lightbridge firmware PROBLEM (and solution) - firmware version not found by dji assistant tool

Joined
Jun 30, 2016
Messages
1,022
Reaction score
385
Location
PL
[ 1.370000] UBI error: ubi_io_read: error -74 while reading 512 bytes from PEB 35:2048, read 512 bytes
[ 1.380000] UBI error: ubi_io_read: error -74 while reading 64 bytes from PEB 36:0, read 64 bytes
[ 1.390000] UBI error: process_eb: bad image sequence number 248044948 in PEB 36, expected 1667215269
[ 1.400000] UBI error: ubi_init: cannot attach mtd2
[ 1.410000] UBI error: ubi_init: UBI error: cannot initialize UBI, error -22
Looks like your root file system is damaged as well. I doubt the one from Ph3 will work for you - someone would have to dump the whole NAND on a working device in order to fix such damage.

What happened to that camera? was it left on direct sunlight for a year?
It would seem the whole NAND got de-programmed.
 
Joined
Sep 12, 2019
Messages
16
Reaction score
1
Age
34
Looks like your root file system is damaged as well. I doubt the one from Ph3 will work for you - someone would have to dump the whole NAND on a working device in order to fix such damage.

What happened to that camera? was it left on direct sunlight for a year?
It would seem the whole NAND got de-programmed.
Dji-Pro # nand bad

Device 0 bad blocks:
07f80000
07fa0000
07fc0000
07fe0000
Dji-Pro #
 
Joined
Mar 30, 2019
Messages
2
Reaction score
0
Location
Texas Hill Country
Nope. Have you read all the threads you found?



No manufacturer is obliged for lifetime warranty, and every device has expected lifespan. Planned obsolescence is an issue with some companies; it is hard to prove though. But companies avoid it nowadays due to serious image losses. I see no evidence of planned obsolescence in Phantoms; though the NAND chips used are definitely not top tier, and someone might have noticed during development that they're a weak spot.



I don't think we have DJI support folks here. Write on DJI forum.



If you have the ability to fix electronics - diagnose and fix it.
If you don't - pay for a fix or buy a new device.

I made a diagnosis diagram and fix instructions for the issue on Ph3; you can make one for stand-alone lightbridge based on it:

Hey Quaddamage--Thanks for your reply, I think... You used almost as many words as I did but didn't address anything helpful to my post, which remains a valid one. I believe that the point of these forums is exchanging information and being helpful and supportive of each other, is it not? If I wanted to be chided by someone I would have posted some random point, without specifics, that makes it obvious I have not idea what I am talking about. I am pretty sure I credentialized myself as written. I have had every Phantom since the Phantom 1 came out and have customized and enhanced everything throuh the P4, Pro+, Inspire 1 and F450 and 550 among numerous fixed wing autopilot driven long range FPV planes, so save me the newbie dribble.

YES, I've read more threads on this than most any challenge I have tried to address in the past. I never implied DJI had any responsibility for lifetime support nor that DJI could not obsolete anything they want at any time. As a matter of fact they made it known at one point that they had the exprerssed right to obsolete the N1 Video Encoder at any point. QUITE SIMPLY, I AM JUST TRYING TO DO IS CONFIRM WHETHER INDEED THEY HAVE EXERCISED THAT RIGHT. And if I wanted to post this on a DJI forum where they won't address these things directly, I would have done that. I think that is the final baseline point you put spin on that I will waste further bytes complaining about. Oh wait--one more.... My question was not about fixing broken hardware not requisite of your convoluted flowchart for turning a P3 into a time machine but thanks for that work of art. I'll blow that up in print and cover my wall with it so that I might be able to read it. You must be a frustrated organic chemist.

Again, I am talking about the DJI N1 Video Encoder being used with a DJI bird other than the Matrice, which it expressly states (now) it is meant to work with. For many years it also worked with the Inspire 1 and others to allow non DJI cameras. I am not using a Phantom 3. I have three N1 video encoders. I have plenty of triage and technical skills and build most of my own components myself. I cannot, however override that apparently DJI has chosen to disable the ability for the N1 video encoder to transmit video on non-Matrice platforms.

So, my simple question remains here. Has DJI disabled the ability for the N1 video encoder to allow transmission of non DJI video over the Inspire 1 lightbridge platform. None of us who have been pursuing this need to chase if our non DJI cameras will no longer work with Inspire 1's lightbridge and that no downgrade of Inspire 1 drone or remote control will get us back to transmitting non-DJI camera video again.

Thanks to all and best regards!
 
Joined
Jun 30, 2016
Messages
1,022
Reaction score
385
Location
PL
@TXHillCountryFPVer, you are right in that this forum is very friendly and supportive. I hope we'll both be able to keep that state in the posts we write.

This thread is on fixing Lightbridge in a specific case of DaVnci video processor bootloader de-programming. That's why I provided fixing chart. For speculating on DJI actions in regard to N1 Video Encoder, maybe a separate thread would help?
 
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Hi guys, Lightbridge 1 Air Unit issue here
when trying to run according to steps on page #1 we get this.. (there is mentioned a line containing Target: CORRUPT - what can this mean??)

Code:
-----------------------------------------------------
  TI Serial Flasher Host Program for DM36x
  (C) 2009, Texas Instruments, Inc.
  Ver. 1.50
-----------------------------------------------------


Flashing NAND with ubl1_editedByGaucho.img and u-boot_modifiedByGaucho.img.


Attempting to connect to device COM2...
Press any key to end this program at any time.


Waiting for the DM36x...
       Target:
       Target:
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:   BEGIN
BEGIN commmand received. Sending CRC table...
100% [                                                              ]
                          CRC table sent....



Waiting for DONE...
       Target: CORRUPT
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME

Waiting for the DM36x...
       Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
       Target:  BOOTME
When trying to run without bootpin it gives us this. Appears like the NAND is empty.. what can we try to do?

Code:
Target: DM36x initialization passed!
      Target: UBL Product Vesion : DJI-ABP-SUPER-UBL-1.0-rc0(2014-08-25)
      Target: Dji UBL Version: 1.51(Aug 26 2014 - 16:00:24)
      Target: Booting Catalog Boot Loader
      Target: BootMode = NAND
      Target: Starting NAND Copy...
      Target: Valid magicnum, 0xA1ACED66, found in block 0x00000019.
      Target: Valid magicnum, 0xA1ACED66, found in block 0x0000001B.
      Target: Valid magicnum, 0xA1ACED66, found in block 0x0000001D.
      Target: Valid magicnum, 0xA1ACED66, found in block 0x0000001F.
      Target: No valid boot image found!
      Target: NAND Boot failed.
      Target: Aborting...
THANKS IN ADVANCE
I have the EXACT same issue, 1 old unit and 1 new unit (4yrs old package still sealed) and both have the same error, I will let you know if and when I figure it out.
 
Joined
Oct 23, 2019
Messages
9
Reaction score
1
Age
49
Hi,

i did everything as described in your post to solve the firmware problem by reprogramming the images.
But it failed.
I tried several times, sometimes the CRC is transmitted, one I saw the ULB was transmitted. But usually it comes to this output:
-----------------------------------------------------
TI Serial Flasher Host Program for DM36x
(C) 2009, Texas Instruments, Inc.
Ver. 1.50
-----------------------------------------------------


Flashing NAND with ubl1_editedByGaucho.img and u-boot_modifiedByGaucho.img.


Attempting to connect to device COM11...
Press any key to end this program at any time.


Waiting for the DM36x...
Target:
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BEGIN
BEGIN commmand received. Sending CRC table...
100% [ ]
CRC table sent....



Waiting for DONE...
Target: CORRUPT
Target: BOOTME

Waiting for the DM36x...
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BOOTME

Waiting for the DM36x...
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BOOTME

I tried with to different FTDI units on Vista64 and XP32 systems. All the same results.
The support told me, they will not reapair or exchange the units due to product's end of support is reached.
I need it very urgent in the next week for a cable camera. Otherwise I cannot do my job.

Many thanks for your investigations and further held in advance.
 
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Hi,

i did everything as described in your post to solve the firmware problem by reprogramming the images.
But it failed.
I tried several times, sometimes the CRC is transmitted, one I saw the ULB was transmitted. But usually it comes to this output:
-----------------------------------------------------
TI Serial Flasher Host Program for DM36x
(C) 2009, Texas Instruments, Inc.
Ver. 1.50
-----------------------------------------------------
Hi, that's so interesting, you have almost the exact identical situation I do, you weren't the one who bought those off of eBay by any chance are you?
Even when you buy those units off of eBay brand new, sometimes those are worse because they've been sitting for 5 years.
I have literally the exact same issue as you when it comes to trying to flash the board, my log is literally identical.
I'm going to investigate into the situation further as I also need this light bridge for a remote camera, I believe the chip set may need to be swapped and then reflashed.

Flashing NAND with ubl1_editedByGaucho.img and u-boot_modifiedByGaucho.img.


Attempting to connect to device COM11...
Press any key to end this program at any time.


Waiting for the DM36x...
Target:
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BEGIN
BEGIN commmand received. Sending CRC table...
100% [ ]
CRC table sent....



Waiting for DONE...
Target: CORRUPT
Target: BOOTME

Waiting for the DM36x...
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BOOTME

Waiting for the DM36x...
Target: BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
Target: BOOTME

I tried with to different FTDI units on Vista64 and XP32 systems. All the same results.
The support told me, they will not reapair or exchange the units due to product's end of support is reached.
I need it very urgent in the next week for a cable camera. Otherwise I cannot do my job.

Many thanks for your investigations and further held in advance.
 
Joined
Jun 30, 2016
Messages
1,022
Reaction score
385
Location
PL
I tried several times, sometimes the CRC is transmitted, one I saw the ULB was transmitted. But usually it comes to this output:

Code:
Waiting for the DM36x...
        Target:
        Target:  BOOTME
BOOTME commmand received. Returning ACK and header...
ACK command sent. Waiting for BEGIN command...
        Target:   BEGIN
BEGIN commmand received. Sending CRC table...
100% [                                                              ]
                           CRC table sent....

Waiting for DONE...
        Target: CORRUPT
        Target:  BOOTME
Either your device does not get enough power, or the cable is too low quality (or just too long).
 
Joined
Oct 23, 2019
Messages
9
Reaction score
1
Age
49
Either your device does not get enough power, or the cable is too low quality (or just too long).
I use a lab power supply with thick cables, the air unit pulls just 250 mA. Or do you mean the TX/RX cables? I was investigating this and tried shorter cables, other GND connection and more.
Investigating the RX TX signals, I found something strange...
DS1Z_QuickPrint1.png

The blue curve (CH2) is what the air unit receives, just after powering up. The big block is generated when receiving the CRC data. But the last 2/3 look strange due to the levels, which are not 0V /3.3V, but low is about 2.2V. Actually I cannot explain. It looks like conflicting outputs on this line. E.g. air unit pulls this line up, FTDI cannot pull down to 0. May somebody can confirm. Or, even better, confirm valid levels when programming succeeds.
 
Joined
Jun 30, 2016
Messages
1,022
Reaction score
385
Location
PL
I use a lab power supply with thick cables, the air unit pulls just 250 mA. Or do you mean the TX/RX cables?
I meant both:

- Power cables (and power source) for amperage (usually people supply voltage from USB). Though if it takes only 250mA, even USB shouldn't be a problem. I am more familiar with Ph3, where power consumption is much higher due to Ambarella sitting on the same line.

- And TX/RX cables can cause transmission issues as well - sometimes people use meter long cables or put them near sources of interference.

But considering your expertise, you've probably already checked that. If the voltage is fluctuating from expected levels, you either have short to ground, or resistance in series.
Or the DM36x chip is a little fried.

I did not heard about the UART line being used for anything internally within any of the DJI boards; the proper DM36x lines are just outputed directly to the RX/TX pads without connecting to anything else, AFAIK.
 
Joined
Oct 23, 2019
Messages
9
Reaction score
1
Age
49
I meant both:

- Power cables (and power source) for amperage (usually people supply voltage from USB). Though if it takes only 250mA, even USB shouldn't be a problem. I am more familiar with Ph3, where power consumption is much higher due to Ambarella sitting on the same line.

- And TX/RX cables can cause transmission issues as well - sometimes people use meter long cables or put them near sources of interference.

But considering your expertise, you've probably already checked that. If the voltage is fluctuating from expected levels, you either have short to ground, or resistance in series.
Or the DM36x chip is a little fried.

I did not heard about the UART line being used for anything internally within any of the DJI boards; the proper DM36x lines are just outputed directly to the RX/TX pads without connecting to anything else, AFAIK.
Thanks for your reply, I am still investigating, it seems there is a solution....
 
Joined
Sep 14, 2017
Messages
9
Reaction score
1
Age
29
Hello

Thank you Gaucho for your work. I allready had to reflash many Phantom 2 Vision+ with no image. It worjed always without any problems.

Actually I need to get my RE700 working again. Until now 3 pieces are bricked thanks to the app update.
Is there a way to flash the wm301_repeater_2g_v1 module from the RE700?
I got one working RE700 and 3 bricked RE700.
I think the ddwrt software wasn't flashed correctly during update and now it won't even boot.

I also use a tplink cpe210 with 24v powermodule and 18650 batteries as alternative. But I can't give this solutions to all my friends ;-) there should be an easier way like the RE700 working again.

Thank you in advance

Fabio
 
Joined
Oct 23, 2019
Messages
9
Reaction score
1
Age
49
Hi guys,

I found a solution (partly by accident) when I was searching a reset pin for the board. Just to eleminate the chance of transient power issues short after powering up the unit.

To solve the "CORRUPT" message described above, I pulled the marked pin to GND by a 8.5k resistor. No warranty that this will work and not damaga your board. But pulling pins to GND via a resistor like this has only low risks.

After that continue with the procedure of goucho. In my case the board completes flashing the images with no issues. My air unit works again as a new one.
Please let me know if somebody can confirm this procedure.
Good luck!

btw: DJI support told me that the unit will not be reapaired or exchanged anymore. Thanks DJI!! Never will buy a faulty expansive product with this kind of support!!

20191024_121159.jpg
 
  • Like
Reactions: quaddamage
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Hi guys,

I found a solution (partly by accident) when I was searching a reset pin for the board. Just to eleminate the chance of transient power issues short after powering up the unit.

To solve the "CORRUPT" message described above, I pulled the marked pin to GND by a 8.5k resistor. No warranty that this will work and not damaga your board. But pulling pins to GND via a resistor like this has only low risks.

After that continue with the procedure of goucho. In my case the board completes flashing the images with no issues. My air unit works again as a new one.
Please let me know if somebody can confirm this procedure.
Good luck!

btw: DJI support told me that the unit will not be reapaired or exchanged anymore. Thanks DJI!! Never will buy a faulty expansive product with this kind of support!!

View attachment 115224
Hi LB victim, I will try this on one of my boards today and let you know what happens, Thanks!
 
Last edited:
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Did you try my solution?
Yes I did, thanks!, I thought I posted it here.
It took a while to flash but eventually it went through, I did have to use 16k ohms instead, 8.5k just didn't work for me for some reason.
However now I can't get my second ground unit to flash...lol. I'm thinking about trying your solution on the ground unit as well but I just have to figure out which pads to connect to.
 
Joined
Oct 23, 2019
Messages
9
Reaction score
1
Age
49
Yes I did, thanks!, I thought I posted it here.
It took a while to flash but eventually it went through, I did have to use 16k ohms instead, 8.5k just didn't work for me for some reason.
However now I can't get my second ground unit to flash...lol. I'm thinking about trying your solution on the ground unit as well but I just have to figure out which pads to connect to.
Ok, fine.

I started the programming as Goucho told first. Then I found, the pulling this pin to GND the unit restarts, so I thought it is the reset. Letting the pin float brings the same CORRUPT result as described. Holding it down enabled the unit to be programmed, just with the running instance of the firmware tool.

Good luck with your other unit.
 
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Ok, fine.

I started the programming as Goucho told first. Then I found, the pulling this pin to GND the unit restarts, so I thought it is the reset. Letting the pin float brings the same CORRUPT result as described. Holding it down enabled the unit to be programmed, just with the running instance of the firmware tool.

Good luck with your other unit.
Thanks!
Yeah I'm not really sure about the architecture on how these boards are set up, but I would imagine when they are programmed on the factory line they are pulling a couple other pins to ground as well to clean up the signal, just a guess but that's what I'm thinking.
I have also seen some other post where people use smaller shorter wires and it eliminates the air but who knows.
I was going to do a video and how to go through the whole process but the programming is way too inconsistent.
 
Joined
Oct 14, 2019
Messages
6
Reaction score
0
Age
112
Ok, fine.

I started the programming as Goucho told first. Then I found, the pulling this pin to GND the unit restarts, so I thought it is the reset. Letting the pin float brings the same CORRUPT result as described. Holding it down enabled the unit to be programmed, just with the running instance of the firmware tool.

Good luck with your other unit.
How did you know which pin to to pull to ground?. I also found a reset pin on the ground unit.
 

New Threads

Forum statistics

Threads
137,417
Messages
1,420,338
Members
99,799
Latest member
cwarren2485