Monday, 10 August 2020

REFLASHING BM62 to 2.1.3

P2_0 and P2_4 pulled to ground

EAN pulled high directly to VDD_IO (this is required even that datasheet stated there is pullup!)

22uF cap on VDD_IO (check erata)

5V connected to 5V

gnd to gnd

rx to B10 of stm32

tx to B11 of stm32

run isbtflash.exe

select com port, speed 115200:240 hit connect

browse-> select firmware files:

















hit update button

after the restart of the module identify in EEPROM_tool as IS2064S_012_SPK02_V2.1_BETA

unfortunately, EEPROM_TOOL is not able to upload a new EEPROM patch created from

UITool_IS206x_012_DualModeSPK_v2.1.25_UI_Default_Table.txt or UITool_IS206x_012_DualModeSPK_v2.1.25_BM62_EVB_UartEnable.txt

and IS206X_012_DUALMODESPK2.1_E1.0.4.1_1214.bin

"Device's IC version doesn't match File's IC version: IS206X_012_DUALMODESPK2.1_E1.0"

reading the second post in this thread I just remembered about stupid EAN PIN!

PULL P2_0 and ENA DOWN DIRECTLY TO GROUND! reset! -> IS206X_012_DUALMODESPK2.1_E1.0 !!! VOHO!!

write new eeprom, do not forget check "Load default in into IP" on third screen of MPET.exe

code for stm32:

#define BLUETOOTH_EEPROM_UPLOAD PA4 //goes to P2_0
#define BLUETOOTH_FLASH_UPLOAD PA5 //goes to P2_4
#define BLUETOOTH_EAN PA6 //goes to EAN

void setup() {
  Serial3.begin(115200); //Serial3 => TX = PB10, RX = PB11
//  digitalWrite(BLUETOOTH_RESET, LOW);
  digitalWrite(BLUETOOTH_RESET, HIGH);

void loop() {
//  Serial.println("Start\nsending 0x0120fc00:");
//  delay(1000);
//  Serial3.write(0x01);
//  Serial3.write(0x20);
//  Serial3.write(0xFC);
//  Serial3.write(0x00);
//  Serial.println("responce:");
//  while (Serial3.available()) {    // If anything comes in Serial1 (pins 0 & 1)
//    Serial.write(;   // read it and send it out Serial (USB)
//  }
//  delay(1000);
  while (1) {
    if (Serial3.available()) {    // If anything comes in Serial1 (pins 0 & 1)
      Serial.write(;   // read it and send it out Serial (USB)
    if  (Serial.available()) {     // If anything comes in Serial (USB),
      Serial3.write(;   // read it and send it out Serial1 (pins 0 & 1)

Monday, 22 June 2020

Fixing AcceleDent Aura

Battery with same dimensions and same capacity for the unit can be found here:


If your unit stops working and the LED Indicator is flashing alternating green and orange it means that you reach some stupid hardcoded lifetime of unit. Check FAQ on

There is only one correct response to this if you pay 1000euros for this unit:

Open it, desolder Microchip SPI EEPROM 25LC1024, then use Arduino with this sketch to dump content and then another sketch to clear the whole EEPROM. Solder it back and the unit is again working. My dump is saved here.

Friday, 11 October 2019

connecting JTAG to BK8000L

from datasheet of BK3254 bluetooth module, we know that CPU is BA22, there is posibility that cpu in bk8000L is the same, so I searched "ba22 openocd" and found this post:

from which I take jtag HW and SW part

jtag HW and SW part

 FT232H usb module:

SW: (jtag-r34432.tar.xz)

patch for generic FT232H board

 open ftdi_driver.c

and add this line to ftdi_driver.c file

{ 0x0403, 0x6014, NULL, NULL, l_amontec_jtagkey2, 0 },

/* vendor, product, name pairs of supported devices */
struct ftdi_device_desc supported_devs[] = {
    { 0x0403, 0x6010, "Beyond", "Debug Key", l_beyond_debug_key_v1_2, 0 },          // Beyond Debug Key
    { 0x0403, 0x6010, "Beyond", "Debug Hub", l_beyond_debug_key_v1_1, 0 },          //   (eng. sample 2)
    { 0x0403, 0x6010, "Beyond", "JTAG Adapter", l_beyond_debug_key_v1_0, 0 },       //   (eng. sample 1)
    { 0x0403, 0xcff8, "Amontec", NULL, l_amontec_jtagkey2, 0 },                     // Amontec jtagkey/jtagkey2
    { 0x15ba, 0x002a, NULL, NULL, l_olimex_armusb, 0 },                             // Olimex arm-usb-tiny-h
    { 0x15ba, 0x002b, NULL, NULL, l_olimex_armusb, 0 },                             // Olimex arm-usb-ocd-h"
    { 0x15ba, 0x0004, NULL, NULL, l_olimex_armusb, 0 },                             // Olimex arm-usb-tiny"
    { 0x15ba, 0x0003, NULL, NULL, l_olimex_armusb, 0 },                             // Olimex arm-usb-ocd"
    { 0x0403, 0x6010, "Digilent", "Digilent Adept USB Device", l_digilent_hs1, 0 }, // Digilent hs1 ("adept")
    { 0x0403, 0xac09, "Atomic", NULL, l_amontec_jtagkey, 0 },                       // Atomic programming AP-114 (same layout as Amontec)
    { 0x0403, 0x8220, "DISTORTEC", "JTAG-lock-pick Tiny 2", l_lockpick2, 1 },       // JTAG-lock-pick Tiny 2
{ 0x0403, 0x6014, NULL, NULL, l_amontec_jtagkey2, 0 },
    { 0, 0 }

Connecting JTAG to BK8000L

AD0    TCK
AD1    TDI
AD2    TDO
AD3    TMS
AD5    /SRST
then short SCLK and SO pin on external flash, so CPU cant start up, plug FT232H adapter to USB and fire up jtag sw:
update: starting jp3 shortly after releasing RESET button does the trick also ... 
 nail@sid:~/tmp/jtag_ba22$ ./jp3 ftdi jtag://localhost:1234 -t15M
jp3: JTAG protocol via USB/parallel port for linux.
Version 1.6.3
Assuming debug_if's TAP has an IDCODE of 0x14951185.
Using 'FTDI Single RS232-HS (?)' interface.
Enabling high speed ftdi mode.
Using JTAG clock of 15 MHz
Disabling UART not supported/required on this adapter.
JTAG chain length: 1
Found device 0 (IDCODE: 14951185) to have a debug_if connected to it
  Debug interfaces(s) detected: if3
  Using debug interface 3.
Processor version: BA22 v5.2.2
  Memory interface for debuggers: yes
Remote ba debugging using jtag://localhost:1234
Press CTRL+c to exit.


nail@sid:~/tmp/jtag_ba22$ openocd -f interface/ftdi/um232h.cfg -c "adapter_khz 100" -c " transport select jtag"
Open On-Chip Debugger 0.10.0+dev-00936-g0a13ca1-dirty (2019-10-07-08:06)
Licensed under GNU GPL v2
For bug reports, read
adapter speed: 100 kHz

Info : Listening on port 6666 for tcl connections
Info : Listening on port 4444 for telnet connections
Info : clock speed 100 kHz
Warn : There are no enabled taps.  AUTO PROBING MIGHT NOT WORK!!
Info : JTAG tap: auto0.tap tap/device found: 0x14951185 (mfg: 0x0c2 (Flextronics (Orbit Semiconductor)), part: 0x4951, ver: 0x1)
Warn : AUTO auto0.tap - use "jtag newtap auto0 tap -irlen 2 -expected-id 0x14951185"
Error: IR capture error at bit 2, saw 0x3FFFFFFFFFFFFFF5 not 0x...3
Warn : Bypassing JTAG setup events due to errors
Warn : gdb services need one or more targets defined

 small update: 

so I have few minutes while eating my lunch:
-> GCC
  • GCC 4.9.2
  • Binutils 2.24
  • GDB 7.8,1
  • Newlib C library 2.1.0
so I created account and requested access to beyond studio

GCC is GPL, so it one of these applies:
 - processor supported in upstream

 - they must give as patch

give as some hints, but I do not have time to play with it

Transfer rate in excess of 600 kB/s
30MHz maximum JTAG clock

SW was something special for BA22:
there are patches for gcc and other utils .....  anyone to take look?

then I go back to original site where It all started:
and previous post:

I installed radare2 pyba2 and run ti against bk8000l.bin, but I never use it so, I only try what I find on net :)

git clone
cd radare2 ; sys/
r2pm init
r2pm -i lang-python

git clone
cd pyba2

r2 -I BK8000L

in radare2:

aaaa -AA

Wednesday, 11 September 2019

Audi a4 b5 avant modifying washer fluid reservoir

How to modify washer fluid reservoir on the Audi A4 (b5 platform) to support the headlight washer pump and level warning switch for water level.

Positions of new components after the modification:

 Washing fluid level warning switch

Check the previous picture for the exact position for the switch. The hole needs to be 25mm wide, I use the wood drill (see the pictures). Simply place the rubber seal on the reservoir and mark the center of the seal. Be careful while drilling to avoid damage to the reservoir. After the hole is drilled, clean up edges.

Headlight washer pump

 There is a mark on the water tank at the position where the pump should be mounted. I use a 20mm drill for wood, gently to avoid damage to the reservoir, then clean edges. After drilling and cleaning are done insert rubber grommet for the pump, use some water as lubrication. Then install the pump. Wires for the pump must be at least 2.5mm2. The connector could be the same as for windscreen just connector on the pump must be modified, little notch on the connector must be cut flush.

Friday, 2 August 2019

AUDI concert/chorus stable 12V power for antenna amplifier

If you have a problem with your radio reception it is possible that your radio is shutting down power for the amplifier in your antenna. I do not know why it is, just that output is controlled by the RDS processor, which will shut it after 5seconds.  So simple solution is simply emulating signal from the RDS processor. 1resistor(10k SMD 0805) need to be desoldered and then turned 90 degrees so one side is still connected to the base of the transistor and other side can be connected to ground, check pictures:

 After this mode, 12V DC is on antenna input jack, the only problem which probably raised up is that this setup is probably not protected against short to ground, so be careful and check your cables in your car!

EDIT: radio will not disable 12V on antenna input if around 2.5mA is drawn from it, so 390Ohm resistor should do the trick, use 1W, it can get hot!
EDIT2: soft coding of radio will change nothing.

AUDI concert/chorus fixing missing LCD pixels

Old audi radios concert and chorus (1st gen) are here for almost 20years and flex cable connection to front panel LCD is most likely gone on all of them. Recently I found a cheap solution ( around 10euros). This is not well tested ye, but what can we lose?


- missing pixels, rows or columns

few pictures caused  by bad flex cable connection:

problematic flex cable:

(updated: ) take pictures before you disassemble it

To fix this we need a new cable, I sourced mine at aliexpres (but only 2pcs required). If it will be no longer available simply search for bmw e34 dashboard speedometer flex cable, check this search on aliexpress. You also need T-tip iron similar to this one.  You can buy a combo of cable and iron, but unfortunerly these combos are most time 1cable and 1 iron and you need 2 cables.

When you will have required cable and iron

- while old cable is attached to LCD mark position of tracks on the LCD on both left and right side of LCD, use permanent marker which can be then removed with alcohol, but test if it can be removed or use transparent duct tape. Nex step is to take old flex of lcd and PCB.

 - clean in with IPA. The cable need to by extended, to accommodate for more tracks required. I simply cut one side of cable to close as possible to track, but not damage it. Then I attache this part of the cable to the LCD with fully heated iron, take your time and align tracks of cable to pads on lcd. You will do not have another chance with this probably ....  then count how much more traces you require and cut it from second cable and attache it.

The cable is longer then required, so updated next step is to assamblle it all together, then place cable on PCB side and mark where it needs to be cut and cut it with scissors.