Jump to content
CaddyInfo Cadillac Forum

Understanding Fault Tree for E046 PCM Code


Recommended Posts

Could use some help/advice on diagnosing an E046 trouble code for a '92 Eldorado with the 4.9 liter engine. I also failed my last smog check because of excessive NOx, btw. I'm thinking the two things are related.

E046 is the the 'Left to Right Bank Fueling Difference' that will produce a Service Engine Soon light. My E046 is the single current code but the SES only comes on intermittently.

My due diligence of Googling other people's experience with the E046 on this and other forums has been a bit disappointing.

Many of the E046 complainers don't seem to use a Factory Service Manual for diagnosis, and most of the E046 threads I've seen have the guys replacing O2 sensors and fuel injection parts without really taking a more scientific approach (i.e. not throwing parts at the problem and hoping it will go away).  Apparently the O2 sensors can be bad and produce an E046 without spitting out a more O2-specific trouble code.

Anyway, the fault tree for E046 can be found on my attachment IMG_0379.jpg with some more info on E046 found in attachment IMG_0385.jpg

Going through the steps of the tree, I am witnessing the ED35 (Right O2 sensor cross count) bouncing around from 0 to a high of 34, while the ED34 (Left O2 sensor cross count) varying between 0 to a high of about 24. These values do not seem to linearly increase as the car accelerates or travels at a higher rate of speed. Even just idling in Park in my driveway can produce non-zero cross counts.

What should I do now?

I have seen previous posts with differing opinions, e.g. the actual numerical value doesn't matter, it's the frequency of the numbers changing, the 0 value is no good, a high number is better, etc. and so forth.

Some previous posters don't seem to have a really good grasp of what 'cross count' really means (me included).

I'm taking an educated guess that if the cross count reading is '5' as an example, this means the O2 sensor is dictating 5 changes per second from a rich to a lean burn. Is that correct?

So I have many more questions than answers, is there somebody out there that can help me with this dilemma?

I feel that just blindly replacing O2 sensors as historically been done might be my last resort otherwise.

Thanks, cadillac fellas.

IMG_0379 copy.jpg

IMG_0385 copy 2.jpg

Link to comment
Share on other sites


What was the status of the auto light on the A/C control panel?

Cross counts are the number of times the sine wave signal ( in this case it is O2 sensor voltage ) crosses a base line voltage per second. The PCM uses this "count" to determine fuel/air mixture. For example a "lazy" or slow O2 sensor would not cross the base line a given number of times in that second. So, if the PCM wants to see the signal cross the base line 5 times a second and it is only crossing 2 times in a second, the PCM detects that as a problem.

Looking at the flow chart it appears the PCM is looking for something above 3 for normal operation.

 

Edited by OldCadTech

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

hello and thanks, I really appreciate you smart and experienced folks here on the forums.

 

I will double check the Auto light tonight.  I just realized there are two 'Auto' light icons on the A/C display!

In the attachment below, the Service Manual shows an 'Auto' icon on the left side that indicates Closed or Open Loop Mode status.

The right side has an 'Auto' icon that will say if the car is in Park/Neutral or not P/N, correct?

Which one should I be focusing on?

Doesn't Open or Closed also mean when the thermostat opens to let the hot coolant flow through the upper radiator hose too?

 

IMG_0387 copy.jpg

Link to comment
Share on other sites

Personally I prefer using a scan tool.

Anyway....It may make more sense if you think of the engine as 2 different 4 cylinder engines attached at the bottom. Each bank being 1 of the 4 cylinder engines. The computer is thinking....huh...they should both be pretty close in readings. But that is not the case with this code.

Also on the chart (Figure 9 above)....ECON is the right bank O2 sensor.

                                                          DE FOG is the left bank O2 sensor. 

 

Likely a sleepy O2 sensor.

 

   

Link to comment
Share on other sites

The one on the left under econ, it indicates the engine is running in closed loop.

The flow chart wants to make sure the engine is running in closed loop before proceeding with the trouble tree.

Basically, Open loop means the PCM is running on a block learn or a preset group of engine parameters and is ignoring the O2 sensors until they should be hot enough to operate correctly and the engine no longer needs a richer mixture, such as when the engine is cold. Closed loop is when the PCM is using the data from the O2 sensor to control fuel mixture. There are a LOT of variables such as block learn, long and short term fuel trim and integers. Way to much detail for a code like this.

Please take note of the " Likely causes of an E046" in the FSM, you need to be certain that all these possible causes are RULED OUT before proceeding with the E046 trouble code flow chart. The FSM is assuming NONE of those conditions exist, otherwise the test is NOT valid.

 

 

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

great information all around!

i'm certainly learning something beyond what I got just from flipping through the service manual.

 

So to answer the previous question, the PCM is showing 'Closed Loop' running and the Auto icon is lit up.

Assuming that the "Likely causes of an E046" is ruled out (which I intend to double check) wouldn't the fault tree still indicate bad or lazy O2 sensors because both the right and left sensors had stretches of zero value cross counts during my test drives?

Link to comment
Share on other sites

That is a probability, but, due to the fact both of them indicated 0 cross counts, I'm thinking more likely a fuel injector problem or a vacuum leak. It could still be an O2 sensor but don't throw parts at it until you run out of money.

If you have a vacuum gauge, install it and note whether the needle is steady or fluctuates, what is the reading at idle, how does the engine react if you snap the throttle, what is the reading and how does the needle react at a steady 2000 rpm? See the attachment...

 

 

vacuum-readings.png

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

981026GM03-229.JPG

You can get some vacuum hose and a "T" fitting and connect to 102 or, if the other ports above it are capped off remove the cap and install the vacuum gauge there.

The gauge will fluctuate like the picture on the bottom row in the middle...

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

The other thing I kept forgetting and was going to mention a long time ago before I got side-tracked on the E046 is that the emission test failure for NOX is more likely an EGR problem. Remove the egr valve and make sure the ports are clean and not restricted. The E046 and a restricted or plugged vacuum port could all be related as well.

There are vacuum passages under the throttle body that are infamous for plugging also....

How's that for confusing the issue?

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

Sounds like a good comprehensive plan!

So it looks like I should start with checking the vacuum in the throttle body first, correct?

I'm almost sure the end of my vacuum gauge is threaded.

Is using a 'T' connection my only option?  How would I block off the open end of the T, please?

Link to comment
Share on other sites

You will need some different sizes of vacuum hose & you'll have to do a little plumbing if it is threaded.

Disconnect the elbow from the line and use the long side of the T in the line, hook the gauge into the short fitting.62637_zzz_500.jpg

Yes check the vacuum first...

Edited by OldCadTech

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

Hi, I dug around in the garage and discovered I have a Sun brand vacuum gauge that looks very similar to the black one in your picture.

Is it okay to simply push the end of the rubber vacuum hose onto the nipple located at 102 ?

The hose isn't threaded, it looks like yours.

 

 

Link to comment
Share on other sites

You need that vacuum line connected, that is why you need the T and another piece of hose, so you can "tap" into the line. Unless one of the other nipples has a cap on it, then you can connect directly.

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

You can fit vacuum hose inside a larger vacuum hose and it will seal, but it has to be a tight fit.

The best place is to disconnect the vacuum hose from the # 91 fitting and connect the vacuum gauge there but it requires an adapter like the one to the left of the gauge and hose in the vacuum gauge picture above or several different sizes of vacuum hose, the hose is cheap.

 

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

it took a while to locate the right attachment for the big nipple # 91 (brake booster hose fitting) but I finally did it, and here are the results:

At a coolant temperature of 207 deg F, after 20 minutes of warming up

1) Vacuum at idle (650 rpm) was a fairly steady 18.5 " Hg

2) Vacuum at 2000 rpm was 21.5 " Hg

3) Vacuum with snapping throttle to about 2000 rpm caused the needle to fall to zero, then quickly jump up to 23 " Hg then go back to the idle vacuum of 18.5" Hg

Link to comment
Share on other sites

That's pretty good, a little on the low end but not too ba.

40 minutes ago, area310dude said:

was a fairly steady 18.5 " Hg

 

When you say fairly steady how much fluctuation was there?

Here is a more detailed chart, which one is the closest to your condition?

vacuum.GIF

 

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

At idle, there is some rapid waver that is plus or minus maybe 0.20 " Hg about the 18.5 " Hg reading.

The wavering is less than a quarter inch in either direction, in other words.

If that waver is acceptable, the Normal engine picture seems to be what I'm seeing.

Link to comment
Share on other sites

My only concern is, it is on the bottom of the range. If it was me, I would check base timing and if it was okay, I'd dial in a couple of degrees of advance and see where the needle points after that. Do you have a timing light? They're a little archaic now but I think Sears still carries them. If you do have a timing light, there is a method to put the PCM in set timing mode and there is a special wrench that makes loosening and tightening the hold down bolt a breeze, so let me know before you try to make any adjustments to the timing.

The bolt is a 15mm head machine screw. You can use a crows foot wrench on it but it takes a swivel ( universal joint ) & an extension and it takes a bit of finesse.

wrench.jpg

Lets move to the EGR valve for a minute. To check the EGR operation you will possibly need a mighty-vac, hand vacuum pump. The first check is to check if the pintle can even be moved. With the engine at idle wrap your fingers around the valve and "lift" the diaphragm, the engine should die. If the diaphragm will not move, Replace the EGR  valve and gasket. If it stays running replace the EGR valve and gasket. If it dies, you will need to apply vacuum to the valve. IF you don't have a hand vacuum pump, you can disconnect the vacuum line from the vacuum gauge, leaving the adapter connected at the #91 fitting. Plug the line with a golf tee or whatever will work, start the engine, remove the plug and cover the hose with your finger. Momentarily connect it to the EGR vacuum port. do this as quickly as you can keeping the vacuum leak time as close to zero as you can, and the engine should die. If it stays running or simply runs rough replace the EGR valve.

mvac.jpeg

egrvlv.jpeg egr2.gif

In the picture above the diaphragm is the silver looking plate above the brass colored brace.

Let me know what happens :paint2:

Edited by OldCadTech

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

Yes, probably needs a new EGR and gasket.

Before you buy the new EGR, remove the old EGR, leaving the gasket as it was when you removed the valve. Make sure the pintle opening is not plugged with carbon first. If it is plugged, clean it real good with carb cleaner. After it is clean, lift the diaphragm by hand again and watch the pintle, if it moves okay by hand, apply vacuum to it and again, make sure it moves freely. If it passes all those tests and the intake port is not plugged, re-install it, run the engine again and retest it.

If you lift it this time and the engine still continues to run. then the intake manifold MAY need to come off and be hot-tanked. Cross that bridge when you get there.

 

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

Okay, I can see that a complete EGR system check is really essential to passing a smog test with low NOx.  I will definitely put that on the agenda, and check if the EGR solenoid is working well too.  As well as cleaning out the throttle body and the EGR tubes.

Since the previous vacuum gauge testing seemed to eliminate a fuel injector problem or a vacuum leak, should I also plan to replace the two O2 sensors in order to eliminate the E046 code and help improve my chances of passing the smog test?  Remember I was getting some extended lengths of time with zero (0) cross count readings on the right and left O2 sensors?

I got a lot of things on my plate with the old Eldorado I plan to take care of, but for the immediate future I think I had best try to pass that darned smog test first and make my car street-legal !

thanks very much for your great advice so far, OldCadTech

 

Link to comment
Share on other sites

I would fix the EGR system and clean the T-body first.

Then, put some miles on the vehicle if you can without getting pulled over. It's about a 10 minute cycle for the cross counts to set the E046 so drive it for 10-15mins if you can.

Retest the O2 sensor response cross counts again, and see if they wake up or improve. Wouldn't hurt to do a few "hard accelerations" either to help "push" them out of bed.

THERE IS ALWAYS ENOUGH TIME TO DO THE JOB RIGHT - THERE IS NEVER ENOUGH TIME TO DO THE JOB AGAIN !!!

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...