When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Why don't you use a custom OS with a designated SD mode? Not that this won't work but there are easier ways with the custom operating systems that are available. I also don't have a stock 09 6.0 suburban file to compare against but why are your prediction coefficients zero'd out? I've only ever seen that cause problems and weird ones at that, kind of like your airmass dropping out. You also don't have to change your dynamic airflow disable and re-enable when the MAF is set to fail so I would put those back. if you can post up the stock file it may be more helpful since I would be able to compare and see exactly what you've changed.
So maybe I am missing something here, but in the log you posted "cylinder airmass" is not being logged. I really wanted to see what the advance table was looking like during this event, but first thing I saw was the single row across the top. Not seeing it even had me checking other logs from a bunch of years to make sure I wasn't missing something. All use that parameter to populate the y axis.
The irony here is the log has "airmass" in the file name (sorry couldn't help but say that lol). I know sometimes things don't translate to other scanners so well, but I was really interested to see what this thing is really doing and then I noticed that, and now I'm a little confused. Then again we also miss stuff from time to time too. From the log, I am still trying to figure out the problem exactly. Pretty tough without being able to see the cylinder airmass (load) applied. I do see the dynamic and so forth, but without plugging in some custom math to convert the dynamic air to load, then it's hard to tell what's happening. Also, I don't feel like redoing my charts at the moment lol.
Last edited by ChopperDoc; 06-17-2019 at 11:29 AM.
So maybe I am missing something here, but in the log you posted "cylinder airmass" is not being logged. I really wanted to see what the advance table was looking like during this event, but first thing I saw was the single row across the top. Not seeing it even had me checking other logs from a bunch of years to make sure I wasn't missing something. All use that parameter to populate the y axis.
The irony here is the log has "airmass" in the file name lol. I know sometimes things don't translate to other scanners so well, but I was really interested to see what this thing is really doing and then I noticed that, and now I'm a little confused.
Okay, so I went ahead and turned your Dynamic Airflow value into G/CYL (cylinder airmass) by making a custom PID. For those that are having the same problem as I am not seeing the cylinder airmass, plug this into a math slot and log it on the y axis.
Math is: DYN / (RPM/15)
I didn't adjust the values to the table ones, but this paints a much better picture of what is going on... Honestly, seems normal.
I do see the spot in question though, airmass or not. That is even weirder than not seeing the airmass in the log. Straight to 0 and back. That is fkn scary.
Cylinder airmass pid is maf based so it isn't being scanned as I'm in VE only mode. Instead, dynamic airflow is logged and I used that pid along with math to give Dynamic Cylinder Airmass which is populated in 2 places (chart vs. time area as "Dyn Cyl Air, and on the far right graph area "Dyn Cyl Air").
Why don't you use a custom OS with a designated SD mode? Not that this won't work but there are easier ways with the custom operating systems that are available. I also don't have a stock 09 6.0 suburban file to compare against but why are your prediction coefficients zero'd out? I've only ever seen that cause problems and weird ones at that, kind of like your airmass dropping out. You also don't have to change your dynamic airflow disable and re-enable when the MAF is set to fail so I would put those back. if you can post up the stock file it may be more helpful since I would be able to compare and see exactly what you've changed.
Unfortunately there was no OS options for this specific 09 e38 ecu, the only close option was "[ECM] 120 - Speed Density Air Mode Patch: OS patch for Speed density air calc mode, not returning to normal."
Tried that and same thing was happening before and after.
As of yesterday, I am using a 2013 e38 ecu that I transferred all of the 09s tune data to, that one has a 2 bar OS option. But I'll see if the issue is gone after testing the new ecu after the owner welds up newfound hot side cracks 😁
Prediction coefficients were zerod out for tuning purposes, I've been instructed to do just that for many years by seasoned tuners to avoid external fueling changes to the tuned VE platform. If it is the cause of the issue, it will be very clear with the new ecu if the issue repeats itself. Then I'll adjust those to stock, so far there hasn't been any solid conclusions from anyone that has heard about this airmass drop but a lot of suggestions, all of which I've tried with no luck. This may be another one of those "ah that's it" things that may look odd but isn't the issue. Time will tell as everyone is confused by this.
My understanding of SD is, VE essentially becomes the dynamic airflow calculation, as MAF is disabled. Dynamic is basically the "result" of calculations, be it MAF, VE, or both which is the normal setting. There is a VE bias at low RPM and a MAF bias in higher RPM up to 4k (stock) and then straight MAF barring any throttle transitions. But yeah, in SD it's all VE lol. That doesn't mean there isn't a cylinder airmass parameter, which is the final calculation, used to determine load. But anyway... I digress.
Earlier I didn't zoom in enough and I was doing idle stuff before that, so until I changed my charts layout it didn't stick out right away. Did that and can see the dynamic, IPW, and cylinder airmass (from dynamic, custom math lol) all drop off, to include some weird activity on the O2's, one goes leaner, the other gets richer. It looks like one of them may have been removed for the wideband, but it's still logging voltage, and happens to log the most right after the "dip" in airmass. For some reason it does this every time the other bank goes lean, but basically sits there at 3 mv the rest of the log. Just looked weird.
Like I said, definitely see the spot. Very weird issue. It's a very short "blip" like event but dangerous as all hell. Glad I took some more time to look, that's a brain teaser and a half I'd say. What is the PCM voltage doing? Any chance you have been logging that? Any other weird voltage stuff or spikes/dips going on? It might be worth a shot logging and trying to replicate again. Seems a lot like an electrical gremlin or indeed a bad PCM.
Also, this isn't the first time I've seen a log not "save" inputs, where they can be viewed on one computer but not another. Rare, but it does happen.
I do see the spot in question though, airmass or not. That is even weirder than not seeing the airmass in the log. Straight to 0 and back. That is fkn scary.
Yes sir it is wild isn't it....for diagnosing purposes I put the 3000+ rpm 0 g/cyl area up to 0.60 g/cyl timing to a lowish 13 to 14 so when it does act ridiculous, at least the timing won't be 50 degrees in boost as the injector pulse also seems to eventually reduce greatly when this all happens.
Last time this happened I luckily didn't see knock as the split second lean condition from this "event" wasn't nearly as bad as the timing spike.
My understanding of SD is, VE essentially becomes the dynamic airflow calculation, as MAF is disabled. Dynamic is basically the "result" of calculations, be it MAF, VE, or both which is the normal setting. There is a VE bias at low RPM and a MAF bias in higher RPM up to 4k (stock) and then straight MAF barring any throttle transitions. But yeah, in SD it's all VE lol. That doesn't mean there isn't a cylinder airmass parameter, which is the final calculation, used to determine load. But anyway... I digress.
Earlier I didn't zoom in enough and I was doing idle stuff before that, so until I changed my charts layout it didn't stick out right away. Did that and can see the dynamic, IPW, and cylinder airmass (from dynamic, custom math lol) all drop off, to include some weird activity on the O2's, one goes leaner, the other gets richer. It looks like one of them may have been removed for the wideband, but it's still logging voltage, and happens to log the most right after the "dip" in airmass. For some reason it does this every time the other bank goes lean, but basically sits there at 3 mv the rest of the log. Just looked weird.
Like I said, definitely see the spot. Very weird issue. It's a very short "blip" like event but dangerous as all hell. Glad I took some more time to look, that's a brain teaser and a half I'd say. What is the PCM voltage doing? Any chance you have been logging that? Any other weird voltage stuff or spikes/dips going on? It might be worth a shot logging and trying to replicate again. Seems a lot like an electrical gremlin or indeed a bad PCM.
Also, this isn't the first time I've seen a log not "save" inputs, where they can be viewed on one computer but not another. Rare, but it does happen.
Pcm voltage stays the same as before the occurrence (around 13.0 volts).
Oh yeah don't worry about the narrow bands as one of them is toast (bank 2). Has since been replaced.
I could imagine viewing different logs would have some issues with missing data but here's a snapshot in case. This is what I see lol.
Yes sir it is wild isn't it....for diagnosing purposes I put the 3000+ rpm 0 g/cyl area up to 0.60 g/cyl timing to a lowish 13 to 14 so when it does act ridiculous, at least the timing won't be 50 degrees in boost as the injector pulse also seems to eventually reduce greatly when this all happens.
Last time this happened I luckily didn't see knock as the split second lean condition from this "event" wasn't nearly as bad as the timing spike.
New ecu trials are coming up.
Very wild, and that had to be some pucker factor there for you seeing 50 degrees into boost, aaaaaand of course the injectors basically going into DFCO on you too. Good lord. Looking forward to the update to this story as I don't think any of us here have seen anything quite like this.
Pcm voltage stays the same as before the occurrence (around 13.0 volts).
Oh yeah don't worry about the narrow bands as one of them is toast (bank 2). Has since been replaced.
I could imagine viewing different logs would have some issues with missing data but here's a snapshot in case. This is what I see lol.
This is the stuff nightmares are made of... Yeah I see the voltage, nothing unusual about that, other than your injectors are in DFCO and dynamic is 0.0 at a MAP of 140... Nothing to see here though. This looks like it went on a lot longer in this log. Good thing you have that timing pulled now.
On a side note... Does that chart say "Horsiepower"? LOL. Well, it's negative, add that one to the list of **** that is wrong with this picture...
Edit: 5 minutes later and I am still laughing at this for some reason. I'm probably just tired, but now I'm afraid to sleep because I'm pretty sure nightmares are coming after having seen all this.
Last edited by ChopperDoc; 06-17-2019 at 04:35 PM.
On a side note... Does that chart say "Horsiepower"? LOL. Well, it's negative, add that one to the list of **** that is wrong with this picture...
Edit: 5 minutes later and I am still laughing at this for some reason. I'm probably just tired, but now I'm afraid to sleep because I'm pretty sure nightmares are coming after having seen all this.
Haha yes horsiepower 😁🐴. I used the delivered engine torque pid to make a hp graph, so when the airmass drops in this situation along with other related data, the trq and hp go super low.
Definitely a nightmare of a situation but hopefully the new 2013 e38 ecu ends this bad dream for good.
Very wild, and that had to be some pucker factor there for you seeing 50 degrees into boost, aaaaaand of course the injectors basically going into DFCO on you too. Good lord. Looking forward to the update to this story as I don't think any of us here have seen anything quite like this.
All I can say is these dang LS engine are tough, I would have guessed the first time something like this happened, that would be the end of pistons/piston rings.
Unfortunately there was no OS options for this specific 09 e38 ecu, the only close option was "[ECM] 120 - Speed Density Air Mode Patch: OS patch for Speed density air calc mode, not returning to normal."
Tried that and same thing was happening before and after.
As of yesterday, I am using a 2013 e38 ecu that I transferred all of the 09s tune data to, that one has a 2 bar OS option. But I'll see if the issue is gone after testing the new ecu after the owner welds up newfound hot side cracks 😁
Prediction coefficients were zerod out for tuning purposes, I've been instructed to do just that for many years by seasoned tuners to avoid external fueling changes to the tuned VE platform. If it is the cause of the issue, it will be very clear with the new ecu if the issue repeats itself. Then I'll adjust those to stock, so far there hasn't been any solid conclusions from anyone that has heard about this airmass drop but a lot of suggestions, all of which I've tried with no luck. This may be another one of those "ah that's it" things that may look odd but isn't the issue. Time will tell as everyone is confused by this.
Whoever is instructing people to zero out prediction coefficients have no idea what they are doing, it causes all sorts of weirdness so don't do it. I would be swapping operating systems to a more commonly supported one with a custom OS available, it really is worth it but since you have a new ECM with the 2 bar OS option just go to that after you finish troubleshooting.
All I can say is these dang LS engine are tough, I would have guessed the first time something like this happened, that would be the end of pistons/piston rings.
HUGE pucker factor yes 😑
The drop out isn't going to cause any damage what so ever, it's not spraying fuel when it happens since airmass goes to 0 and when it returns it's back to "normal".
06-17-2019, 10:53 AM
