Inconsistent misfire around 2000 rpm

[Nicholas H]

Hi there everyone,
I did notice another user posted a similar issue who was having a similar issue around 3000 rpm but i just thought i would document and get some ideas to my particular problem. I have recently gotten an Ecotec 2.2l running on MS2 with 2.4l COP wasted spark, mototron 80 lb injectors, 6+1 crank trigger, borg 259 turbo, custom intake, divided header into the twin scroll turbo, secan interooler, meth injection setup, in a fiero (450 whp). Anyway i am having a misfire issue around 1400-2500 rpm. Some days it is non existent and it runs perfectly fine, other days without even having attached a computer to it since the day of running well it will misfire and buck constantly at those rpms then clear up everywhere round them. I have attached the current tune file and 2 data logs. One is of the car running smoothly (smooth rpm) after having done some tuning and another is of the car running as bad as it has (rough 2000rpm) and this was without having messed with the tune at all. Tonight i did change the injector pwm current limiting from 75 to 100 but did not help any that i could notice. The problem seems to progressively become worse as i drive the car almost like a heat issue but i have swapped coil packs out with fresh new cold ones when the problem arose only with no effect. The injectors measure 12 ohms when cold and 13.4 ohms after misfiring (hot). I have added fuel and removed fuel from the trouble zones only to no improvement.

One note i do need to make is when i first assembled this MS2 i did make the mistake of soldering one diode in backwards frying a circuit path just below U4 and taking R56 with it. I did sent the MS2 out for getting repaired and the company did repair the components but they actually did not catch the fried circuit path which powered the injectors causing the car to not fire up the first time until i found it and repaired it. It is because of this i am questioning if the CPU could have been damaged and is acting erratically. Is this at all possible or is it such a long shot? It has taken firmware when i reflashed it but i have not tried recently and i am contemplating trying this.

The other user i saw with a similar problem was solved by going to an older version of firmware. I am currently running 3.3.3 firmware. This is just such an odd issue i can't pin point it to an injector issue or ECM issue or possibly something else i haven't though of yet. I'm sure i have missed some pertinent information and apologize for the long winded paragraph.

Thank you in advance for any of your ideas as to what this could be.
-Nicholas

[Nicholas H]

Attached is a picture of the installed engine in the fiero. It has been a year and a half in the making. Just this last hiccup really before actual tuning progresses and the fun can begin. So any help in eliminating this misfire would be appreciated. If building a new megasquirt has the possibility of fixing it i am willing to do it.

-Thanks again,
Nicholas

[elutionsdesign]

I just took a quick look at the log, in the area that seems to be affected you've got a lot of ignition advance. 33 at 2000 rpm in a modern combustion chamber DOHC seems excessive.

[Matt Cramer]

The only thing that really stood out to me in the logs is that it goes pretty rich when acceleration enrichment triggers.

I don't know what you're using for coils, but 2.2 ms is not very much dwell.

[Nicholas H]

The acceleration enrichment i haven't completely figured out just yet and do need some work with that definitely. I have a slight stumble if i lightly tap the pedal, but not if i lightly roll into the throttle. Just some playing that i will need to learn and experiment with. I had questioned the timing around that area before but received the suggested map from a company that builds turn key ecotecs. I will try to take a little more timing away from those areas and see if it helps at all. I did remove some before with me trying to be conservative but i will try to take out maybe 5-8 deg and see what happens. I really appreciate you guys looking into this issue. I hope this issue is really just a timing problem but I guess my one track mind felt as though it was a fuel issue.

Thanks again and I will report what I find later today after i turn the timing down,
-Nicholas

[Nicholas H]

Unfortunately dialing back the timing was not the issue this engine is having. Even after cutting the timing back to around 17 under cruise the popping and missing is still the same as it was before. I have attached today's trial tune and the data log of the car at its worse moment today. i did think that it ran more smoothly with less timing in it when it was not misfiring so i will keep my timing dialed back i think but i am really unsure what else could be causing this issue. If you can think of anything else to try just let me know. I appreciate your time looking into this. I would like it to turn out to just be a tuning issue but it just feels like a physical problem with an electronic to me. This is the first car i have ever tried to tune so i could be very wrong as is why i am hesitant. The misfire is rather faint during warm up and early on in the full warm temps and then slowly progresses to feeling the misfire as the car bucks.

Thank you,
Nicholas H

[Matt Cramer]

Have you confirmed the timing you see in TunerStudio matches what you read on a timing light?

What sort of coils are you using?

[Nicholas H]

I did confirm timing when first setting up the megasquirt but I have not checked this in a while so I may triple check this to be 100% sure. The idle is alright and the misfire did not change with removing 17 deg timing so it did not indicate to me that the timing is off. The coils are from any of the 2.4l or 2.0l ecotecs and have similar characteristics to the ls coils. they are a logic level triggered coil. I do not have a cam sensor so this engine is running batch fire and wasted spark. I have kept the coil dwell to a minimum just be sure i am not over heating a coil (I plan to slowly bump this up to 2.5 ms). I am not straining the ignition much with only 5 lbs of boost i figure and do not experience any issues with more rpm or more load.

Thanks,
Nicholas

[Matt Cramer]

If they have the same sort of dwell characteristics as LS coils, I'd turn the dwell up to 3.2 ms.

[Nicholas H]

Alright, so i triple checked timing with the light and it is correct and i tried upping the dwell on the coils to 3.0 then to 3.2 then to 3.5 for a little bit only to have no effect on the issue. I am feeling more confident on something getting warm as the car runs and causing the issue though. When the car is first started the misfire is extremely faint and can hardly be noticed. Even after the car gets up to temp around 195 deg F the misfire is rather faint but sometimes lightly audible. After driving for while and parking the car for 3-4 hrs and driving again, the misfire is extremely evident. This amount of time allows the engine to cool to around 130 and intake temps possibly 8 deg over ambient. Have you ever heard of an injector causing something like this? I have tried swapping coils after everything is hot without having an effect so the only other thing i can think that could be causing this would be an injector or something in the megasquirt. I have left the Megasquirt cover open to cool whatever may be getting hot but it did not have any effect either. I could not feel anything getting warm on the board but i have not tried to use one of those canned air products to freeze the megasquirt after running yet.

-Thank you,
Nicholas H

[Nicholas H]

Just as an update, I have not been able to find the problem that i am having. I built a brand new megasquirt to eliminate a variable but has been unsuccessful in solving the problem.
Symptoms: misfire at 2000-2500 rpm. More prominent as temperatures rise in engine bay.
What i have tried: adjusting fuel and timing around problem areas to zero effect, new plugs, new coils, new megasquirt, adjusting fuel pressure up and down, ohmed the injectors, mototron 80 lb/hr before and after problem occurs (12.2 ohm cold, 13.5 ohm after problem arises), and i am still considering buying good Injector Dynamics 1000s to see if maybe that is the issue.

What i have seemed to notice is that after letting the car sit for a couple days without starting, the car will run just fine without a hick-up no matter how hot i get everything. After starting the car approximately 5-6 times over 2 days the problem with the misfire comes back and every time after that when i start up the car the misfire persists. This issue just does not make any sense to me. Hopefully one of you has an idea that i can try.

I suppose i cannot rule out mechanical engine problem but the problem does not behave as such.
The megasquirt grounds are all tied to a single stud on the head of the engine with the battery ground directly attached to the same stud. The wideband heating side is grounded to chassis to reduce noise. If there is anything you can think of that could be causing this issue please feel free to chime in. I am at a loss.

-Nicholas H

[billr]

Whenever I see a problem that is only in an rpm range like that, I think of vibration of something that is going into resonance at a frequency related to the rpm range. My first guess would be a bracket for the CKP or CMP sensor, but you aren't showing a sync-loss, so maybe a wiring connector somewhere.

That doesn't really correlate to your observation that the problem goes away after a long "rest". The only thing I can think of for that is hydraulic valve lifters that are slowly pumping up too much and finally relax after a long period with oil pressure being at zero. Or maybe something with the hydraulic VVT mechanism (if it has one!)?

Really just trying to give food for thought...

[Nicholas H]

I have wondered the same thing with the lifters myself. It may be something I need to consider trying. They were cleaned up used lifters so anything is possible. I do appreciate your thoughts though. Sometimes it takes a fresh view to see things I have overlooked. On the drive home today the car ran pretty well but might have had a very slight decel popping. If it did it was hardly noticeable.

Thanks again and I will keep the updates coming as I try to find this issue.

-Nicholas

[Nicholas H]

I Have been unable to come up with any solid conclusions but i have done a compression test and all cylinders read 180 psi when the engine was cold. Plugs looked as i would expect from the afr i am running, dark around the body, white/tan porcelain, dark area on the ground strap just after the radius towards the threads, and heat color from the electrode just after the tip of the ground strap. All indicating that there is not a severe problem with timing or fuel (rich on idle but i will deal with that later) I have sprayed carb cleaner all around the intake checking for vacuum leaks, played some with timing with no real effect yet, and closed the gap on the pugs to 26 thou. The misfire has become much more consistent now over the last few days and is consistently there but is not nearly as bad as it once was. It still does have good days and bad days but on its good days it is still audible but cannot feel it. I will be trying a leak down test on the engine to see if there are any abnormalities there but with the compression test having read alright i am not really expecting anything. One note that i may add is that the exhaust on this engine gets very hot. I am new to turbo cars and this being a mid engine without the benefit of getting all the front ram air a normal car gets I am unsure what is normal. I built an under car scoop/spoiler that sits maybe 3-4 inches from the ground angled upwards to try and move some air over the header and turbo and i have also installed a small 6" spal fan in the quarter panel to move air when sitting. After driving for about 25 mins on the highway at 2500 rpm at 10-12 inches of vacuum, the header got hot enough to soften and pop a hose off of the waste gate (rubber fuel line type) which lies on the underside of the collector away from the header as much as possible but still subject to radiant heat, everything around the header was way too hot to touch as well. I am wondering if this faint misfire is causing the excessive heat or if it is just just that naturally hot around a turbo. I will be adding a vent scoop right above the engine to cool it down further back there. I will upload my most recent data log and tune later tonight when i get back home if any one is interested.

As always, thanks for reading and any ideas that you think may be the issue,
Nicholas H

[Nicholas H]

Sorry for the delayed post with my tune and data log. The car has been running much more consistent for the last few days. I have not changed the tune just to see what it does. I am hesitant to say that it is missing per say any more but rather a very weak fire around 1500-2500. I will probably try a little more fuel and maybe a little more timing around this area to see if it changes. I know i need more work around the accel enrichment area, maybe a little less initial enrichment and more fuel above the idle cells or vise versa. I don't have a huge TB on the engine and it responded well to less accel enrichment last time i tuned.

I will continue to tune later this week and see if it changes if any and hopefully for the better.

Thanks,
Nicholas H

[Matt Cramer]

The only thing that really stands out to me in that log is that the battery voltage is a bit jumpy.

[Nicholas H]

That is something i have not taken a look into actually. I noticed this a while back but could not find any information about the typical voltage fluctuation seen by the megasquirt to know if it was abnormal or not. I may try to unplug the alternator and see if anything changes having a constant batt voltage. A short drive of course. I am currently using a GM CS130 D alternator with a resistor inline for the exciter signal. I will have to look into what resistor value i used for this but maybe there is something to this. This engine never did have this type of alternator to begin with and neither did the car so there is definitely some unknowns in the charging system. Are voltages seen by the megasquirt normally a flat line for the most part?

Thanks for looking,
Nicholas H

[Matt Cramer]

The MS's grounding can also affect noise - where is it grounded?

[Nicholas H]

All 5 of the grounds were put into a single connector that is grounded directly to the aluminum head and on that same stud is a very heavy ground wire that is connected directly to the neg on the battery. I will grab a picture that shows how it is routed. Basically all the ground wires from the megasquirt are directly connected to the battery terminal. It is a short run probably 2-3 ft in total wire from the megasquirt to the battery. There is also a chassis ground that is part of the battery neg wire that connects just below the battery on a cleaned spot of the chassis and there is a ground strap that is on the same post as the megasquirt grounds and the battery neg cable. I will post with a picture as soon as i arrive back home.

Thanks,
Nicholas H

[Nicholas H]

In these pictures u can see the grounding scheme I have going. The black loom has all of the ground wires from the megasquirt that are not spliced coming from just in front of the engine past the firewall (short run). The large black wire is bolted directly to the battery and the other end directly to the head. The other ground wire leading from the battery terminal is bolted to the chassis. And another ground strap is from the deck lid hinge to the ground bolt with the megasquirt and battery ground lead.

-Nicholas H