Mitsubishi 3000GT & Dodge Stealth Forum banner

Lean Misfires at 2700-2800 rpm

102K views 831 replies 58 participants last post by  mateuszyo 
#1 · (Edited)
Misfires and/or Hesitation between 2500-2900 rpm

Here is a brief summary of the 19 pages of posts in this thread:

If I am accelerating/cruising/revving in neutral, in the 2500 - 2900 rpm range, the car misses/hesitates, epecially @ 2600 rpm. I have made numerous logs of the issue and here is what I've found:

O2 Trim:
At 2450 rpm, the O2 trim starts to increase from normal (~100)
By 2600 rpm, the O2 trim hits 168 (max).
By 2800 rpm, the O2 trim starts to come back down
By 2950 rpm, the O2 trim is back to normal (~100)

If I add more fuel (14 - 17%) to bring the O2 trims back down to ~ 100, it still misses.

RPM & Timing:
If I rev the car in neutral, timing and RPM will plateau briefly once it gets to 2700 rpm. If I have the exhaust in sport mode, I'll hear a loud pop when it misses.

Knock:
While revving in neutral, more often than not, I'll see 1 - 2 counts of knock when it misses. During cruise, but more often during light accel, I'll occasionally see sustained knock counts of 1 - 2 counts from 2600 - 3000 rpm (total knock sum of ~ 10 - 12).

WOT:
Despite adding extra fuel, I'll still see a brief lean spike in the 2600 - 3000 rpm range when it misses. The lean spot, secondary to the miss, occurs simultaneously on the front and rear banks. The miss is less noticeable at WOT, but it is definitely still there.

Race Gas:
While revving in neutral and/or cruising between 2600 - 3000 rpm I see occasional counts of knock (ie. in neutral 1 - 2 counts and while cruising 1 - 2 counts continuously with a total sum of ~ 10 - 12). I believe this is "phantom knock" and perhaps the knock sensor is picking up the misses and confusing it with knock?? Alternatively, maybe this is related to my noisy rear valvetrain?

Methanol:
1) While revving in neutral, normally the rpm and timing plateau @ 2700 rpm when it misses. While manually spraying meth the car revs smoothly and there is no plateau in the rpm and timing.

2) Normally, while cruising @ 2700 rpm, the car misses constantly and the timing fluctuates +/- 1* at a fairly rapid pace. As soon as I start spraying meth the missing stops and the timing holds rock solid.

This information points to a fueling issue/air and not a spark issue.

Things I've tried to fix the problem:
1) Hotwired Supra fuel pump.
2) Disconnected stock FP relay
3) New PTE 580 cc injectors. Matched & flow tested x 2
4) Aftermarket FPR in a dual feed configuration
5) Aftermarket Fuel filter & SS braided lines
6) New ECU (used) - known to be good.
7) New OEM coils & PTU
8) Wires x 3 (Accel, NGK & now MSD)
9) Plugs gapped to 0.028 x 4 sets (NGK Coppers, NGK non-resistors, NGK iridiums and now Denso IK27s)
10) GM MAF & MAFTransltor (No longer use stock MAF)
11) Checked resistance of CAS
12) Checked resistance of PTU
13) Checked Resistance of Fuel injector resistor pack.
14) Checked compression. All cylinders are between 142 - 149 psi
15) Sent oil away for analysis x 3. Everything checks out except iron was slightly higher than average (20 ppm). June 2008:High levels of iron resolved on their own after using Amsoil 20W50. Therefore, this is not related to the missing.
16) Vac. line reduction & emissions delete
17) Disconnected TPS. No effect. Subsequently reconnected.
18) Disconnected O2 sensors and cruised around in open loop. No effect on the missing. If anything it seemed to get worse.
19) New OEM ISC
20) Increased base timing to 10*. Made the miss _feel_ less noticeable. Definitely still there, no change in the logs.
21) HKS DLI - No effect on the missing.
22) New Optima Yellow Top battery
23) All Rubycon caps in ECU & ECS changed. Traces in the location of Rubycon caps verified.
24) ECS disconnected. No change.

Things that made the problem worse/more noticeable:
1) Lightweight aluminum flywheel.
2) GM MAF in blow through?

Things others have tried that I have not:
1) Changed engine bottom end
2) Changed out noisy lifters.
2) Swapped fuel injector resistor pack

Strange things I've noticed:
1) While cruising @ ~ 2700 rpm it misses continuously. This is coupled with very slight pressure fluctations, (ie. +/- 1 cm Hg). However, once in a while there is an unusually large "miss" and the car lurges a bit and I'll see a relatively large manifold pressure spike, measured with an electronic boost gauge (ie. from -10cm Hg to 0.00 and then back to -10cm Hg). These pressure spikes are seen by the GM MAF as changes in air flow, ie. during the small pressure fluctuations AF readings oscillated +/- 30hz and during the large pressure spikes AF readings oscillated +/- 100Hz. I believe this could be a separate issue, ie. surging? It has been described here http://www.3si.org/forum/f1/all-surging-problems-316380/

2) If I cold start the car and then rev it up in neutral, it revs smoothly through the entire rpm range. Once the car gets warm, ie. coolant temp of > 120 - 130* F it starts to miss. On the other hand, if I try and drive the car right away when it is still cold, it'll still miss in the 2500 - 2900 rpm range. In fact it misses horribly when I drive it when it is cold. At times it will literally fall on its face once it gets to 2600 rpm. I think this has to do with the ECU adding extra fuel while the car is warming up and this effectively masks the problem.

3) Hot start problem. Car has a hard time starting when hot. However, if I run the pump, for ~ 30 seconds using the check connector), it starts up easily, however, it stills wants to run extremely lean and rough (as if its missing). If I try and rev it up in neutral, I can't get the rpms above 2600 rpm because of the missing. Makes me wonder if there is something under the hood, which is sensitive to heat, causing the missing and/or lean condition. This issue probably isn't helped by the fact that I'm using a MAFTranslator that feeds a fixed temp signal and I have deleted the FPR solenoid.

4) Jumpy OEM tach in the 2500 - 3000 rpm range (esp. @ 2600) while accelerating. Also noticed the tach in logs looks jumpy @ WOT throughout the entire RPM range. Unaffected by installation of HKS DLI.

5) Occasional low load sustained knock while gently accelerating @ ~ 2600 rpm. Occurs with/without the EGR connected. See up to a 12 counts of sustained knock, which continues despite 7* of timing being pulled by the ECU. Seems to continue until I take my foot off the gas. I believe this is phantom knock as it was not cured by running race gas.

7) Loud ticking from my rear valve cover, in the area around cylinder 2. I have 99 lifters, so it shouldn't be lifter tick. In addition, it sounds louder than lifter tick. I've sent my oil away for analysis and they've verified that it isn't rod knock.

8) When I rev the car in neutral, while keeping the throttle position constant, the rpms gradually increase until it gets to ~ 2700 rpm. Then it stumbles, the rpms drop to 2500 rpm and then the cycle repeats itself. If I give it enough gas, it'll eventually make it over the hiccup, until it reaches 3000 rpm and then the same thing happens again. It also wants to do it at 3500 rpm. Its only once I get over 4000 rpm that it revs gradually and smoothly.

9) Rich idle (new for 2008). When it is hot outside or if the car is really hot I get a terribly rich idle. The instantaneous O2 trim pegs out a 32 and the AFR sits at around 10.5. The result of this was 2.3% fuel dilution in my oil, which effectively made my Amsoil 20W50 into 10W40.

July 2008: Lifted my heads and pushed all of the coolant out around piston #3. Piston jammed itself in the cylinder, the bottom of the piston was torn off and the free rod bashed the block to bits. Currently rebuilding and estimated date of completion is May 2010. Hoping this problem doesn't continue.

Things I'm planning on trying when I get the cash:
1) Buying & installing an AEM EMS.

Any suggestions/ideas would be _greatly appreciated_.
 
See less See more
#2 ·
Do a search as I am sure there is a thread on this topic from 2004 year.
Before going to the AEM ems, mine did the same at that exact rpm under boost...stuttered at 2800 rpms under slight accel. can't recall why the stock ECU did that, but I also had the S-AFC in there at the time too, and the ecu was getting old at 100,000 mileage.
 
#3 ·
I tried searching already and I came across a few threads, however, none of them were very helpful. One thread suggested adjusting the base timing to help solve the problem. I've tried this to no avail. The way I see it, a bad ECU is a diagnosis of exclusion and I'd like to exclude all other possibille causes before I run out and buy an AEM.

Declan
 
#384 ·
Was going to read all these pages, but I read the first 11 and skipped to the end to find the same issue still happening.

I am of the belief that the ECU in your year car is CRAP first before any testing of IT or other testing of any other parts are done. If your ECU is CRAP then how can ANY other testing provide accurate results? Since the car is controlled by electronics (the ECU) then all test results before repairing or replacing the ECU are in effect worthless.

CAPS are not the only things wrong with your year ECU and your thread is the product of a broken ECU..... random tests/problems ALL OVER THE BOARD which I again say/point to broken ECU.

In trying a different ECU ensure the following:

From Hans:
You do not have to install the exact same ECU part number in your car BUT you do have to install a new ECU
with the same E2T number!
Pull out the ECU and under the part number on the label there will be a number starting with E2T.
Those numbers on your original ECU have to match your new ECU.

Live by this rule:
A KNOWN GOOD ECU = ECU that has been refurbed by a respectable shop especially when its a 1st gen ECU including some hybrid years.

IE: All 1st gen ECUs are BROKEN before PROVEN otherwise.
IE: A known good ECU may NOT be good just because the caps were replaced.

If the IAC is bad it can short the ECU which presents f'ed up situations when putting in a so-called known good ECU and the wonderful flip side of this is that a bad ECU can also cause a bad IAC.

FIAV is inside the throttle body and if you've ever had anyone run carb cleaner or the like through your throttle body chances are the FIAV was ruined in the process. I'm also playing with an idea that an engine that overheated a bit (not enough to kill it) could screw up the FIAV too.

Also, when you are POSITIVE (IE: your ecu has been diagnostic tested by a repair shop that handles these ECUs; send it to them no matter what they say over the phone) that your ECU is good then EVERY TIME you make a change to your system to perform testing/analysis you need to give the ECU some real drive time -days or miles running errands n such- so it has a chance to learn the new setup before you know for sure if the prob is gone. My rule of thumb is at least 1 week of daily driving.

Also be aware at least in the manual that I have the directions for testing the coolant temperature sensor are reversed from what they should be. I believe I wrote about it somewhere if you do a search for "coolant" and posts by "ProwlerGT"

Also, click my Hesitation Tracking List in my sig to see what has fixed this problem for others. Majority of the time it has been ECU and coincidentally as often as people talk about the IAC it hasn't been the fix most of the time... but it still is the fix some of the time. :) Ain't life grand? LOL - All these variables drives me nuts.
 
#5 ·
is your MAF draw-thru or blow thru?

Stock PCV system?

How is the BOV configured?

If draw-thru do you have that Y shaped BOV silencer installed in the intake?

I am trying to figure if there is air bypassing the MAF somehow.

Bob
 
#6 ·
I have a blow through system with the stock PCV system + Krankvents. I have both the front and the rear krankvents installed.

My BOV is located before the MAF and its configured to dump to the atmosphere.

I'm going to check for boost/vacuum leaks today and I'll let you know if I find anything.

Declan
 
#7 ·
I found a few small boost/vacuum leaks and I fixed them. This didn't help the lean spot at 2800.

I didn't get the chance to test for exhaust leaks, but I think there must an exhaust leak on the front bank, before the O2 sensor, because according to the NBO2 sensors my front bank is always slightly leaner than my rear, but when I pulled my downpipe last month, the front was black (rich) whereas the rear was white (lean). This would also explain why I'm getting misfires in the 2200-2800 rpm range (closed loop), but not in the open loop 2200-2800 rpm range.

Declan
 
#8 ·
Declan, could you post picks of your blow through setup please, interested in having a go at this myself.
 
#9 ·
I have the same problem running draw thru. I really have given up on it. (rough running in closed loop between 2500 and 3000) Wot thru these rpm ranges are just fine. I had worked it out a while ago by fiddling with the maft - (making the 3000 rpm range pretty rich) but have not been able to do it with my 560cc injectors.

Honestly - I am just buying time until I get the funds togather to buy the maft-pro.. I am really sick and tired of letting the stock ecu run closed loop. people running maft-pro "closed loop" with the wide-band have had really good luck with it.

sam
 
#12 · (Edited)
Yeah, that looks right. Did you cut the wire going to the FP (capping off stock relay output and using aftermarket relay output) and disconnect the resistor from the circuit?

Come to think of it, if the stock switching is still connected, the voltage spike might tweak out the system for a second.
 
#13 ·
AudioPmp23 said:
Yeah, that looks right. Did you cut the wire going to the FP (capping off stock relay output and using aftermarket relay output) and disconnect the resistor from the circuit?
I tapped into the line after the relay and/or resistor so as to run the line in parallel to the stock circuit. I figured since I have stock wiring running to the back, why not take advantage of it and it use it also.

The stock switching/relay system is still more or less connected. However, I disconnected the input (5) to the stock relay, which results in the relay being bypassed at all times. The relay can still switch back and forth, but when it does, it doesn't do anything because there is no input power source. This also has the effect of eliminating/bypassing the the resistor.

Declan
 
#14 ·
keithmac said:
Declan, could you post picks of your blow through setup please, interested in having a go at this myself.
Here are the pics you requested. I apologize for the somewhat ghetto looking engine bay. :)

If you're wondering why I have a breather filter on the HKS SSQV, its because at one point in time, I tapped a line in to the bottom of the SSQV and I was using it as a "turbo bypass" in order to help with compressor surge. This didn't work so well with the MAF only a few inches away as it created a load of turbulence and it would really screw with my AFRs. I have since done away with the "bypass" but I just haven't gotten around to removing the filter.

Declan
 

Attachments

#16 ·
I tried disconnecting my EGR to see if this would help and it did not help. I'm still of the opinion that I have an exhaust leak on the front bank. I'm going to put some combustion chamber cleaner in tonight to see if I can see a leak.

Declan
 
#17 ·
Take the BOV off and cap the fitting as a test. There may be air sneaking in at part throttle and messing with the MAF....

Bob
 
#18 ·
Thanks for the suggestion, but I did away with the fitting when I switched to blow through (a number of months ago) because of the problem you describe. Air was sneaking in at idle/part throttle which was causing turbulence.

You'd think that turbulent air would cause the MAF to see more air than there really is, but for some reason, it did the opposite. The turbulencec made me run really lean (the MAF saw less air than there was). I no idea why it works this way, does anybody else know why this would be?

Declan
 
#19 ·
Initially, I thought this problem was due to an exhaust leak on the front bank, because when I pulled the DP a few weeks back, I noticed that my rear exhaust manifold was white (lean) and my front was black (rich). However, I'm starting to wonder if its something else.

Today, I drove home to visit the parents and while doing some short WOT pulls, I noticed that I was getting, what looks like, heavy spark blow out on my rear bank. My rear O2 voltages were dipping as low as 0.60V, @WOT, in between 4600 - 6000 rpm and then it would start to knock.

At WOT, despite the voltages being as low as they were, I couldn't feel/hear any spark blow out and the voltages weren't bouncing around like they do with spark blow out.

In addition to the above mentioned issue, I still have the lean misfires anywhere below 3000rpm, but it is definitely worse between 2700-2800rpm. Contrary to what I thought before, it doesn't seem to make much difference, if I'm in closed loop or open loop, so long as I'm under 3000rpm it wants to miss.

Does this sound like spark blow out? Or it could it be an exhaust leak? Could it be bad wires, or does it sound like something else?

At the moment, I'm running ngk iridiums gapped to .028, new stock coils, NGK resistor wires, @ 17.5 psi of boost.

I'm almost at my wits end with this problem and _any_ help would be greatly appreciated.

Thanks.

Declan
 
#22 ·
I had really crap part throttle miss-fire, changed the plugs and leads but that didn`t make a difference. My realtime o2 correction was pegged at 38% when it happened (max lean?). Pulled the ECU and replaced the capacitors, the difference is night and day! totally different car to drive and no missfires at all.

I don`t think you`d be able to tune though an ecu problem such as this with the SAFC because the values make no sense when the ecu goes tits up, plus the caps performance may change while driving if they`re on the way out.

Seriously pull the ecu and change out the caps, I`ve read a service life of between 7 and 10 years for the caps so even the newer cars may start to suffer from this problem?
 
#24 ·
I also have new caps.

I question whether or not the problem is ECU related. If it were ECU related, one would think that changing the % overclock would shift the misfiring rpm point up or down. I have tried different speeds of crystals and they have no effect on the Rpm point at which it misfires.

I'm also wondering if this issue of "spark blow out" on my rear bank is related. I've never had issues, to this degree, with spark blow out. At WOT, my rear o2 voltages are dropping to 0.60V between 4600-6000rpm and then they go back to 0.91V from 6000rpm - 8200rpm. Like I said before, it doesn't feel like spark blow out and the o2 voltages aren't bouncing around. However, I have noticed that my timing will start zig zagging a bit in the 4600-6000rpm range when it is blowing out. The strangest part about it all, is that if I do back to back runs, on one run it'll bounce between 0.89-0.91V and the next it drops all the way down to 0.60V.

Any ideas?

If I was to try swapping in an ECU from a 93 VR4 (with a 95 engine), would this mean that he is using a 95 ECU or a 93 ECU and would I have to make any changes to my car or his ECU, so it'll work with the CAS?

Thanks.

Declan
 
#25 ·
DoctorDex said:
I'm also wondering if this issue of "spark blow out" on my rear bank is related. I've never had issues, to this degree, with spark blow out. At WOT, my rear o2 voltages are dropping to 0.60V between 4600-6000rpm and then they go back to 0.91V from 6000rpm - 8200rpm. Like I said before, it doesn't feel like spark blow out and the o2 voltages aren't bouncing around. However, I have noticed that my timing will start zig zagging a bit in the 4600-6000rpm range when it is blowing out. The strangest part about it all, is that if I do back to back runs, on one run it'll bounce between 0.89-0.91V and the next it drops all the way down to 0.60V.

Declan
I had a similar issue with the my front o2 going very lean at those rpm points, but I never dared pushing it above 6000rpms. My problem turned out to be the coils on the front bank.
 
#26 ·
Just how lean or rich are those O2's when you think it's spark blow out? What are the AFRs exactly?
The 'bouncing timing you see, are you positive it is not just the ecu pulling timing when knock is coming on?
And what the heck is "TurboBob ecu"?
One of the first symptoms of an ECU (Mitsu uses) going south is the fueling...for what ever reason, a crappy ecu will often start to have problems controlling the fueling..then too lean..knock or like a bad O2 sensor run with leaded fuel, it starts to read rich so the ECU starts to lean her out terribly.
Best bet_ find a stock 1st gen ECU and try that.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top