69GTOby

The car is a 1990 Cadillac with a 350 Chevy TBI. It currently has about 70k on it, original engine. I have been trying to fix this issue since I have owned the car for about a year and a half now. I have replaced almost everything at this point. It surges when under load, and seems to lack power going up hill - the power comes in and out, and just doesn't seem to have full power. Other than that, it runs smooth as butter. No misses, backfires, pops, or anything!

I just replaced the fuel pump, again with a TPI pump this time. I rebuilt the TBI and did the Ultimate TBI mods. IT has 14-15 psi all the time. I put on a Caprice y pipe that flows much better (2 1/4" downpipes, 3" collector, vs 2" downpipes, 2 1/4" collector), stock 2 1/2 system w/o cat (thought it might have been clogged). Big open air cleaner. New distributor, cap, rotor, plugs, timing set to 0* base w/ wire disc. New EGR, MAP, O2 sensor. I have searched high and low for vacuum leaks and found none.

No codes to speak of. It had one for the fuel pump relay that has come back a couple times, but it always has 14-15 psi, set with adjustable regulator. I drove with a gauge on the windshield for a while, and it never goes below 14 with the new TPI pump. I think it just throws that code during starting sometimes because of the OP switch or something. No CEL ever! It had a lean O2 code once, but I haven't seen that one since I replaced the O2 last summer, with an AC one, of course.

I collected some data (AC off):
Idle in gear: RPM 550-575; BLM 108; INT 125-130
Idle in park: RPM 625-700, 675 avg.; BLM 108; INT 114-118
Deceleration: BLM 108; INT 120 +/-5
WOT: BLM 128; INT 128

While cruising the BLM and INT seem to be all over the map.
At idle the O2 seems to swing pretty well, and swings okay while cruising.
At WOT, which is where the power is lacking, the BLM is flatlined at 108 and the O2 is pegged at .9V.

I am running out of ideas. I was wondering if the smog system could be diverting some of the exhaust (oxygen) at WOT, making the O2 volts peg, thinking it's rich when it's really not... But, does it even use the O2 for feedback during WOT?? And the BLM and INT must be based solely off the O2 right?

Please help! I am at the end of my rope now. I thought for sure the TPI fuel pump would fix it. I replaced the pump with a stock replacement Delphi one, but the pressure was only 12-14, and down to 9 or 10 at WOT. With the new TPI pump it is 14-15 all the time, but the problem remains! Sorry for the length. I just wanted to cover all my bases and provide as much info. as possible. Thanks for any and all help!

RBob

iTrader: (1)

The fuel pump relay code is due to lack of voltage from the FP relay. I'd replace it and check the contacts in the relay socket.

RBob.

69GTOby

I swapped the relay with another one a long time ago when I first saw the code, and I guess that wasn't it. I also measured for power and ground at the fuel pump to make sure it was getting all it needed before I replaced the pump a second time, and it had 14V with it idling. I really don't think this code it related to the problem. It has adequate fuel pressure. I really think there is something else going on.

69GTOby

Come on... anyone? I need some ideas. Something is going on here...

69GTOby

The problem does happen at WOT... If it was an oxygen sensor or a bad O2 signal because of an exhaust leak or something, would it happen at WOT, or does the engine just run off the program at WOT and not look at O2 feedback or BLM/INT????

Schurkey

.9 volts on the O2 sensor is a rich (lack of oxygen) reading. The computer is pulling fuel to try to lean the mixture, as shown by your block learn readings. (Well, in one place you say block learn is 128, and in another you say it's 108. 108 would mean the computer is trying to lean the mixture.

I think it's unlikely that you are truly too rich at WFO. Somehow, you're getting a false-rich indication. Most faults--misfire, plugged injectors, cracked exhaust manifolds, AIR system incorrectly routed "upstream" instead of "downstream"--tend to create actual or false lean conditions. I chased a false lean indication on my TBI pickup for two friggin' years because I was too stupid to change a faulty O2 sensor (it seemed to read fine at low speed, but went false-lean at high speed.) You've already changed the O2 sensor, though.

What do the plugs look like after a WFO blast?

Do those engines use a backpressure-sensing EGR valve? Removing exhaust restriction could affect EGR operation--but I fail to see how that creates a rich mixture at WFO.

Does the coolant temperature sensor seem to produce accurate data? How about the MAP sensor?

Any chance you've got leakage past the fuel pressure regulator or past the injector O-rings into the engine, causing actual richness?

69GTOby

I installed the original air cleaner assembly. I thought maybe it was just throwing off the system. Anyway, it runs the same, but I can hear an exhaust leak. It has always had a little leak where the y pipe meets the intermediate pipe, and right before the muffler, but those are downstream and should not affect the O2 volts. But, I hear one near the engine. It ticks under load. It was hardly noticeable before over the big air cleaner sucking.

I looked and looked but couldn't find. It is definitely leaking under the hood somewhere and it isn't that one year old y pipe. I removed the whole AIR system. The cat has been long gone, and the pump is loud anyway. The only thing remaining is the 8 tubes in the manifolds. They are not coming out easily, even when hot. I need to bring a line wrench home from work, remove the tubes, and remove the heat shields. The manifolds have all their bolts and I do not see any cracks, but the shields cover a lot, especially on the driver's side.

I get how this exhaust leak is a definite issue throwing off the O2 feedback and making the computer fuel based on false info. But, at WOT the O2 is ignored... That means that there must be something else going on, right?

The plugs look pretty good. They are not ghost white, nor sooty black. They're tan on the electrode on one side and more white on the other. They look pretty normal.

The EGR seems to function normally, which it should, since it is new, lol. It even kind of threw me. I was cruising at 60 on the highway when I noticed the O2 volts suddenly plummeted. I thought what the heck?? My O2 just died! I got off at the exit and immediately saw it swinging again. I realized... the EGR must have been circulating! Cool.

The temp on the scanner always reads about 205. However, I did see today it was 225 when I got off the highway on the way home and sitting at a light. It didn't increase more, but that's a little high. Would a 180* t stat be okay for the ECM, or is that too low? I have not changed it yet, but I am thinking I should. Is there only one temp sender, for the ECM? Is it the same as the temp I see on my scan tool?

The MAP seems to be good. It is new as well. It reads about 11 at idle in park, and 8 or 9 in gear. I measured 21"Hg with a manual vacuum gauge on the PCV port of the TBI at idle in park, warm.

At this point, I would not rule anything out. Maybe it actually is rich. It always smells rich, but it always has since I removed the cat. It just smells like an old car now, lol. There are no external fuel leaks, I can tell you that. The spray patterns always look "normal" although I am not thrilled about the description of a "normal" spray pattern. I will look at them again for excess fuel leaking. How would I see fuel leaking past the regulator into the engine though?

Schurkey

So an actual gauge shows 21" of vacuum while the MAP shows 11"???

Fix that first.

O2 suddenly plummeted? First Guess: Misfire. I'm not thrilled about the EGR operation, but that's clearly secondary to the MAP sensor at this point.

69GTOby

My manifold gauge is "manifold pressure" and reads about 21"Hg. The MAP sensor reads ABSOLUTE pressure, which includes atmospheric pressure. It reads 8/11"Hg in gear/park, which equates to about 20-22"Hg manifold pressure, so no problem there.

It DOES NOT have a misfire, ever. It loses power smoothly, randomly. It seems to be heat related. It seems to happen on hotter days, 70*F+ with a hot engine 200*F+. It has been cooler lately, around 60*F, and it really has not happened. The more this goes on, the more patterns I notice, and the outside temp. and engine temp. really seem to play a role.

Oh, and the O2 plummeted while cruising because the EGR started circulating. It took me a minute to realize what was happening. That is what I was trying to say... It was cool to watch. The volts just flatlined... It would have thrown a code, except that the ECM was expecting the O2 volts to do that because it was commanding the EGR to recirculate inert exhaust gas. So, no problem there either!

69GTOby

I am getting more and more convinced the problem is indeed heat related. See my post from Monday? It was about 84*F here on Monday, and that is the last time I have experienced the power loss issue. Ever since, this week, it has been in the 60's, chilly and less humid, and the car has been running great. I don't think it has acted up once since Monday. Oh and the coolant temp. was 225 on Monday when I got home, as I described... The last couple days, it has only gotten up to 205, like normal.

If it was getting up to 225 or so, would that cause the ECM to lean out the mixture, or over richen it, or remove a lot of timing advance or something causing this sudden power loss and then surging power loss thereafter????

69GTOby

So, I drove the car Friday, when it was really hot out, and it started acting up. I looked at the data, the temp was about 205-210, and it was showing spark knock every time it acted up!

Over a few days, I confirmed that every time the car has low performance and surges under load, it: 1) is hot (hot engine and hot outside); and 2) has spark knock. I put it in graph mode... Driving normally, it's a flat line indicating "NO" but once the car warms up, under load it will have power loss and surges and EVERY time it does the graph will spike indicating "YES" continually until I back off the throttle.

The hotter it is, the faster the car warms up, and the sooner it has the symptoms. Could this be caused by a faulty knock sensor? Could running premium fuel solve the issue? I set the base timing to 0* with the wire unplugged like it should be, but could I try backing off even more timing, like running it at -2* or -4*? HELP

Schurkey

If the EGR works PROPERLY, and you don't have a heap of carbon in the combustion chambers...it seems you need more octane.

My long-block would be the same as the '91--'93 Fleetwood, (9.6:1 compression) but installed in a K1500. I have no octane problems like that. I had more knock sensor activity when I ran a 205 thermostat. I don't know what the difference in the engine is from '90 to '91.

I've never understood why the O2 plummets with the EGR operation.

Electric fans? I'm mildly concerned about the temp being so far above the thermostat rating.

69GTOby

All these SBC's are the same, but different LOL. Mine is VIN 7 which is basically a truck engine with a roller cam, and perhaps higher SCR? Based on my research, when people have this problem, power loss with a spark knock, higher octane fuel does not fix it, however, I suppose most people have trucks with lower SCR and timing or other issues.

I got a 180* thermostat to help keep the temp down. It should still get into closed loop no problem, hopefully. And I will fill up with 93 next time!

My EGR does operate properly according to all my tests, plus it is new. There shouldn't be much carbon in the chambers, at least not anymore. It only has 75k miles, 59k when I got it, and I give it plenty of full throttle to blow it out, especially out of frustration over this issue, which might well have been as simple as the fuel I am running, LOL!