AFR question for the pros

[Breeze1]

I don't know if I have been influenced by the ignition timing vids I have been watching lately (and the risk of knock it involves when badly set) but that made me think about my recently blown up engine.

The question might sound dumb but is quite worrying to me. Basically on my old M52TUB25 I had put a M54B30 intake manifold and a big a** filter on a home made CAI (see pic attached).

The engine felt literally transformed by the IM, not so much after the CAI, but still, I could feel a difference and I immediately noticed that the exhaust sound had dramatically increased at low rpm. All that without engine management tuning.

Now that that one is gone and has been replaced by the M52TUB28, a question came to my mind: is it possible that these mods have created too lean an AFR which could have caused knock? I know that the old engine was not in top condition so maybe it was too weak for the increase in air flow it received and the new one wouldn't mind, but I still feel like asking...

I remember that after going for a test drive just after putting the "CAI wall" in place, when I came back I did hear a knocking sound coming from the engine and I found out that the oil separator hose wasn't attached. I put it back in and all went back to normal. Might be just a coincidence, I don't know...

P.S.: my engine bay is much cleaner now

[bmwdirtracer]

In a word, no. The absurd (sorry) air filter didn't make your engine run lean. Nor did it make it faster, nor did it change the sound of the exhaust. The noise difference was the unmuffled air filter. (That's one of the reasons the factory puts the air filter in a box.... another is that it helps to prevent water ingestion, and therefore hydrolock. The fact that the intake roar is unmuffled makes people think the car is faster.)

Now, it's possible that the intake manifold made a small difference in horsepower, often with a similar loss of torque.

You see, the engine is a SYSTEM. While you can achieve incremental improvements with nothing but an intake, the cylinder head can still only flow the same amount of air....ditto, the cams, exhaust, etc. To get a large improvement in torque and horsepower, the entire system must pump more air.

What throttle body did you use? The M52 computer and throttle pedal can't run the M54 throttle body, which is the primary restriction in the intake, anyway.

Now, certainly, the disconnected CCV system would be cause for your engine to run extremely lean....but you still have the O2 sensors to correct the mixture, and the knock sensors to retard the timing, if they detected pinging.

The engine died for a different reason; what did a postmortem reveal?

[Breeze1]

In a word, no. The absurd (sorry) air filter didn't make your engine run lean. Nor did it make it faster, nor did it change the sound of the exhaust. The noise difference was the unmuffled air filter. (That's one of the reasons the factory puts the air filter in a box.... another is that it helps to prevent water ingestion, and therefore hydrolock. The fact that the intake roar is unmuffled makes people think the car is faster.)

Now, it's possible that the intake manifold made a small difference in horsepower, often with a similar loss of torque.

You see, the engine is a SYSTEM. While you can achieve incremental improvements with nothing but an intake, the cylinder head can still only flow the same amount of air....ditto, the cams, exhaust, etc. To get a large improvement in torque and horsepower, the entire system must pump more air.

What throttle body did you use? The M52 computer and throttle pedal can't run the M54 throttle body, which is the primary restriction in the intake, anyway.

Now, certainly, the disconnected CCV system would be cause for your engine to run extremely lean....but you still have the O2 sensors to correct the mixture, and the knock sensors to retard the timing, if they detected pinging.

The engine died for a different reason; what did a postmortem reveal?

I'm not going to contradict you on the technicalities of it, but I can assure you that the difference was impressive (I'm talking about the IM) The engine went from choking at 5000rpm to happily revving to 6500rpm and the acceleration was waay better.

About the CAI, I felt that it gave a touch more torque at low rpm's but nothing at the top, and mostly a much more audible soundtrack. And again I assure you that it did increase the noise level of the exhaust at low rpm's with the throttle opened just slightly, which could very clearly be heard coming from the back of the car, whereas with more opened throttle and higher revs I would clearly hear the sound of the intake coming from the front of the car. I did have both resonators deleted as well. That does not make it ridiculous to me...

I used an M54 throttle body with an adapter aluminium bored plate.

[Breeze1]

Sorry I hadn't seen your question about how the engine died.

We didn't bother opening it since it was going to be replaced anyway (have to source the parts new from the dealer and spend a lot of time on it for a result that is not guaranteed).

But there were aluminium particles both in the oil pan and in the intake manifold, which was also full of engine oil.

It appears that a ticking sound I had heard for more than 3 years, which I thought was coming from the hydraulic pushrods, was in fact the sound of a slight piston knocking.

It probably overheated at some point in its life before I got it, because we found a missing bolt on the block and the car had a new radiator when I got it while having had a poor maintenance...

[bmwdirtracer]

How are you running the M54 throttle body? It has no cable, and is drive by wire from potentiometers in the throttle pedal unit. The M52TU uses a cable.

If the engine was choking at 5000 rpm, and this changed when you got the new intake, I'd have to say there was likely something seriously wrong with the last intake. Or did you maybe delete the resonators at the same time? Maybe they were blocked? Maybe the injectors were clogged ? I've driven a large number of 323i (M52TuB25) cars, they don't choke at 5k rpm..

[Breeze1]

The adapter plate is made to convert the drive by wire into a drive by cable. I'll have to dig for the detailed procedure since it's been a while now, but quite a few people have made this modification succesfully. It's one of the most cost effective ways of gaining decent hp on an NA setup, along with headers (and tuning).
And both M52TU's (b25 and b28) benefit in the same way from this upgrade since the stock -narrow- one is restricted by comparison. That is one of the main differences between M52TU's and M54's apparently...

The same upgrade is applied between non-TU M52's and M50's manifolds, with the same results.

[pbonsalb]

Could you have suffered a failure unrelated to the “performance” mods? Like a failed Disa valve? I think you would have a CEL for fuel beyond adaption range if you had really added much performance.

[Breeze1]

I had the CEL before anyway with p1397 code diagnosed so I couldn't tell. Not failed because of the performance which is nothing mindblowing (around 15hp at the crank from I have read on the subject) but because of the much larger air intake without a DME tuned accordingly to balance the increase in air intake.

Although it's indeed highly unlikely I thought I would ask just in case. It could also have been a combination of several factors, one of which being a "in a not so good shape" engine.

Before that mod I did not even bother revving it past 5000rpm but after it's got much more enjoyable and I was relatively frequently visiting the 6000/6500rpm mark...

[bmwdirtracer]

Is there a computer in this adapter plate? I really can't begin to imagine how an adapter plate mimics two extra potentiometers and a different computer. Are you sure that you're not using an M52TU throttle body on that M54 manifold? That, at least, would be electronically feasible.

At this very moment, I have a fellow instructor, good friend who needs a throttle body for his '99 M52 528. I've got BUNCHES of spare M54 manifolds, throttle bodies, and injectors around, but there's no way I know of to make his computer and wiring harness run this combination.

[se93]

Is there a computer in this adapter plate? I really can't begin to imagine how an adapter plate mimics two extra potentiometers and a different computer. Are you sure that you're not using an M52TU throttle body on that M54 manifold? That, at least, would be electronically feasible.

At this very moment, I have a fellow instructor, good friend who needs a throttle body for his '99 M52 528. I've got BUNCHES of spare M54 manifolds, throttle bodies, and injectors around, but there's no way I know of to make his computer and wiring harness run this combination.

My E46 328ci had the M54B30 intake manifold conversion done using an adapter plate to fit the Factory M52 throttle body onto the M54 intake manifold. As bmwdirtracer says, i too wish the M54 throttle body would function on the M52 with just an adapter plate considering there are electronics differences involved and not just a physical throttle body.


Sent from my iPhone using Tapatalk

[Breeze1]

My E46 328ci had the M54B30 intake manifold conversion done using an adapter plate to fit the Factory M52 throttle body onto the M54 intake manifold. As bmwdirtracer says, i too wish the M54 throttle body would function on the M52 with just an adapter plate considering there are electronics differences involved and not just a physical throttle body.


Sent from my iPhone using Tapatalk

That's what I meant by "drive by wire to drive by cable conversion". The adapter plate makes the M52TU TB function with the M54's IM. Care to explain why the M54's drive by wire is better than the M52's?

I have read that the M54's added some lag in throttle response by comparison with the cable solution but it's just hearsay and it's beyond my mechanical knowledge (or lack of lol).

And where you satisfied with the mod?

[328 Power 04]

You didn't run lean with M54 manifold. MAF and oxygen sensors make sure of that.
I have the massive M54B30 manifold swapped on my M52TUB25 Z3 with a mounting plate to get the M52tu throttle to mount on. I have done a bit of extensive research into these manifolds and sacrificed a few to a bandsaw to see how they worked internally.
With correct tuning (of the VANOS and DISA) the low rpm torque losses of a huge manifold can be completely erased, and allow greater airflow at high RPM.




I love bringing this picture up when I can, but this is how much variability there is when VANOS cam angles are edited. The units are airflow and load per rpm. Mass Air Flow directly correlates to hp, and load directly correlates to torque.

This also shows the 2 different intake runner profiles offered by the DISA system. This is an M54B30 manifold on the 2.5. The 2.8 will be even easier to mitigate the torque losses if you do correct tuning. *This tuning takes time, understanding, and lots of logging*.


Test v0.14 mod.png





Getting to the throttle body. The M52tu throttle body is very much a drive by wire system. Yes it has a cable... but that cable is connected to a dual potentiometer inside the throttle body itself, and then the potentiometer guides the MDK motor function.
It is a hybrid-ish type DBW throttle, but the control is electric. Yes you can move it physically from the pedal with engine off.. but disconnect it to throw it in limp mode and then see what happens.
I also tune the throttle body response behavior, and I can make the throttle body open to 100% at 25% throttle, or inverse: 25% open throttle at 100% pedal. The electric control is primary in the throttle.
With tuning, both the M54 and M52tu throttle can be made to be brutally direct. BMW programmed them to have a soft delay on purpose. It's not a limitation of it being electric. It's all programmed that way.


In the best sense, the M52tu throttle is a mechanically assisted drive-by-wire system. But the electric part is dominant.

[Breeze1]

You didn't run lean with M54 manifold. MAF and oxygen sensors make sure of that.
I have the massive M54B30 manifold swapped on my M52TUB25 Z3 with a mounting plate to get the M52tu throttle to mount on. I have done a bit of extensive research into these manifolds and sacrificed a few to a bandsaw to see how they worked internally.
With correct tuning (of the VANOS and DISA) the low rpm torque losses of a huge manifold can be completely erased, and allow greater airflow at high RPM.




I love bringing this picture up when I can, but this is how much variability there is when VANOS cam angles are edited. The units are airflow and load per rpm. Mass Air Flow directly correlates to hp, and load directly correlates to torque.

This also shows the 2 different intake runner profiles offered by the DISA system. This is an M54B30 manifold on the 2.5. The 2.8 will be even easier to mitigate the torque losses if you do correct tuning. *This tuning takes time, understanding, and lots of logging*.


Test v0.14 mod.png





Getting to the throttle body. The M52tu throttle body is very much a drive by wire system. Yes it has a cable... but that cable is connected to a dual potentiometer inside the throttle body itself, and then the potentiometer guides the MDK motor function.
It is a hybrid-ish type DBW throttle, but the control is electric. Yes you can move it physically from the pedal with engine off.. but disconnect it to throw it in limp mode and then see what happens.
I also tune the throttle body response behavior, and I can make the throttle body open to 100% at 25% throttle, or inverse: 25% open throttle at 100% pedal. The electric control is primary in the throttle.
With tuning, both the M54 and M52tu throttle can be made to be brutally direct. BMW programmed them to have a soft delay on purpose. It's not a limitation of it being electric. It's all programmed that way.


In the best sense, the M52tu throttle is a mechanically assisted drive-by-wire system. But the electric part is dominant.

Sounds all good! Wish I could say I tuned my DME accordingly as well lol. I'll have to dig deep into that stuff to be able to understand ignition timing and what not... and that doesn't seem to be simple to say the least!

[Breeze1]

Is there a computer in this adapter plate? I really can't begin to imagine how an adapter plate mimics two extra potentiometers and a different computer. Are you sure that you're not using an M52TU throttle body on that M54 manifold? That, at least, would be electronically feasible.

At this very moment, I have a fellow instructor, good friend who needs a throttle body for his '99 M52 528. I've got BUNCHES of spare M54 manifolds, throttle bodies, and injectors around, but there's no way I know of to make his computer and wiring harness run this combination.

How aBout swapping in M54's Intake manifold+ throttle body + gas pedal system and flashing the DME to make that work? That is if his M52 is a TU...

If it works I bet he's going to be very satisfied with the result

P.S.: if his M52 is a non-TU I think it would be even easier: swap in M50's IM+TB since both TB are "by cable" (need to confirm that the M52's TB does not have potentiometers as the TU's does, based on 328 Power 04's contribution.

[328 Power 04]

The conversion as mentioned above would require using the MS43 dme. MS42 is a perfectly fine ecu.