MSD80. Advanced ‘features’

MSD80 is a very complicated engine management system. Times, when the engine management system was performing specific, easy to understand algorithms, are gone for good. The engine control units of present moment are specialized, high performance computers, which performs several tenths of tasks at the same time. Of course, the software for these computers is written by humans (programmers), who also can have mistakes. And even if the programmer writes the perfect code, the manufacturers of processors/micro-controllers can have faults (it means, the processor performs the exact task incorrectly). Faults happen also during development of algorithm, performance tests of units. The manufacturers of sensors and actuators also can have faults (sensors send incorrect data and the actuator executes the order incorrectly). Below  – the summary of TOP 10 most common MSD80 bugs.


Place 10. Misleading self-diagnostics report. This problem is typical also for several next records, but here are more general cases included, for example: due to incorrect individual adaptations of cylinders error messages regarding Lambda probes and/or CO catalytic converters are recorded; due to specific defect of NOx sensor – the error message regarding defect of temperature sensor in the exhaust system. Needless to say, that you can change the sensor mentioned before, it’s connecting wire for hundreds of time – the result will stay unchanged: the error message regarding this sensor (inappropriate temperature of it).


Place 9. This is a hardware problem, but, as it is directly connected to MSD80, it’s also included in the list. Defect: improper control of ignition coils, which results in high risk of damages of IGBT transistors, which are controlling the ignition coils and are working in severe temperature and voltage conditions. Also this inaccurate closure is a cause for often defects of ignition coils (the insulation of ignition coil is being hit through). There is no information, that this problem would be solved for newer MSD80 releases – this problem is one on the top of N43/N53 engine problems. In addition, when the defective IGBT is replaced, the cause of problem is not solved – after some time problem can repeat. Description of problem here and here.

Place 8. Uneven road surface can be the reason of error messages regarding Rough run/Stratified charge. The place in top is controversial – from one side, this error message (together with ISTA notices regarding misfire of exact cylinders and suggestion to replace the spark plugs, ignition coils and injectors) is one of the main reasons for expensive and baseless repair works without visible improvements. From other side – if we know the causes and effects of this error message (this error message has no significant impact both in passive or active mode), the situation makes you smile – nothing more. More about the problem read here. For the sake of justice it should be noted, that  several later software updates eliminate this problem.


Place 7. Incorrect and even misleading algorithm of adaptations. For example, injector openings for bank X are highly corrected with “+” sign (prolonged), but in the same time temperature corrections for all/this bank are with “-” sign. The result: total corrections are small (the situation actually is close to perfect), thou even the menu “picture” shows very incorrect performance of engine. This situation is mainly eliminated with updates of software.

Place 6. Like diesel engines M57/N57, N43/N53 and other petrol engines don’t perform the self diagnostics of thermostat (decreased engine temperature). If the mechanical part pledges in opened position (very often defect), the engine, not even keeping in mind, that the water-pump performs in lowest possible mode, can not reach the nominal temperature. Consequences: overheated/damaged CO catalytic converters (because they are increasingly heated to reach “necessary” work temperature), disturbed desulfations sessions of NOx catalytic converter, increased fuel consumption.


Place 5. Coarse logic errors in algorithms of individual corrections in cylinder’s fuel mixture. Two most known examples:

1) Let’s suppose, that cylinder X misfires time by time. When performing tests of efficiency, MSD80 very simply counts a total sum of all cylinders (X) summary efficiency (Y). Due to misfire, the average efficiency (Y) is below the norm, as a result the fuel mixture is enriched in the current cylinder, but other cylinder get only lean mixture. The problem only increases.  The solution would be (not included even in latest MSD80 releases) to ignore the measurements, which differ from average values very strongly;

2) Let’s suppose, that due to some  external circumstances the performance of cylinders was dis-balanced (for example, due to AC, angular play of two-mass flywheel, some misfire), and MSD80 has decreased the amount in current cylinder to balance the efficiency. In the moment, when normal performance of engine summons, would be self explanatory, that if the fuel amount to cylinder is decreased, the efficiency of it would be below the average value. It would be normal to increase the fuel amount (at least to average value), but instead of that the MSD80 records the error message regarding inadequate low cylinder efficiency and stops the creation of individual fuel corrections. Very simple, but at the same time a gross logic error.


Place 4. Mathematical errors during adaptations in some specific cases. For example, if NOx system (NOx sensor and/or NOx catalytic converter) has problems (related error messages are stored in error message list), MSD80 performs totally incorrect fuel mixture adaptations and usually following error messages appear:

29D1: misfires of 5th cylinder;

29F5: efficiency of CO converter below threshold, bank 2.

Also error messages regarding fuel trim (especially, for bank 2) usually appears. None of these messages (in exact situation – when/if NOx system is damaged) has a real cause, all of them usually are the result of incorrect operation of MSD80.


Place 3. Third place is dedicated to adaptation problems after a series of specific (for example, NOx catalytic converter  and/or sensor) error messages. Even if the cause of these error messages has been eliminated, even if the error message memory of MSD80 has been cleaned and there are no new/other error messages, even if all symptoms (learnbits, integrators, registers of individual adaptations) initially shows correct performance of MSD80, in the course of time adaptations are created incorrectly and after some (or some tenths) motor-hours the engine starts to loose the power, starts to vibrate, shiver (misfire appears), the error messages regarding Lambda probes and/or CO catalytic converter defects. The dealer’s diagnostics set ISTA D will not show the cause of problem – it, falsely, will suggest to replace injectors, CO catalytic converters, Lambda probes. Only using INPA loader 2.023 and observing specific problems it’s possible to recognize this situation. Unfortunately in 99% of cases the situation is not being recognized, the car owners spend large amounts of money for replacements of injectors, ignition coils. More about his you can read here.


Place 2. In the moment of variable conditions (variable opening of throttle, variable load of engine) MSD80 suddenly (unjustified) detects group misfire of some cylinder (in case of such conditions the control system of uneven run of flywheel has to be turned off), disconnects the “damaged” injector, goes top open loop mode and continuous to work with 5/3 cylinders (N53/N43), turns on the EML signal lamp. The engine performs with decreased power, vibrates very strongly. This defect is very rare – on 100’000 km confirmed for one time (for several more times – it’s possible, that this defect was the cause for following problems). The good news – allow the engine to run idle for several minutes, and it will successfully restore deleted bank and offset adaptations and will start to create new and correct adaptations. And again – there are no suggestions (for example, appropriate suggestions in CCC monitor), a non-technical driver will not continue the driving, because the engine during restoring of adaptations vibrates very strongly, RPM in idle changes etc.

And finally:


Place 1. It’s rightly assigned to hot restarts (it means, arbitrary restart of MSD80 without losing power supply). Defect recurrence – rare. For the current car on 100’000 this defect observed 1 .. 2 times. But the effect is impressive: the engine simply switches off for half of a second… Even the reading of tachometer disappears (but the engine continue to turn because of inertia). The good news – this restart is quick, and MSD80 “recovers” without error messages or other symptoms. Considering the very rare observance of this defect and very unclear causes (the problem can be caused by fault in program code, for example, stack overflow, addressing if not existing memory cells etc., some mismatches in RAM/EEPROM, mistakes in calculations, for example, division with 0 and incorrect processing of situation), it’s unknown, which software versions and in what amount are affected. The EML signal lamp is not switched on in this situation, MSD80 error message memory don’t contains any error messages. Extreme feelings guaranteed. Corresponding error message is stored in info memory list and is not available (visible) via ISTA D.


Related entries:

Error messages. Intro

N53B30. When adaptations failed

N43B20 with strange behaviour

Uneven run. Rough run. Smooth run