Audi A8 1998: geluidsprobleem met originele audi concert installatie

Dit is volgens mij het probleem:(bron www.erta.ru/review/chorus-problem_eng.shtml)

Problem of Some AUDI Chorus and AUDI Concert Autoradio Models,
or "Delayed-Action Mine" from Blaupunkt Company
Autoradios equipped with standard details usually have a considerable reliability margin in comparison with "mass consumption" devices. So it caused our interest when in 1999 several AUDI Concert autoradios from Blaupunkt Company with the same defect (the volume control system became disable) were brought to our workshop. The initial analysis of this problem took us to an inauspicious conclusion. Only in one case the reason was a loosen screw of the volume control knob. In all the other cases the reason of the volume control disability was the failure of the internal non-volatile memory (EEPROM) of the system main microcontroller. To solve this problem we had to replace the chip by a different one. To our great disappointment those autoradios had a MC68HC05 Motorola microcontroller (main MCU) although it's universally recognized as a highly reliable one. By the way, this microcontroller is actually the most expensive electronic detail of the autoradio.

Having decided to find out the reasons of the internal non-volatile microcontroller memory failure, we have comprehensively investigated the circuit design and the microcontroller software. The results were quite unexpected! The main and the only reason of the microcontroller failure was its software error in the part where the non-volatile memory record was realized. The program slowly killing the microcontroller was "kindly" designed by Blaupunkt Company.

Let us consider this problem in details.

The internal non-volatile microcontroller memory (EEPROM) is meant for the storage of different data, which determine the system functioning. The memory data record is stored even if the voltage is completely off. Usually stored in the non-volatile memory are the tuner frequencies and all the control levels at the moment of turning off (to bring the device to the same condition as it was before the off-moment, even if it was disconnected from the car voltage). There can also be stored the beginning and the end of the tone control interval and other relocation "hardware-dependent" data. At the modern development stage of electronics the non-volatile memory always has the limit of the recording cycles amount. The EEPROM wears out (molecular changes take place) during the "write EEPROM cycle" procedure and equals 10000 cycles for the MC68HC05 family of microcontrollers as Motorola Company guarantees. The real possible amount can be several times bigger but this resource is limited anyway.

The generally used design practice is to take into account the EEPROM resource finiteness and to write current levels of volume, fader, balance and tone at the device-off-moment into the internal non-volatile memory. Instead of it the Blaupunkt designers decided to write the data into EEPROM after each minimum step of control (each regulator step) but not after at least the whole user control. So if there are for example 32 steps from the volume minimum to its maximum, the microcontroller will accomplish 32 "write EEPROM cycles" during the changing from the volume minimum to maximum. Let us suppose that an average user changes the volume level by 5 steps and the EEPROM resource is not 10000 but for example 50000 cycles. It is easy to count that the microcontroller will "die" after 10000 times of the volume changing (or after 2-3 years if we suppose the volume is changed 10 times a day).

We have located this volume controlling and "slowly dying" EEPROM cell in the 01A8h microcontroller internal address space. It is easy to make sure of extra EEPROM cycle records, checking the output VPP1 voltage (EEPROM charge pump) during for example the volume changing.

The 01A8h cell begins "to forget" the data when the volume control EEPROM resource is exhausted. It becomes apparent as the autoradio turns on with an absolutely casual volume level and the control process becomes unpredictable. At first the volume level malfunctions can be seen at low temperatures only and the operating becomes normal with the device heating. But the EEPROM molecular destruction process occurs with each volume knob turning, so in several weeks or months the problems will take place independently of the device temperature. The further usage of the fault system will cause the failure of cells nearby and the device can be completely out of operation.

You can the Blaupunkt autoradio models with such software error see in the following table:



Autoradios Part number
AUDI Chorus 7 646 243 380 ***
AUDI Chorus *** 4B0 035 152A
AUDI Chorus A8 7 647 243 380
AUDI Chorus M4 7 648 245 380 4B0 035 152B
AUDI Concert 7 646 248 380 4B0 035 186
AUDI Concert A8 7 647 248 380 ***
AUDI Concert M4 7 648 247 380 4B0 035 186C
AUDI Concert M4 7 646 248 380 4B0 035 186B
AUDI Concert TT
AUDI Concert NAV 7 647 247 380
The full model name is usually marked on the right autoradio side.

Of course this error could not be ignored and the AUDI Chorus model production was just stopped when they became to fail on a mass scale. The AUDI Concert models (mentioned above) were also taken out of production but replaced with the AUDI ConcertPlus ones, which didn't have such a problem.

It is very easy to diagnose this defect. If you see the fault of all the controls ("Volume", "Balance", "Fader", "Treble", "Bass"), it just means that the control knob is out of order. If the problems take place with the volume control only (all the other controls operate well), it indicates that the EEPROM microcontroller has worn out and you have to replace the chip.

To liquidate the recurrence of this failure after the microcontroller replacement we continued our investigation. We managed to unassemble the internal program of the MC68HC05 microcontroller, which is used in these autoradio models. All the operations with the 01A8h cell were located. Then the program was changed, and as a result all these EEPROM cell operations were successfully readdressed to the random-access memory (RAM), which resource is unlimited. Now we buy the "clean" one-time-programmable microcontrollers and program them with the object code of the revised program. After the assemblage of the reprogrammed microcontroller to the autoradio the recurrence of volume control problems is absolutely out of question! The improved program operation can be easily checked by the microcontroller output VPP1 voltage during the volume level changing.

We didn't change the Balance, Fader, Treble and Bass controls. Because they are not used very often and this EEPROM cells resource (as we estimate) is several times bigger than the possible period of autoradio usage.

In conclusion. Be careful! We know about a lot of cases when some dishonourable "specialists" fix the worn EEPROM chip by its heating. Heating can prolong the term (may be for some weeks) but is not able to solve the problem completely. The real fixing is only the microcontroller replacement with the one, programmed with the improved software.