Xbox 1.6 orange light max fanspeed no boot

[Mr.Kovacic]

When pressing the power button, nothing happens. Sometimes it would make the fan reach max speed 3 times and flash orange for a split second. Sometimes it would do something in between and slightly nudge the fan to spin and nothing more. I should mention my xbox has a modchip.

 

Based off the information i have gathered from forums and reddit posts, this is what i know; 
- IDE cable and disc drive are not the cause of the problem, everything is disconnected from the motherboard

- power button and eject button work

- checked the board for corrosion, cleaned with iso. alchohol

- fuse on power supply is good

- checked voltages on power supply ATX connector (connected it to the motherboard, stuck the probes inside the connector where the wires go and pressed the power button. This probably isnt the way to do it i know). The standby and power on voltages match, but after a few seconds the drop back to zero (excluding the orange wires which do have standby voltage) because i presume the xbox doesnt boot

- checked the transistor at Q7C2 and i am not getting a match for the voltages. 1.1v on the collector and 0.6 on the emitter. I might have read the values wrong, but the still dont reach what the are supposed to

 

Not losing hope, open to all suggestions

[Bowlsnapper]

7 hours ago, Mr.Kovacic said:

When pressing the power button, nothing happens. Sometimes it would make the fan reach max speed 3 times and flash orange for a split second. Sometimes it would do something in between and slightly nudge the fan to spin and nothing more. I should mention my xbox has a modchip.

 

Based off the information i have gathered from forums and reddit posts, this is what i know; 
- IDE cable and disc drive are not the cause of the problem, everything is disconnected from the motherboard

- power button and eject button work

- checked the board for corrosion, cleaned with iso. alchohol

- fuse on power supply is good

- checked voltages on power supply ATX connector (connected it to the motherboard, stuck the probes inside the connector where the wires go and pressed the power button. This probably isnt the way to do it i know). The standby and power on voltages match, but after a few seconds the drop back to zero (excluding the orange wires which do have standby voltage) because i presume the xbox doesnt boot

- checked the transistor at Q7C2 and i am not getting a match for the voltages. 1.1v on the collector and 0.6 on the emitter. I might have read the values wrong, but the still dont reach what the are supposed to

 

Not losing hope, open to all suggestions

Sounds like the ADM1032 temperature sensor. Replace it with a known good one and boot up

 

 

[Mr.Kovacic]

On 1/3/2024 at 4:47 PM, Thairanny said:

Is it unmodified?

No, it has a modchip

[Mr.Kovacic]

On 1/3/2024 at 7:56 PM, Bowlsnapper said:

Sounds like the ADM1032 temperature sensor. Replace it with a known good one and boot up

 

 

That was what i tried to do initially, but then i found out that my 1.6 xbox does not have a seperate temperature sensor, it is intergrated into the GPU (?) or some other larger component

[Bowlsnapper]

7 hours ago, Mr.Kovacic said:

That was what i tried to do initially, but then i found out that my 1.6 xbox does not have a seperate temperature sensor, it is intergrated into the GPU (?) or some other larger component

Oh wow. I had no idea the 1.6 integrates that also... I would try replacing the encoder chip at that point, but that may be a bit much. Dang the 1.6 is a pain!

@KaosEngineer I feel like I wasted this poor man's time. What would you do in this scenario? You are well familiar with this issue.

[KaosEngineer]

On 1/5/2024 at 12:48 PM, Bowlsnapper said:

Oh wow. I had no idea the 1.6 integrates that also... I would try replacing the encoder chip at that point, but that may be a bit much. Dang the 1.6 is a pain!

@KaosEngineer I feel like I wasted this poor man's time. What would you do in this scenario? You are well familiar with this issue.

I believe the temp sensor on v1.6 Xboxes is part of the custom edit: Xcalibur Xyclops chip. 

The custom edit: Xcalibur Xyclops chip replaced the TSOP, PIC processor (System Management Controller) and temp sensor all in one package.

[Bowlsnapper]

Just now, KaosEngineer said:

I believe the temp sensor on v1.6 Xboxes is part of the custom Xcalibur chip. 

The custom Xcalibur chip replaced the TSOP, PIC processor (System Management Controller) and temp sensor all in one package.

... MAN That's problematic...  Can the chip be replaced and remedy temp control issues? You think that's what this is?

[KaosEngineer]

On 1/5/2024 at 10:39 PM, Bowlsnapper said:

... MAN That's problematic...  Can the chip be replaced and remedy temp control issues? You think that's what this is?

One or two of the traces from the CPU temp sensor diode to the edit: Xcalibur Xyclops may be damaged resulting in a FALSE positive overheating condition.

[Bowlsnapper]

1 minute ago, KaosEngineer said:

One or two of the traces from the CPU temp sensor diode to the Xcalibur may be damaged resulting in a FALSE positive overheating condition.

I'm trying to find the location of the senser. Looks like it may be the southbridge, basically? I wonder what could damage traces id clock caps aren't an issue on those things... not leakage anyway.

[KaosEngineer]

On 1/5/2024 at 10:58 PM, Bowlsnapper said:

I'm trying to find the location of the senser. Looks like it may be the southbridge, basically? I wonder what could damage traces id clock caps aren't an issue on those things... not leakage anyway.

The CPU temp sensor diode connects from the CPU pads  to the edit: Xcalibur Xyclops chip.  I've not traced down which pins of the edit: Xcalibur Xyclops they connect to.

CPU BGA Pads:

• THERMDA = AA15
• THERMDC = AB16

The traces make their way moving from top to bottom side of the PCB several times along their journey to the edit: Xcalibur Xyclops chip.

Image of the CPU BGA pads (Edit: Opps I forgot the D in THERMDA and THERMDC). And, these traces are already making there jump from the top to the bottom side of the PCB moving at an angle down and to the right of each PAD to a via.

[Mr.Kovacic]

6 hours ago, KaosEngineer said:

The CPU temp sensor diode connects from the CPU pads  to the Xcalibur chip.  I've not traced down which pins of the Xcalibur they connect to.

CPU BGA Pads:

• THERMDA = AA15
• THERMDC = AB16

The traces make their way moving from top to bottom side of the PCB several times along their journey to the Xcalibur chip.

Image of the CPU BGA pads (Edit: Opps I forgot the D in THERMDA and THERMDC). And, these traces are already making there jump from the top to the bottom side of the PCB moving at an angle down and to the right of each PAD to a via.

Okay i have the board right infront of me. Basically i should be looking for broken traces that lead back to the Xcalibur chip?

[Mr.Kovacic]

44 minutes ago, Mr.Kovacic said:

Okay i have the board right infront of me. Basically i should be looking for broken traces that lead back to the Xcalibur chip?

So, underneath the CPU i have found what seem to be like corroded traces. I know you cant tell from the picture (because iPhone), but they look very brown and rusty (red arrows). The blue arrow points to a trace that is discolored. It looks slightly brown

[Mr.Kovacic]

10 minutes ago, Mr.Kovacic said:

So, underneath the CPU i have found what seem to be like corroded traces. I know you cant tell from the picture (because iPhone), but they look very brown and rusty (red arrows). The blue arrow points to a trace that is discolored. It looks slightly brown

Here is a slightly better picture

[KaosEngineer]

Wait a minute.  I traced the THERMDA and THERMDC signals from the CPU to the IC U3B1 pins 2 and 3.  

The pictures I used were component-free hi-res scans from the now defunct web site diy.sickmods.net.

What is the IC at U3B1 on a v1.6 motherboard?

Is it an ADM1032 just like used on v1.0-1.4 motherboards?

The chip is simply at a different location on a v1.6 motherboard - U3B1 instead of U6F1.

[Mr.Kovacic]

38 minutes ago, KaosEngineer said:

Wait a minute.  I traced the THERMDA and THERMDC signals from the CPU to the IC U3B1 pins 2 and 3.  

The pictures I used were component-free hi-res scans from the now defunct web site diy.sickmods.net.

What is the IC at U3B1 on a v1.6 motherboard?

Is it an ADM1032 just like used on v1.0-1.4 motherboards?

The chip is simply at a different location on a v1.6 motherboard - U3B1 instead of U6F1.

 

If i understood you correctly, the chip does exist just at another location. I went to find where it is, and if can read the numbers and letters on the motherboard, its supposed to be here... but it's not? Its not underneath also

[KaosEngineer]

On 1/6/2024 at 10:15 AM, Mr.Kovacic said:

 

If i understood you correctly, the chip does exist just at another location. I went to find where it is, and if can read the numbers and letters on the motherboard, its supposed to be here... but it's not? Its not underneath also

Okay, I'll have to do a bit more tracing the signals to see if they go beyond there.  

You could use a multimeter to measure resistance from pin 2 and 3 to pins of the edit: Xcalibur Xyclops chip.  I believe they go into it instead of having the ADM1032 chip. I'd never tried to figure out where they connected before. Odd...

 

Edit: Oops, not thinking straight it seems.  The temp sensor not integrated into the video encoder, MS custom Xcalibur chip, but the Xyclops chip.

[Mr.Kovacic]

2 hours ago, KaosEngineer said:

Okay, I'll have to do a bit more tracing the signals to see if they go beyond there.  

You could use a multimeter to measure resistance from pin 2 and 3 to pins of the Xcalibur chip.  I believe they go into it instead of having the ADM1032 chip. I'd never tried to figure out where they connected before. Odd...

Googled the schematics for the ADM1032 chip and found which are pins 2 and 3 (sorry if this is obvious i didnt know). I have continuity and am getting 0.4 ohms on both pins to the Xcalibur chip

[Bowlsnapper]

18 minutes ago, Thairanny said:

This may be incorrect and I know its unrelated, but the 1.6 I have now with no power, If I remember correctly, had power not long ago and only after plugging it in, something popped, then no power. Pretty sure the transistor next the the xcalibur chip is bad, BUT i think it may have been caused by the caps first being bad. All the large caps were popping, and I think because of it the transistor got fried. I would expect the Xcalibur to be a less developed solution to previous revisions, I’m not surprised if many 1.6 specific issues stem from the chip.

So maybe bad caps create conditions that cause failures in the Xcalibur? 

Perfectly plausible. I am starting to see cascading failures from SMD components going bad and causing other components down the line to go bad as well. Or from electrolytic caps not being replaced proactively. I am starting to get VERY worried. @KaosEngineer Would you agree with that assessment?

[KaosEngineer]

On 1/5/2024 at 10:38 PM, KaosEngineer said:

I believe the temp sensor on v1.6 Xboxes is part of the custom edit: Xcalibur Xyclops chip. 

The custom edit: Xcalibur Xyclops chip replaced the TSOP, PIC processor (System Management Controller) and temp sensor all in one package.

Oops wrong chip name - the MS custom Xyclops chip, not the Xcalibur video encoder.

21 hours ago, Mr.Kovacic said:

Googled the schematics for the ADM1032 chip and found which are pins 2 and 3 (sorry if this is obvious i didnt know). I have continuity and am getting 0.4 ohms on both pins to the edit: Xcalibur Xyclops chip

 

[KaosEngineer]

Interesting...the THERMDC and THERMDA (thermal diode signals) from the CPU only go to that missing chip's location (U3B1).

I know that the dashboard's reported CPU temp on v1.6 Xbox's is always higher than on previous Xbox revisions. 

And, the temp sensor's I2C (SMbus) address is different on a v1.6 motherboard.

 

I need do more research and track down any documentation on the Xbox's GPU. 

 

Questions to be answered:

Does it have a thermal diode?

If so, which BGA pads of the GPU are they?

And, do they connect to the Xyclops chip?

[Bowlsnapper]

12 minutes ago, KaosEngineer said:

Interesting...the THERMDC and THERMDA (thermal diode signals) from the CPU only go to that missing chips location.

I know that what was the CPU temp on v1.6 Xbox's has been noted to be higher than on previous Xbox revisions.  And, the temp sensor's I2C (SMbus) address is different to read the temps on a v1.6 motherboard.

 

I need do more research and track down any documentation on the Xbox's GPU. 

 

Questions to be answered:

Does it have a thermal diode?

If so, which BGA pads of the GPU are they?

And, do they connect to the Xyclops chip?

I think this is literally the first time I've ever seen you not know something, Kaos.

Are you saying that the diode is built into the IC of the CPU?

[KaosEngineer]

22 minutes ago, Bowlsnapper said:

I think this is literally the first time I've ever seen you not know something, Kaos.

Are you saying that the diode is built into the IC of the CPU?

Yes, the CPU has a thermal diode in it that connects to the ADM1032 dual temp sensor for the value to be read from it across the SMBus.  The other temp sensor is built-into the ADM1032 which simply reads the ambient air temperature around the chip.

Edit: I do not know if there's a similar temp sensor in the GPU.

Originally dashboards stated that the two temps were that of the CPU and GPU.  But the GPU was actually the ambient air temperature inside the Xbox by the ADM1032 chip - U6F1.  Later revisions of replacement dashboards started labeling this temp as the motherboard temp.

[Mr.Kovacic]

51 minutes ago, KaosEngineer said:

Yes, the CPU has a thermal diode in it that connects to the ADM1032 dual temp sensor for the value to be read from it across the SMBus.  The other temp sensor is built-into the ADM1032 which simply reads the ambient air temperature around the chip.

Edit: I do not know if there's a similar temp sensor in the GPU.

Originally dashboards stated that the two temps were that of the CPU and GPU.  But the GPU was actually the ambient air temperature inside the Xbox by the ADM1032 chip - U6F1.  Later revisions of replacement dashboards started labeling this temp as the motherboard temp.

Seems to me like they sacrificed ease of repair for more precise temperature readings by removing the ADM1032. Anyhow my understanding of xboxes does not go beyond this point. I could try and troubleshoot stuff that you folks reccomend. I doubt this information will help but i can share it:

I found the xbox in my closet and wanted to try it out. Played Flatout for 3 hours and it just died. Didn’t wanna power up anymore. A few years back (before it died completely) my dad sent it in for repair. No idea what they did there, dad said they changed some capacitors. The reason why it went in repair is because it also didnt want to power up, but last time you could see the green light for a second. I still have the post-it note they stuck on the xbox, it says “power supply” so i guess that was it.

[Bowlsnapper]

2 hours ago, KaosEngineer said:

Yes, the CPU has a thermal diode in it that connects to the ADM1032 dual temp sensor for the value to be read from it across the SMBus.  The other temp sensor is built-into the ADM1032 which simply reads the ambient air temperature around the chip.

Edit: I do not know if there's a similar temp sensor in the GPU.

Originally dashboards stated that the two temps were that of the CPU and GPU.  But the GPU was actually the ambient air temperature inside the Xbox by the ADM1032 chip - U6F1.  Later revisions of replacement dashboards started labeling this temp as the motherboard temp.

Oh sorry, that is actually what I meant, is the GPU... although I did wonder about the CPU too. I was never really sure of where the "motherboard" temperature sensor was.

[KaosEngineer]

3 hours ago, Bowlsnapper said:

Oh sorry, that is actually what I meant, is the GPU... although I did wonder about the CPU too. I was never really sure of where the "motherboard" temperature sensor was.

U6F1 on a v1.0-1.4 revision motherboard.  I'm not sure what's being used on a v1.6 motherboard.  The location I traced the CPU's THERMDC and THERMDA (thermal diode signals to - BGA pad AA15 and AB16) end up at an unpopulated SMD 8-pin set of pads at location U3B1.  

i know that the SMBus address where the supposed temp readings are being read from for a v1.6 motherboard are different than for v1.0-1.4 motherboards . The value used as the CPU temp is 5 to 10 degrees C higher on a v1.6 than a v1.0-1.4 motherboard.