Typical situation in case of damage of the NOx sensor

Engine: N43B20

Problems: increased fuel consumption, uneven running, increased CO/HC (the car is not able to pass TI).

Error messages, recorded in MSD80:


The error message 2A99 regarding sensor of the crankshaft – consequences of the performance of aftermarket immobilizer.

Other error messages:

a) 30DA; 11005 = 2AFD; 11000 = 2AF8 – directly related to problems of NOx sensor;

b) 2C7F; 2C6C – consequences of incorrect individual adaptations (they are disconnected due to problems of NOx sensor).


Here, freeze frame of the error message for Lambda probe:

As we see form attributes of the error message:

a) wideband probe indicates a lean mixture (voltage above 2.00 V);

b) control probe, instead, indicates a rich mixture (voltage above 0.8 V).

These totally opposite indications – not damages of probes, but consequences of incorrect distribution of fuel between cylinders (see description below).


Problems with NOx sensor are confirmed by live data:

The engine was warmed up, the car drove at least 5 minutes, the temperature of exhaust gases reached over 220 oC, while driving (all conditions to switch in the NOx sensor).

As we see, the NOx sensor has not turned on – it doesn’t report Lambda (both values – narrowband and wideband), not the content of NOx.

If the NOx sensor has not turned on, MSD turns off Stratified charge – increased fuel consumption is self-evident consequences.


Due to problems with NOx system, MSD is not able to maintain the correct fuel mixture:

As we see, a fuel mixture of both banks is rich (lambda <1.0). DME “see’s” it, but has not corrected.

Rough run data menu is even more catastrophic:

For this SW version:

a) cylinders are displayed in firing order (it means, 1/3/4/2);

b) bar indication +0.200 corresponds to a non-working cylinder.


As we see from the menu:

a) mechanical efficiency of cylinder No.1 in idle is around -11%;

b) mechanical efficiency of cylinder No.4 in idle is around +11%.

Despite such huge differences in efficiency and persistent permanent vibration, MSD in no way tries to solve this problem! Such a permanent difference of efficiency is observed for a minute, five minutes, and even ten minutes – functionality of MSD was seriously disturbed, no doubt about it!


Obviously – the first cylinder has fuel shortage (lean mixture), cylinder No.4, instead, largest fuel amount (comparing to other cylinders).

To make sure about it (and also perform preventive repair and cleaning), the engine “head” has been taken off.


Here – the engine “head”, on the right side – cylinder No.1.

As we can see, the cylinder No.1 has the leanest fuel mixture (highest temperature – exhaust valves have been overheated), but in cylinder No.4 – very rich mixture, strong sediments observed. The state of the engine burning cameras confirms, what we saw in diagnostics – the fuel mixture distribution between cylinders has been very incorrect and corresponds to data, indicated in the Rough run menu!


It’s clear, that such fuel mixture distribution between cylinders means:

a) increased load for CO catalytic converters;

b) increased CO/HC in the exhaust, problems with passing TI;

c) permanent vibration;

d) possibility of misfires in cylinders with the leanest mixture;

e) as we see – the error messages regarding fuel mixture and Lambda probes;

f) incorrect fuel mixture also within the limits of the bank.


Of course, the engine with such problems will vibrate, shiver. And all these problems – due to a defect of the NOx sensor!


Installation of NOXEM



The same engine after repair.

Idle. Homogeneous mode.


Idle. Stratified charge.


In both injection modes, measurements confirm – mechanical efficiency is equalized perfectly: within +/-0.002 units (+/-1%) without remaining offset.


Fuel mixture in both banks is kept with Lambda 1.00 (correctly);

huge multiplicative correction (and the difference between banks) is disappeared!


Injectors flowrate measurements (in idle, Stratified charge) confirms –  injectors are in perfect condition!