USB SPI Flasher with PIC18F2455 - now with source and binaries

[maximilian0017]

Device is working like a charm but i've hit a snag.

The new Nandpro20b.rar wont work with the pic board
Error:
Looking for USB interface device
Incompatible ARM hardware for this version of NandPro!
Can not continue

I was trying the new version because of the bugfix mentioned by Tiros.

[raidenxtribe]

Same problem here !
 

[neprikon]

to charge jasper will approach////

[Arakon]

dark the other side is, young padawan.

...what?

[Lstudentik]

for suit jasper

[jskew]

[gupek]

I have build art2003 programmer for pic and flasher...

firstly I program PICFLASH.hex, then connect to pc and no usb device was detected...

secondly I program PicXBoot_12.hex and some usb was detected (pin 17 not connected to ground) and I was install driver and my flasher was visible in PDFSUSB.exe...

so I load the PICFLASH.hex and want to program... then try to connect 17 pin to ground... program was not successful

now when I get back to winpic800 ant try to write anything from < 0x0000 to 0x0400 ) I have error
so there is no possibility to write picxboot_12.hex but picflash.hex i write normaly...

non usb device is detected right now and from 0x0 to 0x0400 there are ony "FFFF"

please help me how to make this flasher work right now?

[Grim187]

@gupek program the picxboot_12.hex via art2003, then wire it up like this and use the included programing software to program picflash.hex (rc6>ground dident make a difference for me and i still cant get it back in to programing mode).

what pic are you using? (model and version (sm, plcc, etc.)

[B1N4RY]

Has anyone considered to use this as a starter kit?

[gupek]

@Grim187: problem is when NOW I try to program (winpic800 (sw), art2003 programer (hw)) picxboot_12.hex i have error at 0x0 ... write 0xef99 read 0xffff

my pic is PIC18F2455-I/SP 0939VA4

i Can program anything from 0x0400, but not from 0x0 to 0x0400 any one know why ?

[cory1492]

gupek: I have this in my picboot config bits

Code:

#pragma config WRTB     = ON       // Boot Block Write Protection 000000-0007FFh
#pragma config WRTC     = OFF  // config register write protect 300000-3000FFh
#pragma config WRTD     = OFF  // EEPROM write protect

and in datasheet...

Memory that is not code-protected can be erased using either a Block Erase, or erased row by row, then written at any specified VDD. If code-protected memory is to be erased, a Block Erase is required. If a Block Erase is to be performed when using Low-Voltage Programming, the device must be supplied with VDD of 4.5V to 5.5V.

Check your VDD if using LVP? Tried full device erase first instead of just write? Why only half of your bootloader seems to have protect issues...? Blocks seem to be 64bytes.

(rc6>ground dident make a difference for me and i still cant get it back in to programing mode).

There is a reason I posted binaries a while back (which I compiled for 2455 with lvp but tested on 4550 with lvp disabled in winpic soft and had no issue); though initially it was a stopgap so others wouldn't have the problems I did with programming mode jumper it looks like DarkstarTM hasn't corrected this.

[gupek]

when i read my pic i have WRTB=ON and all code "0xFFFF"

erasing all didn't help, also when i change WRTB=OFF winpic says ...
" writing addesss 0x30000c
written : 0xe00f read: 0xa00f " -> still protection

lvp is ON in settings

between vdd and vss is 2.75V, so i need to give at vdd 5V and try to erase? -> yes, it works again

program via pdfsusb.exe success

------------------------------------------------------------------------------------

and now I have bad flash config...

Looking for usb interface device
FlashConfig:000000FE
Starting Block:0x000000
Ending   Block:0x0003FF
Press any key to begin reading nand
Error:  40 reading block 0
Error:  40 reading block 1
Error:  40 reading block 2
Error:  40 reading block 3
Error:  40 reading block 4
^C05

why?

every time is different flash config...
00024000
00012000
0077A6E2
0012FE54

I have done last DarkstarTM scheme...

is this something wrong wired?

------------------------------------------------------------------------------------

when I connect usb flasher to pc when xbox has no power plug in then it is visible as memory access...
but when power is plugged in to xbox and I plug usb flasher to pc it is not discovered at all, is it normal?

[cory1492]

Which files did you use for the pdfusb.exe? If you used the latest ones in the OP the pinout changed and doesn't seem to be reflected between the binary and the diagrams for V3, use the bin and pinout I posted a few pages back (pay attention to the pull ups/pull downs, MCLR is especially important) if possible and see if that improves the situation. And no, it's not normal if you can't detect the PIC with either the xbox on or off it's power - it should always detect in windows.

I have gotten occasional block write issues (I assume it is the same issue that caused nandpro2b to be released), but have only had flash config ID issues when one of the nand data lines was poorly connected after having lifting the NAND off the board.

[gupek]

i use binaries v3 for the first page of topic "picflash.hex", I can't see your binaries in topic... only diagrams... can u post link to correct binaries?

[maximilian0017]

And no, it's not normal if you can't detect the PIC with either the xbox on or off it's power - it should always detect in windows.

As posted some time ago i had some trouble with detection, now i connect the usb connection first and then connect the xbox360 powerbrick to the mains, then start the program, this works every time.
(usb powered, original Darkstar firmware with bootjumper problem)

[gupek]

I am doing the same but still have bad flash config and also conecting rc6 to ground make no different...

[cory1492]

Bins linked with the discussion around the problems I had with the posted binaries in post 91 of this thread, tagged to the word 'here'
http://www.xboxhacker.net/index.php?topic=12306.msg83216#msg83216
Provided the mclr pullup is in place and the USB regulator capacitor is a decent value, the only other stability issues with detection should be clock (sometimes needs an added resistor, it's possible also for a resonator to be really inaccurate as well) and having proper noise filter caps close to and across chip VCC/VDD connection.

On my own 4550 (which is also using USB for 5V) I have 2 (+) and 2 (-) points (a pair on each side of the chip), and have put a 104 (100nF) ceramic cap on each pair right next to the chip legs (meaning I have basically equal to 200nF to filter line noise off power) and had a 224 (220nF) germanium cap on Vusb to ground (though apparently the 604 (6nF) ceramic I have on there ATM, mistaken for a 104, works just as well.)

Hopefully the built-from-git bins will help, but in the least will put you on common ground with my own setup so I can try some things to see if I can reproduce your issues.

[gupek]

ok, bins success load, different is that now rc6 works now (xbox can't be not conected to power and then device is visible in PDFSUSB.exe)

mclr pullup (10k->5V), capacitor (Vusb) is 220nF, between Vss i Vdd 100nF...
I had change resonator(24MHz->12MHz) with no success, effect like before... I'm using 2 leg resonator and two 10pF smd capacitors between each leg...

I had 8 pin (Vss) not connected... i thought pin 19 (Vss) will be enough ...

after this I had few correct config and now again wrong

[Grim187]

im gonna build 10 more, i wanted to know is there a "easy pin" to do a activity led?
im looking at the diagram and the only thing i can think of is vdd, usbd +or- or transistor one of the rb's.
i haven't done any voltage testing yet.

[cory1492]

gupek: I can confirm, Vss pins are tied internally to each other. Do you have Vss, 360 ground and usb ground tied to each other (common)?

Grim187: there are plenty of spare pins if you are going to make more with epoxy coating (permanent for usb xbox spi only), you could do activity in the code, an led for 'on', one for spi write activity and one for spi read - should be relatively simple.