Print

[ldotsfan]

I needed an eeprom reader to revive a dead xbox with error 7 but my soldering skills suck.. I put this off for a few months until I realized I could use a breadboard instead. I used an old serial cable with the female DB9 connector on both ends. I stripped off one end and used a multimeter to identify which colored wires correspond to which pins of the DB9 connector as there are no standard color scheme for RS232 wiring. As the colored wires inside the RS232 cable are too thin, I used thin wires to force them inside the breadboard holes. Not too pretty and I redid it a few times and tested with the multimeter to ensure proper connectivity.

The SCL, SDA and GND lines (white colored) were made from an old IDE cable by tearing out the individual threads after stripping out the connector. I connected to pin 12 for GND, 13 for SCL and 14 for SDA of the LPC (Mine is a Xbox 1.1 so the LPC holes are not filled with solder). I also had to play with the PonyProg settings as I was using Windows ME on the old Pentium 2 PC which still had a serial port.

Here's the end result:
(IMG:http://photoserver.ws/files/dtdrw3criv9lg658txbf.jpg)

So if you can't solder and need an eeprom reader, this method is an viable alternative.

This post has been edited by ldotsfan: Mar 26 2008, 04:36 PM

[Rocker1]

Hey there Ldotsfan Ive tried this method but I get the message "bus busy or hardware error (-11)" is something wrong with the board set up?

(IMG:http://i266.photobucket.com/albums/ii248/Metalartin/SL380601.jpg)

(IMG:http://i266.photobucket.com/albums/ii248/Metalartin/SL380600.jpg)

(IMG:http://i266.photobucket.com/albums/ii248/Metalartin/SL380580.jpg)

[Rocker1]

How did you go about using the multimeter to test the connectivity and what are the correct readings meant to be?

[ldotsfan]

See if your multimeter has a continuity testing mode, it will beep if there is. Otherwise use the ohm meter setting. Zero if there is good connection, infinity if the circuit is open.

I found another diagram depicting unicorn's design: http://forums.afterdawn.com/thread_view.cfm/661897#4020259. See if this helps.

And the polarity of the zeners matter. The direction of the dark stripped end of the zener should be facing one way.

This post has been edited by ldotsfan: Oct 19 2008, 03:51 AM

[Rocker1]

Thanks the orange wire in that second pic was in the wrong place just for the pic I did have it next to the red where it should be when connecting.

I think I have them zeners the right way up(dark bit at the bottom right?) and Im going to to try and use the LPC header since I missed that part the first time around... much easieir.

So with the setting at 2000 Ohms from the start pin to the exiting wire should read zero for each wire?

[Rocker1]

I've had no luck seriously worried that 'll break it eve more now so I'll take rest from it and try again tomorrow... I'm less optimistic that this is gonna work now though.

Could someone take a look at my circuit Im not sure what else to try.





Here are is the cable I'm using to the serial port:

[obcd]

Hi,

if I remember right with a breadboard, the top and bottom 2 rows of holes are connected horizontally to each other, and the the middle holes are connected vertically to each other.

With this assumption in mind, it looks like the right end of the resistors is not connected to anything else.
The bottom orange wire probably isn't connected to the blue wire and the zener diode as well.

I could be wrong, and maybe your breadboard does connect those wires and resistors.

I also have the impression that the wires are not soldered to the db9 connector. The way the wires are bend around the connector pins is asking for bad connections and shorts.

I have also written remarks already about this type of eeprom reader. It doesn't work with all serial ports. I have 2 pc's. One works perfectly, and the other one refuses to read a thing with the same eeprom reader circuit and xbox.

regards.

[Rocker1]

Thanks obcd Ive made those adjustments and you are correct they werent connected but it doesnt seem to make a difference, I even soldered the wires onto the the connector.


For a 1.6 board the positions are different but colour code was the same for the LPC pinholes right?

Heres how I connected it:
-------
|0 - A|
|0 - 0|
|0 - 0|
|C - 0|
|0 - G|
|0 - 0|
|0 - 0|
|0 - 0|
-------

A = LPA
C = LPC
G = Ground

So if theres nothing else wrong with the circuit or my choice of connections... a soldered connection always works?

This post has been edited by Rocker1: Oct 19 2008, 09:46 PM

[obcd]

If a soldered connection is done the right way, it should always work.

If your wires are oxidated, the solder will refuse to stick to it. Such a solder joint can give bad contacts.
If the items soldered together were moved during the cooling of the solder, the soldering can be bad as well.

If the soldering looks mechanical solid, it usually is electrical in order. Solderings can become bad after a couple of years if they are mechanically stressed too much. If the soldered component heats up a lot, it's soldering can become bad as well after a couple of years. Those failures can be visual inspected.

I am unsure about the position of the SCL and SDA signals on the lpc bus of a 1.6 xbox.

I do know (from hearing) that a 1.6 xbox switches the power off in case of a frag. If you connect the eeprom reader, the xbox normally frags. You can avoid that by leaving the pc site disconnected until the xbox booted (into an error screen or wathever)

You can find the SCL and SDA signals on several chips of the xbox. The eeprom chip has them, the clock chip has them and the pic microcontroller has them. If you check the pins of those chips and the lpc bus holes, you should measure zero ohms between them. If you have that, you know you are on the right holes.

To know which pins of the chip are SCL and SDA, you can google for the datasheets of that chip. I recently saw a good pin description of the pic microcontroller in the xbox on this forum.

Maybe the LPC holes of a 1.6 mobo are color coded, but thoses of my 1.0 and 1.1 surely aren't.

So, I would suggest you check the holes to make sure they are the SCL and SDA signals.
If you switch the xbox on, with the reader connected, your xbox should frag.
(this is the only case where a frag is a good sign.)
If you disconnect the pc connector before switching things on, make sure you connect a gnd wire between the pc case and the xbox shielding. Leave that wire connected all the time. As the xbox supply cable doesn't have an earth wire, there might be a voltage difference between the xbox shielding and the pc shielding. The wire will ensure they both have the same voltage.

regards.

[ldotsfan]

QUOTE(obcd @ Oct 20 2008, 05:01 PM)

You can find the SCL and SDA signals on several chips of the xbox. The eeprom chip has them, the clock chip has them and the pic microcontroller has them. If you check the pins of those chips and the lpc bus holes, you should measure zero ohms between them. If you have that, you know you are on the right holes.

To know which pins of the chip are SCL and SDA, you can google for the datasheets of that chip. I recently saw a good pin description of the pic microcontroller in the xbox on this forum.

http://forums.xbox-s...&...t&p=3036207
There's for the PIC16 chip but I'm not sure if the 1.6 has the same chip as the older xbox versions. Testing for continuity between the right pins of the the eeprom chip and the pins of the PIC should confirm if the SCL and SDA signals are there.

[Rocker1]

Just a question about the multimeter readings... notice you said they should read zero?

On my multimeter I have a diode setting(which stays at 1 for both blue and red which I thought would mean open circuit) but no continuity setting, so if Im right the Ohms setting at 20k is the most suitable for a 4k7 resistance. I get 4.60 for both the red(SDC) and blue(SDA) wires and zero for the black(ground), also the blue and red wires look to exit before the Zener in the circuit diagram so thats where I put them... after checking it didnt make a difference to the resistance anyway.

I check the LPC header against the eeprom and Im confused about my readings:

I tried a few different setting first at 20k Ohms and the only 2 that came close to a reading of zero were as follows:



-------
|0 - 0|
|5 - 6|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0| <
-------

5(0.02) and 6(0.00) refer to the pins on eeprom I connected them to as noted in the eeprom guide or SDA(top left) and SCL(second down) respectively:
(IMG:http://i65.photobucket.com/albums/h209/1st-unicron/HDD-TUTOR/v1.jpg)

Thats nothing like the diagram says at all for the LPC header connections(going by the above picture second from the left is SDA and the one above SCL by my readings) ... am I looking for the wrong thing here or does this suggest not many people have used the LPC header on a 1.6 board? No wonder I wasnt getting anything if thats correct. I will try again tomorrow thanks for the input guys... so you suggest the ground wire doesnt go in the LPC header but attached between the closest screw and srew pad or something?(I may attach the wire between the PC case and sheilding regardless of when I plug the reader in as an extra safety measure... thanks for the tip)



This post has been edited by Rocker1: Oct 21 2008, 01:04 AM

[obcd]

According to the datasheets:

Pin 5 of 24LC02 = SDA
Pin 6 of 24LC02 = SCL

QUOTE

-------
|0 - 0|
|5 - 6|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0|
|0 - 0| <
-------

On my xbox 1.0 and 1.1 I can find SCL and SDA on those pins of the LPC header as well. I never was sure they were still connected on a 1.6 mobo. You just confirmed this. Thanks.

Yes, take your GND connection from a screw. It's one wire less you have to solder.
The extra Gnd wire between the pc case and the xbox shield is important for the safety of the electronics.

Use your lowest ohm range to check for continuity. You will never read exactly 0 ohms as your probe wires have a resistance as well. Everything below 3 ohms can be considered connected to each other.

http://forums.afterd...view.cfm/357863

The picture on this tutorial shows the exactly same points on the LPC for SCL and SDA. (You are looking from the back to the front which is a bit confusing)

Your picture seems to show the way a lpc bus has to be rewired to install a modchip. (1.6 and 1.6b mobo's)

Where did you get the information of the SCL and SDA pin's being on another location?

regards.

[Rocker1]

If you look on the photo given for a 1.6 board from the original eeprom reader guide(the same one is in my last post under the eeprom chip) its colour coded, the problem could have been that as he said he borrowed the picture that the LPC header was already annotated... maybe he just failed to mention that the LPC part should be ignored? Id assumed that he'd marked them on with the same colour code as the 1.1-1.5 board example.

[obcd]

I agree it is misleading with the colors. They are not the colors for the SCL SDA and GND wire. They just described what connections are needed to rewire the LPC bus for a modchip. Microsoft removed some of the signals on the 1.6 LPC holes to make it more difficult to install a modchip.

So, what you measure are the correct points, the same as with the other xbox models.

Are you succeeding to read the eeprom? You can check the eeprom.bin file with the pc program liveinfo beta 3.

regards.

[Rocker1]

I still cant get this eeprom reader to work (IMG:style_emoticons/default/sad.gif) Ive tried everything almost.... soldered version too.

This post has been edited by Rocker1: Oct 22 2008, 01:46 AM