Slightly lean idle & miss

[curt-s]

So I've run into an issue lately and I'm not entirely sure when exactly it started. I know this is a novel in itself, but I want to make sure all the points of what I've experienced and done are there.

Let me first say that only a month or so ago, I would idle around 14.7 most of the time but randomly at idle, my AFR would go to 16+ when my O2 would read 0 volts and the engine would want to die out. If I just -tapped- the throttle, barely even pressed it and only increasing my TPS % by 1, O2 would come back to life and the engine responded accordingly. On the opposite side, my low fuel trims reached almost 120% and my mids were almost 110%. I spent most of my time in the low, apparently.

Now, however, it's a different story and I'm trying to eliminate possibilities.

The problem? At idle, it ranges between 15.2 and 15.7 with an occassional blip to 16.5+ causing by what I'm figuring is a miss based on slight engine stumble, exhaust pop, and sudden lean spike. Does this fairly often after it's warm.
I've got phantom knock right off idle on acceleration, ~2000RPM consistently, sometimes 2500rpm, and touches here and there up until about 3.5k. Really pisses me off because I have to let off the gas until it goes away otherwise I end up with 40+ sums and MMCd won't stop beeping. Everything below 2500rpm I believe is phantom knock just because it's reproduceable almost every single time and I'm always off boost at those points with low load. It seems like if I give it lots of load, then the knock isn't nearly as bad.
WOT in 2nd and 3rd will put me at 10:1 or below AFR. I will get knock up high and around 15psi, hitting 1500Hz on the MAS. Pulls like a raped ape, though, and I lose traction in second after 3.5-4k when hard down. Of course, they're M+S tires I just found out recently.

Build:

2.0 std bore 6 bolt w/ 1g head
1g pistons and rods
crank 0.01" over
King XP mains
Clevite rod bearings
3g lash adjusters
stock turbo cams
14b
evo 3 manifold w/ 4 layer gasket
megan O2 and downpipe
3" stovepipe straight-through exhaust
stock intake manifold
fiav delete w/ blockoff plate
egr blocked off
throttle body cleaned
450cc injectors (~140k on them) recently cleaned with pressurized carb cleaner + AA battery. 1 entire bottle for all four
new upper injector o-rings, lower seals
fuellab afpr, set to 37-38 psi off vacuum, does that damn leak-down after pump turns off
recently rewired walbro 255lph
fidanza 3.2 w/ pp + aluminum flywheel
new 1g knock sensor
prothane full mounts all around (no inserts)
new PCV valve + plastic check valve (properly set direction)
new valve seals, SS valves, seats cut and lapped
new TB shaft seals, but they *still* leak on the throttle cable side if I move the shaft up and down
new Denso O2, only has 1200km on it.
AEM UEGO installed after flex section
1g MAS and stock intake, 1g bov uncrushed and recirculated
full hard intercooler piping, some generic wide fmic
no exhaust shields (want some desperately though)
j-pipe and upper elbow piping does get pretty damn warm after a little while
Autometer 30psi boost gauge
Autometer 100psi mechanical oil pressure gauge with -4AN braided, tee'd off OFH w/ stock sender
OFH port (12psi at idle)
Full BSE
NGK BPR6ES gapped to 0.025 (tried 0.028 as a test; no change, but I haven't changed back)
Manual boost controller, bleed-type, steel ball, from RTM Racing. Set to 15psi, but the BOV leaks around there and I'm a little weary of my MAS Hz causing cut.
Ground straps to intake elbow, intake manifold support, downpipe (Do I need more?)
Replaced ISC o-ring with a home depot part, closest I could match, but I've ordered the correct ring and am waiting on it.
MMCd to log only, non-EEPROM ECU

It's a mainly stock build at this point, with some foundation laid for more down the road. I've got about 1,150km on the engine and for the most part it's been acting pretty good except for this bit.

Note: idle vacuum is between 19inHg and 20inHg at 750rpm, dead constant with no needle fluctuation.

I've also got a set of BPR7ES brand new in box purchased today that I'm going to swap just to see if maybe there's a plug issue.


What I've done:

Cold BLT off compressor housing when 30 degrees ATDC. Slight bubbling heard out of valve cover, crank case does pressurize slightly but not from PCV. Must be from valve seals. Pressurized to 18 PSI, takes 1.5 minutes to leak down to 0 and loses first 8 psi in 30 seconds. Only place I can hear a leak besides VC holes is the cable side of the TB. Damn seal. Still I figure 1.5 minutes is pretty good and shouldn't be causing TOO much of an issue. Fixed one tiny leak at a coupler, but other than that, soapy water found nothing. I use a regulator and ball valve off the coupler to turbo so that I can stop the incoming air charge and let it bleed out of the system on its own.

Reset the BISS to idle at 750. The funny thing is while I was doing this, my AFR was right where it needed to be -- hovered around 14.7. I needed to back it out about 2-2.5 turns. Shut engine off, removed jumper wire, turned it back on, and it idled high right off the start my AFR was back at mid-15s. I don't recall hearing any miss at the time but I wasn't paying 100% attention to that. I was in shock that my AFR was spot on. Now that it's set, if I come off cruise and clutch in, RPMs will stick to around 1000-1500 until I come to a complete stop. After a couple seconds the ISC will bottom out and I'll be about 750-800.

Reset TPS by disconnecting throttle and backing out idle position switch. Verified butterfly closes entirely. Set TPS to 10.9% via MMCd, reattached throttle cable and set idle position switch. Verified TPS is still 10.9%. However, once the car is running, TPS jumps up to a bit over 11%. I will redo this using voltages right off the sensor rather than trusting MMCd values. Verified resistance values throughout range.

Reset idle position switch. Verified functionality with MMCd via FLG2 value.

Swapped transistors to a spare. Caused stumble at idle, and shut the engine off once while on the highway. Swapped back.

Swapped coil packs. No change. Cleaned contacts, even sanded the spark plug wire contacts and shoved them in good and deep. No change.

Verified ignition timing is constant at 5 BTDC.
Verified mechanical timing is correct.

Reset ECU because my low fuel trim was 82%. Yes, 82%. After operating temperature was reached, fuel was pulled continually until I was under 90% again.

Resoldered injector plugs, TPS plug, O2 plug, coil pack plug. Wires were pretty frayed, so I used better condition plugs and soldered (I know how to properly solder with my Hakko and do not doubt my connections).

Tightened up the exhaust manifold nuts, all turbo related bolts were unmoveable.

I do not see any carbon buildup from an exhaust leak pre-WBO2.

During low-RPM cruise, my AFRs will hover around 14.7 as expected and LFT will be pulled to around 90%.

I have not discounted the possibility that there is a pre-WBO2 leak, maybe in the flex section or otherwise. I will not negate the possibility, but having never dealt with a leak like that before, is it possible to have a leak indicating a fairly consistent AFR? Maybe the frequent miss is another unrelated problem. The idle even "feels" more rough than when it used to be at ~14.7.

I sat at watched my low fuel trim actually oscillate between 82 and 81 yesterday after it warmed up and my AFRs were still mid-15. So it looked like the ECU was -still- pulling fuel, but they jump back up during real cruise.

I have been able to get the ISC to hold 750rpm idle with 30-50 value and I have no surge.

I will also entertain the notion that my fuel pressure gauge plumbed into the fuellab AFPR may be faulty, causing less pressure to be read. It must be linear then because I'm 37-38 off vacuum and 30 at idle at ~19inHg.

 

[1davho]

I don't know anything about mmcd, but it sounds like your injectors aren't dialed in. And 90% fuel trim means its adding that much fuel, not taking it away. I would start by changing global and deadtime to get your fuel trims close and see if that gets your o2 sensor to start acting correctly.

 

[curt-s]

With a 1g, it's the other way around. 100% trim means it's not adding or subtracting. Our trims revolve around 100% mark, with maximum being 139% (adding 39% more fuel) and the minimum being .. I think 80% (subtracting 20% less fuel).

I've had 139% trim before when I witnessed what the removal of the lower honeycomb in the MAS does (by previous owner).

These are stock M/T injectors for the stock M/T ECU without DSMlink.


A note about my oxygen sensor: it does appear to operate correctly. On cold start, WBO2 reads about 13, and then slowly advances higher to where it sits once the O2 heater is up and running and oscillation begins. Oscillations at idle before 190F coolant temperature are slow, but the sweep the range. I've noticed at hot idle, though, oscillations are much quicker but they also don't seem to sweep as much unless my MMCd refresh rate is too slow (between 3 and 4Hz). Entirely possible.

I don't recall at all what I did between when it was wonky before with proper AFR but intermittently dead O2 at idle causing stall-out unless I ever-so-slightly tapped the throttle to keep it alive, and now, where there's no stall out but a miss and slightly lean idle. I'm running the same transistor and coil pack, just cleaned injectors and swapped resistor pack.


My Q is how could the WBo2 be reading 15.3+ lean, when it's subtracting fuel, not idle surging, is missing, showing steady 19-20inHg vacuum at 750rpm, and passing a BLT rather well. Add to this that I can get it to center the ISC fairly reasonably to hold that idle too and when I do play with the ISC enough to do it, RPMs drop to 750 right at clutch-in without any lag time waiting for me to come to a stop (where it would then bottom out). That sounds right to me. It's the black plastic ISC, btw, with resistances of 43ohm on each coil which I've read is correct with these particular ISCs.

I've also replaced the caps in the ECU a couple years back, using high quality parts.

 

[1davho]

Ok. Gotcha. Maybe your fpr is bad, letting too much fuel in at idle. Or like u said, the gauge might be off

 

[curt-s]

That could definitely explain so much fuel being pulled, and the downpipe flange or flex section, or O2 housing could have a leak. Not sure about the stumbling and miss, though.

However, that might have been fixed.. I put in the BPR7ES gapped to 0.028 (best guess, my wire gauge had 0.030 which didn't fit, and 0.025 which did so best effort ) and she has more pick-up off the line, low fuel trims went up a bit to just under 96, and just feels smoother across the entire RPM range. I think either something fouled the old plugs, or one or more were bad. Attached picture of old plugs:

You must be logged in to view this image or video.

Maybe something was hindering a good grounding. Looks like anti-sieze but I don't recall putting any on.

I've still got a slightly lean idle, but it's possible there's an exhaust leak somewhere. I'll have to check it out and put some more kilometers on these plugs to see if it's fixed. Still appears to miss at idle every now and then, but don't these engines usually have that? Maybe I need to get in the plug holes and clean up the threads.

*added*
Youtube video, as proof..

You must be logged in to view this image or video.

If anybody is wondering, I "routed" the spark plug wires like that on purpose to keep 1&4 and 2&3 as far away from each other as possible. It's not a permanent solution, just checking. They're Taylor 8mm from the their Sport Compact series.

I should also note that I gapped the previous BPR6ES to 0.025 as a test only with no change. I normally ran 0.028, as I am with the 7ES right now. I'm not saying I needed the colder plug, just wanted to get new ones to see how things were going to act with brand new ones. I figure 7ES and 6ES at my power level should be relatively similar in functionality.

 

[curt-s]

Update:
Interestingly enough, my low fuel trims have returned to hover around 100%. If I idle for a while, they start pulling fuel, but so far so good.

Another side effect is if I go to 10psi or above, my AFR reads just above 10 now in the 10.1-10.3 range and doesn't really bottom out. On the flip side, it doesn't "feel" quite as fast. Maybe my clutch is starting to slip already. Next time, full faced street disc -- none of this 4 puck kevlar catchphrase.

Still have the slightly lean idle, perfect 14.7 at cruise, and semi-frequent miss now. Can feel the engine miss at idle, almost like it quickly stuttered and resumed.
PK down low still, more so off the line if I clutch out below 1500rpm. PK almost always at 2000rpm, regardless of load or gear/neutral, with 2500-3000 sometimes as well. Once I get on it, it won't hit 4 counts or higher until I'm in the upper bands in 3rd gear. Fourth gear is just a crapshoot.

I'm going to look at my exhaust and knock sensor this weekend.


Oh, and another thing I've witnessed ever since the ~500km mark is it sounds like idle detonation. The tiny 'snap' sound every so often when the engine is up to operating temperature.

 

[curt-s]

Still have the miss, but I fixed the slightly lean idle.

O2 housing to downpipe gasket (both Megan parts):

You must be logged in to view this image or video.

Well dang, didn't know it could throw off the reading that much. Threw in a spare 2.25" I had laying around to get it up and running and it seemed to work but after a couple hours, it crept back to low 15. Picking up a 3" anyway and it'll be torqued properly and retorqued after a heat cycle or two.

 

[Simpleman1]

I noticed that I have almost the same issue with my '90 Talon. At idle the AFR reads between 15.6 - 16.5, alot of time it pegs to 17 even in normal cruise. At WOT I get 11.5 most of the time, ocationally I see 10.2. It has been like this for a year or more, seems like it is getting worse. I have DSMLink and tuned few years ago. I am interested on what the DSM community can advice. The leaky downpipe is something that I can check when time permits. Thanks.