Lowering frame to change rear HO2S, pull on front head bolts?

[hjb981]

1. Bank 1 keeps reverting to the 20's while bank 2 is reverting back to 12 to 13. I wonder why. Maybe members familiar with this will comment. I need to look the manual over when I get a chance, but I am running in an out right now
   

Much appreciated.

I think that it is strange that the long term fuel trim jumps around so much. I would have thought that the long term trim should change rather slowly, and that the short term trim should change more quickly.

[hjb981]

The values for fuel trim above are for approximately 55 mph cruise, with varying torque (see the edit in the above post, just over the table). In the beginning of "Engine Controls" in the FSM, it says "Varies" for both the long and short term fuel trims, so maybe they should vary?

[BodybyFisher]

I am going to do some reading later on these parameters and those codes and see what I can find out. Will let you know. Hopefully someone can shed some light on this issue

[hjb981]

FSM about fuel trim for bank 1 (equivalent text for Bn 2):

Short Term FT Bn 1: Display shows counts from 0 to

255 and percentage from -100% to 99%. The

normal position for the short term fuel trim is

128 counts or 0%. A count greater than 128 or a

percentage greater than 0% indicates that time

is being added to the injector pulse width, increasing

the amount of fuel to the engine to compensate for a

lean condition sensed by the bank 1 (rear) HO2S.

A count less than 128 or less than 0% indicates time

is being subtracted from the injector pulse width,

reducing the amount fuel to the engine to

compensate for a rich condition sensed at the

bank 1 (rear) HO2S.

Long Term FT Bn 1: Display shows counts from

0 to 255 and percentage from -100% to 99%.

The normal position for the long term fuel trim is

128 counts or 0%. The long term fuel trim is based

on short term fuel trim learned values. It is stored in

memory blocks of the long term memory according

to MAP and RPM values. A count greater than 128

or a percentage greater than 0% indicates that time

is being added to the injector pulse width, increasing

the amount of fuel to the engine. A count less than

128 or a percentage less than 0% indicates time is

being subtracted from the injector pulse width,

reducing the amount fuel to the engine.

Given that information, perhaps the reason that my LT fuel trim jumps around so much is that I jump between different values for different MAP and RPM readings. I still wonder if it should vary that much though.

[hjb981]

When I started the car today, I could here a sound from the exhaust, that sounded similarly to tapping an air pressure valve or something like that. It could be described as pffff---------pfff-----------pfff--------------pfff and so on, with one pfff per camshaft rotation (one for every two crankshaft rotations). It was not noticeable enough that I would have heard it from the front -- I only heard it because I purposefully went back and carefully listened to the exhaust. The exhaust smelled a bit like gasoline. I think that means that one cylinder was misfiring, giving the pfff sound when it opened the valve. After a couple of minutes or so, it went away. It was -12C / 10F when I started the car, so pretty cold.

Later, I went to the mandatory safety and smog inspection for this year. I had the following exhaust values at idle:

vol % CO: 0.91

vol. ppm HC: 256

vol. % CO2: 13.2

vol. % O2: 1.84

lambda: 1.051

That would normally have failed a car from 1997, but because I imported it under special circumstances, I go by the emissions requirements from the early 70s (the rules are made to be applicable to cars imported from any country, including those not requiring any emissions control systems). HC, for example, would be about 5 or so for a new car of today (I think 100 is the limit to pass), and it should not be that much different on any gasoline car with a 3-way CAT. HC did drop down when revving the engine a bit (not sure how much) and I could not smell any gasoline in the exhaust.

[BodybyFisher]

If you have any current codes your emissions will be high. Any current emissions codes will fail a car here. Where are you at this point, do you still have emissions codes, did you replace the pre-cat O2?

[hjb981]

There were NO codes present when I did the check-up, and that was after a cold start, 10 minutes of driving to a lunch place, and then driving the car around for 20 minutes, including a couple of stretches on the highway at about 75 mph. It is impossible to do WOT:s to clean things out, since there is not enough traction at the moment, so all of it is rather "easy" driving. Even before, the codes did not always appear immediately, but they would usually have surfaced after that amount of driving.

The pre-cat sensor should arrive on Monday (two of them, actually, so post-cat as well). Current codes would fail a car here also, but even if I would have had them, the normal rules do not apply to my car because I imported it as part of moving back to Sweden after living abroad and using the car abroad for an extended period of time. That being said, I still want it to run properly.

I think that there are two issues at hand, that may or may not have the same cause.

First, there is the appearance of those codes, which started to happen after I changed the rear manifold O2 sensor. This possibly also includes the values from the smog check and the different fuel trims for the two banks.

Then, there is the misfiring at cold start-up.

One thing to note is that both manifold O2 sensors give good readings, and at the same time the pre-cat sensor reads rich. We will see if this continues after I change that sensor, but if it does (I am not convinced that the sensor really is bad, because if it is it very quickly went bad after I got the two manifold ones), I have a theory: what if I did not achieve an air-tight seal when I installed the O2 sensor. In an exhaust stream (especially in one bank of a V8 engine) there will be pressure waves and low pressure in between. During that low pressure, outside air could actually leak into the exhaust system. If that happened, it would make the O2 sensor read lean, since it would sense extra oxygen. The PCM would then richen the mixture to compensate, resulting in to much fuel and a larger value for the fuel trim on that bank. If the exhaust stream is hot enough to burn some of that fuel together with the oxygen that was leaked in (I do not know if that is the case), then it could make the pre-cat sensor read rich because there would be too little oxygen left. However, the emissions test value for oxygen (1.84%) seems a bit high (I guess) so maybe the sensor is reading incorrectly after all. The lambda value in the exhaust stream was measured to 1.05, which would indicate a lean condition (too much oxygen in the exhaust stream).

Do you think there is a risk of an untight air seal for the sensor? Maybe a grain of rust or dirt came in between the seal and manifold when I tightened it (it is hard to see and clean properly down there).

Another idea is that the copper paste that was supplied with the oxygen sensor (AC Delco AFS 109 HO2S) as an anti-seize compound traveled down the exhaust stream and poisoned the pre-cat sensor, but that seems unlikely, since they should not supply an anti-seizing compound that could poison the sensors.

[BodybyFisher]

I am sure you have an air tight seal if the O2 sensors are tight.

The misfire at cold start up is an ominous sign. That is a symptom of the head gasket being bad. Are you overheating at all? Are your temps normal?

If you have access to a test kit check your coolant for combustion by products.

If that is good, maybe you have a leaky injector that is relieving itself in your cylinder on shut down.....that would explain your rich mixture also. You can have your OIL checked for fuel at Blackstone Labs or a local shop if you have one that can do that.

Check your ignition wires with an ohm meter to be sure you dont have a bad wire causing this miss.

[hjb981]

One more thing: The drivability of the car is fine. The engine responds well, is powerful, runs smoothly, idles smoothly (sometimes with an occasional very slight irregularity, but that is when I really try to feel it, and I cannot remember it ever idling better than it does now, so I think I should consider it fine). It is hard to judge the fuel consumption, but I do not think that it is anything out of the ordinary, and considering the temperatures we have now, I would expect it to be slightly higher than in summer (which I have seen every year).

Dirty injectors could also be a cause of increased HC (fuel in the exhaust), since the O2 sensors only sense oxygen. They would sense extra oxygen from any cylinder with a dirty injector, and then increase fuel trim. It would, however, NOT sense the extra fuel from the other three cylinders, resulting in a rich-running engine.

[BodybyFisher]

Another way to determine if you have a leaky injector is to gently lift your fuel rail and prop it up and pressurize it by turning on your key. See if you have any dripping taking place. While you had the fuel rail up, I would change the injector o-rings. DO NOT touch or try to clean the injector tips.

[hjb981]

There is no overheating at all, and I do not see any temperature spikes after climbing long hills. Also, the coolant stays in place. The radiator hose pressurizes and keeps pressure on shut-down, but it does not get excessively hard or blown up like a balloon. I have changed my coolant + pellets, (1.5 years since last change, planned next this spring). The temperatures can go up a little if I go from highway to city driving or extended idling, but that is normal and has been that way for the 8.5 years that I have had the car. I think that oil is leaking into the cylinder from a worn valve guide (if it was coolant, I think that it should have cleaned the cylinder and plug rather than cake it up with a very thick layer of deposits). Something has been going on with that cylinder for a long time, such that if I drive the car without warming it up and then let it sit for a long time before starting it again, it can develop a rather loud valve tap or knock. It can do it a little otherwise too, but not very much. And it only happens on cold starts and quickly disappears. I am almost sure that it is not a cold carbon rap. And it is always only one tap per camshaft revolution (one per every other crankshaft revolution). I think that one of the valves slowly leaks oil that goes into the cylinder or on top of the valve if it is closed, and maybe gets on the spark plug when the engine is started and causes the cold misfire. That would also explain the knocking sound. I think that the valve sticks a bit somehow when the engine is only driven a bit and then shut off, and that is why there would be a louder knock if I did that (which I never do for that reason).

[BodybyFisher]

I would lift the fuel rail and check for an injector leak

[hjb981]

Yes, that is good idea, I wrote my previous post before I saw yours. I am also thinking of doing a compression test and a fuel pressure test (although I do not have the equipment for that at the moment). Where do you get the o-rings? The manual talks about two materials in different colors (brown and black).

Is there anything I should think of when lifting the fuel rail?

[hjb981]

Or, rather, any tips on how to make it as easy as possible to lift the fuel rail?

[BodybyFisher]

The guru used to say to use a block of wood to hold the rail up and turn the key on to pressurize the rail. When I needed the o-rings I just went to the dealer and they gave me what it called for, maybe someone else here will comment on that.

Would it help if I sent you a couple of bottles of Techron in the mail? Will that get through customs? Once you replace that O2 sensor, if it still runs rich, you need to tackle that quickly or it will damage your CAT

[hjb981]

Thanks, that is very nice of you to offer! I think that customs would not be a problem, but shipping gets trickier if it is hazardous (I guess that it is flammable). I will investigate that. I found something on UPS about shipping hazardous goods internationally, and I will have a look what the rules are. Hopefully it is just a matter of labeling the package and not too cost-prohibitive.

Off topic: I have the radio on in the background as I am writing, and I just heard about the horrible tragedy in Newtown. What a sad and terrible thing, that someone would do something like that. I feel sad for those involved.

[BodybyFisher]

I could pack the Techron very good and seal it in a separate container in case it leaked. Cadillac/GM recommends using it at each oil change. Let me know what you find, I have not doubt it will be expensive to ship. The guru did not like 'off the shelf' injector cleaners as they could damage the motor windings in the fuel pump and the injectors. How are you going to get the car up to do the pre-cat O2 sensor, ramps or a floor jack?

Yes this is a terrible tragedy, its been on the news all day, 18 children less than 4th grade were killed. More info will filter out as the day goes on as to what caused this nut to do something so horrible.

[hjb981]

Getting the car up:

Alternative 1: jack it up and put it on stands, then crawl under on my back in 3 inches of snow/slush/ice (depending on temperature).

Alternative 2: rent a lift (indoors).

The lift is 15 bucks per hour, which I consider very much worth it at the moment . So it will be a lift (the kind that you drive onto, not the kind where the wheels hang freely, but that should not matter much for changing the O2 sensors).

I made a trip this weekend, and on the way there I had quite strong headwinds and lots of snow coming down. There were also soft snow on the ground that I was driving through, so although it used more fuel than normal (about 20 mpg, 2 hours highway), I was unsure what to make of it. It did look like the instantaneous mileage on the DIC went up very high quite easily, though. On the way home, conditions were more normal, and the 23 mpg I got then is something I would consider normal for the driving conditions, based on previous experience, and the instantaneous mileage appeared to behave normally. I also did not get any codes on the way home (2 hours highway), although after stopping at a store, the P0138 and MIL came back on the 10-minute trip home. The new O2 sensors should arrive tomorrow, and I will see what happens after installing them. Btw, the engine ran on all cylinders right from the cold start on both of these trips, as far as I could tell.

[hjb981]

Yesterday, I removed the rear manifold sensor, cleaned off all of the anti-seize compound from the area where the gasket contacts the manifold, put new anti-seize compound on the sensor threads, and re-installed it, using a torque wrench set to 35 ft-lbs. I used a toothbrush to spread anti-seize all over the threads, but without putting so much on that it was pushed up towards the gasket. That may have been unnecessary, but at least now I know that the gasket seating area was clean when the sensor was installed, and that it was properly torqued. It was an easy and quick job anyways, now that I had a good tool (the crow's foot) and the sensor was not stuck.

Then I noticed that my rear-end was a bit too high. Sigh. The exhaust valve on the refurbished compressor I bought two years ago seems to be stuck (the compressor did not run more than a few seconds on start-up, so it is not the sensing arm that is broken). I just disconnected the air lines at the compressor, then reconnected them, let the car level once, and then pulled the fuse. That will work until I have time to fix it properly (but if I need to put heavy things in the car, I will have to let it level again and later manually let the pressure out again). I actually have two spare compressors, one that does not run, and one that has a frozen exhaust valve, so maybe I can fix a functioning one from those two.

After waiting for the delayed pre- and post-cat oxygen sensors (FedEx considers normal winter (no storm, just slippery roads like most of the time in winter) to be "conditions beyond their control"), and then waiting for myself to have some time available , I replaced the pre-cat sensor today. That was very easy, and it came loose without much effort at all. The post-cat sensor, on the other hand, did not come off, but I was not even able to put much torque on it, since the O2 sensor removal tool did not grip it properly. Instead it deformed the flanges on the O2 sensor. I will try later with a different tool, but at least I got the sensor with the trouble codes replaced. It was actually blacker than the two in the manifolds.

I have not seen the codes reappear yet, but it is too early to say. I will see what happens when I drive home for Christmas tomorrow.

Footnote: after having worked with the manifold sensors (AFS 109) and the pre- and post-cat sensors (AFS 98) I think that the difference between them is the cables are led through a more heavy-duty sock that is also sealed towards the sensor on the sensors that sit close to the CAT. Probably done that way because they are in a more exposed position. Except for that, the sensors look identical.

[hjb981]

I have done some city driving, and 45 + 60 minutes of highway driving now, and the PCM seems happy (no codes). However, looking at scatter plots of the logged O2 sensor voltage readings, the two manifold sensors and the pre-cat sensors all swing back and forth between high and low voltage, but while the two manifold sensors are slightly more heavy on the low-voltage side (lean reading), the pre-cat sensor is noticably heavier on the high-voltage (rich reading) side. I am not sure if this is something that could just be an effect of the fact that while the manifold sensors get individual exhaust streams after the PCM varies the mixture slightly back and forth a few times per second, the pre-cat sensor gets the combined exhaust stream of the two. Not sure if that should cause more rich readings for the pre-cat sensor. I guess that I am just saying that I cannot really say if it is normal or not. I think I will let the PCM decide .

[BodybyFisher]

No codes, that is great good for you I think you have solved it if you drove that far. If you want me to send you a couple of large bottles of Techron, PM my your address, Ill pack it up really secure.

Merry Christmas

[hjb981]

Thanks again, I will get back a few days into January. I am on vacation (and also away from my car) until then.

Happy New Year (a little in advance)

[BodybyFisher]

Happy New Year!

[hjb981]

I have not had any codes reoccurring after changing the pre-CAT HO2S, so I guess that when the manifold sensors were changed, the PCM could vary the fuel mixture more quickly, and then the sluggish old sensor could not keep up anymore. The only one left to change now is the post-CAT. I could not get it loose when I changed the pre-CAT, and the tool actually would slip on the sensor. I think I will get it loose if I go to the lift again and do it immediately, when the exhaust is still warm (I did the pre-CAT first the last time I tried). I will also bring my crow's foot, because it is a very tight fit on the sensor, tighter than the HO2S tool that I have.

I think that the car runs fine now, but that it does use somewhat more fuel than normal (and this may have slowly evolved over the last few years). Dirty injectors could contribute to that, and possibly also a bad post-CAT sensor. According to the FSM, the post-CAT sensor has a small influence on fuel trim.

I have investigated shipping of hazardous materials, and it seems that for one, it is very expensive. The minimum fee for FedEx (they had an online price list) was quite a bit north of 100 dollars (and that does not include the standard shipping charges, which get added. It also appears that the shipper would have to attend some kind of seminar to be qualified to pack the materials. I am not 100% sure of that, but to sum it up, shipping Techron internationally seems to be very expensive and complicated.

I read up on Techron, and it seems that the active compound is polyetheramine, PEA, (http://www.techron.com/what-is-techron/proven-science.aspx). Luckily, Redline SI-1 is also contains PEA (http://www.redlineoil.com/content/files/tech/MSDS%20SI-1%206-11.PDF), and is available in Sweden. Naturally, there may be some other differences, since PEA is not the only ingredient. Further, "poly" means several, but does not state how many, so the molecular length (weight) could be different. If I had it available, I would just buy Techron, but I think that SI-1 is likely to have a similar effect, and I will try it (I have used it before as preventive maintenance, but it has been a few years now).

Mike, thanks once again for your offer to help with shipping! At the moment, I think that it looks like too much hassle for you and too much money for me.

[BodybyFisher]

I am glad its running well and the code stopped popping up.

Good job on researching the Techron. We have had extensive discussions regarding injector cleaner and its detrimental affect on the fuel pump and injector windings. Techron according to the guru does not have these negative affects when used correctly.

I have used Techron when I take a trip to run it through the system as quickly as possible so the fuel pump and injector dont marinate in the stuff.

Glad to help, let us know how we can help. By the way its never a hassle helping a fellow Cadillac owner around the world..... Mike