I'm getting some spark knock that gets worse on long highway runs - it doesn't just depend on being in closed loop. I've retarded my timing and it still happens. It mostly goes away on 93 octane fuel but I would rather not spend money on premium since I know the engine is not designed for it. I noticed my INT values are low - can someone take a look to see if my engine parameters look appropriate?

Data

BLM values:


I'm mostly concerned about my low INT values - shouldn't the BLM values gradually change to compensate?

INT values:


Additional info - EGR commanded


Rich-Lean indicator - What is this?


INT values again


thanks in advance.

[This message has been edited by masospaghetti (edited 05-30-2012).]

The BLM values can't go any lower than 120, that's the minimum allowed by the factory chip. 128 is neutral, so your car definitely is trying to run lean.

I'm not sure about the "rich/lean" indicator but it looks like it's the value that counts how many times the O2 sensor has cycled between indicating a rich or lean condition. In order to achieve stoich, the ECM tries to adjust the mixture such that the O2 will keep cycling between those readings. That counter is an 8-bit value which wraps around to 0 when it reaches 256.
If you look at the O2 sensor you'll see a jagged graph going back and forth between low and high voltage (high voltage is rich, low voltage is lean). Those transitions are what I believe is being counted by the "rich/lean" indicator.
[edit]after reading dodgerunner's post, I checked my logs again and agree that the jaggedness of your "rich/lean" graph is a problem. Mine is smooth but gets rough like yours only when it's misfiring. The roughness of that graph indicates the O2 sensor isn't cycling at consistent time intervals as it should be if things are running properly. It implies to me that your ECM is having to hunt around to find the right mixture.[/edit]

It looks like you're probably running too lean, you might see evidence of that if you pull the plugs.
If you disconnect the MAP sensor or the O2, it will run richer.

I don't know if a bad O2 sensor can give a reading richer than it really is - I've normally heard of them failing in the opposite direction as they age.

It's possible you have an exhaust leak that is tricking the O2 sensor. I don't know if the effect of that would be real dramatic but people say it can be.
I'm having similar low BLM/INT values and found that some plugs were lean, others weren't. In my case I'm suspecting bad injectors giving a mismatched flow rate. The ECM can't adjust the injectors individually so if they're way mismatched it might have this effect in the leaner cylinders.

[This message has been edited by armos (edited 05-31-2012).]

As was said your INT says your running lean most of the time. It's trying to add fuel but not having much luck.
The rich/lean counter should be a very smooth saw tooth shape. Yours is ragged so things are not right.
In addition to the above ideas have you checked the fuel pressure? If you pressure is not up around 40 psi you could get those results also.

Harbor Freight has theirs on sale right now. " TARGET=_blank>http://www.harborfreight.co...ter-92699.html[/URL] (Yes it works on fieros)

Or borrow one from a parts store if you don't have one.

[This message has been edited by Dodgerunner (edited 05-30-2012).]

Thanks for the replies.

I don't have an obvious exhaust leak anymore (I don't hear it, anyway) after I fixed a failed manifold gasket. O2 sensor is only about two years old, a Denso part. I have a spare NGK that I can try to see if it helps.

I actually already have that Harbor freight fuel pressure gauge, will check that as well. MAP sensor is about a year old, a BWD product. No obvious intake leaks, EGR tube was replaced with a good one, idle is pretty steady and not too high. Any other potential causes for my symptoms?

EDIT: I am running junkyard Multecs out of a 2.8 Beretta. The coil impedences were checked good before installation, but I wonder if my injectors are also not working 100%.

BTW Dodgerunner - your ALDL cable is top notch. Thanks for a great product.

[This message has been edited by masospaghetti (edited 05-31-2012).]

Before getting into injectors, I'd suggest pulling some plugs and see if it looks like you have an uneven mixture between cylinders. If fuel pressure is good but you see some burned plugs and some healthy, then I don't know what else to suspect other than injectors.

quote Originally posted by masospaghetti: On a related note, FIC sells a 17# injector set for $130 - would this work on the 2.8? It's listed as for 3.1 and 3.4 engines, and I noticed a Fiero owner commented on them. FIC

Presumably they should, the ECM will just adjust the BLMs downward to compensate for the higher flow rate. However, it can't adjust below 120 on the stock chip.
I received refurb injectors today from "Precision Auto Injectors" and the attached note said they are 17lb/hr. He says they'll work fine even though the originals are 15lb. It appears this substitution is common practice. I still suspect 15lb/hr would be more ideal since it probably makes it easier for the ECM to accurately control them. But lots of people on the forum have commented on using 17lb injectors so it's probably not a big deal.
I'm beginning to wonder if 15lb/hr Multecs are really available, maybe 17lb is all anybody really has for sale.

If you go toward the $150 price range, you can get Bosch type III 17lb injectors which are modern and supposedly better than pretty much anything out there.
http://fuelinjectorconnecti...ewProd&productId=186
The way people rave about those Bosch III's, I was tempted by that listing but the PAInjector Multecs were better for my budget so I didn't look into the compatibility any further.

FWIW, I've read an opinion on the forum that Multecs might not work as well with the stock ECM as the original pintle style injectors that came with the car. But many people do use them, and after watching a video from FIC of the lousy spray pattern on the pintle injectors I just couldn't convince myself to bother with them.
http://www.youtube.com/watch?v=VfynoxL2RZI

armos, thanks for the reply.

I have been thinking about the data today and it doesn't really make sense to me - Why would the INT go low, and keep going low? I thought that once the ECM sensed that the engine was running lean (via the O2 sensor) it would increase the INT value until that was no longer the case. The INT values on my engine, on the other hand, are continuously bouncing between 128 and in the double digits. Does this mean that the ECM is receiving bad data, possibly from the O2 sensor?

quote Originally posted by masospaghetti: armos, thanks for the reply. I have been thinking about the data today and it doesn't really make sense to me - Why would the INT go low, and keep going low? I thought that once the ECM sensed that the engine was running lean (via the O2 sensor) it would increase the INT value until that was no longer the case. The INT values on my engine, on the other hand, are continuously bouncing between 128 and in the double digits. Does this mean that the ECM is receiving bad data, possibly from the O2 sensor?

I'm not sure about that either. My thinking is that if one cylinder is lean and the other is not, then no matter what it does with the mixture one cylinder or the other isn't going to be right. In that situation, I don't know whether the O2 sensor will be happier on the lean side or the rich side.

When you check your fuel pressure, also make sure it isn't leaking down with the key turned off. If you have a leaking injector or cold start injector, that might explain the problem. If it's dumping fuel somewhere then the ECM may try to go lean to compensate for what it sees at the O2, causing some lean cylinders.

armos: While I hate "throwing money" at a problem, I went ahead and bought a set of injectors on "wholesaler closeout" from Rockauto - I figured my injectors are probably tired, and they only had six left. $130 for a new (not rebuilt) set of Standard FJ27 injectors.

I'll replace the O2 sensor and take it out for a drive before replacing the injectors and log the data, to see if the O2 sensor was the culprit.

Will provide an update this week. Waiting on some heater hoses to finish R&R'ing the heater circuit.

I have my fuel injectors out, waiting on the new ones, but I found a lot of what looks like rust in the fuel return line, and also in the injectors themselves - how could this happen? The filter is only a couple years old and when I took it down, not a lot of dirt came out of it. I'm going to replace the filter as insurance, but this seems strange to me.

Cut the filter open and see what’s inside. If you’re getting rust into the system it will either be stuck in the filter or in the injectors.

As to the BLM (I can’t see the picks so I can’t see what it’s at), if it’s low then the ECM is seeing rich and trying to go lean. If for some reason the ECM thinks the car is running rich when it’s not, it will try to lean the mixture out when it’s already lean causing a condition that will “ping”. The O2 sensor is the most likely cause. If one or two cylinders are running rich it could also cause what you’re seeing but they would have to be pretty rich (like a stuck open injector).

Car isn't running yet, but I did install six new Standard FJ27 injectors and have the motor all back together. New Wix fuel filter installed. BTW, NAPA sells Wix filters still made in the USA, unlike most of the other chain stores that sell products manufactured offshore, and the prices are completely reasonable. Fuel filter was $7.47, a Wix Gold oil filter was $3.40.

Only thing left to get the car back running is the front two heater hoses which have been discontinued. I'm rigging up new hoses using the Unicoil shaping coils. The hoses that are sold at parts stores as "Heater to Pipe" are the wrong parts, maybe they fit a non-A/C car, not sure.

For reference, the FJ27 injectors are 4-hole disk-type injectors. They are approximately 1/8" shorter than the old Multec/Tomco injectors as measured between the two o-rings - They seem to seat in the rail and manifold fine, but i'll know for sure when I try and run the engine.

Once the car is drivable, I will record a run with the new injectors and fuel filter. if still not right, I'll install a new oxygen sensor and do another run.

[This message has been edited by masospaghetti (edited 06-12-2012).]

Update

I drove the car on the same loop with the new injectors, filter, and o2 sensor with the same results - INT values low, BLM values stuck at 120.

What else could cause this condition? Would a plugged exhaust cause this?

EDIT: I haven't yet checked fuel pressure...my harbor freight gauge bit the dust. But the pump is a new Delco unit, and now it has a new Wix filter as well. The car runs good when cold, I feel like it would run worse when cold if it were a fuel supply problem.

[This message has been edited by masospaghetti (edited 06-16-2012).]

Any other thoughts on this? I have a new fuel tester on order but haven't received it yet.

I noticed a flag in WinALDL called "Run rich flag" and it is checked for part of the time - what does this mean?

Are you still pinging now? Or just seeing low BLMs?

Although the fuel pump is recent, it's also possible you could have a leak somewhere. Make sure it doesn't leak down pressure when you get your tester. Even if it maintains pressure while running, a leak could cause other problems. If for example the cold start injector was leaking, it would cause some cylinders to get more fuel than others.

One possibility for the low BLMs is that for whatever reason the engine simply isn't pumping air as efficiently as it should. Less air getting through it, so it needs less fuel. But I don't know how much BLMs are expected to decline for a worn engine.
So it seems logical to me that a plugged exhaust could have this effect, but I'm just being theoretical here. I don't have direct experience with how much a car's BLMs will change from something like that.

quote I noticed a flag in WinALDL called "Run rich flag" and it is checked for part of the time - what does this mean?

There's a "rich flag" on the 'flag data' tab on the bottom left side. I see that turn on and off while my car is running also. I believe it's normal, it probably just means that the O2 sensor is reporting a rich condition at that moment. Or more accurately, at the moment of the last data packet that was sent over the cable by the ECM. If that's indeed what it means, it would cycle much faster than what we're able to see over the cable.
If somebody knows for certain what this is please say so.

The low BLM and INT numbers mean the ecm is PULLING fuel, making it run leaner. That tells me the ecm thinks you are running too rich.

The rich/lean indicator is how the narrow band O2 sensor operates. Unlike the wide band O2 that can report the air/fuel ratio exactly, the narrow band O2 swings from rich to lean as the ecm adjusts INT and BLM. The ecm is looking for equal time rich and lean (usually).

My guess is a bad O2 sensor. It sounds from your description that it's actually running lean because the ecm is removing too much fuel.

Thanks armos, and Hudini, for the replies.

The car is still pinging and even after replacement of the injectors, O2 sensor, and fuel filter. My fuel pressure tester is on the way. I will test my fuel pressure when that arrives but I am kind of stumped otherwise.

I'm also confused because clearly the ECM is having trouble keeping the engine running with the proper A/F ratio, as noted by the jagged rich/lean indicator, bottomed-out BLMs, and low INT values. It does seem like the ECM is receiving bad information, or maybe as you suggested armos, the fuel flow to each cylinder is inconsistent. If engine wear was the culprit, wouldn't this tend to make the engine run rich?

Not sure whether to buy a new muffler to rule that out but I hate throwing parts at a problem to fix it (even moreso than I already have).

[This message has been edited by masospaghetti (edited 06-19-2012).]

While I can see clogged exhaust contributing to low BLMs, I don't see how it would cause pinging. I would think that if exhaust is clogged it would affect the throughput of each cylinder pretty much the same, so the ECM should still be able to keep the mixture decent on all cylinders.


Is your pinging happening only at steady highway cruising, or does it happen at times of heavy throttle? The latter isn't healthy but I'm curious if you've noticed it.
During heavy throttle or high RPM (>3600) the ECM should go into "power enrichment". In that mode it will ignore the O2 sensor and try to blindly achieve a 12:1 air/fuel ratio. When doing this on a car with low BLMs, it temporarily bumps them back to a required minimum of 128. So because of the boosted BLM value, in your case the resulting pulse width should be even richer than 12:1.
I believe the reason it bumps BLMs back to 128 is to protect against cases where the ECM is confused by a faulty sensor. Due to the risk, GM decided not to trust BLMs below 128 when in power enrichment. So between that behavior and the 12:1 target, it would be difficult for even a confused ECM to run lean in PE mode.

If you've noticed it still pings when it should be in PE then I think it's most likely a physical problem, not an ECM sensor issue.
I just reread your comment about finding rust in the fuel lines and it makes me wonder if there's a flow restriction somewhere.
But this would not be consistent with low BLMs.. unless the restriction was only affecting some but not all injectors. If all of the injectors were seeing less flow then BLMs ought to go up.

The pinging occurs during acceleration only, not cruise, both light and heavy throttle. I haven't run it past 3600 rpm though.

Thats great information to have though regarding power enrichment. Does it also move INT values back to 128? Or could INT still be leaning the mixture out?

I will try another seafoam treatment through the brake booster line and see if I may have not completely cleaned the carbon out of the cylinders.

quote Originally posted by masospaghetti: The pinging occurs during acceleration only, not cruise, both light and heavy throttle. I haven't run it past 3600 rpm though. Thats great information to have though regarding power enrichment. Does it also move INT values back to 128? Or could INT still be leaning the mixture out?

I'm not sure about INT. According to an old thirdgen.org forum post supposedly INT is locked at 128, which makes sense, but I haven't seen much to confirm that. BLM is either 128, or it's learned value, whichever is greater.
From looking at the parameters in TunerPro it appears 60% throttle would trigger power enrich. There are other conditions for MAP and RPM. I don't know if you have to meet *all* conditions or just *any one* of them to trigger PE. It would seem sensible that it would be triggered by meeting any single condition.

But there's easier ways to go rich without messing with PE mode.
If you unplug the MAP sensor then from what I've seen in my logs, it will run rich. If you unplug the O2 sensor it should definitely do that, maybe more severely.
It might be worth doing one of these things to force the ECM to constantly command rich, and see if it still pings.


One other thought about your low BLMs - if you had a slight EGR leak it might cause that. The exhaust would displace usable oxygen, so the fuel pulse would have to drop in proportion to how much oxygen there is. That doesn't explain pinging though.

Looking at EGR from another direction - when the ECM has *commanded* the EGR, it will also increase the spark advance. If something isn't right with the EGR, maybe that extra advance is playing a role in the pinging. It would be easy enough to try electrically unplugging the EGR solenoid to see if that has any effect on the problem. I'm assuming that when you get a code 32 (EGR fault) it will disable the EGR advance, but I don't know for sure if the ECM is that intelligent.

armos, thanks again for the suggestions, will try.

I did receive my fuel pressure tester and pressure holds at 40 psi when the ignition switched on - and after a few minutes it had barely dropped, maybe a couple psi.