My Rover hates me.

I'm sure this isn't a new statement, but it's where I'm at. Let me go over the saga:

4 years ago I was looking for an old truck to drive around town. I'm a wedding photographer, and work from home most of the time, so I don't need a true DD. I have an odd way of doing things sometimes and so i had decided that this car could cost no more than $1,500. I had no good reason for this other than it seemed like fun.
I couldn't find an f100 or c10 that would touch the budget her in south florida. So one day while perusing craigslist, I saw a 90 range rover classic for sale about 2 miles from where I was sitting. I drove over, it started up and drove (valves sounded awful though) and paid the man $900.
After replacing the exhaust, fixing the power windows and a few other things (alternator) I started driving it around. A week later I was driving back from dropping my son off at school and something broke, bad. I got it home and started searching for a motor. I Found a local guy who would sell me a supposedly low miles 4.2 to replace my 3.9 for $300. I bought it and went to work at planning a rebuild of the 4.2 and then a R&R of the rover motors. It's at this point that I figured out it was going to be basically impossible to pull the motor in inclined driveway :) so I took it to my rover mechanic and had him do the heavy lifting. I got a call after it was in, with bad news that the wiring harness was cracking all over the place. You can't buy an engine harness, so I pulled one out of a Discovery 1 and rewired it to fit the classic. Then I spent a long day rewiring the engine bay. After that, we couldn't get the oil pump primed, and finally figured out that we had a timing cover gasket that was missing one hole. Ok, new gasket problem fixed. Finally I got the truck back and things were glorious until the next day. I noticed that when the truck was warm, and I pulled up to a stop light, it would drop idle to around 500 and run terrible, like it was missing a cylinder or the timing was far off. However, if it did that long enough, and the ecu stepped in by going out closed loop and into open loop (limp mode fueling) the motor would run perfect. This has been going on for 2 years now and I'm at a loss. Here's what I've done. New ECU, multiple TPS, Multiple IACV, AFM, new injectors, new Fuel Pressure regulator, new coolant temp sensors (both), new fuel temp sensor, new fuel pump, new fuel filter (3), new battery, new alternator, cleaned up all engine grounds, new factory 02 sensors. reset base idle, reset timing, new plugs, new 8mm wires, new coil. new ignition amplifier. I've tried different cap and rotor combinations. If I run the engine with the Maf unplugged it runs well, but smells terrible. It shouldn't run well with the maf unplugged, but it seems to be fine.
Has anyone else had and fixed this problem with a classic range rover? Any ideas? the truck runs great 95% of the time and it's been on a couple 2k mile road trips. Runs fine most of the time at 75mph. but this is driving me nuts. I hate to give up on something, but it almost makes me want to sell it.

So the thing that jumps out at me is that you've replaced the AFM but it still runs better when you unplug it.

That suggests to me that the unit isn't bad, but it's providing bad data, since it runs better without that input.

Bad ground for AFM? Bad power supply to AFM? I don't know these in particular, but you said MAF... It needs good power and ground to heat its element as well as solid ground that's sensible relative to the ECU, right?

That's what I've got...

Vacuum leak?

wvumtnbkr wrote: Vacuum leak?

I was just thinking this while driving the kids home. Unplugging the MAF 'should' richen the mixture which would even out the fuel/air if you had a vacuum leak and make it run better. If there is a leak with one cylinder maybe that cylinder is misfiring at idle when things are very lean and for some reason the 02 sensor isn't seeing it as a lean condition because it's only one cylinder? Maybe?

Ransom wrote: So the thing that jumps out at me is that you've replaced the AFM but it still runs better when you unplug it. That suggests to me that the unit isn't bad, but it's providing bad data, since it runs better without that input. Bad ground for AFM? Bad power supply to AFM? I don't know these in particular, but you said MAF... It needs good power and ground to heat its element as well as solid ground that's sensible relative to the ECU, right? That's what I've got...

yes, that makes all the logical sense in the world- but I've tested it with a meter for operating volts at idle as well as ohms when sitting and all is within spec.

wvumtnbkr wrote: Vacuum leak?

I've thought the same but I just can't find one. I've replaced nearly all of the vacuum lines and checked the others, and I've done the starter fluid test as well....just can't find a leak.

I have wondered about purge valve and charcoal canister for a while now. What would the canister being saturated, or the purge valve sticking do?

Don't take it personally, they hate everyone.

I have no experience with that particular system but in true internet form I'm not going to let the prevent me from offering advice. I think the first thing to do would be to determine if the rough running is the result of the low idle or because of it. If you open the throttle a bit when it's acting up does it smooth out? If so I'd be looking at a possible base idle or throttle position sensor adjustment. If it still runs rough I'd start with a compression check.

APEowner wrote: Don't take it personally, they hate everyone. I have no experience with that particular system but in true internet form I'm not going to let the prevent me from offering advice.

seems reasonable to me- The low idle is from the rough running I believe. If I go WOT it ALWAYS fixes the rough running. Sometimes I pull away from a stop light and it runs fine, other times it's missing until the ecu catches up.

My theory today is dirty advance weights or a broken spring in the dizzy.

grover wrote:

APEowner wrote: Don't take it personally, they hate everyone. I have no experience with that particular system but in true internet form I'm not going to let the prevent me from offering advice.

seems reasonable to me- The low idle is from the rough running I believe. If I go WOT it ALWAYS fixes the rough running. Sometimes I pull away from a stop light and it runs fine, other times it's missing until the ecu catches up. My theory today is dirty advance weights or a broken spring in the dizzy.

That's a possibility. Do you know for sure that it's the switch back to open loop that "fixes" it or is that an educated guess? One other thing I've seen in cars of that era that can cause those symptoms is a sticking EGR valve.

I would suggest EGR valve as well, but don't even know if that motor has one :-)

APEowner wrote: </cite <

That's a possibility. Do you know for sure that it's the switch back to open loop that "fixes" it or is that an educated guess? One other thing I've seen in cars of that era that can cause those symptoms is a sticking EGR valve.

Well, I'm suppose it's an educated guess. It throws an 02 code and then heads to another fuel mix.

I have no idea about the egr valve, would that be the same as a purge valve?

O2 code how? Lean?

mazdeuce wrote: O2 code how? Lean?

I don't think it states lean or rich. just that the lambda is reading wrong. code 44 for me.

OK, let's assume it's lean because closed loop usually defaults to rich and you say it smells bad. You're chasing lean. Lean at idle. Just idle or part throttle too?

I don't believe it's at part throttle, it just doesn't seem to "catch" at the right rpm when you slow to a stop after about 10 min of driving. It's fine before that. If the code pops up, I.e. the ECU cuts in, then it idles perfectly. If it doesn't then it pulls away at about 30% power.

OK, so something is telling it to lean out. Or mechanically leaning out, until the computer says f-this and goes off the tables. Is there a way to check the actual voltage output of the MAF at idle vs. the expected readings, say with a multimeter and a needle probe? What else controls fuel with the throttle completely closed? Is the throttle completely closed? (Does it have a TPS that indicates closed? Does EGR act differently at idle?
I'm just trying to mentally sort through this and hope that my asking questions allows you to stumble onto sometjing.

mazdeuce wrote: OK, so something is telling it to lean out. Or mechanically leaning out, until the computer says f-this and goes off the tables. Is there a way to check the actual voltage output of the MAF at idle vs. the expected readings, say with a multimeter and a needle probe? What else controls fuel with the throttle completely closed? Is the throttle completely closed? (Does it have a TPS that indicates closed? Does EGR act differently at idle? I'm just trying to mentally sort through this and hope that my asking questions allows you to stumble onto sometjing.

In reply to the MAF readings, yes- and i've checked and adjusted it into spec, seems to be working correctly. The ecu should be reading the tps (working and checked with a mm) coolant and fuel temp sensors and adjusting the IACV. Also the 02 sensors. They should all be working, but I've not checked the 02 sensors physically. They have been replaced with no change though and to be honest I don't love the idea of doing that again. They're also not cheap.
I read the last part before I got on the plane to nola this weekend and many beverages over the weekend didn't bring any additional clarity. I agree that something must be leaning it out, I think i'll take the distributor off tomorrow after I take the kids to school and see what I can find.