Cadillac Owners Forum banner
13K views 62 replies 14 participants last post by  Sevillian273 
#1 ·
I'm changing the plugs soon in my 4.5. I have an ACCEL super coil, MSD cap and rotor, and MSD 8.5mm wires all feeding my current AutoLite double platinum plugs. They only have 20K miles on them but I'm having a 'stumbling upon acceleration' issue and I cant find anything wrong with the fuel system so now I'm on to the ignition....

I'm not trying to start a 'my plugs are better than your plugs' debate.:stirpot: Im just curious what everyone is running in their 4.x and how satisfied are you with your choice of spark plugs?
 
#2 ·
Take a hard look at NGK Iridium plugs.

We run those in our heavy duty Olds 455 marine engines - not cheap but they stay clean and last forever.

Stumble on acceleration is usually not spark plugs unless you have the coil wired backwards or the distributor module is on its last legs.
 
#3 · (Edited)
I'm not brand loyal as far as spark plugs. I never noticed performance differences between brands (except on two-strokes). I just replace the plugs with whatever type it came with using whatever brand is on sale at the time. I'm all about minimizing cost of ownership with my cars and I think being brand loyal can cost too much (especially when you keep multiple cars as long as I do!). Of course there are cases where I make exceptions and use only AC Delco. However, plugs aren't one of them.

I'm pretty sure I have regular single tipped platinums in my 4.9...I think Bosch. They've been in there since the first month I've had the car, maybe 7-8 years. No problems.

I can't remember what my '90 has anymore...but they both run perfect.

I agree with Sub, put your concentration elsewhere.

Little petty problems like this get especially hard to diagnose when things have been modified. If I remember your car correctly, you have Bosch injectors, deleted EGR, deleted AIR system, and an aftermarket ignition system...

Sometimes you just have to let the problem get worse before you can accurately diagnose it. The problem will be annoying while it's there, but try to diagnose too early and you could end up sticking more time and money into it than the problem is really worth.
 
#5 · (Edited)
EGR is not deleted. I only do that when diagnosing a problem and want to rule it out. Timing is at 10deg.

I recently compared the 'fuel used' on the DIC to the actual fuel used. I started with 17.3 gallons and drove until I had 3.5. That's 13.8 gallons but the DIC said 10.3. A difference of 3.5 gallons. Somehow fuel is getting into the engine unaccounted for. Problem is, when I pulled the rail up and turned the key on with 40+ psi to the rail there were no drips. I watched the guage slowly lose pressure while still observing no drips at all from the injector tips. According to the FSM the leakdown is at the fuel pump check ball... No leaky FPR either.

I've done every fuel system test in the book and everything is in spec. Yes, the bosch injectors are .2lbs higher flow rate than oem but I had no issues when I first got them. It seems as if the engine gets choked off with fuel accelerating off-idle. Its even worse when cold which makes sense because the ecm is running the car richer until the engine reaches closed-loop.

And another thing: When sitting at a light the integrator counts slowly reduce down from 128(ideal) all the way down to 89. A lower value means the computer is taking time off of the pulse width to get the air/fuel ratio where it should be. This further confirms rich operation which leads me to believe that this is what fouled up one of my o2 sensors in a previous thread.

Everything points to leaking injectors but as I said, they did not leak before my eyes. And 3.5 gallons is ALOT of gas to have simply dripped away. PLUS: All the computer parameters that indicate rich operation are identical for both banks and stayed that way when I swapped the right injectors for the left 4. Its is inconceivable that ALL 8 injectors began to leak the same amount at the same time. Something is affecting the fuel system globally. Only the fuel pump, FPR, and the ECM can do that as far as I know.

After all the testing and monitoring I've gone through I'm at the point where I'm ready to just start blindly throwing parts at it. The DIC says I'm getting 21.4 mpg but the actual mileage is really 16.5mpg. The car loses about 25% of its fuel per tank. SO I can keep wasting money on gas until something breaks all while running the risk of ruining another set of o2 sensors OR pretty much flip a coin and buy parts. Either way I lose. :banghead:

Even though the FPR does not viibly leak, and both static and dynamic pressure test indicate that it is ok, that's the first part I'm gonna try. I ordered the cheapest 'KEMPARTS' FPR from rockauto for under $30. If that doesnt change anything, I'll try another ECM followed by another set of injectors. No change after that and I'm getting a Honda.
 
#10 ·
Wow thats a rare thing to hear! Most forums frown upon those plugs but I've never seen any real evidence as to why they 'are no good'. Most praise AC Delcos as gospel and conversely I haven't seen any evidence as to why.... But I do think that recommendations come from real experience and that is tangible enough for me so when I'm done blindly throwing parts at the fuel system, my next set will be Delcos. Changing plugs on a transverse V8 is no fun otherwise I'd take my chances with the +4's but I've heard alot more negative rhetoric about them than positive... Whether it be fuel or spark, I still feel I'm in a no-win situation. :banghead:

And to add insult to injury, my latest calculation on my most recent tank of gas has revealed 14.4 mpg while the DIC says 21.1mpg...
 
#11 ·
Sorry, I'm out of ideas on this one. You can try plugs but I doubt those will help any. If the plugs were bad enough to return you 14.4mpg, that thing would be chugging and misfiring so bad...

Let me ask you this, are you sure of the non-stock injectors you have installed? Did you physically check the number stamped on them to ensure they are the "correct" 19# ones?
 
#12 ·
Let me ask you this, are you sure of the non-stock injectors you have installed? Did you physically check the number stamped on them to ensure they are the "correct" 19# ones?
What resource could I use to check them? And it ran good and got normal mileage for a few months after I installed them. The driveability problem gradually set in along with the bad mileage.

I got my $30 FPR in the mail (looks like the $50 BWD unit that you recommended) so I'll put that in. After all, the rubber diaphragm inside IS 20 years old and I'm sure the rubber compound wasn't designed for 10% ethanol... I doubt this will fix the problem. This is becoming more and more like like the madness I went through with code E041 which turned out to be the ECM after hours and hours of diagnostic work and parts. I hate to do it, but I'm gonna drop another $80 on a ECM if the FPR has no effect. After that, I'm gonna sell all of my tools, lease a Yaris, and get a desk job. :cookoo:
 
#13 ·
I have the Accel plugs and 8.8 wire, my tires were chirping when I had it fresh installed

Stumble on tip-in is usually EGR. Plug off the vacuum line from the valve to the solenoid and see if the problem goes away.
 
#14 ·
Done - no dice. I even removed the valve and blocked the passage with a plate to rule it out completely. All it got me was a E048 and a light knock at cruise. :banghead:

I need a way to SAFELY monitor fuel pressure while driving at any speed. My guage is to short to reach the windshield. Cant seem to find an extension for it.... I'd rather not hack the hose. Fire sucks.
 
#18 ·
When I decided to clean up my 85 I spent a couple of years chasing down problems. One of them was intake gasket that seems to have been the source of a perpetual oil consumption problem. Another was a hairline crack in the throttle body at the butterfly valve that was allowing unmetered air into the engine causing lean/rich cycles. I wouldn't be surprised if you have some kind of structural problem like this.
 
#20 · (Edited)
My '85 indicates from 24 to 27 MPG, but the indicated average seems OK at 18.6, mostly, and sometimes 19, depending on highway or city/highway driving. Maybe it's not accurate, but it's acceptable.
I rebuilt the TBI a few years ago because the regulator housing was seeping fuel, but no change in mileage etc.
I don't usually run any other plugs than the ACs in the daily drivers. In my performance cars I use Accel plugs, which are actually Nippendenso U-groove types.
 
#21 ·
Put in the new Kemparts FPR today- No changes. Wasn't really expecting anything either. The old one was pretty rusty inside but still held vacuum. So thats one thing off the list... One thing I noticed while monitoring the fuel integrator is that around 2500rm at 45mph or so, the car surges with power and the integrator goes way down below 60! That means the computer is taking ALOT of time off the pulse width under those conditions. Still have no idea what is going on.

Next random part: Allante fuel injectors! I found a set of new (doubt it) OEM replacements on ebay for 142 w/ shipping. http://cgi.ebay.com/ebaymotors/ws/e...50411510683&ssPageName=ADME:X:AAQ:MOTORS:1123 Not sure what the flow rate is on those but at least they are for a 4.5. They'll go in next weekend. Meanwhile I'm still getting around 14mpg. Sucks.
 
#22 ·
Got the Allante injectors in the mail - They came in factory sealed packaging and not a scratch on them. I installed them and she runs. The stumble is gone and integrator counts are a bit higher but now I have an extended crank issue when hot... Every hot start is an extended crank now :cookoo: Fuel psi is in spec during the hot start and FPR is brand new and does not leak.

Also the integrator differs from left to right now. I dont recall if there was a difference from L to R with the previous Bosch III's but I had 89 counts at idle which was rich. I now have 111 for the left and 98 for the right. Both should read the same. Something still isnt right but I suspect that the lesser flowing stock injectors have alleviated the driveability issue. As for the extended crank, I'm clueless. I'll give it some more time while I calculate the new mileage before I go for a new computer.

I dont see what else could cause one side to run richer than the other. The only kind of sensor that is specific to each bank is the O2's. Other than that you have injectors and sparkplugs. The rest of the system affects both banks the same. Fuel pump, rail, FPR, MAP, MAT, ISC, TPS, ECT - none of those can influence just one side of the engine.

Usually when things just dont make sense it's the computer. And I have a hunch... When one of my OEM Multec injectors crapped out it was on that side of the engine, #6 if I recall. It read just 1.2 ohms when It showed symptoms and I probably drove home about 5 miles before I removed it. I wonder if the heat from such low resistance did damage to the injector driver circuit in the ECM? I'm clutching at straws here...:ill:
 
#37 · (Edited)
I dont recall if there was a difference from L to R with the previous Bosch III's but I had 89 counts at idle which was rich. I now have 111 for the left and 98 for the right. Both should read the same. Something still isnt right but I suspect that the lesser flowing stock injectors have alleviated the driveability issue. As for the extended crank, I'm clueless. I'll give it some more time while I calculate the new mileage before I go for a new computer.

I dont see what else could cause one side to run richer than the other. The only kind of sensor that is specific to each bank is the O2's. Other than that you have injectors and sparkplugs. The rest of the system affects both banks the same. Fuel pump, rail, FPR, MAP, MAT, ISC, TPS, ECT - none of those can influence just one side of the engine.

:
I have another question. Is there anything that says it can't be the opposite? What if the 98 bank is running correctly and the 111 bank is lean vs the 98 being rich and 111 correct? My 4.9's integrator counts reside around the same ranges you describe in this thread (usually <128, varying depending on throttle given/conditions), but yours differ per bank where mine are the same across both banks. This does tell you that the engine runs a little rich overall (maybe by the influence of one of the global controls), but it can't tell you which bank is the problematic one in the difference, right?

We can't just assume that the value closer to 128 is the bank that runs correctly, can we?

If the scenario where the 98 bank is actually the correct one, then it's possible that there's a vacuum leak on the 111 bank, making it run more lean on the 98 bank.

I guess overall what I'm trying to say is if you have a global control reducing pulse width that's fine as long as it's working correctly and it's keeping a 14.7:1 ratio. But the difference in banks is certainly an indication of a problem and I'm not quite sure if the integrator values give enough information to tell us which bank is the problem.

Now that you've swapped out injectors, is there a way to reset the block learn value and start fresh? Disconnect the battery? The car is now running on block learn values for the old injectors and will take a while to change (I think).

Also, I noticed you only gave 1 value for the block learn, 118. Keep in mind that the pcm stores multiple block learn values for different conditions. It doesn't use the same block learn value for idle , part throttle, and full throttle. It's a "staged" value.
 
#23 ·
The integrator counts are only a short term fuel trim, so may vary at different conditions. Recording the values at different conditions could help pinpoint the problem. Have you checked it during varying conditions? Idle, low/high rpm, low/load?

What are the values for the block learn? Does the ECM go into closed loop like it's supposed to?
 
#24 ·
Each differing condition ( acceleration, cruise, deceleration) sets the integrator back to 128 and from there (in my case) it dials the fuel back lower than 128, correcting for richness. This happens for both sides but the right bank integrator goes lower indicating that it runs richer than the left. L and R differ by 10 to 20 counts at any condition. The difference is most pronounced at idle. Block learn is 118 at present.

It does reach closed loop. That's when the ECM starts to dial the integrator back from 128.

The new FPR does perform better than the old one. A hard stab on the gas in D would shoot the pressure to 50psi for a second with the old unit. The new unit hardly lets the needle move at all.

The new injectors have definitely helped out the richness and there is more available power. I'm an idiot for not taking readings before removing the bosch III's. I cant remember if there was a L to R difference with those. Getting sick of pulling the rail but now I can RnR it in under an hour.

After I run through a tank of gas and calculat mileage and fuel data accuracy, I'm putting in new plugs. If the rear plugs look anything like the oxy sensor did, then they could use a change. If the mileage is still horrible, then I will get a new ECM.
 
#25 ·
Each differing condition ( acceleration, cruise, deceleration) sets the integrator back to 128 and from there (in my case) it dials the fuel back lower than 128, correcting for richness. This happens for both sides but the right bank integrator goes lower indicating that it runs richer than the left. L and R differ by 10 to 20 counts at any condition. The difference is most pronounced at idle. Block learn is 118 at present.
so if I'm reading that correctly, the integrator is normal during all conditions except idle?
 
#26 ·
No, the integrator just resets to 128 and then counts back to dial in the mix for that particular driving condition. I think each condition is called a cell - not sure. For instance, if I sit and idle in P the integrator counts down to 89, shift into D and instantly back to 128 again and then counts back. So if I sit at a light in D it counts down and when the light goes green and I hit the gas its back to 128 and then counts down while I accelerate. During acceleration when it gets to around 70 to 60 counts there is a surge of extra power which makes sense because by then the fuel mix is close to where it should be.
 
#27 · (Edited)
Well out of curiousity, on my way to work earlier this afternoon, I checked my own block learn and integrator counts to compare to the info you gave. They were both the same at all times, not differing bank to bank. The block learn values were slightly above 128, in the low 130's but I wonder if this is due to the fact that I'm using a set of Ford 19# injectors...

ANYWAY, on to your problem Sevillian, something occured to me while I was chugging away at work (I think about anything and everything at my physically active, but mind-numbing job). This might not be your problem, but is one of the only theories I can attempt to help you with.

The lower integrator values you are stating don't necessarily say if the car is running rich or not. Only the O2 sensors can tell you that and a O2 code would be tripped if that was the case. A cars fuel management is always trying to achieve the correct stoichiometric ratio, right? That's what your car is doing by adjusting the values per bank. Your car isn't running rich right now. If it was, you'd have major driveability symptoms that, from what I understand, you really aren't getting. It's running "correctly", as in the air/fuel mixture is correct, but it's using more fuel to achieve that.

So what does that mean?

Well, the integrator is calculated from the engine management gadgets you listed, with large emphasis on the MAP and O2 sensors. The MAP is a global measurement while the O2's are per bank. What this tells us is that the O2 on the effected bank is detecting a lean condition, and the computer is adjusting the integrator accordingly to compensate.

How is there a lean condition on one bank?

VACUUM LEAK!!! My money is on an intake gasket leaking vacuum on whatever bank has the low integrator counts.

One thing I'm curious about is to what it takes to set a E046 code, because you'd think something like this would set that. Perhaps the banks don't differ enough yet to set it?

If I remember correctly, you have an FSM (I never got one yet otherwise I'd do this homework for you as I'm interested in this). Does it have diagnostic charts for the E046? If so, take a look to see what it uses trip this. Does it use integrator values? Then follow the chart and see if it leads you to an intake gasket leak diagnosis. If so, I'd consider that to the true confirmation that this is the problem.
 
#28 ·
There is no lean condition on either bank. Both are richer than they should be. Any integrator count lower than 128 means the ecm is correcting for a rich condition. Less integrator means time is taken off of the pulse width to achieve stochiometric mix. If I had a lean condition, integrator counts would be higher than 128 meaning time is added to the pulse width. Im OK with higher or lower integrator counts because they designed this system to be able to operate at all possible conditions(sea level, high altitude, hot weather, cold weather etc) but what makes me uncomfortable is that there is a fueling difference between both banks.
 
#30 ·
I wish you were right, I really do. i would gladly change the intake gaskets 50 times than go through these mental gymnastics... I will report back after I calculate fuel mileage and change the plugs. I'm starting to really miss driving my Ford. Thanks for taking an interest in this madness, lol. It's nice to see that your values are the same for both banks. I seem to remember mine being the same also back before I started having issues.
 
#34 ·
What about a restriction or leak in one of the exhaust manifolds or one side of the y pipe. If I remember correctly, that car has a crossUNDER pipe. Is it collapsed/crushed?
Yep a crossunder pipe just like your eldo. It's not crushed or damaged in any way. I'm not too comfortable being under the car on ramps while it is running but there are no audible leaks as far as I can tell.
 
#32 ·
Hmm.. you may be on to something... I checked the rich o2 charts but not E046. I'll do the tree tomorrow but what's interesting is:

"Likely causes of code e046 are:
1. Lean or faulty injectors on one side of the engine.
2. Cracked or fouled spark plugs.
3. Exhaust or intake manifold leak."

#1 is unlikely since I've replaced so many injectors, #2 MIGHT be possible but it runs pretty smooth to have a fouled plug or misfire issue... #3 would give me at least one LEAN bank so I really doubt that one.

Definitely will change plugs after I calculate mileage. After that I replace ECM... again.... Another 'CarDone' reman unit.
 
#35 ·
So as it stands now:

1. Continue driving and calculate mileage and fuel info data accuracy.

2. Change plugs (AC Delco standard)

3. Check plug wire resistance and perform HEI performance test per FSM

4. Replace ECM

5. Install 2 brand new oxygen sensors of same brand

6. Purchase full coverage and theft insurance.

7. Drive to remote area.

8. Place rag in gas filler tube.

9. Ignite rag.

10. File a claim.

11. Lease a CTS-V and enjoy. :D
 
#36 ·
Man, I'd like to help more...I'm out of options. This isn't even my car and it bugs/intrigues me.


I'm just a 25 year old with a box of hand tools and some knowledge I learned on the go.

I'm a civil engineering student, not even remotely related.

Where's all the diehard Caddy tech-heads? Sub, Ranger, Jake? Not Northstar related but basic concepts are all the same!
 
#38 ·
Before the single oem injector failed it always said 128 give or take a few for both banks but at that time I had no idea what that parameter even stood for. I only take the time to read the FSM and learn about this madness when there is a problem. I understand what your getting at and it may be possible but I tend to doubt it because I never saw such lower values before and I'm running OEM 4.5 injectors now. But the idea does make sense. There are still driveability issues but they are better than before presumably because the system had to 'correct' with higher-flow-than-stock injectors(19#).

128 is what the computer wants and that's what it resets to for each 'cell' (or whatever its called) and intends to keep it that way but the output from the o2's are telling the computer that 128 is too much fuel. The FSM says that 128 is ideal and means that the computer does not need to adjust the fuel mixture so for now I'm taking that as gospel. So I wouldnt say that the one that runs closer to 128 is the one that runs correctly, just the one that needs less correction for whatever reason. I dont know enough about the system to say whether or not it is acceptable for the engine to run so much less than 128 at every different throttle condition but I know for sure that there shouldnt be a fueling imbalance.

Btw block learn is 108 not 118, my bad. From what I understand block learn is like an average of the integrator counts but I'm not sure how it comes into play in the system. Funny, its 108 for both banks despite the fueling difference. There's only 2 parameters for block learn, L and R, So maybe it changes for each 'cell'? I'll have to take a look. Either way I'm still not sure what it means. I need to talk to the folks who DESIGNED this system. I bet they would have a good laugh at all my speculations.
 
#39 ·
Btw block learn is 108 not 118, my bad. From what I understand block learn is like an average of the integrator counts but I'm not sure how it comes into play in the system. Funny, its 108 for both banks despite the fueling difference. There's only 2 parameters for block learn, L and R, So maybe it changes for each 'cell'? I'll have to take a look. Either way I'm still not sure what it means. I need to talk to the folks who DESIGNED this system. I bet they would have a good laugh at all my speculations.
Yes, I think that's correct. If you leave block learn displayed while driving, it changes per conditions. IT uses a value when you are at idle, another while accelerating, etc.
 
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