[FIX] Issue antminer l3+ (boards not recognized, bad fw flash , etc)

[cryptonitro]

Here's the issue you may have , hashboards are not recognized, anymore



/!\ please read carefuly /!\

**************************************************************
Firmware Flash
**************************************************************

Here's a fix solution to rewite PIC's of L3+ after some of us flash bad firmware to their L3+.

Many has recovered 3 boards , some all of it

Thanks to Blissz for this fix (please do not pm him for this )

1. Download File
(re-up)

WARNING: Don't flash this to anything else as a bricked L3(+) caused by flashing a wrong firmware version, as this might harm a working L3(+).
MEGA
or
1fichier

2. Plug only 1 board from chain 0 (white cable on the left) as you can see under , you can plug upper hashboard 1st



3. Go to system -> Upgrade -> "Flash new firmware image" choose the file downloaded to flash new firmware and click flash image
wait for reboot.

If kernel logs get this ... 0xcc , wait after you see it (5 minutes, to see again 0xcc or most of time 0x3 and then it will fix your board) other way try to connect an other boards and restart from "3"

Dec 17 14:09:30 (none) local0.warn cgminer: Started cgminer 4.9.0
Dec 17 14:09:30 (none) local0.notice cgminer[377]:  detected at /sys/class/gpio/gpio51/value  chain 0
Dec 17 14:09:30 (none) local0.notice cgminer[377]: detect total chain num 1
Dec 17 14:09:30 (none) local0.notice cgminer[377]: i2c init ok
Dec 17 14:09:32 (none) local0.notice cgminer[377]: Chain 0 PIC Version 0xcc
Dec 17 14:13:45 (none) local0.err cgminer[377]: Blisssz: Chain not found 1
Dec 17 14:13:45 (none) local0.err cgminer[377]: Blisssz: Chain not found 2
Dec 17 14:13:45 (none) local0.err cgminer[377]: Blisssz: Chain not found 3
Dec 17 14:13:51 (none) local0.notice cgminer[377]: Chain 0 PIC Version 0x3
[...]

you will see some logs like this , PIC's is currently writing.

Code:

Dump IIC Data:[41][98][0][0]@ChainID:[0], Chip[C9]

wait logs "miner start and switching pool" the board has been flashed with success.


After fixing all or some of boards , you can finaly flash back bitmain l3+  original Firmware.

**************************************************************
Manually Flash
**************************************************************

/!\ I will not assist with "how to" here cause i didnt experience it yet
but if you have stuff (pickit3) and knowledge to manually flash chip of your miner here's the files

 ! warning ! have a look to the pic version you have (writed on the chip).

there is PIC16F1704 or PIC16F1705

Here's the hexa files for each version

MEGA
PIC16F1704.hex
PIC16F1705.hex

1FICHIER
PIC16F1704.hex
PIC16F1705.hex

Thanks to martin_fryc for sharing files, after succefully fix his L3+ by manually flashing PICs

**************************************************************

[omshree]

How did you get this error? Did you flashed fw before that?

[Bulletdodger]

Weird, havent seen it before. :/

[omshree]

Had the same issue, you have to write the PIC controller manually, extracting the working hexa file from a working hashboard, needs to be done with special chip reader tools.

[cryptonitro]

Had the same issue, you have to write the PIC controller manually, extracting the working hexa file from a working hashboard, needs to be done with special chip reader tools.

hi thanks for answers

how can i do that if boards are not recognized ?
if you have that error ? may be you have file and how to ?

yeah i flash before have that , may be something went wrong.

i changed control board , nothing better

i have only blackcat usb at this time capable to flash , but its for jtag

[blissz]

I am pretty sure it's indeed the pic on each mining board that needs to be reprogrammed. Very annoying.
What you can try:

https://shop.bitmain.com/support.htm?pid=007201604120945219855QObfk20066C

flash this image to a micro sd card (with win32diskimager) and boot that image. Then check the kernel log. Then flash again the l3 image from there and see if it helps. It might force flashing the pic's on the hashing boards.

[cryptonitro]

I am pretty sure it's indeed the pic on each mining board that needs to be reprogrammed. Very annoying.
What you can try:

https://shop.bitmain.com/support.htm?pid=007201604120945219855QObfk20066C

flash this image to a micro sd card (with win32diskimager) and boot that image. Then check the kernel log. Then flash again the l3 image from there and see if it helps. It might force flashing the pic's on the hashing boards.

will try to do that tomorrow , but there is no intruction inside the .zip

will copy the .img to SD with diskimager. Then  copy/paste L3+ firmware inside SD ... hope i'm right  with steps and miner will autoboot from SD

[cryptonitro]

Code:

************ board_new ***************
[   40.728138] fix hardware version
[   40.850954] total chain_num{0}
[   40.854283] FPGA detect asic addr err
[   40.854283] 
[   40.859802] 
[   40.859802] 
[   40.859802] !!!!Restart1 detect asic addr!!!
[   40.859802] 
[   40.869138] total chain_num{0}
[   40.872422] FPGA detect asic addr err
[   40.872422] 
[   40.877942] 
[   40.877942] 
[   40.877942] !!!!Restart2 detect asic addr!!!
[   40.877942] 
[   40.887276] total chain_num{0}
[   40.890588] FPGA detect asic addr err
[   40.890588] 
[   40.896110] 
[   40.896110] 
[   40.896110] !!!!Restart3 detect asic addr!!!
[   40.896110] 
[   40.905462] total chain_num{0}
[   40.908742] FPGA detect asic addr err
[   40.908742] 
[   40.914249] 
[   40.914249] 
[   40.914249] !!!!soft set defult asic num 0!!!
[   40.914249] 
[   40.923654] clear FPGA nonce buffer
[   40.965261] FPGA return data error
[   40.969021] FPGA don't code

The firmware to copy on SD seem to be designed for S series, i start from it and flash L3+ firmware , flash was ok but still no hashboards and same error in kernel after

Code:

Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio51/value  chain 0
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio48/value  chain 1
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio47/value  chain 2
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio44/value  chain 3
Nov 21 10:01:33 (none) local0.notice cgminer[5756]: detect total chain num 4
Nov 21 10:01:33 (none) local0.notice cgminer[5756]: i2c init ok
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: in tty_init
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 0
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[0]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 1
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[1]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 2
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[2]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 3
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[3]
Nov 21 10:01:44 (none) local0.notice cgminer[5756]: open device over
Nov 21 10:01:44 (none) local0.notice cgminer[5756]: send cmd to get chip address
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 0 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 1 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 2 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 3 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Need to recheck asic num !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 0 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 1 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 2 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 3 ASIC 0 !!!

sad

[blissz]

Code:

************ board_new ***************
[   40.728138] fix hardware version
[   40.850954] total chain_num{0}
[   40.854283] FPGA detect asic addr err
[   40.854283] 
[   40.859802] 
[   40.859802] 
[   40.859802] !!!!Restart1 detect asic addr!!!
[   40.859802] 
[   40.869138] total chain_num{0}
[   40.872422] FPGA detect asic addr err
[   40.872422] 
[   40.877942] 
[   40.877942] 
[   40.877942] !!!!Restart2 detect asic addr!!!
[   40.877942] 
[   40.887276] total chain_num{0}
[   40.890588] FPGA detect asic addr err
[   40.890588] 
[   40.896110] 
[   40.896110] 
[   40.896110] !!!!Restart3 detect asic addr!!!
[   40.896110] 
[   40.905462] total chain_num{0}
[   40.908742] FPGA detect asic addr err
[   40.908742] 
[   40.914249] 
[   40.914249] 
[   40.914249] !!!!soft set defult asic num 0!!!
[   40.914249] 
[   40.923654] clear FPGA nonce buffer
[   40.965261] FPGA return data error
[   40.969021] FPGA don't code

The firmware to copy on SD seem to be designed for S series, i start from it and flash L3+ firmware , flash was ok but still no hashboards and same error in kernel after

Code:

Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio51/value  chain 0
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio48/value  chain 1
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio47/value  chain 2
Nov 21 10:01:33 (none) local0.notice cgminer[5756]:  detected at /sys/class/gpio/gpio44/value  chain 3
Nov 21 10:01:33 (none) local0.notice cgminer[5756]: detect total chain num 4
Nov 21 10:01:33 (none) local0.notice cgminer[5756]: i2c init ok
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: in tty_init
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 0
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[0]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 1
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[1]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 2
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[2]
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: tty_init 3
Nov 21 10:01:43 (none) local0.notice cgminer[5756]: Start A New Asic Response.Chain Id:[3]
Nov 21 10:01:44 (none) local0.notice cgminer[5756]: open device over
Nov 21 10:01:44 (none) local0.notice cgminer[5756]: send cmd to get chip address
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 0 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 1 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 2 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Chain 3 ASIC 0 !!!
Nov 21 10:01:46 (none) local0.notice cgminer[5756]: Need to recheck asic num !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 0 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 1 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 2 ASIC 0 !!!
Nov 21 10:01:48 (none) local0.notice cgminer[5756]: Chain 3 ASIC 0 !!!

sad

Hmmmz I'm afraid you need to flash them by hand on the board itself. Let me know if you need any help.

[cryptonitro]

i miss the hexa file , i dont know if pic.txt could be the one

[Egorrichminer]

Hi Folks!

Absolutely the same stuff I'm having now with my L3+ too.

I had a well working L3+, then by mistake flashed it with D3 firmware (instead of different ASIC device... to fix the "x" issue solved in the 20th of November 2017 formware for D3). After that I immediately re-flashed back to the L3+ firmware but got what i got, Zero chips on each of the four hash-blades...

I tried all the stuff including Micri SD image as did here and the same it didn't help.

I also need this suggestion on which hex file to take and from where (the path) on the correctly working hash blade from another L3+. So I "ssh" the L3+ and should "scp" the file ?? to the same location on the malfunctioning L3+ (do that for all 4 blades) and then reboot. Cirrect? Which path to the file then guys?

Please help.

Thanks,
Egor

[cryptonitro]

Hi Folks!

Absolutely the same stuff I'm having now with my L3+ too.

I had a well working L3+, then by mistake flashed it with D3 firmware (instead of different ASIC device... to fix the "x" issue solved in the 20th of November 2017 formware for D3). After that I immediately re-flashed back to the L3+ firmware but got what i got, Zero chips on each of the four hash-blades...

I tried all the stuff including Micri SD image as did here and the same it didn't help.

I also need this suggestion on which hex file to take and from where (the path) on the correctly working hash blade from another L3+. So I "ssh" the L3+ and should "scp" the file ?? to the same location on the malfunctioning L3+ (do that for all 4 blades) and then reboot. Cirrect? Which path to the file then guys?

Please help.

Thanks,
Egor

need to plug programmation stuff in 6 holes on the right of eeprom texas , and extract hexa file from a working l3+ hashboard, then flash it manually on the fail board $

and share the hexa file for some of us have only one fail l3+ but no other to fix

[gians82]

Hello,

I activated today in my new antminer l3 + but I have the exact same problem, it does not recognize any Hashboard.

I opened ticket to bitmain. (*the product is new has never been used.)

how did you solve this problem?

Thanks

*I can not open antminer because I lose the guarantee.

[cryptonitro]

Hello,

I activated today in my new antminer l3 + but I have the exact same problem, it does not recognize any Hashboard.

I opened ticket to bitmain. (*the product is new has never been used.)

how did you solve this problem?

Thanks

*I can not open antminer because I lose the guarantee.

bitmain will told to open , and sent them hashboard ... where come from your miner ?

[gians82]

Hi,

I bought the miner directly from bitmain in China.

But are not the miners tested before shipping?

[gians82]

Hello,

there are no other solutions to solve the problem?

Bitmain does not respond to my ticket.

Thanks

[omshree]

No other solutions apart from sending it back, I am the one who discovered the manual flashing of the PIC controller, and I can confirm it works.

[CryptoWatcher420]

what rom flashing tool did you use?

[gians82]

Hi, you tell me how I can do, do you have a guide? Thanks

Bitmain does not respond to my ticket, but I can not understand, the products before being sent are not tested.

[omshree]

what rom flashing tool did you use?

PICkit3