Fixing flash in DM365 chip within Ph3 Pro gimbal

I'm on 1.7.60 on my AC, and a firmware from matching time period for RC (can't remember the version, but it is the one released closest to the AC one).
The version I have the most trust in is 1.6; but I upgraded years ago, and it works - so I'm sticking to 1.7.

In regard to range vs firmware version - I once worked on a chart about that, but I'm not exactly sure if I did everything right. I never had time to properly re-check the values:
Please help me understand...
 
Just a heads up.
Fixed the 8 FPS FPV problem due to nand flash replacement on WM320 gimbal top board.
Took Option B - replace encode_usb only: using P3X_FW_V01.11.0030_m0800-encode_usb-mod1.elf on P3P with V1.11.0020 FW installed. Did the job.
Thanks to all responsible for this hack.
 
  • Like
Reactions: quaddamage
Hey mate,
I have tried this fix but it doesn't seem to flash all the way. When I power the board without shorting the bootloader pins I do not get any output through putty.
When I power the board with the bootloader pins shorted then I get BOOTME though.
When I run the flash command it only gets a small distance in before stopping at: Waiting for SENDIMG sequence...
I can restart the flashing command but it does not get any further..

Any help? I believe the NAND is corrupt and needs replacing.
Thanks


"

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


Flashing NAND with UB1.img and UB2.img.


Attempting to connect to device COM4...
Press any key to end this program at any time.


Waiting for the DM36x...
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: DONE
DONE received. Sending the UBL...
100% [ ]
UBL sent....


Target: DONE
DONE received. UBL was accepted.
UBL transmitted successfully.


Waiting for SFT on the DM36x...
Target: Starting UART Boot...
Target: BOOTUBL
BOOTUBL commmand received. Returning CMD and command...
CMD value sent. Waiting for DONE...
Target: DONE
DONE received. Command was accepted.
Sending the UBL image
Waiting for SENDIMG sequence...
Target: BOOTME
Target: BOOTME
Target: BOOTME
"
 

Attachments

  • P3X_FW_LOG_AB_Latest.txt
    14.6 KB · Views: 145
Any help? I believe the NAND is corrupt and needs replacing.

Maybe. But in such cases, there's usually error message.

Flashing NAND with UB1.img and UB2.img.

The extracted files have usually different names, are you sure those files are ok?
You didn't switched them by mistake?

If only the first image is flashed correctly, the bootloader will be able to display something on screen.
So either there's hardware issue, or your first image (the 14kb one) is bad.
Or you're not setting proper options in PuTTY - that's also an option.
 
I am not sure if I really do have the correct files...
Its a bit above my head.
I got all files from the description box of:
 
I have uploaded a video of what I am doing here:

Alternatively I can change some resistors to demote the unit to P3H but I don't know which..
Thanks, Reuben
 
Your images are OK.
Your FTDI setup works - you've confirmed that by seeing BOOTME in PuTTY.

Everything goes smoothly, but then the chip does not react on next command. It returns to BOOTME instead - maybe it reboots?

So:
1. Your files are OK (or at least not the issue)
2. Your method is OK, you're doing everything correctly
3. the 650 mA is quite low - is your power supply capable of giving more? Maybe the chip restarts due to power shortage?
4. The protocol used to program that chip is sometimes affected by the PC used, maybe try another PC?
5. Are your commections good quality? Maybe it gives interference or resistance?

If you won be able to make this work after checking 3,4,5 - then you proably have a hardware issue. It isn't possible to easily say which chip is at fault - DM365, NAND or RAM.
 
My power supply is capable of supplying 3A but it only draws max 700ma.
My cables are questionable - they are pretty long.
I am currently using my own laptop but I tried the same steps two days ago on my work laptop with the same results.
I will try my desktop shortly.

It is not easy to shorten the wires as the adapter is integrated into the USB plug.

I will try a different FTDI unit if I can find one.

Do you think if there is an issue on the board it would be most likely the NAND?
The chip does seem to respond to commands from the TI Flashing program..

How about the resistor modification to essentially turn the unit into a P3A (so I can use it in the mean time until parts arrive)

Thanks
 
Is there a chance your nand flash memory chip is physically damaged due to yaw arm strike because of AC crash ?
Need more info.
 
Last edited:
Is there a chance your nand flash memory chip is physically damaged due to yaw arm strike because of AC crash ? This was the problem with mine.
It could be possible that the chip is physically damaged. I have looked at it closely and doesn't LOOK damaged but I know it could have sustained a hit.
I bought the whole drone on trademe for cheap with this issue. The moisture sensor is red so that could mean something too...
 
My chip was cracked in a half moon shape, but could only see it under the microscope.
Where in NZ are you located ?
 
Last edited:
Do you think if there is an issue on the board it would be most likely the NAND?

I only repaired a few drones myself, but also many people contacted me about these repairs. From HW failures, there was only one confirmed case on issue being somewhere else than NAND (and the issue was that DM365 itself required rebailing).

How about the resistor modification to essentially turn the unit into a P3A (so I can use it in the mean time until parts arrive)

I've seen a post somewhere with a person claiming that it worked. Never did that myself.
Besides resistor, you need to flash modified firmware to tell the camera that it's now P3A.
 
Do you have the listing number ? The 'P' number you gave me is the final sale number or whatever they call it.
The reason I ask is I might be able to glean a bit more history about the craft from the auction Q&A etc.
 
I only repaired a few drones myself, but also many people contacted me about these repairs. From HW failures, there was only one confirmed case on issue being somewhere else than NAND (and the issue was that DM365 itself required rebailing).

I will continue to try and flash uboot today and if still no luck then tomorrow I will attempt NAND reball
 
I cleaned the NAND chip and it is broken.
This seems to be the opposite of what the seller said. He put "Has not been crashed" but I can't imagine the NAND breaking in the same half moon shape without physically getting hit..

Ah well, I can now stop wasting my time and will try to source another chip.
The crack was very small, unable to see until I cleaned the top and used a light. Did not have to use microscope (but I do similar things for a living so I am used to spotting these cracks in tracks etc)

Thanks for the help guys, I will continue to update once I get some NAND
 
I remember this auction, but the bidding went way too high for me.

From Auction Site:
"The videos and photos save to the memory card but it doesn't show images on the Samsung S8+ phone when connected to the controller. The video plays from off the memory card"

So camera is OK, which is a good thing and camera gimbal appears in good shape, so doesn't look like it's been crashed.
 
I remember this auction, but the bidding went way too high for me.

From Auction Site:
"The videos and photos save to the memory card but it doesn't show images on the Samsung S8+ phone when connected to the controller. The video plays from off the memory card"

So camera is OK, which is a good thing and camera gimbal appears in good shape, so doesn't look like it's been crashed.

Yes, I have had the drone flying and they do save to SD

He said that it wasn't crashed but it must have as the NAND is cracked.


There is also a small crack in the end of one of the arms (which points to a crash also)
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,110
Messages
1,467,709
Members
104,999
Latest member
intertronixlabel