockquote>

Don't happen to have one of those handy but I know that most of the pre-OBDII cars that fail for NOX usually have a malfunctioning EGR system or a vacuum leak of some sort (not always enough to cause driveability problems but enough to raise combustion temps- ex. leaking throttle body gaskets, cracked vacuum lines, etc.). One nice thing about OBDII systems is that they have the ability to check their own systems, like Fords DPFE that checks to see that the EGR is not only opening but functioning by looking for a pressure change in the EGR tube. After working on OBDII systems I hate to get stuck with an old OBDI car.
 
#59 ·
Thanks DM, I did exactly what you said today and all is normal. Fuel pressure is normal during engine running conditions too. The O2 sensors are operating properly and go full rich during WOT, it's in the recovery phase that lasts for a good 10 minutes after that I am having the problem. I am thinking now I may have a bad injector on that bank. I am going to get the graphing multimeter from work and test each injector, thank God for these Mitchell manuals on CD that I have or else I would be up a tree without a stick. Funny thing I found while perusing the specs of the car was a listing for a supercharger that was never installed on these vehicles from the factory.

The other possibility is I have a bad plug wire that is misfiring or even a set of bad plugs, I have to machine down a socket to pull them so I intend to do that tomorrow. I have a feeling it is a mechanical fault not electronic the more I look at it, the all OK KOEO/KOER code 111 says it all.

I still can't find a PCV on this thing because that was the first thing I suspected, I don't think it has one? Weird? <img src="confused.gif" border="0">

[ March 23, 2003: Message edited by: 4 Jaw Chuck ]</p>
 
#60 ·
You guys have me sitting on the edge of my chair with all this state of the art fuel management info. D Morris, I have a question: back in the late 80s early 90s I was in an un-named state that required sniff tests on all cars and light trucks, regardless of year or equipment. My 36 PU with SBC 350 would pass with flying colors, no CAT, EGR, Pump etc. They went to a centralized computer system to where there was no cheating and it still remained well in the green. They finally determined that only smog equiped years were required to test so did not do this but about 3years. Just why was a plain Jane motor able to pass these tests?

Trees
 
#62 ·
Just why was a plain Jane motor able to pass these tests? <hr></blockquote>
Well I can only speak from the experience I have. Growing up in Jersey everything got what we called a standing idle sniff test for inspection. The main difference is the standards varied according to vehicle year. A 1968 vehicle did not have to meat the same requirement as a 1988 vehicle. This being the case a 1968 mustang for example could be outputing 475 HC's and still pass inspection while a 1988 mustang might fail at only 150 HC's. The 1988 is running cleaner but still failed state inspection. This was one big reason why many people tried to hold on to pre-1975 vehicles, their emmissions standards were much more relaxed because they were produced before cat converters were required by law. The states could not expect peoples cars to meet strict emmissions requirements that weren't in effect when the cars were produced. All my cars have always passed state inspection, even my mustang has a sticker on it and passed for many years in Jersey before I moved to Maine. It's also allowed like 5 or 600 HC's and a whole lot of CO's because of it's age. Hope I answered your question well enough.
Steve, most likely your sensors are working or you would have a check engine light on. They may not cycle as well as when they were new (we call it 'getting tired' in the field) but they are most likely still getting the job done. :D
 
#65 ·
Quick update for everyone, the O2 sensor has solved most of the problems. The swing readings I was getting on recovery appear to have been "re-learning" related. I tried yesterday to reproduce the symtoms and was unsucessful no matter how hard I tried. Interestingly enough the CID code came up which indicates a bad cam position sensor during my testing (tach stopped reading), so that sensor is next. Cylinder balance tests came out fine and my remote tach indicated near perfect idle drop on each cylinder while the computer was running it's balance tests.

Looks like I should have driven the car for awhile to let the computer get cozy with the new components, everything is good now except for the CPS. I might add that the rad leak turned out to be a poor seal on the block heater (someone pulled out the o-ring and siliconed it in?) which in turn sprayed the rad with coolant and caused the starter to get flooded which destroyed the brushes. The domino effect in it's finest form.

It looks like as long as I can keep POLICE CARSfrom hitting the darn thing I should be OK. :p Insurance adjusters look at it Wednesday.
 
#66 ·
Glad to hear things are going ok chuck. bc brought up a good point that is easily overlooked. Even with all the advances in the systems the speed that the info makes it to the scanner is still rather slow. What I have found is that on scanners like my Snap-On or Fords NGS if you reduce the amount of PID's you're looking at the info travels faster. If I only bring up O2 sensors I get a much more accurate view of what they're doing but if I bring up twenty sensors the rate at which I see them operate is greatly reduced.
 
#67 ·
That makes a lot of sense DM, I can't thank you enough and the others on this thread who have helped me with this problem. The learning curve was steep but I feel closer to the summit than before. It certainly has illustrated how hard a job mechanics have trying to fix these cars in as short a time as possible (especially without an extended test drive), I can certainly understand why some just start throwing parts at it when you consider the diagnostic time involved on some of these things. It seems you either pay for diagnostic time or parts, at least with parts your getting actual components with your money even if your chasing your tail half the time. I guess the happy medium is somewhere in-between.

One thing is for sure, Ford not publishing the MAPS and circuit logic has to be some kind of crime. Having to guess or use intuition to determine actual operating parameters and logic takes a lot of experience and time looking at bad components and how they affect the system as a whole.

My hat goes off to those of you who do this for a living. The profession is certainly under appreciated especially when you consider the number of different makes and models and powertrain combinations, the possibilities are truly endless.
 
#68 ·
Having to guess or use intuition to determine actual operating parameters and logic takes a lot of experience and time looking at bad components and how they affect the system as a whole. <hr></blockquote>

In this business this is what seperates the men from the boys and what seperates the A,B, and C techs in the field. It's also a big reason why even with education there is still so much that just can't be taught, it must be learned in the field. I thought I knew alot when I got out of college until it came time to apply what I had learned with my first hard drivability repair. Unfortunately there is also no lesson anywhere for correctly diagnosing that chaffed wiring harness hidden behind the firewall ;) . Either way glad we were all able to help, had a lot of fun with this one. :D
 
#71 ·
Originally posted by 5.0Towncar:
<strong>I don't think I have check engine light...All I have is a dummy display. The basic "check engine temp" "oil pressure low" "charge system" and so on...Never seen it say "you screwed up your motor" ;) ....Steve</strong><hr></blockquote>

Steve, just to let you know that for whatever reason certain year Fords didn't have a Check Engine light in the dash. These cars still retain fault codes just like any other car only they don't turn on any light so the customer doesn't know. I have had to explain more than once to customers who don't want to pay for diagnostic code reading because they didn't see any warning light. The other posibility is your car has a light but the bulb is blown. You may want to see if you can find someone who can hook your car up to a scanner and check for any codes.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top