Ремонт прошивка восстановление ICOM

Шаурмэн

спасибо

Ferrum

Канал на телеге

Изображения

IMG_5316.jpg (6.5 Кб, 36 просмотров)

PavelZ

Здравствуйте! Недавно попал мне в руки ICOM A2 (внешне) - неисправный, историю никто не знает, перестал читать и всё. Мигает system красным.
По сети работает, к машинам не подключается.
Разобрал, все чисто, ни горелого, ни паяного.
Залил на флэшку ICOM_Restore.bin, подержал кнопку, нажал 3 раза, замигали огонечки, дождался пока все красные станут зелеными, он сам перегрузился.
Зашел в веб-интерфейс, залил ICOM-BootImage-01-28-00c.bin и ICOM-ApplicationImage-01-44-00c.bin из ICOM-Image-03-15-04_unprotected.7z ([Только для зарегистрированных...] (ICOM_Restore - тоже отсюда)) И ВУАЛЯ!!
К машине подключился, но красный огонек так и мигает.

Куда копать, как отключить FAN TEST, какие тесты провести? Почему в веб написано A1 если внешне A2? Или это китайцы так сделали, типо корпус А2 а начинка А1? Есть еще A1 рабочий, но его трогать не хочется, в веб они идентичны, даже серийник, это нормально?

Логи: [Только для зарегистрированных...]
Скрины и фото прилагаю на яндекс-диске.

[Только для зарегистрированных...]

Ferrum

Ошибку по FAN нужно сбросить подключив вентилятор c сигналом скорости. Тут есть фото точек куда подключать. Все китайские ICOM прошиты как А1 с одинаковым серийником 1120134 для отключения контроля вентилятора нужно прошить в А2 с другим серийником и это тоже я объяснял как делать.
Даже у оригиналов А1 и А2 платы одинаковые и отличаются только прошивкой

simkard

===== English (I apologize for the usage of english) =====

Dear All,

I'm writing here as I want to upgrade my chinese clone ICOM Next to the latest firmware available ... which means I will need to do some electronics first.

From what I understood here is my situation :

• My ICOM is working fine
• It doesn't report 256MB RAM but only 128 (so I will need to replace some resistors)
• Serial number is actually 1120134 (I will need to modify it for converting it to A2 and avoiding FAN self test problem)
• Actual IMAGEs on board are default ones : PACKAGE "03.15.00" ; SYSTEM "01.26.00" ; APPLICATION "01.41.00"

I've ordered the necessary resistors in order to replace the wrong ones but it seems that my ICOM has some weird differencies with the ones described in this thread :

• No white markings on PCB but the reference is "FT_icom next_plus"
• Crystal Y4 installed seems to be the good one : 33000
• R15 is already soldered/available on PCB (but what is its value ... this is another question)
• R119 measure 1k (soldered ... so pretty sure it is a 10k // as I read from other people)

Please see attached screenshots for more informations.

I have defined my route like this :

1. Modify serial number to something like : 1801201 (or higher like 1801202, 1801203, etc ...) // This will require unsoldering the EEPROM and reading/writing using an EEPROM R/W
2. By modifying serial number, this will change the model from actual "ICOM A1" to "ICOM A2" (in web portal port 58000) and will avoid having the FAN self-test error
3. Replace necessary resistors on the PCB (like R119 and many others to be sure values are the correct ones to get full 256MB RAM and correct power supply to RAM chips)
4. Proceed with USB RESTORE method at some point in order to get a fully working ICOM
5. Proceed with "Update Firmware" from web portal port 60080 in order to get latest firmware/IMAGEs (SYSTEM + APPLICATION)

Am I correct or am I missing anything ?

Thanks for your great work, time and help @Ferrum

Вложения

ICOM informations.zip (319.9 Кб, 6 просмотров)

simkard

IMGBB picture codes seem to fail so here they are again :
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]
[Только для зарегистрированных...]

simkard

My measurements with components still on board (without unsoldering each of the resistances // I'll do it when I'll receive the new resistances)

RESISTANCES

• R15 : 26 ohms (should be 52 ohms)
• R101 : 7 kohms (should be 22 kohms)
• R102 : 6.9 kohms (should be 10 kohms)
• R119 : 1 kohms (should be 1 kohms) // This one is tricky as it shows correct 1k value, but it is not ! As someone else reported, once unsoldered, the correct value shown is 10 kohms so it should be replaced with a 1 kohms
• R156 : 230 ohms (should be 70 ohms)
• R155 : 70 ohms (should be 230 ohms)
• R178 : 1.48 kohms (should be 10 kohms)
• R179 : 10.33 kohms (should be 1.4 kohms)
• R195 : 0.5 ohms (should be 38 kohms)
• R233 : 7.66 kohms (should be 47 kohms)
• R280 : 12.86 kohms (should be 38 kohms)
• R285 : 12.81 kohms (should be 38 kohms)
• R292 : 230 ohms (should be 70 ohms)
• R293 : 70 ohms (should be 230 ohms)

VOLTAGES

• RAM Vdd = +2.96v (should be +2.5v) // Well too high as from datasheet it is supposed to be +2.5v
• RAM Vref U18 : +1.48v (should be +1.25v)
• RAM Vref U19 : +1.48v (should be +1.25v)
• RAM Vref U20 : +0.27v (should be +1.25v) // This is the one driven by R119
• RAM Vref U21 : +1.48v (should be +1.25v)
• RAM Vref are lightly too high based on datasheet informations : it should be 0.5 X Vdd ... so if powering RAM chips with +2.5v ... it makes +1.25v

Ferrum

You won’t be able to fix everything because there is no error information for this clone, and the one I wrote has only partial data for your clone!

One PCI device is not working
PCI 0000:00:10.0: IRQ 5 for idsel 16 pin 1
PCI 0000:00:10.2: IRQ 5 for idsel 16 pin 1
PCI 0000:00:19.0: IRQ 6 for idsel 25 pin 1

One kline device is not working

The firmware is damaged in the part of the program responsible for initial initialization and this needs to be corrected by full programming of flash drives

simkard

Hello Ferrum,

Thanks for your feedback on my posts (and answering in english // you can answer in russian for the sake of this post and the future searches).
By the way, I have some questions regarding your comments :

What do you mean ? CPU has some burned IRQs ?
Or is it a problem of solder balls/joints under the chip/IC ?
I found these 3 lines in my DMESG (but I have WIFI card so I should find 4 logically ?)
[ 3.274879] PCI 0000:00:10.0: IRQ 5 for idsel 16 pin 1
[ 3.274927] PCI 0000:00:10.2: IRQ 5 for idsel 16 pin 1
[ 3.274963] PCI 0000:00:19.0: IRQ 6 for idsel 25 pin 1
A little bit after, I can see these lines too :
[ 3.350915] NAND device: Manufacturer ID: 0xec, Chip ID: 0xda (Samsung NAND 256MiB 3,3V 8-bit)
[ 3.351002] Scanning device for bad blocks
[ 3.351130] Bad eraseblock 1 at 0x00020000
[ 3.351205] Bad eraseblock 2 at 0x00040000
[ 3.351280] Bad eraseblock 3 at 0x00060000
[ 3.351355] Bad eraseblock 4 at 0x00080000
[ 3.351430] Bad eraseblock 5 at 0x000a0000
[ 3.357214] Bad eraseblock 134 at 0x010c0000
[ 3.357437] Bad eraseblock 138 at 0x01140000
[ 3.358293] Bad eraseblock 157 at 0x013a0000
[ 3.359106] Bad eraseblock 175 at 0x015e0000
[ 3.361002] Bad eraseblock 218 at 0x01b40000
[ 3.370709] Bad eraseblock 438 at 0x036c0000
[ 3.374442] Bad eraseblock 523 at 0x04160000
[ 3.378217] Bad eraseblock 606 at 0x04bc0000
[ 3.401228] Bad eraseblock 1135 at 0x08de0000
[ 3.414803] Bad eraseblock 1447 at 0x0b4e0000
[ 3.440873] NAND flash size = 256 MB
...
[ 5.583137] block 2 is bad
[ 5.583199] block 3 is bad
[ 5.583222] block 4 is bad
[ 5.583246] block 5 is bad
[ 5.583269] block 6 is bad
[ 5.589182] block 135 is bad
[ 5.589367] block 139 is bad
[ 5.590225] block 158 is bad
[ 5.591038] block 176 is bad
[ 5.592988] block 219 is bad
[ 5.602919] block 439 is bad
[ 5.606757] block 524 is bad
[ 5.610480] block 607 is bad

Why are there some "Bad eraseblock" ??? Is the NAND chip dying but still working ?

Strange ... on all BMW models I have used this ICOM with (Exx series + Fxx series), I never had any troubles at all.
Can you please give me additional details ?

Are we talking about the main ICOM firmware // The one I'm looking to upgrade/update ?
Should the procedure [Modify Serial Number + USB Flash Recovery] can/will resolve this issue ?


I tried searching in this thread, but all I could find related to the PCI messages from dmesg is this thread :
[Только для зарегистрированных...]


Thanks

Ferrum

Один PCI абонент отсутствует он не инициализирован процем, причин может быть несколько как отсутствие связи с процем так и неисправность микросхем. В таких случаях с бедами в рабочей части прошивки я полностью переливаю флеши


Данное исправление не является полным решением проблем вашей копии. У нее много других ошибок которые я не буду показывать