Select a technical image to view it in detail.
At this point I have driven several hundred kilometers in different driving conditions, and I am ready to share my first impressions of the MSD81.
General behavior
Subjectively, the DME works correctly. Switching between injection modes and NOx converter regeneration sessions is soft and imperceptible. I observed no misfires or rough running when switching between injection modes.
Differences from MSD80
The comparison is with MSD80 running the last ZB release and DIRU.
- At idle, Stratified charge tries to hold Lambda around 1.80, slightly lower than on MSD80.
- Regeneration sessions happen more often than on MSD80, even when the NOx data are perfect. BMW engineers seem to have lost some of their optimism about the real performance of NOx catalytic converters. In addition, after a period of higher required torque, the DME assumes the NOx converter is partially contaminated (a quite logical assumption). A first regeneration is then required soon after switching to Stratified charge.
- The math for creating the injector (multiplicative-type, flowrate) adaptations of the banks has changed. Previously there were several problems with injector flowrate, even when the multiplicative-type bank adaptations were correct. MSD81 creates correct flowrate adaptations for the injectors:

- The flowrate data measured in Stratified charge mode (../F5/Shift+F6/5, lines 7 to 12) are no longer applied immediately to the injector flowrate (../F5/Shift+F6/F1, lines 4 to 6) during Stratified charge. More precisely, they are not applied to the 2nd impulse, the one that can be seen with INPA. These test results may be applied to the 1st impulse (which cannot be seen with INPA), or uniform coefficients for Stratified + Homogeneous mode may be applied instead. I will explore this in more detail.
- The Rough run data (../F5/F7) indicate a very stable idle, although subjectively the idle stability is not very different from MSD80. This means the “sensitivity” of MSD81 in this menu is reduced. Obviously, BMW engineers continued the fight against the false 3104 error messages.

In the image above, cylinder No.1 is switched off and the engine runs in Stratified charge. On MSD80 in this situation, the bar of the 1st cylinder indicated around +7 units. Here it is only around +3.5 units.
Rough run in Homogeneous mode


In the two images above, cylinder No.1 is switched off and the engine runs in Homogeneous mode. Unfortunately, the Rough run algorithms work incorrectly in this mode. Very often the DME identifies an increased(!) mechanical efficiency of the switched-off cylinder, which is not real. It feels as if code from some very old MSD80 release had been copied and pasted. This means that identifying misfires will be difficult at idle in Homogeneous mode. At the moment I have no information about other modes.
Results that are close to perfection
It is not all criticism. Below are some images that are close to perfection.

The 2nd bank has relatively new injectors, and both the Offset and Multiplicative LTFT are perfect. The 1st bank has injectors with a service life of at least 200’000 – 300’000 km. Here, too, the Offset LTFT is perfect.
Injector data in Stratified charge:


The injector of the 2nd cylinder is slightly dripping, but the cold start was perfect at the same time. The injector flowrate adaptations are perfect.
Injector tests in Homogeneous mode:

The test results are close to ideal.
Idle smoothness:

The image above shows Homogeneous mode.

The image above shows Stratified charge.
Note: MSD81 has reduced sensitivity in this menu. A value of 0.1 units corresponds to a mechanical efficiency difference from the ideal of around 3%.
A final note: the service life of this engine is more than 800’000 km.