Random misfire, possibly injectors? (warning, long post, I’m desperate)

Attachment 50836Attachment 50837Attachment 50838

I made the changes GHuggins suggested above. Stoic to 14.1, O2 switch point to 550 with 500 low and 600 high. Multiplied the air flow tables by .80 too. Then went for a drive...

I just flashed the changes, let the engine warm up and off I went. Dot - 02 is the first log. Dot - 20 MAF is another log with speed density disabled to experiment a bit. Dot -04 is SD enabled again.

No reduction of note in the miss. AFR still seems to want to find 14.7'ish and call that stoic. I suspect there is no way trick this ECM either. Maybe, but I dunno....

A very interesting note however, and I'm calling it a minor victory. And that is the trim disparity between the banks. It was reduced significantly, like 30%.

Also, please look closely at the MAF only log. I drove up my 'test mountain', and put a load on the engine in fifth gear, keeping rpm at about 1800. Time between 10 and 12 minutes in the car is climbing the grade. The miss completely goes away. I'm not really seeing anything different than throttle position, up a couple percent. Then I just drove back down the hill.... Trims go to crap without load and the engine just hates that.

Spark and KR: Yep, I know..... I am constantly chacing KR. Crappy assed 90 octane fuel. The tables still need work. Problem is, I'm not sure what else I can do. I take spark out, KR pops up somewhere else. I have the PE engagement set pretty aggressive. I suppose I can just move the TPS enable lower, but it's pretty low right now. Disabling DFCO is an option too I suppose.

I was working on transients. I think that's probably what I need to spend a lot more time on. They are somewhat of a mystery to me though.

I can see in the logs what's going on and I took a look back at the other tune on the other "like" problem I was referring to earlier that I was able to fix, but your not going to like it....

We have made some pretty decent strides I would say making your problem better, but without the O2 tables, I'm not exactly sure on how to work around it.

I haven't checked out your latest logs, but if you watch your post prior to these, slow the display down to a 2 or 3 setting on your datalog and watch the dial and bar graph display screen when your holding it at steady throttle in the troublematic rpm areas pay close attention to your O2's switching corresponding with timing and the tach gage. Doesn't show all the time, but there were some areas where it was everytime...

The O2 goes lean "AF goes to around 15 to 15+ at times", timing starts jumping rapidly +/-.5 degrees and the tach starts dancing about 25 or so rpms. This is that "dragging burlap bag" sensation... Then the O2 switches rich "AF goes to around 14.1", timing smooths, tach smooths.

I dialed in the O2's then MAF on the other like problem car that cured ALL of it's problem...

You can try really increasing your min airflow table now until you get self cruise then back back off a little, but this again will only lessen the problem. You can also try shifting the baro correction table - this is used to calculate all fueling - may or may not help. Usually helps more on boosted cars...

Basically it's exactly what you thought from the getgo. Misdialed O2's combined with natural occuring reversion of the cam you've got.

Go to HPtuners.com and email tech support asking for the tables to be added. At that time Bill should email you back asking for the tune. If it's possible, hopefully they can get them added and you can fully fix your problem.

You'll even find after dialing in the O2's it's easier to richen or lean out the O2 settings and get them to actually do what you want them to do.

Sorry can't be of any more help right now.

Maybe someone knows more than I do on the subject and will perhaps know a work around?

Well thanks anyway for all your time and input! I'll review the logs some more to see what you are describing.

I did email tech support, along with my tune file and the screen shots you posted earlier, asking if it was possible to add the O2 tables. Reckon I'll just wait and see.

You can still try increasing the min air flow table in the 1000+ rpm areas. I would set that back to stock then add whatever you added to idle to get it that high - looked like you might have added somewhere around 3 or so...

You can even try shifting your baro table - it's not going to hurt anything and see what happens.

Curious to see if HP will add the tables.

Sorry couldn't help more.

You helped me quite a bit actually. I learned a few things too. I owe you a few beers. I'll mess with air over the next couple days. I have to head back to the salt mine and earn a pay check..... What baro table? The one under the VE buttons?

I'm curious what this VE Reversion button does. "VE Reversion Vacuum: Below this vacuum, reversion conditions will be set".

What in the heck are the "reversion conditions"?

Not the vacuum. I don't ever mess with that one. I typically change "still under the VE section", the Baro Correction table. I attached a stock table.

This is used for base fueling. Don't be surprised if you change it and your fueling is shifted slightly or in other words you have to add or take out of the entire area.

BUT on boosted cars, I have seen tremendous gains from playing with this and have not seen anything af ratio change durring wot after the MAF curve was corrected. Don't know exactly how it works other than it's used for base fuel calculations.

Change it, if nothing happens benefit wise, change it back... Simple enough.

GHuggins,

Bill got back to me on the table request. He said he needs me to list the "Table ID Numbers" that you find in the help menu, you get when you scroll over the button for the table. I hate to ask for more help from you, but do you think you can pull up that tune file that had the tables and let me know what their ID number is, along with the official name they are called? Once he gets those ID numbers, he said they would add this request to their work queue.

I suppose now would also be a good time to ask for any other table that might be needed for this car & ecm, if anyone hasn't done so already......

Thank you, thank you.....

OK, these are the only ID numbers I could find and they weren't in the help section, they were right under the tables themselves, so I hope they're what you need.
I got all of the missing O2 tables that I could find. I would recommend going to the tune repository and opening up a different engine platform like the 6.2 ls3 and comparing your tune to it for any other missing tables.

[ECM]-12470 Closed Loop Integrator Delay vs. Airflow Mode
[ECM]-12481 Closed Loop Proportional Gain vs. O2 Error
[ECM]-12484 Closed Loop Proportional Mult. vs. ECT

After comparing tunes and you see any in a grey color and you can click on it and find only one vehicles table present under the compare tabs, let me know and I'll see if I can get those for you too.

Thanks!

Michael I have the exact same problem you have after installing trickflow heads, and a eps cam with 12* overlap. Been chasing it for over a year, tried everything you guys have mentioned and still have a slight lean miss around 1800-2100rpm. Hope someone finds a cure.

Quote:

---

Originally Posted by arsautorepair1

Michael I have the exact same problem you have after installing trickflow heads, and a eps cam with 12* overlap. Been chasing it for over a year, tried everything you guys have mentioned and still have a slight lean miss around 1800-2100rpm. Hope someone finds a cure.

---

Even with TF heads eh? Hmmmmm.....that's great feedback because I have been considering those heads, thinking the port velocity and retention of the swirl dam might help.

I am playing with spark again, and LTFT boundaries (someone gave me some tips offline to try) right now.

I am fairly certain that the problem is due to a lack of control over the O2's and we need the tables noted above. Bill has added them to their work queue, but no estimate when they will be added, or if its even possible.

So have you observed a high bank to bank trim imbalance?

Quote:

---

Originally Posted by Michael_D

Even with TF heads eh? Hmmmmm.....that's great feedback because I have been considering those heads, thinking the port velocity and retention of the swirl dam might help.

I am playing with spark again, and LTFT boundaries (someone gave me some tips offline to try) right now.

I am fairly certain that the problem is due to a lack of control over the O2's and we need the tables noted above. Bill has added them to their work queue, but no estimate when they will be added, or if its even possible.

So have you observed a high bank to bank trim imbalance?

---

Yeah the swirl port in the trickflow heads won't help most likely. I get lots of lean misfires exactly like your describe, mostly on driver side bank and cylinder #1 get 3x the amount. Lowering the timing down to 25 and running open loop at 14.1 around 1400-2000rpms, and injection timing is the only thing that has helped, problem is I only get around 12mpg on the city and 20-22mpg on freeway. I think the cam revisions just like cylinder #1 and the drivers side

Let us know how the ltft boundaries does...

One other thing that you can try which again will "help" but not necessarily cure is increase dwell time on your ignition parameters. I wasn't able to completely elliminate it on one today, but was able to reduce it drastically and nearly completely down to around 1200rpms. I increased dwell in all warmer temp areas to 2000 rpms by 10% then 5% to 2200 rpms. It helped, but that car was running tr7 plugs too. Just trying to provide a better spark in the light throttle cruising areas if I'm thinking of that correctly?

]I must have different dwell tables. I'm not really seeing how I would adjust dwell for temp/rpm....

I don't know if changing spark helped for made things worse. No real change in the miss to speak of. The table certainly 'looks' better though, so I'm sticking with it and will bump up spark in small increments. Chasing KR and smoothing/blending made the tables look like an angry beaver built them.

LTFT boundary change may have helped, but if it did, it wasn't this miss thing... Seems kinda stupid to have LTFT cell boundaries with an enable rpm set to 8192. What's the point in having them, if they are essentially disabled?

I took a couple more logs after flashing the new file. I did a complete write this time, and not just a write calibration. One log is with SD disabled, along with the fuel adders / cuts disabled, and the other is just normal with everything enabled.

MAF was off a bit, so I tried to re-cal it this evening, and just gave up. I get it so trims look great in say, 2nd gear, but then they go to the crapper in fourth, or any other gear. I just don't understand what's going on at this point. Something is just simply wrong. The same Hz cell, but in different gears shouldn't be so far off with MAF, should it???? Load changes, but the air flow is in the same cell. What am I missing??? Seems like once air flow gets to about the 5000Hz cell, trims are pretty good in any gear, but lower than that.....they go batshit.

Someone suggested that the injector data is questionable. I copied the pulse adder data to a new excel sheet and attached it. I'm beginning to think he's onto something there...... Maybe it's because pulse length changes with rpm?? I dunno, just guessing at this point. I'm getting close to accepting defeat and giving up.

Attachment 50941

Attachment 50939
Attachment 50940

When your maf is in the same hz cell and you are only using the maf for fueling the injector should be squirting the same amount of fuel. What the o2's see may not be the same at different loads and rpm though. The reason is because of ignition timing. If the spark is not at the optimum time and you don't get a complete burn it will show different on your o2 than when the timing is at the optimum time and you get a complete burn. Try logging afr error against your cyair and rpm like your timing table. See what areas it is rich and lean in. Then look at what timing you had in those areas. Generally rich will be retarded and lean is advanced. But the best way to set your timing is on a dyno, moving it to find the most torque.

Quote:

---

Originally Posted by openwheel25

When your maf is in the same hz cell and you are only using the maf for fueling the injector should be squirting the same amount of fuel. What the o2's see may not be the same at different loads and rpm though. The reason is because of ignition timing. If the spark is not at the optimum time and you don't get a complete burn it will show different on your o2 than when the timing is at the optimum time and you get a complete burn. Try logging afr error against your cyair and rpm like your timing table. See what areas it is rich and lean in. Then look at what timing you had in those areas. Generally rich will be retarded and lean is advanced. But the best way to set your timing is on a dyno, moving it to find the most torque.

---

Very good point....

There is no doubt that spark affects burn time. But I do not think that's what is causing this MAF problem. I've reviewed the logs a bit, and spark is not that different, in fact, it's the same amount in several gears within the same Hz cell. What is different though, is injector pulse width. I'm fairly convinced that the data supplied with these injectors is not right. Specifically, the pulse width adder table. I have an email to the distributor of these Infinity brand injectors, with my logs asking them to review the logs and PW data. I feel at this point, I'm just chasing my tail - if in fact the injector data is fubar.

I may just suck it up, and buy another set of injectors. The whole reason I went with these injectors, is they are touted as being supplied with "correct data".... I probably should have just bought some from ID. At least then, I'd know the data was good and could just focus on the tune.

I had similar problems dialing in MAF with the previous injectors from FIC, but not nearly to this extent.

A quick update for interested persons....

After a few emails back and forth with the tuning school, I was sent revised injector data. Plugged that in and had to completely redo VE. The VE table actually looks much better. I had some odd valleys and extremely low values in the sub 1400 rpm areas before. No cell with ST error over 2% now. This seemed to help the miss a tiny bit. Acceleration is much smoother now too..... While in MAF-less configuration, I bumped spark dwell time up 15%. This did help some. I bumped it up another 10%, but did not feel any difference, so moved it back to the 15% increase value and left it there. The misfire is almost gone. It is completely gone above 2200. Still there under 2200, but not irritating. I can live with it I think.

I am still struggling with MAF cal though. I can dial it in, almost perfectly, in any gear. With all the fuel adders/cuts off disabled, SD disabled, and LTFT's disabled, I can get the car rolling with a steady engine speed of around 1800 rpm. Then hit the scan button, and slowly ramp rpm up to 4200 (before PE kicks in). ST error is vastly different in each gear. 3rd and 4th are the closest, but still about 5% different. I've logged spark, and it's pretty close in the different gears. So I ended up just calibrating it in 3rd gear. I then enabled LTFT's and all the other fuel adders/cuts, and took a long drive. LTFT's hover about +5 in fifth gear, to a low of -5% is second gear. In 3rd and 4th, they are close to zero.

What I find odd, is even though dynamic disable is set to a 4000 rpm, MAF is obviously still referenced, because LTFT's are darn near zero with MAF disabled.

I also got a response back from HPT support. The tables I requested cannot be added. "Your operating system does not use these tables and they cannot be added, sorry."

Well darn.... I will say this on your MAF. Yes Maf is still referrenced. BUT even though your timing is the same it should be traveling through different areas of the timing table. Try adding timing where it's going in your lower gears. This should help dial in fueling.

Glad to hear things have come along and doing better.