PDA

View Full Version : Here are logic 7 repair instructions for every 1


PAPER
06-26-2012, 02:11 AM
OK I AM NoT TAKING ANY CREDIT FOR THIS .
I FOUND IT WHILE READING ON A GERMAN FORUM .

HERE IS A COPY & PASTE OF THERE INSTRUCTIONS WITH PIC'S
DO AT YOUR OWN RISK !!!


Repair instructions and ASK Logik7 from the E65!
Good day,

as you all know I am going to repair a defective Logik7.
I just wanted to show my work and presentieren because of the many questions that I'm not yet as far : Thumbsup:
Come rare in the hobby room at my in-laws.

Picture # 1: Here's the Logik7 board, you see four stages
1x Center speaker
1x both rear doors
1x bass left
1x bass right
Picture # 2: I've removed the old broken amplifiers and exposed contacts.
Image No. 3: The same close-up
Picture # 4: These two stages are to replace the old
Picture # 5: I was looking something Sprittig ;-)) do not have hands trembled, but I did not want to hide from you!

I
I also soldered the amp already part way to the board (or
no photo), for the speakers I have 0.5 mm thick and 3cm long lines used for
the signals of the thin black? mm .

Will keep you here with that Fred up to date.

mfg odd ....

Edit of 19/03/2012: As I already wrote yesterday, I was
sprittig something and so I said instead of 0.5mm , 1.5mm to
the TDA soldered ;-( (Now I know why it was so hard * LOL *

This will, of course, correct and add new pictures!

PAPER
06-26-2012, 02:13 AM
THIS IS HIS 2'ND POST


So

I was now again in the basement and I studied the board.

If you want to use an external amplifier, for which I have marked the low-frequency signal.

If you simply want to repair his Logik7 that will please the TDA1562 as specified soldered to the motherboard.



TDA1562 datasheet: Page 3
http://translate.googleusercontent.com/translate_c?hl=en&prev=/search%3Fq%3Dhttp://www.7er-online.de/forum/die-modellereihen-der-bmw-7er-serie/bmw-7er-serie-modellereihe-e65-e66/so-wird-s-gemacht/191-aux-im-vfl-und-fl-nachr-sten/%2523post609%26hl%3Den%26client%3Dfirefox-a%26hs%3DSpP%26rls%3Dorg.mozilla:en-US:official%26prmd%3Dimvns&rurl=translate.google.com&sl=de&twu=1&u=http://www.datasheetcatalog.org/datasheet/philips/TDA1562.pdf&usg=ALkJrhjpff9J7SugGX3IDCfVr9q-GX5M0g


I assume no warranty, all work is done at your own risk!


Who is a benefit sional repair wants, which is aimed to please the user: "King.reparatur" from the other forum!

I feel the effort is too great and I do not want to operate commercial is.

I only went about creating it and prove that a repair is possible in our Logik7


ALL REQUESTS FOR JOY WHEN REPAIR

FIG 1: Instructions

FIGURE 2: it should look like it! (Thermal grease and do not forget to drill)

mfg odd ....

More to follow ....... O lines





I hereby publicly thank you for the user from the Russian dj0vitamin BMW forum ( www.bmwclub.ru ), who has developed a similar guide own un told me his knowledge

PAPER
06-26-2012, 02:16 AM
HIS 3'RD POST



I have now a new pair of images from a user.

Have now labeled this.

So much fun with it : Thumbsup:

HP. Here you can see how difficult it is to enfernen the old amps without damaging the contacts

PAPER
06-26-2012, 02:20 AM
4'TH & LAST POST


http://www.reichelt.de/index.html; ACTIO 7d09cd0be611e9a ...?


The connections are labeled with base, emitter, collector (abbreviated diagram by the letters B, E, K)

english or B, E, C!

PAPER
06-26-2012, 02:25 AM
Important: Do not forget before discharging the capacitors. With a light bulb or a resistor.

Way to prevent it before it transmits when soldering

dude_hot10
06-26-2012, 04:15 AM
thanks

dgoodsell
06-26-2012, 07:01 AM
Nice find paper....someone needs to be the guinea pig and let us know how it goes.

Rookie2008
06-26-2012, 11:51 AM
Excellent find bro!

PAPER
06-26-2012, 01:10 PM
WoW , considering all the amp problems .
I thought people would be all over this topic .

132 views & only 3 replys
Sent from my GT-N7000 using Bimmer App

pass427
06-26-2012, 01:36 PM
WoW , considering all the amp problems .
I thought people would be all over this topic .

132 views & only 3 replys
Sent from my GT-N7000 using Bimmer App

That's because I have no clue what was written but I will take it to my electronics guy and I am quite sure he can decipher it. It funny as I just received my Logic 7 amp in the mail today after sending it out for repair. It it works then that will save me on future repairs

mrbmwx5
06-26-2012, 02:07 PM
I just checked around and seem like no one have them chips...
http://www.arrownac.com/parts/description/audio-amp-speaker-2-ch-stereo-7w-class-ab-17-pin-dbs17p-tube

PAPER
06-26-2012, 08:42 PM
All I can tell you is keep looking , they are available & Google search is your friend

Sent from my GT-N7000 using Bimmer App

II Kings 9:20
06-26-2012, 08:51 PM
Nice find. I did a similar repair to my SID (Saab Info Display) albeit way less complex. I hope I never need this info but nice to know it is out there. I hope mine lasts since it is not on the CPO dime.

pepe65
06-26-2012, 08:56 PM
dont know what 96% of any of that means but i sure wish i did.

PAPER
06-26-2012, 09:39 PM
dont know what 96% of any of that means but i sure wish i did.

basically it's instructions on which chip to remove from
the logic 7 , & which chip to replace it with .
Also the soldering points are given .

GregHolla8
06-27-2012, 04:25 AM
I did some research the BEK transistor KT3107 it's Russian
TDA8564 and TDA1562 u could find them all on eBay

spyders 745li
06-27-2012, 05:25 AM
Nice find. Just sent mine off to Becker for repair. Now we have instructions on the repair. Very good find

Sent from my R800at using Bimmer App

GregHolla8
06-27-2012, 06:31 AM
I bought all the caps all the transistor all amplifiers can't wait to get back to NYC to dive on this put the marine corps electronics knowledge to work

second 7
06-27-2012, 07:44 AM
Yes , I seem to have ran into that same wall. A company on the away coast just informed me that they can't locate a chip that is needed . I believe Paper is onto something. Must investigate into it further. Mean while no sounds ! :mad:

Sent from my PC36100 using Bimmer App

PAPER
06-27-2012, 01:02 PM
I think you mean the WEST coast .
& as for the chips , they are available .
Just use Google search or buy them on Ebay

Sent from my GT-N7000 using Bimmer App

PAPER
06-27-2012, 01:04 PM
Also you guys might want to bookmark
This topic before it gets barried .
After all this topic ain't stickied
Sent from my GT-N7000 using Bimmer App

second 7
06-27-2012, 04:01 PM
This is some crazy info. I wish I really knew what to do with it. The shop on the west coast has yet to tell me what chip it was that they couldn't get but i would think just like I found it in eBay and the site you posted I would think by them being a big company. However if I need to get the oart and get it to them Ao be it. I may need that unit if its still availabe. I'm 3 ask units in the hole. Lol

skybmray
06-27-2012, 11:16 PM
Great find, I will be definately ordering the chips as my drivers side bass seems to crackle a lot.

PAPER
06-29-2012, 11:30 PM
Bump

PAPER
07-03-2012, 10:21 PM
Bump

rickmckee730d
07-04-2012, 05:20 AM
Great find :thumbup:, maybe my electronics shop can follow the instructions, i certainly can't :rofl:

PAPER
07-08-2012, 06:49 PM
bump

AceFX
07-09-2012, 03:31 AM
Top effort PAPER. Being an IT Engineer myself this all makes good sense. Great find by you man..... :thumbup:

olegmedinskiy
08-07-2012, 06:13 PM
Hay cool stuff I found a lot about thet on Russian forums I'm fixing my l7 soon ordered all the stuff I need

Sent from my HTC VLE_U using Bimmer App

DriftGirl
09-02-2012, 10:05 PM
Found a guy today that got me the chips needed....my L7 (Subs) seems to work in the mornings, like the board needs to be re-flowed...i dunno :(

They were free, so I may give this "fix" a try...Just wish there were clear pix of the solder points.

http://i250.photobucket.com/albums/gg274/DriftGirl2/DGL7.jpg

TylerZ06
09-19-2012, 05:53 PM
how can I tell if the amp is bad or not? Some of the speakers crackel and some dont.. It seems to cut out at times and I thought the speakers where blown but now I just dont know.. I dont want to go buy replacement speakers since people say the stocks are pretty good. I also dont get much if any bass from the subs under the rear seats. I was thinking about replacing them with some rockford shallow 8's. Input is appreciated!!!

second 7
11-11-2012, 08:06 AM
Hey paper. That was a great fine on the logic seven repair. I tried to go on the Russian link that you included in the forum but I was in able to translate . However the repair was a success. I'm looking to repair my ask unit. Any pics , suggestion or links you can fwd me to out me in the right direction. Preferably type of chip and chip pin location ? Thanks in advance for your assistance ...

PAPER
11-28-2012, 08:51 PM
Hey paper. That was a great fine on the logic seven repair. I tried to go on the Russian link that you included in the forum but I was in able to translate . However the repair was a success. I'm looking to repair my ask unit. Any pics , suggestion or links you can fwd me to out me in the right direction. Preferably type of chip and chip pin location ? Thanks in advance for your assistance ...

this is the pic of the ASK Unit chips that need to be replaced & the chip #

second 7
11-29-2012, 04:56 AM
Thanks you very much. Very appreciative.... For everything .;)

Sent from my PC36100 using Bimmer App

BimmerBeast01
01-06-2013, 02:18 PM
Does anyone know a good place in NYC that does repairs to the logic 7 amp?

stuartjohn24
01-07-2013, 04:51 PM
Nice information!

My AMP finally blew tonight.... Well at least I think it has, my idrive has been playing up intermittently every now and then for a month or so.

Tonight I got in the car and there is music playing but only for a second, then stops for a second then plays for a second etc... at very low volume and crackly. music only come out the front door speakers.

The idrive keeps reverting back to the radio menu, the TV functions dont work it wont select a channel, I can select ipod tracks with a few attempts but they wont play unless you keep trying! and the ASK wont play CD's! Also the MMI wont shut off after I have removed they key!

INPA cannot communicate with the AMP so im assuming thats the issue? It sounds more serious than just the capacitors or amplifier IC's going down and more like the actual MOST interface itself is rebooting. Or could the AMP be rebooting as one of the caps are short and its PSU is overloading and trying to restart?

Will pull it out and loop the MOST bus and see if it is indeed the problem!

Stuart

stuartjohn24
01-09-2013, 11:11 AM
Pulled the AMP last night, looped the fibre bus and idrive is now working correctly and I have sound again! (Sounds completely crap as expected!)

Anyway took the amp apart to see if I could identify anything obvious.

http://i886.photobucket.com/albums/ac66/stuartjohn24/618DCE95-D422-486C-83D2-0146D8D816AB-213-0000002878721839.jpg

Caps don't look suspect visually, which I kind of expected as it was working perfectly up to the point when it suddenly failed.

I will get the caps on the ESR tester at work to confirm if they are good or not.

Interestingly the amp has already been replaced at least once, this one was made in Q1 of 2006.

Upon closer inspection the amplifier PCB shows evidence of water damage, this has caused corrosion between the pins of the IDC board connector.

http://i886.photobucket.com/albums/ac66/stuartjohn24/B97EAB11-DA5D-41D4-BB8B-6BD96C159F3D-213-000000286977079C.jpg

http://i886.photobucket.com/albums/ac66/stuartjohn24/22A28A3F-E965-44FF-9869-66EA8397FF00-213-000000286196B6EB.jpg

The symptoms I am having is resetting of the MOST bus every second or so, with audio only coming out of the speakers driven by the ASK unit (front door mids and tweeters, and the rear mids) also after removing the key from the ignition the MMI will stay powered and not shut down.

The amp is constantly interrupting the MOST bus as if the MOST controller is continually resetting. This leads me to believe the amplifier part is ok and it's a control issue, which almost certainly means a new amp!

I have been inspecting the controller PCB and I have noticed a Maxim watchdog/brown-out supervisor IC, I'm wondering if this is what's resetting the controller every second because of a low supply voltage condition possibly caused by the corrosion on the board connector?

I will power it up on the bench and get a scope on it see what's going on tomorrow, clean up the corrosion and hope it sorts it out!

I will update on progress for those of you interested, may be useful for some of you.

Stuart

PAPER
01-11-2013, 04:03 PM
I hope by cleaning the corrosion your Amp
Starts functioning properly .
Good luck

Sent from my GT-N7000 using Bimmer App

stuartjohn24
01-12-2013, 03:54 AM
Thanks PAPER,

I have cleaned the board up but no luck as yet. One of the processors is not kicking the watchdog so it keeps resetting.

The reason for this is either a duff processor or a fault elsewhere on the board. There is a MOSFET on the amplifier card that's getting very hot, (in my first pic of the PCB its the one on the lower left in the corner) it looks like its controlled by the processor card and is used to power the amplifier up once the processor card has been woken up by the MOST bus.

I'm guessing there is a fault elsewhere with either of the amplifier ICs that's drawing excessive current. Examination with a thermal camera has yet to determine anything.

Pretty gutted really, seems I've been really unlucky and got a much more serious problem. It won't be as simple as replacing caps and amplifiers as I had hoped!

Stuart

PAPER
01-13-2013, 04:25 PM
Sorry to hear cleaning didn't solve your problem .
Keep us posted on your progress

Sent from my GT-N7000 using Bimmer App

stuartjohn24
01-14-2013, 11:11 AM
Right then....

Been messing around with this at lunchtime today, removed all inductors and capacitors from the amplifier card to eliminate anything there.

I fired the amp up and still a fair bit of current being drawn, reset line on processor card still being pulled low by the watchdog IC every 1.2 seconds.

http://i886.photobucket.com/albums/ac66/stuartjohn24/B43B9ECD-550D-4893-B381-CE4EFFDEE948-2388-000001F0E30795C5.jpg

Decided to have a poke around the board with the scope, noticed the processors were doing "something" so tried the thermal camera on the processor card.

http://i886.photobucket.com/albums/ac66/stuartjohn24/130CAD7E-DFDA-45B4-8CA7-10C4A98BF51E-2388-000001EFD88CBBA2.jpg

Noticed the EEPROM cooking itself! Checked the 5V rail and found it to be over 9V!

Anyway long story short... The component getting hot on the amplifier card in the corner was the MOSFET for the 5V power supply. When I originally started testing the card the output voltage from the MOSFET circuit was 12V, I didn't think this was abnormal. Little did I know it should have been 5V!

Luckily the three processors and RAM all run on 3.3V and have not been over voltaged.

I therefore tried running the processor card from a separate 5V bench supply.

This provided some hopeful results, the processor is now kicking the watchdog IC and it's no longer resetting every second or so.

I will investigate whats going on with the 5V supply and take it from there.

There is a slim chance it's repairable if the EEPROMS are not damaged. (I don't have the code, and I can't imagine it's easy to get hold of!)

Stuart

PAPER
01-14-2013, 11:39 AM
Right then....

Been messing around with this at lunchtime today, removed all inductors and capacitors from the amplifier card to eliminate anything there.

I fired the amp up and still a fair bit of current being drawn, reset line on processor card still being pulled low by the watchdog IC every 1.2 seconds.

http://i886.photobucket.com/albums/ac66/stuartjohn24/B43B9ECD-550D-4893-B381-CE4EFFDEE948-2388-000001F0E30795C5.jpg

Decided to have a poke around the board with the scope, noticed the processors were doing "something" so tried the thermal camera on the processor card.

http://i886.photobucket.com/albums/ac66/stuartjohn24/130CAD7E-DFDA-45B4-8CA7-10C4A98BF51E-2388-000001EFD88CBBA2.jpg

Noticed the EEPROM cooking itself! Checked the 5V rail and found it to be over 9V!

Anyway long story short... The component getting hot on the amplifier card in the corner was the MOSFET for the 5V power supply. When I originally started testing the card the output voltage from the MOSFET circuit was 12V, I didn't think this was abnormal. Little did I know it should have been 5V!

Luckily the three processors and RAM all run on 3.3V and have not been over voltaged.

I therefore tried running the processor card from a separate 5V bench supply.

This provided some hopeful results, the processor is now kicking the watchdog IC and it's no longer resetting every second or so.

I will investigate whats going on with the 5V supply and take it from there.

There is a slim chance it's repairable if the EEPROMS are not damaged. (I don't have the code, and I can't imagine it's easy to get hold of!)

Stuart


VERY NICE !!! Glad you have a clue of what it might be . I hope it all turns out
good & repairable for you . Must be Nice to have all those tools at work :thumbup:
As I recall you made yourself a bench tester for the MMI also
( that's a great helpful tool you made , I need to try & find some time to make
a bench tester or even buy 1 if I can find 1 for sale . Some 1 was selling 1 they
made last year but I contacted him to late :(

stuartjohn24
01-14-2013, 03:47 PM
Thanks PAPER,

Yeah it has its benefits, most of this car is electronics so im comfortable with that!

Im busy reverse engineering the entire amplifier card, I'm tempted to re-lay the PCB and fit the replacement TDA8563 and TDA1562 amplifiers properly, been playing around quickly and it looks pretty straightforward. Could even design in some proper heat sinking and cooling or fit some higher powered amps for the subs!

http://i886.photobucket.com/albums/ac66/stuartjohn24/34D9406D-EAE2-443C-AC29-B6EFE1287726-2388-0000020C99BCF2A1.jpg

Will think about doing that if I can fix my amp! one step at a time! Hopefully the EEPROMS aren't completely fried! I may be begging for someone to read the code from the IC's for me!

Anyway...

The test box thing is great, been running the webasto park heater with it in the garage, works well.

Thats interesting, be nice to know what that one was like compared to mine, Ideally it would be best to fit a used CAS if I can find one cheap enough! so I can start coding modules too.

Stuart

PAPER
01-15-2013, 02:41 PM
The unit was small & had a few options .
Oh & he had a CAS hooked up to it . Not
Sure what else as everything was in a housing ,
But he did say it was used to code modules .
As I recall 1 pic had a cluster hooked up &
Another had an MMI

Sent from my GT-N7000 using Bimmer App

stuartjohn24
01-19-2013, 05:21 AM
Hi Paper,

Yeah I will try and get hold of an old CAS or working LM, I guess he's using the CAS to 'wake' modules. I just send the terminal status CAN message using an arduino and CAN shield.

Anyway, I am still working on this! I have been busy reverse engineering the amplifier card to try and work out what has failed.

Here is the schematic I have drawn of the power supply and control part of the amplifier card:

http://i886.photobucket.com/albums/ac66/stuartjohn24/823E2F96-FD67-4CD9-B059-2854A6ED15A7-5850-000004FF290E045D.jpg

It roughly works like this:

Constant battery voltage regardless of ignition status is present at T1, Q1 and pin 2 of the ribbon cable connector.

When the optical bus is woken, the previous module transmits light to the amplifier, the circuit on the processor card 'wakes' the amplifier (more on that later) by pulling pin 15 of the ribbon cable connector to ground/0V.

When this happens T5 switches battery voltage through two resistors (R14 & R17) and to the zener diode D3 which regulates the voltage to 10V. This now switches on Q1 the main power MOSFET.

Q1 now switches on and the output quickly tries to reach the battery voltage, however D2 starts conducting at around 5V, this switches on T6 which in turn tries to switch off Q1, this has now formed a regulator. This is what generates the 5V for the processor card.

T6 was shorted so the regulator circuit didn't work. I have replaced T6 and I now have 5.1V!

More to come...

stuartjohn24
01-19-2013, 06:06 AM
T1, T2, T3 and T8 appear to form part of a protection circuit, I'm sot sure exactly, I haven't been clever enough to work it out yet, analogue electronics is not my expertise!

The circuit is connected to the MOSFET's on the output stage by the main connector.

The amplifier card receives pre-amplified analog audio from the processor card via the ribbon cable.

I have powered the amplifier card up without the processor card and it fires up OK when pin 15 is pulled low, all looking good.

However connecting the processor card it seems to be powered up all the time! Measuring pin 15 and its pulled low all the time.

Inspecting the processor card pin 15 is pulled low by the circuit highlighted and labeled '7' in the image below:

http://i886.photobucket.com/albums/ac66/stuartjohn24/40F4B48F-7F44-4575-82C8-FCDB336326ED-5850-0000050170174B8C.jpg

Circuit - Description
1 - 4 channel digital to analogue audio converter, this surprisingly only appears to be controlling/decoding the center speaker! audio wise anyway.
2 - 4 channel digital to analogue audio converter, also appears to be controlling the fibre interface.
3+4 - EEPROM
5+6 - Static RAM
7 - 'Wake Up' circuit and schmitt trigger inverter, possibly to filter MOST data and send to processor (2)
8 - DSP (Digital signal processor) this performs the decoding of the digital audio data to determine which channel the sound comes from.
9 - System clock circuit, this generates the timing for all three processors.
10 - This is a DC-DC voltage converter, it supplies + and - supplies to the pre-amplifiers 12 and 13.
11 - Watchdog and brown-out supervisor IC, this is the IC resetting the processors every 1.2 seconds.
12 - Audio pre-amplifier, IC on the right is the pre-amp for the subwoofers, the IC on the left is the pre-amp for the rear door speakers.
13 - Audio pre amplifier for the center speaker.
14 - I2C serial EEPROM, My guess is this stores the current equaliser settings that you set using the idrive.

The IC next to 14, is an I2C temperature sensor, its output drives the fan circuit on the amplifier card. The fan circuit is a mess, one forum stated that the fan speed increases with volume, I cant see that being the case, the advised fix if this is faulty tells you to fit entirely the wrong type/gender transistor!

Anyway...

The circuit was covered in a green sticky glue, I removed that and immediately smelled something burned!

http://i886.photobucket.com/albums/ac66/stuartjohn24/875B490F-54BB-4BB5-9E16-A27BBEB0E7A0-5850-00000503D6796AFE.jpg

I found the transistor labeled '1BW' looking a bit crispy, this is the transistor that pulls pin 15 to ground to switch the amplifier on when it receives the optical data.

The transistor was short causing the amplifier to be powered all the time, this may be why the idrive wouldn't shut down when the key was removed and kept the car awake!

It looks as though some over voltage or surge condition has occurred and damaged a number of components/circuits on both boards. Im still hopeful this can be repaired!

Anyway I replaced the transistor and the amp now shuts down and powers up only when light is present at the fibre connector.

Processor is still resetting every second or so when powered up. Interesting, when running the processor card from only 5V and 3.3V it seems to work OK and the processors are running.

I suspect some feedback from the amplifier card is shutting it down or there is something up with the 12V rail on the processor card.

Stuart

stuartjohn24
01-21-2013, 01:20 PM
For those of you following...

I have been playing around with this again today, to cut a long story short the amplifier card power supply circuit works ok now after the regulator repair.

However the processor card has still been resetting, I have been playing around checking address lines and data lines on the processor marked '1' in my previous post and suddenly the card started working again.

Probing the board with the scope in the area of the RAM IC (5 in the picture above) seems to fix it, it looks like a dry solder joint fixed by slight pressure of the probe on the board or the added capacitive influence of the scope probe wiring on one of the address lines stabilises it.

The amp stayed powered up for at least two minutes with a torch aimed at the fibre connector!

The picture below shows the processor (2 in the photo) kicking the watchdog IC (11 in the photo) every 976.8 micro seconds, that's basically 1000 times a second! The watchdog will wait a maximum of 1.2 seconds before resetting the processors if no pulse is received from the processor.

http://i886.photobucket.com/albums/ac66/stuartjohn24/EB8BF1CC-417A-420F-BA46-048B19F855F2-7637-0000069402F1EF2C.jpg

I will investigate what exactly is causing the board to 'fix' itself, I'm hoping its a dry joint or corrosion under the RAM IC that I can clean and touch up with a soldering iron.

Anyway will have another look tomorrow, I hope to report some good news and get some bass back in the 7 pretty soon!

Stuart

PAPER
01-22-2013, 01:04 PM
Wow Stuart you been busy . I need to make
a mental Note to myself to remember to check
this topic & section of the site more often .

Sent from my GT-N7000 using Bimmer App

stuartjohn24
01-23-2013, 03:03 PM
Well... Things started looking up! But... I noticed following further testing that some of the address lines between the processor, EEPROM and RAM were held low during some power-ups on random occasions. Sometimes the address lines would work intermittently, or do nothing for several power-ups. This became more and more frequent, I tried freezer spray on areas of the board to see if that changed things.

Spraying freezer spray onto the processor instantly makes it start working, sadly this points to the processor (1 in my picture above) itself being at fault. Pretty annoying, however being determined to fix this to save myself the cost of a new AMP I tried some other things as I have nothing to lose now.

Considering the failure of the main power supply regulator circuit and the subsequent over voltage of the 5V rail I started again with the EEPROMs, they seem to be working as some code is being read so that CPU kicks the watchdog. I removed the EEPROM labelled 3 in the picture and tested again, The address lines now have clean signals on them and I have tried 30+ power-ups using the fibre port and cycling the main power supply and it works perfectly every time.

Im wondering wether the EEPROM is heavily loading the processor address or data lines causing it to hang or over current or the chip select and output enable function of the EEPROM is not working and its writing data when the processor or RAM is trying to. Hopefully the EEPROM may be readable using a bench programmer, fingers crossed and I will program a new one.

Otherwise I will be on here begging for someone to read the code for me!

Stuart

stuartjohn24
01-26-2013, 03:55 AM
I have pulled the EEPROMS and managed to successfully read the code from them.

To my surprise I've had the EEPROM code on my laptop all along! The E65 data folder containing all the software for flashing modules in WinKFP has the amplifier software hidden away!

I have managed to extract the binaries from the files and successfully match them up with the HEX data pulled from the EEPROMS. I have used a program to compare the data to see if the data in the EEPROMS are damaged. Converting some of the HEX lines of code reveals the same software version and build date, however there are a few very minor differences....

http://i886.photobucket.com/albums/ac66/stuartjohn24/866F4B3B-8DA2-4F29-883C-F49DD4270181-9741-000009254F5F29FD.jpg

You can see on line two that the first byte is 55 in the EEPROM, but it should be FF as shown from the disassembled winKFP file. Also address 105 has 16 bytes different, oddly the EEPROM just has 07 in every location, the other EEPROM is the same in that respect.

The rest of the file is identical, there are 208 pages comparing a total of 128 thousand bytes, would be time consuming doing that manually!

Anyway there is one exception, the winKFP file only has data at the beginning and very end of the EEPROM, there is a huge empty block in the middle, however most of this is occupied on the EEPROM, the reason for this appears to be a holding space for updating software, when you try and update the amp software using winKFP it will load the data to the free space an once all the data has been transferred correctly and the checksums match the amp will reboot and copy that data to the start of the EEPROM, this is the application code. The code at the end off the EEPROM I think is the boot loader. Anyway now I've sussed out the EEPROM software I'm going to flash some new EEPROMS and hope for the best!

I'm after a favour though, the newer amps don't have the epoxy covered processors, but have proper packages, does anyone have a picture of the newer amplifier PCB's or know the part number written on the processors?

Cheers

Stuart

DriftGirl
01-26-2013, 11:57 AM
WOW! Great work Stuart!

stuartjohn24
01-28-2013, 05:35 PM
WOW! Great work Stuart!

Thanks!

Been playing with this again today, I removed the EEPROMS and connected a logic analyser to the first 8 bits of the EEPROM address bus.

http://i886.photobucket.com/albums/ac66/stuartjohn24/84BA0D0C-E874-4079-BCE9-0214BA820F25-11486-00000A189F3C02FE.jpg

The logic analyser connects to A0 - A7, there are 16 address lines in total. I only have an 8 bit logic analyser though :(

Anyway... when the processor powers up it will instantly address the EEPROM and start executing code. The image below shows the address lines after power up.

http://i886.photobucket.com/albums/ac66/stuartjohn24/6E0B9726-3375-4382-A45F-B507755C0AB8-11486-00000A1844A76C3D.jpg

For those of you familiar with HEX, the processor first addresses 0x00000001 of the EEPROM and increments the address by 1 every 500 micro seconds. There is no code there as the EEPROM is missing, so the processor just counts up until its reset by the watchdog IC.

This does prove now that the processor is working, before it would randomly hang at power up. I am suspecting an issue with the EEPROM or RAM addressing that is causing the issue.

Its a very strange setup, the RAM and EEPROM do not share the same address or data bus like conventional micro processor systems, and the upper address bits are controlled by flip-flops.

I hope to post the 'FIX' soon so I havent just filled this thread with worthless rubbish! wish me luck! :D

Stuart

PAPER
01-29-2013, 04:37 PM
I must say your doing a great job on your
Little project there Stuart :thumbup:

Sent from my GT-N7000 using Bimmer App

stuartjohn24
01-29-2013, 04:52 PM
Thanks PAPER,

My mate reckons the amount of time I have spent on this I could have just filled out a load of online surveys instead and just bought a new amp! he's probably right!

Anyway, I fitted some PLCC32 sockets today so I can swap out EEPROMS easily, The two I removed from the card originally erase OK but one of them fails to program with an addressing issue. This definitely means one of them was faulty. The remaining one does program OK but still causes the processor to hang every now and then. When it does work the EEPROM sends out the right data.

I had a PLCC32 1Mbit intel EEPROM kicking around at work, I plugged that in and the address lines work fine, unfortunately its so old (1993) the new programmer wont program it with the software!

A correction to my earlier post, the EEPROM is addressed from A0 to A16, the start address or reset vector is at 0x00010000, thats where the processor starts reading the code from the EEPROM.

Anyway, I have ordered some new EEPROMS today, but some 3.3V ones instead, on the board you can swap a link so you can use 5V or 3.3V EEPROMS, the system is 3.3V but it uses resistors to protect the CPU from the 5V logic output when using the 5V EEPROMS. If this does ever work again at least the EEPROMS wont be damaged if the PSU goes wrong!

I will update when the new EEPROMS arrive and I get them programmed.

Stuart

stuartjohn24
01-30-2013, 11:10 AM
Well, at last I have some good news! :D

Im trying not to get too carried away but i'm 99% sure Ive fixed it! :)

I dug out the old EEPROM programmer today and managed to flash some new EEPROMS with the code I managed to read from the old ones.

I fitted them to the board in then new IC sockets and switched ON, the processor giving the problems was addressing the EEPROM OK but the code was not presented on the output. The other EEPROM worked fine :confused:

After a bit of head scratching the new EEPROMS have a pin to make them go into a low power shutdown state, this was floating as the original EEPROM does not have this feature. I therefore shorted this to 5V and it started working!

http://i886.photobucket.com/albums/ac66/stuartjohn24/908440B6-7425-40D9-B9A4-9ABD975AB552-211-0000001C65D36B3D.jpg

This shows the processor card with the PLCC32 sockets fitted and the new intel EEPROMS (or FLASH memory to be correct). you may be able to see the little link on the left EEPROM. The other EEPROM has a slightly different part number, it seems to work OK with the shutdown pin floating. (this is only temporary till the proper 3.3V FLASH chips arrive).

http://i886.photobucket.com/albums/ac66/stuartjohn24/77DA589C-6BC6-4388-8500-013D3279609C-211-0000001C5CF1CB6E.jpg

The watchdog being kicked by the processor, this is the image from the scope thats connected in the previous picture. This is a healthy state and indicates both processors are running and executing the software stored in the memories.

http://i886.photobucket.com/albums/ac66/stuartjohn24/65766EC0-33A3-420E-A3A5-3BECDD544637-211-0000001C6E6634B7.jpg

MOST activity is now present at initial power-up for about 15 seconds, it then shuts down and waits till light is present at the fibre input.

http://i886.photobucket.com/albums/ac66/stuartjohn24/617B8B68-2E0C-4863-8040-3F2B228DA110-211-0000001C75C6CA3B.jpg

And finally a thermal image showing the two processors warming up indicating they are running the code and on the right the DSP processor is also running now communication has resumed between the three processors.

I didnt have time this evening to assemble everything and try it in the car, the amplifier card need re-assembling which I will do tomorrow and try it out in the car. I hope this has fixed it at last! :dunno:

Stuart

hamfei
03-13-2013, 01:12 AM
You are genius!!

vernon3rdw
04-13-2013, 07:10 PM
Hey all I'm no where near as tech savey as u but I pulled mine and I see same corrosion but the rest of the board is fine
Did u ever get it to work

Sent from my SAMSUNG-SGH-I727 using Bimmer App

oddysseus3
04-18-2013, 10:59 AM
Hello @ all:D

Stuart you are great:thumbup: Now i have learn somthing more about the L7 System:D
Before i was replace only the TDA.

odd.....

mpmaster
04-23-2013, 01:12 PM
thanks

Verberne
04-26-2013, 02:57 AM
Great info on this forum! Priceless.
I'm from Holland and i drive a 2004 735i with logic 7.
Few months ago my right sub was making noise and crackling. My right back door speaker didnt work for a longer time.
For About a week the whole logic 7 shut down and only the ASK works perfect ( replaced a new about a half year ago )
So i ordered the TDA 1562 for the right sub and the TDA 8563 for the back door speakers and soldered it on the board as the instructions on this forum.
I tested the amp but still doesnt work. So I think there is another problem in my amp because the whole amp is not working.
I see a lot of smart people on this forum reparing and experimenting with the logic 7 amp so I hope someone here can help me with my problem.
I'm sorry for my bad english hope you guys know what I mean and can help me with my problem.

Mattman645
05-01-2013, 08:09 AM
Stuart - i have no idea what you are on about in your posts, but it sounds might impressive and hope it all works!
I've just bought a 645 cab and found the rear right speakers buzzing so looks like i might have a similar amp issue.
I plan to remove it and check for water damage first - whats the best/safest way of cleaning off any corrosion?
Thanks

Salmantheman
05-30-2013, 04:40 PM
STUART was abducted by the BMW team for knowing too much LOL :P

lucio_dohc
06-26-2013, 01:09 PM
Hi! Im having a Inssue with my L7 amp, but my problem Isn't the amp chips...its a short in another component.
I has marked it in the picture, I cant find what this component is or the name to search it.
Please if any body knows tell me!...thanks a lot!

Lucio

stuartjohn24
06-26-2013, 01:53 PM
Hi! Im having a Inssue with my L7 amp, but my problem Isn't the amp chips...its a short in another component.
I has marked it in the picture, I cant find what this component is or the name to search it.
Please if any body knows tell me!...thanks a lot!

Lucio

Fairchild Semiconductor
HUF76107D

It's a power transistor that regulates the 5V power supply for the processor card.

How you know it's faulty? The amp is basically a write off if this has gone see my previous posts. I managed to get mine back to a semi working state but its intermittent and crackles now and then so I have given up with it.

The design of this power supply is very poor, it's a major weakness in the design, no surprise it was ditched in the later revision of the amp for a switch mode DC-DC power supply, much more robust for this type of application.

Also the later amps manufactured after Q2 of 2006 have proper encapsulated processors (OS8805) not the chip on board crap the first ones had!

In summary the later ones should be much more reliable than the older ones with respect to the actual clever processing bit. Unfortunately the actual amps aren't any better!

Stuart




Sent from BimmerApp mobile app

lucio_dohc
06-26-2013, 02:39 PM
Thanks a lot for your answer Stuart,..I think that is this component becuse its extremely hot when the board is conected.
I have the same simptoms as you...the fiber optic board is reseting it self all the time, and when i turn of the car te I-drives continues working and non stop.
Sorry I had not noticed your previous post...I read them..and I have the same problem! :S

stuartjohn24
06-26-2013, 02:48 PM
No problem, yeah it gets very hot if you run the amp with the card removed from the metal chassis!

You will most likely find that the component is actually OK, the fault lies with either the Zener diode or the little three pin transistor controlling the voltage at the base input of the big transistor that's getting hot.

I would get a multimeter and check the voltage on the processor card, there are two oblong brown chip things near the ribbon connector on the main processor board. These are inductors on the power supplies, one should be 12v another 5v and there should be a pair just above the left processor which should be 3.3v one is analogue supply and the other digital supply for the two oasis processors.

You should also find a -12v supply for the pre amp IC's I have highlighted most of it on a previous post.

Stuart


Sent from BimmerApp mobile app

ddd535
06-27-2013, 02:05 AM
hello . somebody knows how to read and write internal flash ? chip xs8805b0aq ( os8805). dug the entire Internet . nothing found

ddd535
06-27-2013, 02:19 AM
I decapsulated processor in the old Board of logic7

stuartjohn24
06-27-2013, 03:23 AM
Hi,

Are you trying to flash a blank X/OS8805?

I think it can be programmed directly from the MOST interface when blank, the flash files are in the winkfp data folder, try using winkfp to program.

To read I'm unsure, I have seen options in INPA to read flash files from modules not sure if the AMP supports it, be worth checking.

Otherwise you may need a MOST development kit to be able to talk to the processor at low level via the fibre interface.

Interesting work you have done decapsulating the processor! Have you found any other information on the dies? I assume they are somewhere between the OS8804 and OS8805, the original obviously had external EEPROM and a basic D to A for the audio as it required separate pre-amps on a 24V rail to actually drive the main power amplifiers.

What I find interesting is the OS8805 has built in DSP's but they use the Motorola DSP? Perhaps the system designers were more comfortable using the Motorola DSP instead of the ones built into the oasis chips?

Interestingly the older Becker BM54 radios uses a OS8804 which is MOST capable but its not used! However it's there for its DSP and D to A functions.

Stuart


Sent from BimmerApp mobile app

ddd535
06-27-2013, 10:08 AM
I decapsulated processor to show what is written on the chip . wanted to find a replacement , but in Google no information found . I have not worked with the program winkfp . winkfp be able to find and do the update logic , when a blank flash ?

First_745Li
07-02-2013, 06:18 PM
Bump for good find!


Sent from BimmerApp mobile app

albthat
07-03-2013, 08:36 PM
Stuart, You are not from Earth man! LOL....My goodness you are an electronics genius. There is no way on GOD's green earth I would have/could have done or gone through all of that. Thanks for all the information....It was quite entertaining and I was rooting for you man. I had to laugh out loud at another comment that said you had been abducted by the BMW people. Good stuff though man. GREAT JOB!

shukys
08-11-2013, 04:13 AM
Maybe someone has wiring diagram of Logic 7 and could tell original amp chips markings?


Thanks in advance

thunderkyss
10-05-2013, 03:24 PM
StuartJohn24.... thank you.

I don't understand everything you're saying. I've got a little experience in electronics & amplifiers, but what you've cataloged here has given me a bunch of jumping off points to study & increase my knowledge.

Thank you.

lucio_dohc
10-11-2013, 10:21 AM
Right then....

Been messing around with this at lunchtime today, removed all inductors and capacitors from the amplifier card to eliminate anything there.

I fired the amp up and still a fair bit of current being drawn, reset line on processor card still being pulled low by the watchdog IC every 1.2 seconds.

http://i886.photobucket.com/albums/ac66/stuartjohn24/B43B9ECD-550D-4893-B381-CE4EFFDEE948-2388-000001F0E30795C5.jpg

Decided to have a poke around the board with the scope, noticed the processors were doing "something" so tried the thermal camera on the processor card.

http://i886.photobucket.com/albums/ac66/stuartjohn24/130CAD7E-DFDA-45B4-8CA7-10C4A98BF51E-2388-000001EFD88CBBA2.jpg

Noticed the EEPROM cooking itself! Checked the 5V rail and found it to be over 9V!

Anyway long story short... The component getting hot on the amplifier card in the corner was the MOSFET for the 5V power supply. When I originally started testing the card the output voltage from the MOSFET circuit was 12V, I didn't think this was abnormal. Little did I know it should have been 5V!

Luckily the three processors and RAM all run on 3.3V and have not been over voltaged.

I therefore tried running the processor card from a separate 5V bench supply.

This provided some hopeful results, the processor is now kicking the watchdog IC and it's no longer resetting every second or so.

I will investigate whats going on with the 5V supply and take it from there.

There is a slim chance it's repairable if the EEPROMS are not damaged. (I don't have the code, and I can't imagine it's easy to get hold of!)

Stuart

Hi Stuart, do you know from what part or the board I can run the processor card from a separate 5V bench supply, in order to know if my procesor card is still in working condition?...thanks a lot!

priitcce
04-12-2014, 04:37 PM
Hi

The subs on my car went dead (rear doors and center still works). Took the amp apart and tested the caps. All seem ok. Replaced the sub amps with tda1562q chips. Still not working. No corrosion damage or damage to other components. Does anyone have any ideas?


Priit

tovi
04-22-2014, 07:22 AM
I have 03 745i i replaced emergency brake module and after 5 minute of driving all the light comes on the dash anyone


Sent from BimmerApp mobile app

vernon3rdw
04-22-2014, 08:18 AM
Was it the ABS module You replaced

Sent from my SAMSUNG-SGH-I727 using BimmerApp mobile app

thunderkyss
04-22-2014, 08:22 AM
Whatever it was, probably had nothing to do with the Logic7

tovi
04-22-2014, 08:22 AM
Yes it was the abs module


Sent from BimmerApp mobile app

nelutu
05-11-2014, 02:39 PM
I think you mean the WEST coast .
& as for the chips , they are available .
Just use Google search or buy them on Ebay

Sent from my GT-N7000 using Bimmer App
Hi I have my bmw 730d 06 plate how much cost the map navigation ? thanks nelutu

Romey-Rome
06-27-2014, 02:01 AM
Am I reading that right? pin 4 & 16 go to the same pin? On the amp...

Edit: Found the answer. Clearer pics here. http://www.drive2.ru/b/288230376152265355/

james69z28
08-30-2014, 11:49 AM
great info, does anyone know who the guy is that does repair on the logic 7 amps reasonable, I have read on the net about him but can't locate his e-mail address?

send reply to james69z28@aol.com, and THANKS for all the info we need more guys like this!

whytry
09-05-2014, 07:07 PM
I just did the repairs and used the kit from pixelfixltd and I got this screen below. Any ideas on how to get this thing back up n running?

mrnyc
10-21-2014, 05:13 PM
Anyone know of ask and L7 repair shops in NY or east coast?