1990 L98 stroked to 383 with Superram/heads/cam/headers. Narrowband in the driver side collector, wideband in the passenger side collector. When I start the car cold, the wideband shows it's running rich, as it should be - I want to say lambda is around 0.85. This gradually increases toward lambda = 1, but when the car hits closed loop it starts running really rough and lambda spikes. In closed loop, the idle is very rough and lean with lambda around 1.08. The narrowband says it's running about stoich, although BLMs are pegged at the minimum of 108. Revving the engine in neutral causes some transients, but if I hold it at any RPM it ends up lean at around lambda of 1.08.

The narrowband is brand new, and the wideband has very few hours running in the car (maybe 20?) and was just open-air calibrated as instructed by the manufacturer.

A while back, I accidentally let the #3 (second back on the driver side) spark plug wire sit up against the header, and it melted through the insulation. It wasn't arcing, and I didn't want to spend $40 on a whole new set of wires just to replace the one, so I let it be. I thought this could be causing a misfire in that cylinder, so I disconnected the #3 fuel injector and ran it. I didn't notice a difference, and when I reconnected the injector with it running I still couldn't tell a difference. That points to the #3 cylinder misfiring. However, wouldn't a misfire make that bank appear lean, not rich? The O2 sensor detects oxygen, not fuel, so an unburned fuel/air mix would appear very lean. Even if some of the fuel/air were combusting in the exhaust, you'd expect a little oxygen to be making it to the sensor and indicating lean.

As previously mentioned though, the narrowband (driver side) is indicating rich relative to the wideband (passenger side). If anything, I'd think this would point to a misfire on the passenger side!

I'm all sorts of confused - can you guys help me out?

Here are graphs from two different datalogs. In the first, it's running at hot idle. In the second, the #3 fuel injector is disconnected for most of the log. You see it warming up with rising BLMs before hitting closed loop, and then BLMs begin to drop when I reconnect the injector.

 

I pulled the one spark plug I can get out on the driver side without removing the header, and it looks pretty sooty. Of course, could just be too cold of a plug. Thoughts?



The #3 wire has started arcing really bad through the boot, so I can't really run it to troubleshoot until I can get a new wire and figure out how to get some clearance there... probably a shorty plug if I can find a good one.

 

This is the ONLY 3/4" reach, 14mm thread, gasket style shorty plug I've been able to find locally. I know on a waste spark system having mismatched plugs can be an issue, but the L98 isn't waste spark. I'm sure the heat range isn't quite right either, but there's no way to convert Motorcraft to Autolite to compare. Any reason why this won't work? (Assuming it's actually shorter)


Edit:
I'm seeing lots of recommendations for Champion lawnmower plugs, any thoughts?

 

Forget the WBO2 readings for a moment.

What has been done that would cause the apparent richness that is driving the BLM down to 108? Improper injectors/flow rate/offset settings? Leaking injector(s)? Leaking regulator? Excessive fuel pressure? Excessive VE table settings? High elevation? Does the ecm tuning really fit the hardware (sanity check)? Does it hold fuel pressure?

A NB sensor is unlikely to report a false rich condition. Have you tried leaning it out in order to raise the BLM?

 

Thanks for the reply.
There's no reason it should be running rich - BLMs were spot on when I tuned this a few years ago, and I haven't changed anything on the engine since. I suppose seasonal/regional fuel blends could be different, but that shouldn't be this drastic of a change. Are you thinking the wideband is off?
-

• Injectors are larger aftermarket units, but they've been properly tuned in. I think I had to set their flowrate a little low in the bin to allow >100% VE at WOT, but it's been tuned for good BLMs.
• A leaking or stuck injector seems possible. When I last ran it, I was planning to use a makeshift stethoscope to listen for a stuck injector, but the spark plug issue put a halt to that. I plan to get a mechanical fuel pressure gauge as soon as I can to check for a leaking injector.
• FPR is stock and relatively new, so FP should be stock (40#?) and it shouldn't be leaking.
• It was tuned at about 100' elevation, and I'm currently at 500'.
• I haven't messed with the tune to lean it out - if the PS is really running at 15.5:1, I don't particularly want to go much leaner. I think the DS is running at stoich, though. If it were running rich, wouldn't the BLM and the INT be at 108? Currently, BLM is floored at 108 but INT is floating around 122.

-
I think the tune is not the problem unless both of the following are true:
-

1. It had an issue to make it read lean when I tuned it, like possibly an exhaust leak.
2. My WB is incorrect and reads lean.

Both of these are individually possible, but it seems unlikely that both would be the case.

 

I think you're looking at a double failure situation.

1. An actual rich condition on the driver's side, that is being corrected to stoich by the BLM and INT (-26 points cumulative effect or about 20% mixture error correction).
2. Misfire due to a fouled plug on the passenger side resulting in a false lean indication.

Find and correct the source of enrichment (mechanical or tuning issue) and try a fresh set of plugs in parallel.

 

That makes so much sense, thank you!

If I disconnect the injector on a misfiring cylinder, the WBO2 should show no change, right? I'm thinking I can disconnect one injector at a time to find a misfire. It's a bear to change the plugs with these long-tubes, so I want to be sure first.

 

Well, threw a fuel pressure gauge on there and pressure doesn't even hold a little bit. The pump gets it up to about 42 psi, but it immediately starts dropping once the pump shuts off. In ~6 seconds it's at 20 psi, and by ~15 seconds it's down to 10 psi. Interestingly, the pressure declines smoothly to about that point, but then it starts to drop kind of jerkily. I pulled the vacuum line off the FPR, and I couldn't see or smell any fuel there. Granted, probably not a lot of fuel has to leak out for pressure to drop 40 psi, but I keyed the car on and off probably 20-30 times to see if I could see any fuel there. So, stuck/leaky injector? Any tricks to track that down? I'm thinking key the ignition a handful of times to get some fuel to flow, then pull the fuel rail and see which injector bore has gas in it.

Took out the DS spark plugs on accident (long story, went to replace the #3 plug and wire and got carried away), and they look pretty nasty. #3 is particularly crusty, but that could just be from the wire-related misfire. #5 looks really sooty compared to the rest, but it's hard to tell with different lighting. None of them really jumps out to me as the bad cylinder, though.


(1, 3, 5, 7, left to right)

 

Alright, after a little Googling I did the routine of clamping supply and return fuel lines at the tank in turn. Pressure still dropped rapidly with each clamped, which solidifies it being an injector. Now to pull the Superram... bleh.

 

typically when it dumps pressure real quick, its an injector.

do u have a phone and can you post up some pics during the superam process?

 

That's what I was afraid of. The injectors are new, but who knows - maybe some crap got clogged in one.

Yeah, I can definitely throw up some pics - what in particular are you looking for?

 

anything really. just love the superram! and btw, good job on digging to learn about the crimp method of isolating the injectors in your leakdown test!! u r getting advanced!

 

Haha thanks, you get pretty good at research when it's your job.

Well, I've got the fuel rail out, and there's no discernible leakage out of any of the injectors when I prime the pump. How much fuel would you expect to come out of a leaking injector when you prime the pump?
Did I possibly just not clamp the supply/return lines hard enough when I tested it?

 

if u have the rail out and pressurized then recreate the test.

i assume u checked the fuel pressure regulator for a ruptured diaphram.


maybe put a blanket or board under the rail or something

 

Yeah, I did. Threw some paper towels under each injector and keyed the ignition to prime the pump. Not even a drop of fuel came out of any of the injectors.

Shouldn't crimping the return line rule out the FPR diaphragm? Plus, it's a fairly new reg (5 years or so). I guess I'll pull it apart just for kicks.

At any rate, this seems to rule out the hypothesis that one side is running rich unless the injectors are just super poorly flow matched - seems unlikely. I guess the whole thing is running rich for who knows why, and either a misfire or an exhaust leak on the passenger side is making that side read lean. I guess I'll button it back up and install the narrowband on the PS (wideband won't fit on the DS, unfortunately) to rule out the wideband being off.