• Visit https://www.embeddedcomputers.net/ for Hardware; Software and all other things related to FlashcatUSB

Unable to verify forceware firmware with BCUSB

There was no change using 3.3V, I will have to stop for now as I'm really tired 4:23AM in here, so I need some rest, what was interesting was that I was able to flash most of the 8mb dump you gave me with only 3 mismatches almost at the end when it was the very first time I used Memory.EraseBulk, I haven't been able to flash it back without having many mismatches since the beginning when it reaches data at 0x40000, maybe I have to make the pins perfectly contact, the best will be to solder I know, I will have to get more stuff for that but let's see how it goes.

Will stay away for some time so until next time, thanks again for everything!
 
Ok the errors at the very end will mostly be log mismatches , nothing too serious , but it seems we are getting somewhere, i am putting it down to the clip connection. A direct connection will be a better option.
 
I haven't been able to make any change to the flash for whatever reason, so I will have to try a direct connection, I was considering on cutting the cables from the connectors that are inserted on the clip and soldering this same cables to the chip since it pretty much all set up right or it would be better to get all the cables separately?
 

Attachments

  • DSC02957.JPG
    DSC02957.JPG
    368.3 KB · Views: 49
You could cut up the clip cables and use them , or just solder some new wires from the BCUSB to the chip
 
Before soldering this cables directly into the chip I will like to ask you if I identified them correctly since I don't want to solder them on the wrong place. I make sense of the order by knowing in what way the clip identified my modem, my doubts are with Pin 7 and Pin 8 cables and Pin 15 and 16 cables because of their proximity, I'm not sure if that is the order or if I'm mixing them. Or if you know how the clip pins layout is I could understand it much better, because I don't really know how they are ordered in here, I attached a picture based on how I named each cable and how they connect to the clip pins.

I'm only certain of pin 10 cable because it is the one more separated from the rest, this cable has another cable on the same connector, so if the sequence is by 2 (10,12,14,16) it should be pin 16 right.
 

Attachments

  • clip layout1.png
    clip layout1.png
    700 KB · Views: 47
  • clip layout2.png
    clip layout2.png
    2.1 MB · Views: 47
  • pin layout1.png
    pin layout1.png
    2.6 MB · Views: 48
  • pin layout2.png
    pin layout2.png
    2.3 MB · Views: 49
If what I tried to identify is correct, this should be the clip pins sequence, if that is not right please let me know how the order goes, I have a lot of doubts with 1,3,5,7 and 2,4,6,8, maybe they should be inverted because pin 1 should be the one more to the left but on the blackcat cable pin 8 is the one more to the right (as on the diagram pin 8 is the last right leg of chip), so I'm not sure.
 

Attachments

  • clip sequence.png
    clip sequence.png
    692.1 KB · Views: 41
  • pin 8 cable.png
    pin 8 cable.png
    3.1 MB · Views: 45
I think I have got it right finally, I will just wait for your confirmation if you think this is also the correct layout for the cables that came with the clip.
 

Attachments

  • correct layout.png
    correct layout.png
    5.6 MB · Views: 46
  • sequence for clip cable.png
    sequence for clip cable.png
    2.9 MB · Views: 42
It turns out I was right, this sequence I last thought of is correct as I was able to extract the separated connectors that were inside the plastic cover and try them like that before cutting them, the software was able to detect the modem.
 

Attachments

  • confirmed.png
    confirmed.png
    7.2 MB · Views: 45
I was able to solder the cables directly to chip and get the modem detected by the software, but to end up with the same mismatches errors! when flashing :(. All efforts gone for nothing sigh....

As usual I'm only able to flash my backup without mismatches.
 

Attachments

  • pin 7,8.png
    pin 7,8.png
    3 MB · Views: 42
  • pin 10,15,16.png
    pin 10,15,16.png
    3.4 MB · Views: 39
  • top view.jpg
    top view.jpg
    751.5 KB · Views: 38
I was able to discover how to get "second light" (receive led) blinking and also having the link led blinking at the same time if I have ethernet cable connected. I'm also including my boot sequence from the moment I turn on the modem and having the ethernet cable connected.

The exact way to replicate the second light to start blinking is this:
1. Connect power supply to modem.
2. Connect clip to modem (or now in my case solder cables)
3. Connect clip cable (inverted) to bcusb pins (bcusb does not have to be connected to pc)

Being on this stage I have tried everything as well, with same results, can't enter 192 and can't flash it correctly.
 

Attachments

  • afer a while.rar
    1.5 MB · Views: 16
  • after turning it on.rar
    3 MB · Views: 3
  • boot sequence.rar
    6.1 MB · Views: 4
In the after a while clip , it shows the modem as functioning as normal . Also as your pc gets an IP from the device , you should be able to access the 192 page .
do you have access to a clean pc to see if you can access through that one.
it is weird that it is not flashing properly , we have cut out the clip being at fault , now we need to look at other factors .

1. The pc
2. The BCUSB itself
 
I have flashed SB5100 and SB5101 with sb5100mod and haxorware so I know the BCUSB can flash. And I have tried flashing on other pc and tried to access 192. The factor I want to clear now is software based, if I could hex edit the flash "live" that may let me modify it right, like something that let's me select a byte and change it to what I want. I saw this picture on the internet http://img391.imageshack.us/img391/8416/problem.jpg looking for an SPI flash programmer and that is pretty much what I will like to do.

Also looking at the video for the usb jtag nt http://usbjtag.net/tutorial/sb6120/sb6120shell.php I see it has to run an erase command before programming the new firmware, this is something I can't do with the blackcatusb software. The jtag utility 1.3 has this same commands shown in here but that is only for the blackcat cable version not the usb version.
 
With NT it shows it as erasing the flash , with BCUSB it erases but you do not see it. The problem is your device does not seem to erase properly , if you can get your device to fully erase then flashing it will not be an issue.
BCUSB can not live edit unfortunately.

what you can try is use the NT software to extract all the relevant sectors ie. uboot , ubf1 ,ubf2 etc and then try to load them seperately , using the vmg300 script.

Do you have a usbjtagNT ?
 
No i don't but it seems I will have to buy one in the end. I will get the software for the NT and try what you say.
 
NT is by far the best programmer out there , but if your BCUSB is still able to flash other devices you should still be able to program this device.
 
I want to be able to be able to do it with the BCUSB before considering on buying it. Are there more useful commands I could use to force a complete erase or get commands similar to ldram Image0 and then erase image0 like in the old jtag blackcat cable.
 
The only available commands are the ones in the documentation folder. From that we have seen that one of those commands does not work with your BCUSB , funnily enough the command works on my device, so i find it hard to figure out why it did not work for you.
 
I have every .net framework installed for what I can be sure of, is there any other drivers or software that you will recommend me to have installed?

Also can you erase reply #65 and #66 if anyone else come across this, it would be better to not confuse other people since those pictures where wrong.
 
.Net Framework is the main component for the software to work , apart from the drivers in the release pack. There may be be a conflict on your pc regarding SPI flashing depending on what other components/software/drivers you have installed.
I would rather leave the two posts that way people can see if they went wrong . This could turn out to be a very informative thread.
 
lol ok, yes this thread has expanded more than I expected.

I tried the command Memory.EraseSection(0x0,0x7FFFF0), it definately takes longer time but you know what I'm going to say next...it doesn't erase anything, even though it says "successfully erased section of memory device"

Also could you send me another dump but using the command Memory.Backup, I know is the same thing but I want to keep trying different things.
 
Back
Top