3v AFR help, rough idle

[jakesmithpbr]

Hi,
the other night I installed an afr gauge and sensor, and installed headers. The first start up, it ran extremely rough. Almost like it had a cam. I messed with the tune, mainly turning most my changes back to stock including the MAF table and got it to run slightly decent. My LTFT and STFT have been sky rocketing to like 60+. I have been messing with it ever since and I just cant seem to get it dialed in. I go on data log runs but it only logs for 5 minutes and then stops logging STFT. My gauge ranges from 10-20 at any given time, it might idle at 14.6 or 14.7 for a few seconds but then dips back down to 10. Then it will ramp back up to 20 within seconds. It seems to run just fine for the first 15 seconds after start up when my gauge is calibrating and then once it calibrates it dips to 10 hard. I've tried a lot of stuff but can't seem to get it to run right.

The car is a 2007 mustang gt 5 speed manual, mods are long tube headers and CAI. I ended up connecting the upstream O2 sensors at the collector, then right next to one of the sensors I welded my AFR bung in. Then on my H pipe I installed two plugs.

I know I'm doing something wrong I just don't know what. Doesn't have any exhaust leaks either. Heres my tune and scan:Another header revision mainly stock.hpt
mustang log.hpl

[B E N]

Which CAI is on it? It probably needs maf tuned. This thread is discusses how to go about that on a 3v: https://forum.hptuners.com/showthrea...3V-4-6L-tuning Murfie's posts are very useful. Headers should only result in a change to transport delay, which shouldn't be causing the level of problems you have.

The duration on that log isn't really long enough to get much out of, from what I am seeing you are getting a lot more air than the ECM is expecting, do a basic calculation to get your flow change multiplier through your maf and apply that before you start tuning. I am getting an error when I try to open your file, are you using the most recent HPtuners?

[jakesmithpbr]

Thanks for the reply Ben. I thought all I had to do was change the delay too. Pretty much my car runs extremely lean then runs extremely rich. I've started it with a complete stock tune for my car and it still runs off. I put the stock MAF table back on and it helps a little. and yes i'm using the most recent software. Can you not open the tune file?

[jakesmithpbr]

Which CAI is on it? It probably needs maf tuned. This thread is discusses how to go about that on a 3v: https://forum.hptuners.com/showthrea...3V-4-6L-tuning Murfie's posts are very useful. Headers should only result in a change to transport delay, which shouldn't be causing the level of problems you have.

The duration on that log isn't really long enough to get much out of, from what I am seeing you are getting a lot more air than the ECM is expecting, do a basic calculation to get your flow change multiplier through your maf and apply that before you start tuning. I am getting an error when I try to open your file, are you using the most recent HPtuners?

Also I?m using some random eBay CAI it?s a 100mm. But I have a JLT series 3 intake coming so I can get a real MAF transfer function.

[jakesmithpbr]

Here I saved another tune file, let me know if it works
https://easyupload.io/7xdfp2

[jakesmithpbr]

My bad for spamming, but I changed my tune up a bit, multiplied my entire MAF vs Voltage table by 1.08, turned off LTFT and changed some other stuff I might of been the cause of running rough. Got it to idle right at 14.7, but at WOT its about 13 give or take 0.1 with no issues. Probably going to have to wait until I get my JLT intake so I can change it to a transfer function I know is solid because I feel very little power increase with all of this, but should be about 50+ hp over stock by now.

My bigger concern is that I still cannot log STFT still. It logs STFT for about 2 minutes now, tells me its extremely rich (both banks go up to 60+ then go down to -60 and fluctuates) then it completely cuts out and logs "0" for the rest of my log. Not sure if it's because my MAF table is so far off or what. Going to keep researching but any help would be really appreciated. Thanks

[decipha]

long tubes and cai is no where near a 50hp increase

[jakesmithpbr]

all the places i've looked i should be at 300-310 with long tubes, cold air and a tune

[murfie]

You have a lot wrong going on.
Accelerator should not be at 18% at idle.
Throttle should not be at 13-14% at idle.
Your MAF looks right for Idle, but not for the accelerator pedal or throttle %.

O2 sensors should start to swinging back and forth from .9v to .1v, once closed loop is entered and STFT settle. Yours is lean, like your your MAF is not reporting all the air thats getting in the cylinder. Then with no change what so ever, except for O2 voltage, the trims come back to normal, and slowly trend rich. Doesn't make any sense.


I would suspect the scanner is not reading correctly. It doesn't seem like the car would idle at all like this. Are you sure you are scanning using the latest beta software? If you have both stable and beta installed: Once you install the beta software, you need to navigate to the install folder and make sure you open the beta program. A short cut sent to the desktop helps. With the program open you can go to the help tab, then about, to open a window that says the version and if its beta or not, to confirm you have to correct program open and the latest version.
I use to be very diligent downloading the latest everyday, but it take like 15 minutes to download, and I have become a bit lazy with it myself.

install location.PNG

[jakesmithpbr]

You have a lot wrong going on.
Accelerator should not be at 18% at idle.
Throttle should not be at 13-14% at idle.
Your MAF looks right for Idle, but not for the accelerator pedal or throttle %.

O2 sensors should start to swinging back and forth from .9v to .1v, once closed loop is entered and STFT settle. Yours is lean, like your your MAF is not reporting all the air thats getting in the cylinder. Then with no change what so ever, except for O2 voltage, the trims come back to normal, and slowly trend rich. Doesn't make any sense.


I would suspect the scanner is not reading correctly. It doesn't seem like the car would idle at all like this. Are you sure you are scanning using the latest beta software? If you have both stable and beta installed: Once you install the beta software, you need to navigate to the install folder and make sure you open the beta program. A short cut sent to the desktop helps. With the program open you can go to the help tab, then about, to open a window that says the version and if its beta or not, to confirm you have to correct program open and the latest version.
I use to be very diligent downloading the latest everyday, but it take like 15 minutes to download, and I have become a bit lazy with it myself.

install location.PNG

Hey murfie,
I am going to re-download both the stable and beta version. I just started using HPtuners a few weeks ago so I assumed I should have the current version. And it is very strange, whats even stranger is that it will go into OL Fault almost immediately, mainly because my log shows its running REALLY rich, once it's in OL Fault there are no fuel trims to update and it stays at 0. Been changing a lot in my tune to try to get it to stay in CL just so I can get it dialed in enough to make a little power. I do have it running and driving, but definitely feel like there's a lot of power to be had if everything was running right.

I'll install all of these and see what my logs say. Thanks for the info

[murfie]

OL-fault means its in open loop because it has detected a fault with one or more parts of the closed loop system. O2 sensors being the most common. O2 sensor wiring harness issues. MAF sensor problem or vacuum leaks. Then fuel pump/ fuel pump voltage control. Leaky or clogged injectors. Clogged fuel filter.

The way you describe it, it is normal for the car to be in CL when less than 50% throttle, and switch to OL at 50% TPS or what ever your OL TPS threshold is set to. I dont think it will say OL-fault, just OL or OL-Power enrichment. There are other thresholds that can make it go into OL as well. In order to target the richer lambda than 1, it needs to ignore the narrow band sensor corrections and run in open loop. This is why a wide band sensor is needed to tune that area, because you can't see what the lambda is with the narrow bands. Its perfectly fine to set up a tune to run open loop 100% of the time. It does makes tuning a simple setup like CAI and Headers more difficult. It isn't something that you normally need to do, and if there is a problem its best to find and correct that, than calibrate around it.
Setting the car to run CL all the time, will make it run lambda 1 at WOT, and that can be very dangerous for the engine.

These three changes, some on the axis values, will help it stay in CL longer. I would not advise extended periods of WOT with out PE(10 seconds or longer), or going any more aggressive. Give it lots of cool down time between pulls, don't just pay attention to ECT as thats not the full story of temperature of metals/components inside the cylinder, give it a few minutes or miles of normal driving for heat to dissipate.

stay in CL longer.PNG

Maybe even target .82-.83 to make sure its adding more fuel than needed until you know for certain the wideband is reading where you want it, and you can put it back to .85-.86.

[jakesmithpbr]

OL-fault means its in open loop because it has detected a fault with one or more parts of the closed loop system. O2 sensors being the most common. O2 sensor wiring harness issues. MAF sensor problem or vacuum leaks. Then fuel pump/ fuel pump voltage control. Leaky or clogged injectors. Clogged fuel filter.

The way you describe it, it is normal for the car to be in CL when less than 50% throttle, and switch to OL at 50% TPS or what ever your OL TPS threshold is set to. I dont think it will say OL-fault, just OL or OL-Power enrichment. There are other thresholds that can make it go into OL as well. In order to target the richer lambda than 1, it needs to ignore the narrow band sensor corrections and run in open loop. This is why a wide band sensor is needed to tune that area, because you can't see what the lambda is with the narrow bands. Its perfectly fine to set up a tune to run open loop 100% of the time. It does makes tuning a simple setup like CAI and Headers more difficult. It isn't something that you normally need to do, and if there is a problem its best to find and correct that, than calibrate around it.
Setting the car to run CL all the time, will make it run lambda 1 at WOT, and that can be very dangerous for the engine.

These three changes, some on the axis values, will help it stay in CL longer. I would not advise extended periods of WOT with out PE(10 seconds or longer), or going any more aggressive. Give it lots of cool down time between pulls, don't just pay attention to ECT as thats not the full story of temperature of metals/components inside the cylinder, give it a few minutes or miles of normal driving for heat to dissipate.

stay in CL longer.PNG

Maybe even target .82-.83 to make sure its adding more fuel than needed until you know for certain the wideband is reading where you want it, and you can put it back to .85-.86.

Tried looking for vacuum leaks this morning but I'm going in blind because I can't find ANY diagram of vacuum lines for my car... When I did the headers, I had to lift up the engine a bit so they would fit so it's a possibility I could have pulled a line. It just doesn't make sense that I had zero problems logging fuel trims BEFORE my headers. I have a feeling its something to do with my O2 sensors. How many O2 sensors do I need to properly log? I have two sensors hooked up at the collectors, and behind the driver side O2 i welded on my bung for the wideband. And before the H-pipe I put two plugs, There's a total of 5 bungs on my exhaust.


I do want to try changing those tables although I want to cancel out everything else first because I think it's a simple fix I just have no idea what the problem could be. Should have worked just fine with the tune I had on with stock manifolds.

Another thing, not sure if it's important or not but will still mention.. I installed the headers and started it up and it was running rough, reverted most settings back to stock, got it running OK, then went on a drive. Somehow I broke my USB-C cable and my MVPI wouldn't work at all, so I ended up buying one from a gas station, one of those cheap ones. Could that be an issue too? Not sure if it matters but thought I'd mention. Thanks

[murfie]

You only need the two upstream narrow bands for the ECU to work in closed loop. Make sure you have them plugged into the upstream connectors on the harness. You can't have them plugged into the connectors for the rear O2 sensors.

You need the third external wide band to correct any thing the ecu goes into open loop for. This is only for you, the computer doesn't care about this.

The rear o2's are for monitoring the catalytic converters, They are not needed.

If the USB cable allows the MPVI and laptop to connect with out errors I don't see it being a problem.

[jakesmithpbr]

You only need the two upstream narrow bands for the ECU to work in closed loop. Make sure you have them plugged into the upstream connectors on the harness. You can't have them plugged into the connectors for the rear O2 sensors.

You need the third external wide band to correct any thing the ecu goes into open loop for. This is only for you, the computer doesn't care about this.

The rear o2's are for monitoring the catalytic converters, They are not needed.

If the USB cable allows the MPVI and laptop to connect with out errors I don't see it being a problem.

yeah man that's where I F***'d up. Sent you a pm. Ordered extensions for upstream O2

[jakesmithpbr]

Update:
Figured out the problem; it was the front O2 sensors not being hooked up. Had the cat sensors hooked up instead. All fixed now just waiting for my intake and I'll be all dialed in. Cams next