Jump to content

My v1.0 original xbox doesn't turn on


Recommended Posts

Hi!
Couple of days ago I decided to try repair my xbox which has been sitting on my shelf for at least 15 years after dying suddenly. I opened up a thread on reddit here, and chatted with KaosEngineeer over there.

 

To sum up the current situation:

-The xbox is unresponsive when pressing eject and power button, no sign of life. Back in the days it was repaired under warranty (substitution of the PSU) and then repaired and modded with Aladdin and Evo X.

-Opening it up I noticed that the clock capacitor had leaked, but seemingly not caused any corrosion. I removed the cap and cleaned up with isopropyl  alcohol.

- I tried checking the PSU and it seems to work, giving out 3.3 Volts on the standby contact (brown in my case). I tried short circuiting the standby with the PowON (white) contact by using a piece of wire metal, and I verified that the lines (orange, red and yellow) are more or less in line with the pinout (instead of being 5 Volts there are 5.27 Volts and instead of 12 Volts there are 10.73 Volts).

psu10-11.gif.8868040ea3a6dfb2d0ca0fd5b126b47f.gif

- I checked both visually and with a multimeter that the 4 traces under the motherboard are uninterrupted. In particular, looking at this image:

pe_trace_10-11.jpg.d0b14ac2de40db2a2107e471d1f1851b.jpg

 

There is continuity between the cyan and violet pins on the PIC chip and the cyan and violet contact in the bottom of the image. The 4 resistances in the same image seem to work fine, with a value of about 10 kOhm each.

-I checked that the front panel eject and power button work by verifying that, when pressed, they connect to ground the red and blue traces next to the 10kOhm signal resistances.

 

-Now, the stangest thing: if  I give power, the brown contact is at 3.3 volts (verified while the psu is attached to the motherboard by inserting the multimeter probe in the top space between the plastic plug and the brown wire), but for some reason the PIC pins that should have the same voltage when the console is plugged (before and after pressing the eject and power button) remain at 0 Volt no matter if I press the front panel buttons or not and no matter if I choose a different point for ground ( I tried using all the black cables and the RF cage).

So, basically I'm stuck. Any ideas?

Thank you for help

Edited by Zer0_1Sum
Link to post
Share on other sites
11 hours ago, Zer0_1Sum said:

I tried checking the PSU and it seems to work, giving out 3.3 Volts on the standby contact (brown in my case). I tried short circuiting the standby with the PowON (white) contact by using a piece of wire metal, and I verified that the lines (orange, red and yellow) are more or less in line with the pinout (instead of being 5 Volts there are 5.27 Volts and instead of 12 Volts there are 10.73 Volts).

Those voltages are fine.

 

What voltage are you getting on the blue/purple pins on the PIC?

 

The purple and blue pins marked on the PIC should be approx 3.3 volt unless you press the power or eject buttons..

 

 

Cheers

SS Dave


Soft modding is like masturbating, It gets the job done but it's nothing like the real thing.

Link to post
Share on other sites
Quote

What voltage are you getting on the blue/purple pins on the PIC?

 

The purple and blue pins marked on the PIC should be approx 3.3 volt unless you press the power or eject buttons..

 

I measured it again just now out of caution. It is 0 volts, regardless if I'm pushing the power or eject button or not.

It is as if, even though the 3.3 volts are available, they don't reach the PIC pins

Link to post
Share on other sites
9 hours ago, Zer0_1Sum said:

 

I measured it again just now out of caution. It is 0 volts, regardless if I'm pushing the power or eject button or not.

It is as if, even though the 3.3 volts are available, they don't reach the PIC pins

Can you test to power and eject pins in the 10 pin socket

pe_trace_10-11.jpg.d0b14ac2de40db2a2107e471d1f1851b.jpg.0f2428787fd22e10b449ab03b329f34a.jpg

Both should be approx 3.3 volt

 

Cheers

SS Dave


Soft modding is like masturbating, It gets the job done but it's nothing like the real thing.

Link to post
Share on other sites
32 minutes ago, SS_Dave said:

Can you test to power and eject pins in the 10 pin socket

pe_trace_10-11.jpg.d0b14ac2de40db2a2107e471d1f1851b.jpg.0f2428787fd22e10b449ab03b329f34a.jpg

Both should be approx 3.3 volt

 

Cheers

SS Dave


Soft modding is like masturbating, It gets the job done but it's nothing like the real thing.

 

If I put the black probe of the multimeter on any ground contact and the red probe on those pins it stays at 0 volt.
However, if I put the red probe on the brown standby voltage cable an use the black probe on those 2 contact, I still get 0 volts, but the two contacts above them (which are indicated as ground in the schemes I found) display 3.3 volts.

Am I measuring the tension wrong? I redid the measure you suggested before using this new probes setup, red probe on brown cable and black proble on cyan and purple pins, and this time I get 0 volts without pressing the buttons and 3.3 volts while pressing the buttons.

Link to post
Share on other sites

Quick update:

-After resoldering the bottom of the brown pin I got 3.3 volts wherever they should have been. Xbox still wouldn't turn on.

-I shorted the brown standby and white PowON contacts on on the PSU connector (while plugged into the mobo obviously), and now the xbox turns on as soon as I plug it, but it flashes between red and half green - half orange, with no video output.

-Tried reconnecting the HDD and the DVD reader and plugging the AV into a TV, no change.

Link to post
Share on other sites
  • 2 weeks later...
On 12/9/2020 at 10:02 AM, Crossfader said:

Did you check the Caps in front of the CPU? I have a xbox with no sign of life, after change the caps it works like a charm.

It is not expansive and a chance.

Visually, the caps seem to be ok. The only one leaking was the clock capacitor, and I removed it, but it didn't leave any significant damage to the board or the traces around it.

I also checked those capacitors with multimeter, their resistance increases with time as it should if they are working. Will check them again.

 

Edited by Zer0_1Sum
Link to post
Share on other sites
On 12/15/2020 at 7:33 PM, Zer0_1Sum said:

Is it possible for the 5 capacitors in front of the CPU to be busted even though they visually look fine? What is their role by the way?


Unfortunately I don't have the proper equipment to test them while on board, the multimeter reads tell me nothing

Yes this is definitely possible. I have replaced caps with only a small curved on the top.

  • Thanks 1
Link to post
Share on other sites
On 12/16/2020 at 5:33 AM, Zer0_1Sum said:

Is it possible for the 5 capacitors in front of the CPU to be busted even though they visually look fine? What is their role by the way?

I recently replaced those 5 caps on v1.6, one cap looked fine but had an overly high incorrect capacitance reading., so yes.

As for their role:

"The 3 caps next to the CPU are smoothing the 1.8 volt rail and the 1 near the power connector is for the 5 volt supply." source

 

Link to post
Share on other sites
On 12/15/2020 at 12:33 PM, Zer0_1Sum said:

Is it possible for the 5 capacitors in front of the CPU to be busted even though they visually look fine? What is their role by the way?


Unfortunately I don't have the proper equipment to test them while on board, the multimeter reads tell me nothing

the caps can look absolutely perfect and still be bad.

Link to post
Share on other sites
On 11/23/2020 at 6:14 PM, Zer0_1Sum said:

Hi!
Couple of days ago I decided to try repair my xbox which has been sitting on my shelf for at least 15 years after dying suddenly. I opened up a thread on reddit here, and chatted with KaosEngineeer over there.

 

To sum up the current situation:

-The xbox is unresponsive when pressing eject and power button, no sign of life. Back in the days it was repaired under warranty (substitution of the PSU) and then repaired and modded with Aladdin and Evo X.

-Opening it up I noticed that the clock capacitor had leaked, but seemingly not caused any corrosion. I removed the cap and cleaned up with isopropyl  alcohol.

- I tried checking the PSU and it seems to work, giving out 3.3 Volts on the standby contact (brown in my case). I tried short circuiting the standby with the PowON (white) contact by using a piece of wire metal, and I verified that the lines (orange, red and yellow) are more or less in line with the pinout (instead of being 5 Volts there are 5.27 Volts and instead of 12 Volts there are 10.73 Volts).

psu10-11.gif.8868040ea3a6dfb2d0ca0fd5b126b47f.gif

- I checked both visually and with a multimeter that the 4 traces under the motherboard are uninterrupted. In particular, looking at this image:

pe_trace_10-11.jpg.d0b14ac2de40db2a2107e471d1f1851b.jpg

 

There is continuity between the cyan and violet pins on the PIC chip and the cyan and violet contact in the bottom of the image. The 4 resistances in the same image seem to work fine, with a value of about 10 kOhm each.

-I checked that the front panel eject and power button work by verifying that, when pressed, they connect to ground the red and blue traces next to the 10kOhm signal resistances.

 

-Now, the stangest thing: if  I give power, the brown contact is at 3.3 volts (verified while the psu is attached to the motherboard by inserting the multimeter probe in the top space between the plastic plug and the brown wire), but for some reason the PIC pins that should have the same voltage when the console is plugged (before and after pressing the eject and power button) remain at 0 Volt no matter if I press the front panel buttons or not and no matter if I choose a different point for ground ( I tried using all the black cables and the RF cage).

So, basically I'm stuck. Any ideas?

Thank you for help

try changing the caps in font of the heatsinks. they can look perfect and still be bad. if that doesn't fix it, try finding a new pus for it.

Link to post
Share on other sites

Thank you all guys!

The replacement psu I ordered few weeks ago just arrived, and it seems to work correctly, but using it didn't change anything in xbox behaviour.

It is still dead, unless I short PowOn and Standby, and in that case it flashes with irregular timing between red and half orange / half yellow-ish. I think I can completely exclude the problem is caused by the power supply.

I'll try changing those capacitors. Are these good replacements?

https://eu.mouser.com/ProductDetail/Nichicon/PLG0J152MDO1?qs=%2BxyxD9GtKu0GdHKI1cnZ8g==

Link to post
Share on other sites
  • 2 weeks later...

Hi guys, happy new year!

The capacitors haven't arrived yet, but in the meantime I started desoldering the old KZE capacitors. So far I desoldered two. They both have a capacity of 1650 uF, which is within the 20% margin from 1500 uF.

Their ESR is around 0.23-0.24 Ohms. Is this value good or bad?

Link to post
Share on other sites

I tested some Wurth brand 1500uf 6.3 volt and some Nichion 1500uf caps

Nichion 10 volt  read 1710uf ESR 0.01 ohm

Nichion 6.3 volt read 1521uf ESR 0.01 ohm

Wurth read 1444uf ESR 0.00 ohm

 

2 hours ago, Zer0_1Sum said:

Their ESR is around 0.23-0.24 Ohms

But for a original cap I would say it's on the high side but for a 18 year old cap it's probably ok.

 

Have you heard the saying "Flogging a dead horse" I think this saying applies to this Xbox.

 

Cheers

SS Dave


Soft modding is like masturbating, It gets the job done but it's nothing like the real thing.

Link to post
Share on other sites
1 hour ago, SS_Dave said:

Nichion 6.3 volt read 1521uf ESR 0.01 ohm

That's over 20 times the value I detect with my original old capacitors.

Why do you say such value is ok for a 18 years old capacitor?
I've tried now searching some datasheet (these caps are Chemi-Con KZE series) and I found this:

http://www.chemi-con.co.jp/cgi-bin/CAT_DB/SEARCH/cat_db_al.cgi?e=e&j=p&pdfname=kze

There is no indication for the ESR, but there is the maximum impedance at 100 kHz. I've seen it treated at the same as the ESR (for example by EEVblog), and for a 6.3 volts 1500 uF capacitor it should be 0.023 Ohms.

Link to post
Share on other sites
3 hours ago, Zer0_1Sum said:

Hi guys, happy new year!

The capacitors haven't arrived yet, but in the meantime I started desoldering the old KZE capacitors. So far I desoldered two. They both have a capacity of 1650 uF, which is within the 20% margin from 1500 uF.

Their ESR is around 0.23-0.24 Ohms. Is this value good or bad?

The ESR of KZE caps should be 0.023 Ohms, not 0.23.

https://www.chemi-con.com/upload/files/7/5/32389236352d6c56e8f45b.pdf

Link to post
Share on other sites
14 minutes ago, KaosEngineer said:

The ESR of KZE caps should be 0.023 Ohms, not 0.23.

https://www.chemi-con.com/upload/files/7/5/32389236352d6c56e8f45b.pdf

Hi Kaos.

Ok, so these two capacitors are definetely busted, correct? Tomorrow I'm going to desolder the remaining three, and hopefully the next week the replacements will arrive (by the way, I bought these https://eu.mouser.com/ProductDetail/Nichicon/PLG0J152MDO1?qs=%2BxyxD9GtKu0GdHKI1cnZ8g== , hopefully they are a good enough choice).

Link to post
Share on other sites
13 hours ago, Zer0_1Sum said:

Hi Kaos.

Ok, so these two capacitors are definetely busted, correct? Tomorrow I'm going to desolder the remaining three, and hopefully the next week the replacements will arrive (by the way, I bought these https://eu.mouser.com/ProductDetail/Nichicon/PLG0J152MDO1?qs=%2BxyxD9GtKu0GdHKI1cnZ8g== , hopefully they are a good enough choice).

They should work well.  They have a much lower ESR than the original KZE series caps (0.009 vs 0.023 Ohms) and a much higher maximum ripple current (6300 vs 1820 milliamps).  

Edit: I read the values from the wrong line of the datasheet for the PLG series caps. :(

Link to post
Share on other sites
18 hours ago, Zer0_1Sum said:

That's over 20 times the value I detect with my original old capacitors.

Why do you say such value is ok for a 18 years old capacitor?

Opps for some reason I read 0.024.

 

Cheers

SS Dave


Soft modding is like masturbating, It gets the job done but it's nothing like the real thing.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Similar Content

    • By geohunk
      I messed up. Was fiddling around with configmagic trying to lock the HDD on my chipped Xbox 1.6 so I can transfer and use the HDD on my softmodded Xbox. I followed this guide here: https://www.biline.ca/xbox_uxe_hd2.htm 
       I was able to update the EEPROM as stated and locked the HDD so it can be used on the softmodded Xbox. That worked fine. But now my chipped Xbox turns on but no picture or video whatsoever. I don't know what to do next...
    • By empyreal96
      *Disclaimer: No links to file will  be provided, this is for discussion only*
      Not sure if anyone has yet, but I have been running a diff between Barnabus Kernel Repack and the 4400 leak Kernel Tree..
      I've only just started looking through, and I have next to no C++, C and C# coding skills so bear with me lol
       
      They seem somewhat the same, but there are quite a few files with parts of code removed or added, on both branches. But Barnabus is reliant on precompiled objects (obviously as it's a smaller, almost stripped down and made 'easy' release of the full tree)
      There are some comments in Barnabus code that isn't in the full tree + CPXXUPD code, and some are recorded more 'formally'
      (i.e private\ntos\init\init.c line 244 - 254 in both branches)
      One other thing is the 4400 tree is set to have 'complex's' edits definable where possible, which there are a lot..
       
       
      One thing that really interests me, is Barnabus uses BIOSPack to pack: 2bl, remainder and kernel into a 'Working' rom (I use that word loosely).
      The 4400 tree uses rombld, which requires: preloaded, init table, bootloader, kernel and a rom decode file(I think).. this is a drastic difference to BIOSPack.. But the main issue with Rombld is it will not produce a full rom.. by default it will error on the 'preloader' being to big (9008 bytes not 1664 bytes)..
      To get around that, we can either build for XDK Systems.. 
      Or as I discovered today: modify 'private\sdktools\rombld\rombld.cpp' Line 879 from (2 * 1024) to (21 * 512), to allow rombld to accept the 9kb preloader file.. (I haven't tested a bios rom built with this edit yet)
       
      I will post here anything else I discover! 
    • By eustachy1973
      My modest modification.
       

       

       

       
      128MB Ram

       
      New Fan
       

       
       
      Open Xenium Chip with Ryzee119's XEnium Spi2par2019
       

       

       


       

       

       

      To be continued...
       

Board Life Status


Board startup date: April 23, 2017 12:45:48
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.