MAF issue? Taking its time to learn adaptations?

[Breeze1]

Hello everyone,

So today, after driving my car "hard" (a few pulls, nothing crazy really, and always after letting the oil temp increase and pressure decrease as should be done) for the second time since I flashed it with a stock US 328i map (first time was last sunday), I noticed the following:

- when I come to a stop for the first time after a few pulls, the rpm needle "dips" close to 0, as if the car was going to stall. Then on the next few stops (traffic lights), it dips a little but not as much (500rpm). After that it gets seemingly normal. In all cases, after the dip, the needle goes back to 700rpm and stays there steadily. It did that on two occasions today.

- In third gear and only in third gear, when the rpm's reach 6000, it kind of hesitates and seems to be struggling a bit to get those last 500 (not in second gear for e.g., where it revs happily to almost 7000). Did it on both occasions as well.

- My torque pro app was reading 4.4g/s at 700rpm (750 according to torque pro) at the MAF sensor, after the needle dipped, then goes to 3.4g/s and stays there.

Any idea what that might be? It never did that before the flash and when it was running a 2.5l map. One thing I should mention is that the MAF sensor was not replaced when I swapped the engine. It's still running the 2.5l sensor.

Thanks!

EDIT: an additional piece of info: I recently ran a diagnosis with INPA and it did come up with 4 error codes that didn't seem to be much cause for concern or at least not in a hurry (2 or 3 of them were related to the SAP, which I removed since then), but now I remember that one of them said something about a "capillary leakage <0.5 "something" (will have to run the diag again to get the exact word). But I ran the diagnosis before the flash and it was already there, with no idle issue at all. Same diag after the flash same codes...

[328 Power 04]

The MAF is the same between the 2.5 and the 2.8 m52tu, as are the fuel injectors.

You need to fix your vacuum leaks pretty much.

I don't use Torque Pro at all, so those values don't mean much to me. The car is programmed according to mg/stroke load and/or kg/hr MAF flow. Doing a conversion from your numbers makes it completely implausible.

- - - Updated - - -

It doesn't take much time to learn adaptations. You don't need to repeat the factory flashing, or else you will just cause your flash counter to fill up.


Also, it's worth rebuilding the VANOS seals.

[Breeze1]

Thanks 328 Power 04. I will clean the MAF and check all the hoses over the weekend, for a start anyway. I wasn't intending to flash it again, as I thought the 328i's software might have brought to light some impending defect that was not showing with the less "demanding" 2.5's software.

About the values I couldn't tell really (maybe TP makes some sort of conversion to express the values with the same unit of measure for all cars?), but based on a "generic" article (not specific to BMW's) I read, they seemed in line with the 1g/s/liter of capacity/500rpm/min indicated. Anyway I'll try to make RR communicate with the OBDII port this weekend to get some real logs.

The Vanos seals, I wanted to order Beisan's kit a while ago for the 2.5 but I eventually didn't since the shipping cost to Morocco they quoted me was like twice as much as the seals themselves, for what basically is an enveloppe with a letter in it . And there were none of the symptoms usually associated with vanos failure anyway. But eventually I'll do it on this engine if it doesn't get perfectly stable.

[328 Power 04]

Wait, no, the torque pro numbers are plausible. Not sure how I read it as 4.4g/s at high rpm. (This was after 12 hours of studying)
You can also use INPA to get a quick glance at actual values. That or look into Testo for BMW. Great logging program, which uses the BMW diag infrastructure.

Anyways, 4.4g/s at idle converts to ~15kg/hr. That's fine, I usually see that on a warm engine idle.

The 3.4g/s is maybe a bit low. But a lot of factors can change that value.

Sounds like it may be pulling some unmetered air somewhere, and probably around the idle control valve area. That's usually the cause of rough return to idle. Or your idle control valve is dirty and getting stuck.

[Breeze1]

Wait, no, the torque pro numbers are plausible. Not sure how I read it as 4.4g/s at high rpm. (This was after 12 hours of studying)
You can also use INPA to get a quick glance at actual values. That or look into Testo for BMW. Great logging program, which uses the BMW diag infrastructure.

Well kudos to you for taking the time to reply after that!

Is the ICV the part n°8 shown here (my car has the M54B30's IM)?
https://www.realoem.com/bmw/enUS/showparts?id=AV53-USA-01-2000-E46-BMW-330i&diagId=11_2879#13411744713
I'll clean it along with the MAF sensor.

Another thing I'm going to do which I'm embarassed to say hasn't been done when swapping the engine: replacing the spark plugs; my mechanic told me he doesn't bother changing them when he identifies an engine as being "clean"/ with an apparent low mileage and since all I wanted was to get the car up and running I was fine with that. I don't even know what's in there lol.

So I was thinking about swapping them for colder spark plugs (in anticipation of ignition advance tuning, to help with knock and avoid pre-ignition...), since it never gets nearly as cold in Morocco as it does in Japan (40°F when it gets reaaally cold) but I have no idea what would be the best fit for my car. Opinions differ on what brand / metal to use, and I'm thinking the best would be single copper electrodes. Could you confirm and/or advise please?

Thanks!

[328 Power 04]

Stay with new fresh stock NGK.
If you are not putting 0.9-1.0++ bar of boost, do not change spark plug heat rating. It will only make things worse.

BMW timed the ignition fairly well, and on the moderate to aggressive side. You will only be able to raise ignition advance by 2-3 degrees where it matters before you hit maximum efficiency, advancing spark even more will not get you more power even if you manage to eliminate all knock. I see people attempting all the time to push 28-30 degrees of advance at the same rpms, and it does nothing to power. Even if you use E85 to supress knock.



Most of your power gains on NA will come from adjustment of fueling (not too rich, target 13:1 at full throttle), ignition advance to match the fueling, VANOS+DISA (these 2 especially, for the m54b30 manifold swap), and other non-power refinements that just make it nicer to drive.

[Breeze1]

Stay with new fresh stock NGK.
If you are not putting 0.9-1.0++ bar of boost, do not change spark plug heat rating. It will only make things worse.

BMW timed the ignition fairly well, and on the moderate to aggressive side. You will only be able to raise ignition advance by 2-3 degrees where it matters before you hit maximum efficiency, advancing spark even more will not get you more power even if you manage to eliminate all knock. I see people attempting all the time to push 28-30 degrees of advance at the same rpms, and it does nothing to power. Even if you use E85 to supress knock.



Most of your power gains on NA will come from adjustment of fueling (not too rich, target 13:1 at full throttle), ignition advance to match the fueling, VANOS+DISA (these 2 especially, for the m54b30 manifold swap), and other non-power refinements that just make it nicer to drive.

You just saved my some time and hassle, thanks. About the AFR, given that in Morocco we have no octane rating (and no E85 either), basically what I do is always fill up at Shell or Total gas stations which are renown and said to have "clean" fuel. I often hear Californians complain about their fuel quality; I don't know if ours is better but let's say it's not: is 13.1 at WOT still a good AFR?

I still have miles to go before I can change anything on my maps (haven't been able to even get a proper log yet haha) and I'm still learning at this point (just started reading "Engine Management Advanced Tuning" by Greg Banish) but I would really appreciate if you could point me to some comprehensive tutorial specific to BMW's, if there is one. So far all I could find was "generic" articles or tutorials for other marques (like Subaru on RR's forum).

[328 Power 04]

13:1 good quality or bad quality, is still 13:1, because the oxygen sensors measure the after effects of burning whatever is in there. BMW actually runs it a little richer at full throttle, to keep the cats cooler.
On normal driving you will need around 14.5-14.7:1, which closed loop will already do on its own.

- - - Updated - - -

There is no comprehensive tutorial for BMW unfortunately.

- - - Updated - - -

And by 13:1, I mean + or -. No significant difference in power between 12.7:1 and 13.2:1. BMW wrote the flash themselves to 13:1+- before adding additional fuel for catalyst cooling, and other routines. It's in all their training documents (for dealer techs)

[Breeze1]

13:1 good quality or bad quality, is still 13:1, because the oxygen sensors measure the after effects of burning whatever is in there. BMW actually runs it a little richer at full throttle, to keep the cats cooler.
On normal driving you will need around 14.5-14.7:1, which closed loop will already do on its own.

- - - Updated - - -

There is no comprehensive tutorial for BMW unfortunately.

I thought O2 sensors intervened only in CL, but that must be something I learned about "Subies" or just got mistaken about lol. I'll keep that value in mind, thanks!

[328 Power 04]

Correct, the stock system only keeps fueling in check during closed loop, narrowband only. It uses a simple additive or multiplicative factor to add full throttle fueling.

But, the learned fueling values get factored into full throttle fueling.. so a poorly running car at closed loop will use those adaptations at full throttle. It gets complex then, with all the modifiers.

- - - Updated - - -

And I mean actively tuning to 13.0:1 using a lot of logging with a separate wideband sensor. That takes a while.

[328 Power 04]

But you are getting ahead of yourself a bit.

Car needs to be in its best shape for even remotely thinking about making it better with tuning. Vacuum leaks, aged fuel system/filters, etc etc will only make it more difficult.

[Breeze1]

Correct, the stock system only keeps fueling in check during closed loop, narrowband only. It uses a simple additive or multiplicative factor to add full throttle fueling.

But, the learned fueling values get factored into full throttle fueling.. so a poorly running car at closed loop will use those adaptations at full throttle. It gets complex then, with all the modifiers.

- - - Updated - - -

And I mean actively tuning to 13.0:1 using a lot of logging with a separate wideband sensor. That takes a while.

I did not know about that before ordering a narrow band gauge, unfortunately. I'll have to order a WB one when ready to start tuning, but I've been wondering about something: the WB's sensor gets screwed to the exhaust line and attached to the gauge, fine. But how does it get read, through the OBD port (thus though the DME as well), by a logger such as RR? Will I have to wire it to the DME somehow?

Edit: I'll make sure to get the car in the best possible shape (and I honestly think it already is engine wise, besides this probable leak we discussed and yeah, plugs and whatnot, but hopefully nothing too complex) before I start tuning, but I'm just "getting the theory" while doing so, because well...can't wait

[bmwdirtracer]

https://www.innovatemotorsports.com/products/lc2.php

[Breeze1]

https://www.innovatemotorsports.com/products/lc2.php

Thanks Chris

[Breeze1]

Found the "capillary leakage": a crack had begun to form on the inside of that hose (circled in red). I put some silicone on it as a temporary fix. Also, I found out that when I removed the SAP, I forgot to seal the little hose that was attached to the valve (green).Put some silicon there too. I was not sure if it had to be sealed or not though (maybe it allows some form of breathing/evaporation?).

After about 40 miles of driving, the rpm's needle still "dips" when the car comes to a stop for the first time after a drive, but less than it used to (it doesn't go all the way to 0, more like 300/400 rpm) and then it goes back to 750rpm and stays there. Torque pro's readings for the MAF sensor have gone done about .3g/s, from 4.4 to 4.1/4.0.

I'll perform a diagnosis with INPA and see if the error code still shows up. Maybe I should reset the DME while I'm at it? If I remember well INPA allows to do a hard reset without flashing the DME, but I could be wrong...

[328 Power 04]

If you had those leaks, you need to thoroughly check the intake tract, especially the boots going to throttle body. You very likely have more leaks. Also, the m54 manifold conversion takes an adapter plate, and that can easily be a cause of leaks as well. Smoke machine if you can. The Chinese smoke machine listed in the cool tools thread is perfect for this.
Make sure your ICV is clean.

You should be using the original M52tu injectors, not sure what you have on now.

There's really no such thing as a "reset". If you mean clear adaptations, don't expect it to fix anything.

The internet (many forum users) really really has something against these adaptations, and become fixated on that, instead of actually fixing the real problems.

[Breeze1]

If you had those leaks, you need to thoroughly check the intake tract, especially the boots going to throttle body. You very likely have more leaks. Also, the m54 manifold conversion takes an adapter plate, and that can easily be a cause of leaks as well. Smoke machine if you can. The Chinese smoke machine listed in the cool tools thread is perfect for this.
Make sure your ICV is clean.

You should be using the original M52tu injectors, not sure what you have on now.

There's really no such thing as a "reset". If you mean clear adaptations, don't expect it to fix anything.

The internet (many forum users) really really has something against these adaptations, and become fixated on that, instead of actually fixing the real problems.

I see what you mean, but I think the leak circled in red is more a result of numerous manipulations of the MAF's hose and other stuff around when installing the CAI and its "box". I know that the adapter plate is not a perfect fit but, as far as I know, only because it doesn't allow the Throttle body's "plate" to open 100%, only about 90 ish %. One thing I noticed is that sometimes (really doesn't happen often) when for example I'm cruising in 3rd and accelerate just a bit more (not WOT), there is like a ONE hiccup before it starts gaining speed normally. Besides that and the dip when stopping, the car drives perfectly fine and smoothly.

When swapping the engine, the whole IM, TB and whatnot were thorougly cleaned, and that was like 3 months ago.

I'll run a diagnosis with INPA, see what it says and take it from there...

Besides the slight annoyance it results in, am I into "dangerous territory" if there still is a leak? I'm going to have to travel with it for a week starting next weekend, and would rather leave the thorough investigations for when I come back...

Edit: sorry if my question doesn't make sense, but let's say I found the only leaks that were in the system, do I still need to "bleed" it somehow? I'm thinking no because the "system" in itself consists of airflow and there just shouldn't be vacuum leaks, but I thought I would ask.

About the injectors, they're stock and came with the engine, just like the spark plugs.

[Breeze1]

So I ran a diag with INPA today after performing the abovementionned, and the "capillary leakage" error code has disappeared.

But there is another one that has shown up (or was already there before, I'm not 100% : "signal ignition measuring resistor").

I have attached a pic of the error code and a few analog parameters while at idle, in case any of you guys have an idea about what it is. I also reset adaptations by unwillingly hitting F7 lol

[328 Power 04]

I have to look that code up to see if i can get a better translation somewhere. Seems to speak about ignition coil.

Curious, is your maf the stock one from the 323/328?

[Breeze1]

I wanted to change spark plugs (which I ended up keeping because they seemed fine and I couldn't find the same OEM ref at the shop near me) and I found a little oil in one of the SP's housing, but the oil didn't go through (the plug was bone dry, as well as all the others and the ignition coils).

I did find out, however, that there was a small oil leak probably coming from the cylinder head cover's gasket (hence the oil).

The MAF is stock and was not changed when I swapped the engine. Do the readings seem unusual to you?