Firefly Open Source Community

Title: My Firefly RK3288 is dead. [Print This Page]

Author: adamo.a60 Time: 3/3/2015 22:55
Title: My Firefly RK3288 is dead.
My Firefly ie dead. I followed the "wiki.t-firefly.com/index.php/Firefly-RK3288/Build_Ubuntu_rfs/en" procedure and loaded misc.img and running miniroot successfully. The procedure failed because not enough space in /dev/mmcblk0p1. I tried with an SD card mounted on /dev/mmcblk1p1, but this was unsuccesful too. Now I want to try the recovery procedure, but this is not possible. After various attempt to get the card working, la last situation is that it doesn't respond to any solicitation. The OTG-USB cable is connected and the program is running. I tried both Windows and Linux programs and pcs, but it seems that the otg-usb it is not recognized. I have a second card, but I'm afraid to make any attempt to change somethingand put even that completely blocked.
What can I do to fix this? I spent a lot of time looking at forums in search of a suggestion, but I didn't find a solution.

Author: Jabader Time: 3/4/2015 01:48
Title: Re: My Firefly RK3288 is dead.
May be you can try write images of android/ubuntu to emmc by maskmod?

Author: adamo.a60 Time: 3/4/2015 17:03
Title: Re: My Firefly RK3288 is dead.
I don't know how to enter the maskmode on Firefly and wich utility I should use.

Author: zhansb Time: 3/4/2015 17:52
Title: Re: My Firefly RK3288 is dead.
enter the maskmode, refer to http://wiki.t-firefly.com/index.php/Fir ... update.img

"Erase Flash"

Author: adamo.a60 Time: 3/4/2015 22:58
Title: Re: My Firefly RK3288 is dead.
Sorry, but the problem is: how to enter the maskmode...

Author: busybee Time: 3/5/2015 15:01
Title: Re: My Firefly RK3288 is dead.
Use usb cable to connect the board with an Ubuntu PC.
Hold the RECOVERY button, then power on the the board.
On the PC, run:
sudo lsusb

Please post the result.

Author: busybee Time: 3/5/2015 15:26
Title: Re: My Firefly RK3288 is dead.
And I strongly recommend that you try the sd card boot.

Please follow:

https://plus.google.com/+IanMORRISON/posts/22Vxc6Sr5Ei



Download create-sdcard.7z, then extract it with:
# 7z x create-sdcard.7z

Download the Linux rootfs image
# cat downloadable-rfs-examples
Ubuntu 12.04 https://drive.google.com/file/d/0B99O3A ... Hg3ZWE3LXc
Xubuntu/Lubuntu 14.10 https://drive.google.com/file/d/0B99O3A ... lpTaEdfTXc

You can use either one rfs.
Move the download rootfs image to the working directory and rename it to linux-rfs.img

If you're using 64bit Ubuntu, install the 32 bit libc:
# apt-get install libc6:i386

Then run the script:
# ./create-linux-sdcard

Follow the prompt. The script will repartition your sd card and flash all the images to support sd card boot.

All commands above are run with root user.

Author: busybee Time: 3/5/2015 17:01
Title: Re: My Firefly RK3288 is dead.
The last resort is to force the board into maskrom mode.

Please have a careful look at the bottom of the board.

I've put the images here. You need to short curcuit T13 and C155. Detail steps:

   1. Power off the board.
   2. Plug out SD card if has one.
   3. Connect USB cable between board and PC.
   4. Short curcuit T13 and C155, and keep it.
   5. Power on the board.
   6. Wait some time and release the short curcuit.

The device should enter into maskrom mode.
(NOTE: Short curcuit here means connect the two points using wire or metal nipper.)

Author: solid Time: 3/5/2015 18:55
Title: Re: My Firefly RK3288 is dead.
busybee or adam can you verify the chip being shorted? busybee is he reseting the buck?

also adam, was the board soc pushed beyond 85% load? and what is your power source?

Author: ganggangstyle-e Time: 3/6/2015 15:23
Title: Re: My Firefly RK3288 is dead.
for my experience.
1.in the mask mode is press reset button and quick press recovery button and release rest button and hold the recovery button
2.you can see some debug info in serialport .
3.when your last system is in sdcard , please take out the sdcard .then do 1 , or you'll not in maskmode.

Author: adamo.a60 Time: 3/15/2015 00:11

ganggangstyle-e Posted at 3/6/2015 15:23
for my experience.
1.in the mask mode is press reset button and quick press recovery button and ...

Dear all,
Many thanks for your suggestions, and excuse me for my delay answering. Me and a friend of mine we have been engaged to get operating the dac hardware that is the basis of our project. We worked on the kernel 3:19 and we had to correct many settings, incorrect or inappropriate, that we found inside the Firefly basic kernel 3:10. We added also the driver for our hardware.
Now everything works fine, but booting on an SD card generated with a perfectly working Firefly card. The entering the "mask-rom mode", even after the execution of the procedure suggested by busybee in its post dated 3/5/2015 17:01:50, is no possible. We, still now, cannot write into the eMMC, nor into the SD card. It seems the otg is out of order, but I have not tools for testing it.
Any suggestion?

Author: adamo.angelo Time: 3/20/2015 23:29

busybee Posted at 3/5/2015 17:01
The last resort is to force the board into maskrom mode.

Please have a careful look at the bottom ...

Hi busybee.

I tried several times the procedure you describe here, but nothing happens; no output on RS232, no devices recognized by the 'rkflaskit' utility. Maybe, the program that has to be running while the card is in maskrom mode is some other that I don't know.
Can you help me going out from this situation?
Thank you
Angelo Adamo

Author: busybee Time: 3/21/2015 11:36
Last edited by busybee In 3/21/2015 11:47 Editor

adamo.angelo Posted at 3/20/2015 23:29
Hi busybee.

I tried several times the procedure you describe here, but nothing happens; no outp ...

There's no serial output when entering maskrom.
Also, rkflashkit might not support maskrom mode.

Please run:
sudo lsusb
to see if the device will be correctly detected first.

Then use rk's upgrade_tool in Linux or AndroidTool in Windows to proceed.

If the USB OTG port is broken, then the maskrom method does not work.
You've to resort to SD card boot. Does the SD card boot also fail?

If SD card boot also fails, check the hardware voltage described at the post:
http://developer.t-firefly.com/e ... =224&extra=page%3D1

Author: adamo.angelo Time: 3/23/2015 16:58
Last edited by adamo.angelo In 3/23/2015 17:04 Editor

The problem is this: the OTG is not recognized: lsusb with cable connected and with no cable connected is the same.
upgrade_tool says:
List of rockusb connected
No found rockusb,Rescan press <R>,Quit press <Q>:

The strange thing is that I flashed the Ian Morrison ubuntu 14.04 image into an sd-card and this works perfectly. Using dd command I made an image of the eMMC of a good Firefly board and, through the sd-card, I copied back to the eMMC of the broken board. And this work perfectly!
The only thing I cannot do is enter the maskrom mode, and this is necessary if you want to play with firmware and kernels.
It seems that the OTG is broken: is there any tool to test it?

Really, I need to enter the loadmode to be able to flash firmware and kernels, but this is not possible, due to something missing in the low level firmware of the board or because the OTG port is broken.

Author: busybee Time: 3/23/2015 17:10

adamo.angelo Posted at 3/23/2015 16:58
The problem is this: the OTG is not recognized: lsusb with cable connected and with no cable connect ...

Obviously, the OTG is broken.
Please check the connectivity using a multimeter.
It's very likely the OTG dock isn't sold well, or not well connected with the usb cable.

It's OK to flash the kernel partition by the Linux dd command. But this is not convenient.

Author: adamo.angelo Time: 3/23/2015 19:16
Last edited by adamo.angelo In 3/23/2015 19:19 Editor

Dear busybee,

If OTG cable is in, the voltages I read on the board at OTG pins are:
1.    +5.09Vcc
2.    +0.02Vdc
3.    +0.30Vdc
4.    +2.27Vdc
5.    0Vdc

It seems ok, from the hw point of view! Or not?
What can be the problem?

Author: busybee Time: 3/26/2015 11:53
Last edited by busybee In 3/26/2015 11:54 Editor

Since the system is recovered, we suggest you do some test to the USB OTG in Android system to see if the OTG is really broken.

You can plug a usb mouse/keyboard/harddisk to the otg port via an otg adapter.

1. If the mouse/keyboard/harddisk works, then otg function is considered good.
You can check the RECOVERY and RESET button.

2. If the devices do not work, check the voltage drop of the OTG DM to GND, and OTG DP to GND.
(Use the diode function check of the multimeter). Post the result here.

Author: adamo.angelo Time: 3/27/2015 01:56
OTG doesn't work. I connect a pendrive through an adapter cable, but it is not recognized, while in the perfect working card it is promptly mounted as /dev/sda1.

The voltages I read on the OTG pins  are those I wrote in the previous message:

1.    +5.09Vcc
2.    +0.02Vdc
3.    +0.30Vdc
4.    +2.27Vdc
5.    0Vdc

I have to investigate the reason why the OTG port doesn't work...

Author: busybee Time: 3/27/2015 09:16

adamo.angelo Posted at 3/27/2015 01:56
OTG doesn't work. I connect a pendrive through an adapter cable, but it is not recognized, while in ...

Not voltage, but voltage drop.

2. If the devices do not work, check the voltage drop of the OTG DM to GND, and OTG DP to GND.
(Use the diode function check of the multimeter). Post the result here.

Author: adamo.angelo Time: 3/27/2015 22:46
Last edited by adamo.angelo In 3/27/2015 23:18 Editor

busybee Posted at 3/27/2015 09:16
Not voltage, but voltage drop.

2. If the devices do not work, check the voltage drop of the OTG ...

I think I understood what the problem the otg port has: pin OTG_DP is always at 0V.
If I measure the resistance DP to GND and DM to GND while the power is off, I see 2,5Ohm for the first and hundreds of kilohms for the second. Since I followed on the schematic drawing the OTG_DP wire from pin3 of the OTG plug to CPU pin and no circuit is implied in between other than the CPU pin B26, I guess, either a short circuit on the board, or the CPU pin B26-OTG_DP is broken, showing a low resistance value to GND. If the second is the case, I ask Firefly why, although the schematic drawing does include several protection circuits against electrostatic currents, they haven't been mounted on the PCB: probably, a voltage spike would have been absorbed by those circuits and would not have damaged the sensitive pins of the CPU.

Any suggestions other than send back the card to Firefly for repair?

Welcome Firefly Open Source Community (https://bbs.t-firefly.com/)