Solved

Computer meltdown (literally)

Posted on 2006-06-10
35
1,526 Views
Last Modified: 2007-12-19
I have a PC that was working fine for a long period of time, with no recent changes made.  It hasn't been moved.  Then all of the sudden about an hour after using it last night, it had a meltdown.  Smoke filled the house, and I had to open all the windows and doors to clear it out.  I've never heard of this one before.

Here are pictures:
http://drivingblind.freewebspace.com/

I figure it could be a malfunctioning power supply that caused the motherboard to have a meltdown, however it is strange that it seemed to happen by either the IDE controller or where the LED's plug in.  Therefore, I guess the motherboard is suspicious as well.

Has anyone heard of this before, or have an idea as to how this could happen?
0
Comment
Question by:bleujaegel
  • 12
  • 10
  • 7
  • +2
35 Comments
 
LVL 38

Expert Comment

by:BillDL
ID: 16879501
I've got to congratulate you on some terrific photo's there.  They would be highly valued by any fire investigator or insurance investigator.

I've seen motherboards where individual small areas or one specific component have burned out, but the damage on them was isolated to that small area and was usually caused by a short circuit where eg. one of the solder joints underneath the board made contact with the case through a dropped screw lodging in there or poor fitting.

From that last photo (http://drivingblind.freewebspace.com/images/dvc02829.jpg) it is clear that this had the potential for being the seat of a major house fire.  It's obvious from that how much heat was generated and has caused the hot smoke damage to rise upwards as the board was sitting vertically.  The round IDE cable above (http://drivingblind.freewebspace.com/images/dvc02828.jpg) looks like it has been on the verge of catching fire to the point where it would have required a fire extinguisher.  I would say from that aspect that you were lucky to have escaped a major fire, but it's unfortunate the board (and most likely the whole base unit) is wrecked.

Is it under warranty?  If so, and if you can prove faulty components, you have a good case to claim the manufacturer for selling a DANGEROUS motherboard - not just a faulty one.  They would obviously want to see the board and all wiring looks/hardware that was connected to it at the time.  In that event, it may be worthwhile having a qualified electronics expert to inspect the board and hardware BEFORE sending the whole lot away to the manufacturer's investigators.

What is the Make, Model, and Revision Number of the motherboard?
I would like to find a good hi-res image of the board as it was prior to the fire so I can see what pin headers are under that charred mess shown pretty well in the image (http://drivingblind.freewebspace.com/images/dvc02817.jpg).

I'd like to know what the white 2 or 3-core cable in the above image is for.  It looks like a CD-Rom audio cable, but it seems too thick.  If that is insulating tape around the end where the innner cables extend from the outer sheath, then you could find that any claim could be invalidated if it was a cable you made up - even if it wasn't faulty.  It seems to right at the heart of the seat of fire, and it is more likely to be one of the cables supplying power from the PSU to motherboard via a connector that is now completely melted.

That is your most likely cause, as that is probably a 12 volt supply.  Other connections on the motherboard are more likely to be lesser voltages, and therefore less likely to cause as much effect if there was a short circuit.

If that IS the 12 volt supply line, or any other directly from the PSU, then it would be a very good idea to have an experienced and qualified electonics technician inspect and test it.  Certainly use one who is willing to provide forensic evidence in court if you intend taking the matter further.  In that event, then it would also be vital that you sent the power strip or surge protector block used to connect the mains lead for the computer, and also have your mains supply tested. PSU's usually have their own internal fuse that should blow if an abnormally high power input hits it, but not all PSU's have this.

Modern PSU's are "switching" models that keep a steady and accurate supply level by switching themselves on and off very rapidly, as opposed to the old style stepping "transformers" that used wire coils on a core to reduce the incoming voltage.  It is conceivable that a fault can develop with the PSU whereby it supplies an over-voltage.

A badly fitting connector which doesn't fit snugly around a pin on the socket could cause arcing between the connector and pin, but what cables were connected in that vicinity?  If this is the front panel connector block, then connections like HDD and Power LED's are unlikely to arc enough to actually cause a fire.  That's not to say that they couldn't, because those wires are very thin and may burn out like a standard fuse inside the plastic sheath.

You definitely weren't tinkering with the connections before this, and perhaps put a conector on the wrong pin block?

Bill
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16879583
I've just saved the images and inspected them side-by-side.  Pity they aren't slightly higher resolution.

I'm particularly interested in the component you see on the left-hand dge of the board opposite and just above the half-way mark of the yellow IDE connectors in this image:
http://drivingblind.freewebspace.com/images/dvc02831.jpg

You see the rusty pink coloured rounded object?
I think that is the remains of a mini torroidial transformer that would have looked like the green copper-wound core in this image:
http://home.earthlink.net/~doniteli/cap3.jpg
and as a yellowish core in this one:
http://home.earthlink.net/~doniteli/cap2.jpg

Those images show examples of large batches of faulty leaking electrolytic capacitors that plagued motherboard manufacturers a few years ago:
http://home.earthlink.net/~doniteli/index27.htm
Just do a search in google.com for the key words "leaking electrolytic capacitors taiwan" and you'll see the scope of the mayhem it caused.
The torroidial transformers rely on the power supply being regulated by their associated capacitors.

I am NOT suggesting that this was the cause of your fire, but IF that object WAS one of those transformer coils, then I am surprised to see that the copper wire is melted off it.  It is a fairly heavy guage wire compared with some of the finer wires that are intact on your board, and MAY suggest this as the seat of the fire if this is so.  This would be especially important, because the heat has risen AWAY from that specific area and caused damage above it (ie. Left to right in the image http://drivingblind.freewebspace.com/images/dvc02831.jpg).

I flipped that image 180 degrees horizontally to do a side-by-side comparison with the image http://drivingblind.freewebspace.com/images/dvc02830.jpg
It's not easy to tell, but where the board has melted through into a hole in that image is very close to that mysterious rusty pink component shown on the upperside of the board.
0
 
LVL 91

Expert Comment

by:nobus
ID: 16879950
from your photo, i could not see very well what was melted, the cpu or something that dripped on it.
How was the PC positioned, and what was located above this melted place?  
Can something have dripped down on it?
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16881423
Yeah, it's hard to establish exactly what orientation the board was mounted, and what precise components are burned.
Contrary to what I thought about that thick white cable with the black tape being a soundcard cable, it looks more like a front usb panel cable.  You can see the black and red internal speaker wire coming in at the top left of this image:
http://drivingblind.freewebspace.com/images/dvc02817.jpg
and the other front panel wires (red, black, green and grey) at the top left corner of the board.

The board looks like an MSI one, and judging from the smoke damage in the image linked to above, what you are seeing at the top of the image is most likely to be the bottom panel of the case.
That's not hard and fast, however, because it could have been the top of the case, but the orientation of all the boards I currently have is that way.

Of course, extractor fans would pull some of the hot air and smoke from that burned out part of the board and coat other parts of the board along the path of the air flow, but when you see the edges of the add-on cards, it's clear that the REAR of the board is shown at the:
RIGHT of http://drivingblind.freewebspace.com/images/dvc02817.jpg
TOP of http://drivingblind.freewebspace.com/images/dvc02831.jpg

From what I can see, the 2nd IDE socket from the bottom has deflected most of the smoke and heat as the one at the bottom melted and dripped down onto the area where the fire started, thus perpetuating it and causing more damage there.

I'm still puzzled about what that reddy pink rounded looking thing is that I thought I could see.  Looking more closely, it might just be part of the red motherboard's top surface that has only been partly affected by smoke/heat damage.
It's not easy to tell from the image of the board still in the case, but the board appears to be very close to the side of the case (ie. behind it) as seen here:
http://drivingblind.freewebspace.com/images/dvc02817.jpg
It's obviously had some gap, or the smoke/heat damage wouldn't have had such an effect on the underside of the board here:
http://drivingblind.freewebspace.com/images/dvc02829.jpg
It definitely looks like Right -> Left in the image was Bottom -> Top as mounted.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16881567
Here's my guess (FSB400 + Promise pdc20376):
K7N2 Delta-ILSR
http://www.msicomputer.co.uk/Images/Products/ProductImages/LargeImage/703422_6570_delta.jpg
http://www.msicomputer.co.uk/products.aspx?product_id=703422

Zipped pdf manual:
http://66.96.84.4/support/mnu_exe/mbd_mnu/E6570v1.3.zip
Relevant Layout Images on pages 1, 11, 21, 24, 38, 40, 42, and 44 to 46.
0
 
LVL 38

Accepted Solution

by:
BillDL earned 400 total points
ID: 16881630
It certainly looks like a very similar (if not exact) layout, in which case there are no controller chips in the area below IDE3 where the main damage is located.  The only connector pin-blocks in the vicinity are:
- Front Panel USB Connectors (JUSB2) (Page 44 of manual)
- Optional D-Bracket USB 1.1 and 2.0 diagnostics "card" (JLED1) (page 46 of manual)
JBT1 (bluetooth wireless connection) is just outwith the main damaged area, so I think we can discount that connector.

JUSB2 is what is obviously shown in the burned-out images of the board still in the case and with the grey wires connected to it.  That looks to me like where the fire has started, although I don't think anyone will ever be able to say with certainty from the image.

I would hazard a guess that this would be a very unusual occurrence, given that USB uses 5 volts and is hot swappable, eliminating a possible cause like unplugging or plugging some non-hot-swappable device that could cause arcing.

Either that, or something conductive lodged down the back of the board making contact with the case.
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16882174
I posted another picture on page 1, and added a 2nd page as well.  It appears the USB cable was the one that fried.  I cut the end off of it to take an up close picture.

The mobo is (or was) an MSI 6570 Ver 1.
0
 
LVL 91

Expert Comment

by:nobus
ID: 16883425
>>  It appears the USB cable was the one that fried  <<  strange, since there is only a 5 V - 500mA on it. Must have been a short on the circuit board . .
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16883453
The strange thing is that the computer hasn't been moved, and I never even used the front USB.  If that is the culprit, apparently spontaneous combustion is to blame.

I checked all the power supply connectors, and none of them show any signs of trouble.  I'm wondering is it's possible that dust could have caused a short in the affected area.  You wouldn't think that area would be the one that would fry, though.

The black mark on the underside travels from the area of the onboard USB and IDE connector straight up to the memory slots.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883579
Don't you think it was some smart detective work of mine to figure out your motherboard model before you confirmed it to us?  Yet more excellent pictures to provide clues.

In case you are wondering, I am NOT an Electrical Engineer, Fire Investigator, or a Forensic Scientist, but I DID work for quite a number of years as a Legal Investigator which exposed me to a lot of Fire Investigation reports and Insurance Investigations.  That's the reason I am guessing rather than giving a qualified and definite answer.

I had been wondering what the white circle was at the bottom Right of this image:
http://drivingblind.freewebspace.com/images/dvc02829.jpg
but it is clear when you look at the bottom Left of:
http://drivingblind.freewebspace.com/images/dvc02831.jpg
that it is a plastic clip-down type stand-off.

I remarked earlier that there did not seem from the following image to be a lot of space between the underside of the motherboard and the chassis onto which it was mounted:
http://drivingblind.freewebspace.com/images/dvc02817.jpg
and I notice that the board is screwed down through the hole just to the right of the burned area in that same image.

Looking at one of your new images (http://drivingblind.freewebspace.com/images/dvc02834_1600x1200.jpg) it is clear that you have just placed the board back into the case to demonstrate the orientation, but it obviously isn't secured as it had been before.  The first thing I noticed about that image is the absence of stand-offs in the two tapped holes that line up with the top and bottom rear panel PCI card bays (ie. where the ports poke through).  They are obviously the points where the motherboard mounts onto, because you can see that the board has been pushed back and the mounting holes would line up over them when secured, but where are the stand-offs?

We've accounted for the stand-off at the top right of that same image as having been a plastic one, but it prompts me to ask whether the board may have been screwed down directly onto the chassis holes.  Sometimes these are raised a bit that would allow this, but it's not easy to see from the overhead perspective in that shot.

You can see the head of one of the rivets that hold the case together at the bottom right of that image (still talking about http://drivingblind.freewebspace.com/images/dvc02834_1600x1200.jpg here).
Now turn your attention to the following image that shows the height of the plastic stand-off that was in the corner of the board next to the fire damage:
http://drivingblind.freewebspace.com/images/dvc02839.jpg

It doesn't seem to raise the board very far off the chassis, and IF the board was screwed down directly onto the mounting hole on the chassis through the hole immediately to the left of the fire damage in the above photo, then it could potentially flex the board too much or foul one or more of the connections or tracks against a rivet head if there is one right under where the fire started.

If this is the case, then it would be a puzzle why it didn't happen sooner - unless you had recently been tinkering inside the case and maybe pushed in one of the IDE cables or front panel connector cables to make sure they were secure.  An hour of use might just be enough time for a low voltage short circuit to burn out a thin wire core and ignite the plastic sheath.

Look again to one of your first batch of images:
http://drivingblind.freewebspace.com/images/dvc02817.jpg
The USB Cable is lying as though it still in its original position where it was plugged into the light blue plastic socket which is JUSB2.
The puzzling thing is why the wires (particularly the yellow one) appear to be relatively undamaged, while here the damage is obvious:
http://drivingblind.freewebspace.com/images/dvc02847.jpg

The light blue socket is missing in this image:
http://drivingblind.freewebspace.com/images/dvc02839.jpg
and is melted onto the USB connector in this image:
http://drivingblind.freewebspace.com/images/dvc02847.jpg

Now, the other connectors have been subjected to enough heat to melt them to varying degrees:
http://drivingblind.freewebspace.com/images/dvc02836.jpg
and
http://drivingblind.freewebspace.com/images/dvc02835.jpg

Looking at image http://drivingblind.freewebspace.com/images/dvc02836.jpg which shows the connectors with markings, you will notice that those 4 connectors are melted mostly on the side that the black wire connects to on each, ie. Internal Speaker, Reset Switch, Power LED, and HDD LED.

Look at the other side of them and you will see that the Internal Speaker connector is the least damaged, while the others are damaged to about the same degree (although the Reset Switch does seem a bit more melted).

Look in the bootom right-hand corner of the image:
http://drivingblind.freewebspace.com/images/dvc02839.jpg
and you will see the labelled pin connectors.
Compare this with the pin definitions for JFP1 and JFP2 on page 42 of the manual
http://66.96.84.4/support/mnu_exe/mbd_mnu/E6570v1.3.zip
and it is clear that the black wires would all have been closest to what seems to be the source of the heat, but the pin blocks are NOT actually damaged.  In other words, NONE of those connectors seem to have been the source.

If you look, however, and JDLED2 shown on Page 46 of Manual and immediately underneath the IDE3 socket (just right of centre) in the photo here:
http://drivingblind.freewebspace.com/images/dvc02839.jpg
the left-hand side of that pin block is badly burned.
This corresponds with the bad damage to the cable that went into it.

The cable to JDLED2 is the one serving the diagnostic lights on the USB D-Bracket, while the one that was connected to the blue plastic JUSB2 socket is the one serving the USB port on the D-Bracket.

I would not discount the pin block JDLED2 from consideration as the source of the ignition.  Fanned by incoming air from the bottom front intake fan (or merely from air being dragged through that area by the expeller fan at the rear), the plastic IDE3 socket immediately above it could have ignited quite quickly if that cable or pin block was the seat of the fire.  The air flow would have been from right to left in that image. It may  have started melting the underside of the IDE socket (softer plastic) further across to the left, and dripped burning molten plastic down onto all the components below it (JUSB2).  While burning off, the  excessive heat from the burning plastic drips would be enough to char the motherboard right through and give the impression that the seat of the fire was JUSB2.

Look again at the upside-down view of the board still mounted in the case:
http://drivingblind.freewebspace.com/images/dvc02817.jpg
and you will see how this is a realistic possibility.  The incoming airflow to the case would have been from left to right, which is why the smoke damage on the front of the board is directional, but is straight upwards on the rear of the board.  I can't see if there IS an intake fan at the bottom front of the case (top right of image http://drivingblind.freewebspace.com/images/dvc02834_1600x1200.jpg), but that's where air is usually sucked in.

Remember also that fluff can act as an accelerant where you have sparking or excessive heat build-up.  Tumble Driers can virtually explode if fluff builds up while the clothes are being dried.  It's unlikely that fluff behind the motherboard could act as a conductor and cause a short-circuit though.

I know this doesn't help solve exactly what caused the ignition, but my 3 guesses as to most likely causes would be:

1. Incorrect pin connection or partial connection of D-Bracket connector into JDLED2
OR
2. Rear of board touching chassis close to JUSB2 if stand-offs were too low or board was flexed downwards through pushing in a cable like the IDE cable.
OR
3. A dropped metal object like a screw becoming lodged between the rear of the board and the chassis and causing a short-circuit.

I don't know the release date for that motherboard, or when you bought it new, but the manual tends to indicate that it was March 2003 or before that.  I doubt that you would have a warranty claim after that length of time, but a motherboard actually catching fire and causing a potentially fatal fire is a serious failure that goes well beyond Warranty cover.  The problem is that you would have to prove that the case into which it was fitted, and all hardware connected to it, was fully compliant and therefore authorised or tested hardware.

Bill
0
 
LVL 12

Assisted Solution

by:public
public earned 100 total points
ID: 16883655
The fire appears to begin in the JUSB2 cable.
Evidently somewhere in the cable or the connectors 5V and ground shorted out.
The rest are just secondary effects.
The led contact are not even damaged.
0
 
LVL 91

Expert Comment

by:nobus
ID: 16883677
>>    I'm wondering is it's possible that dust could have caused a short in the affected area  << could be, depending on the amount and nature.
btw - nice analysis from BillDl
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883678
I was still typing and looking at the photo's while nobus and yourself were posting comments.

I tend to agree with nobus about the small voltage and current being unlikely to cause a fire, but you asked
>>> "I'm wondering is it's possible that dust could have caused a short in the affected area". <<<
Household dust is mainly dead skin, hair, and the exoskeletons of the microscopic mites that inhabit it, and which COULD be flammable if exposed to a spark or direct heat.  The danger from dust is mainly the thermal insulation it creates on silicon chips that are heat sensitive.

http://tinyurl.com/ajaub
http://catalog.cmsp.com/datav3/im060005.htm
http://www.friendly-robots.com/images/dustmite3.jpg
http://www.burgesscomputers.co.uk/images/GraphicsCardBefore.jpg
http://techrepublic.com.com/2300-10877_11-5911196-8.html

As I said above though, I doubt that dust would be able to conduct electricity enough by itself to heat up and ignite.

What I can't see in the manual are the voltages and current rating of the JDLED1 socket for the D-Bracket's diagnostics LED's.  I would hazard a guess that they will also be very low at be 3 or 5 volts.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883686
nobus ~ good enough for a civil court case?  ;-)
0
 
LVL 12

Expert Comment

by:public
ID: 16883693
>I am NOT suggesting that this was the cause of your fire, but IF that object WAS one of those transformer coils, then I am surprised to see that the copper wire is melted off it.  It is a fairly heavy guage wire compared with some of the finer wires that are intact on your board, and MAY suggest this as the seat of the fire if this is so

There never was an inductor near the fire.
The round donut things are inductors.

>2. Rear of board touching chassis close to JUSB2 if stand-offs were too low or board was flexed downwards through pushing in a cable like the IDE cable.

No.
Either the cable shorted, or the far end connector shorted.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883695
I'm going to try and melt a wire of the same guage as the usb ones by shorting across a 9 volt battery to see what happens.  I'll let you know the results, unless of course I set fire to my house.
0
 
LVL 12

Expert Comment

by:public
ID: 16883705
>>>I tend to agree with nobus about the small voltage and current being unlikely to cause a fire, but you asked
>>> "I'm wondering is it's possible that dust could have caused a short in the affected area". <<<
Household dust is mainly dead skin, hair, and the exoskeletons of the microscopic mites that inhabit it, and which COULD be flammable if exposed to a spark or direct heat.  The danger from dust is mainly the thermal insulation it creates on silicon chips that are heat sensitive.

Get real. You sound like an NTSB investigator. Or other major government agency investigator.

USB cable has 5V and gnd, and enough current to start a fire. It is the most damaged part of the board. The cable insulation is burned off.
Test the remains, and you will find a short.

0
Threat Intelligence Starter Resources

Integrating threat intelligence can be challenging, and not all companies are ready. These resources can help you build awareness and prepare for defense.

 
LVL 91

Expert Comment

by:nobus
ID: 16883707
don't have the battery exploding in your hands Bill, then we'll have to look for another detective
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16883718
Thanks for the responses so far.  They have been quite informative.  

BillDL good point on the stand-offs.  It doesn't look like they are installed from the picture, but I double checked, and they are all there.  The motherboard was sitting probably about .2 inches off the case.  I didn't notice anything unusual about anything contacting the case.  Again, I hadn't made any changes to the computer for quite a while (including cable swapping).  In fact, it's running Windows Server, so it's been running all the time.  

It's quite a strange problem.  It doesn't make any sense why the 5v usb would cause the problem, yet that is the source.  Go figure.
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16883727
... that is, since I haven't touched the cable at all since it was built.  It probably is the culprit, I'm just mystified as to how.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883729
public, are you able to see something in the photo's that my eyes are unable to see, or are you a forensic fire investigator?

I might be wrong but from what I can see, the JUSB2 socket is the blue plastic one that looks to me as though it would prevent the connector from being pushed into it the wrong way, ie. like the CD Audio Connector or CPU Fan type.

Why do you find my 2nd hypothesis so inconceivable?
>>> "Rear of board touching chassis close to JUSB2 if stand-offs were too low or board was flexed downwards through pushing in a cable like the IDE cable". <<<

If you read through my comment, you will be aware that I THOUGHT I saw a "doughnut-shaped" inductor core (yes, I referred to it as a Toroidial Transformer) with the copper wire melted off it.  On closer inspection, I later stated that it was just part of the red motherboard that LOOKED like such a core from the aerial view.

You stated previously "The led contact are not even damaged".
Do they come pre-distorted then?
http://drivingblind.freewebspace.com/images/dvc02835_1600x1200.jpg
http://drivingblind.freewebspace.com/images/dvc02836.jpg
0
 
LVL 12

Expert Comment

by:public
ID: 16883771
>Why do you find my 2nd hypothesis so inconceivable?
>>> "Rear of board touching chassis close to JUSB2 if stand-offs were too low or board was flexed downwards through pushing in a cable like the IDE cable". <<<

because the ide connectors are barely damaged.
The usb cable has insulation completely burned off, hence high current flowed through the cable. A short to case on the bottom would not damage the cable this way.


0
 
LVL 12

Expert Comment

by:public
ID: 16883774
>You stated previously "The led contact are not even damaged".
Do they come pre-distorted then?

The pins are not even discolored, hence the heat source is elsewhere.

0
 
LVL 38

Expert Comment

by:BillDL
ID: 16883795
public, why do you insist on blatantly criticising me?
I have provided all the reasons for my conclusions.
If YOU are so eminently qualified, then please amend your profile to reflect this, as all I can see is "Hardware Certified Expert - Master Level" and "no special backgroung. generic pc experience".

>>> "Get real. You sound like an NTSB investigator. Or other major government agency investigator". <<<

If NTSB is a reference to The National Transportation Safety Board, then does the name "Lockerbie" and the flight "Pan Am 103" mean anything to you?

I don't suppose YOU have stood next to something like this:

http://upload.wikimedia.org/wikipedia/en/f/ff/PA103cockpit4.png
http://www.nickdidlick.com/recentpix/favoritepix/FavoritePictures/image/lockerbie.jpg
http://www.ncccusa.org/gifs/Lockerbie.jpg
http://www.cbc.ca/news/background/lockerbie/

No, perhaps that experience wasn't really relevant to motherboards, but neither is your flippant irrelevant comment.
0
 
LVL 12

Expert Comment

by:public
ID: 16883817
>If NTSB is a reference to The National Transportation Safety Board, then does the name "Lockerbie" and the flight "Pan Am 103" mean anything to you?

No, I was referring to TWA800.
http://www.twa800.com/index.htm

Rather than inventing unlikely scenarios, just look at the pictures. No special skills are needed. Just use common sense.
These are the eyewithesses. Assuming that the pictures are genuine and not staged, the usb cable short emerges as the probable cause.
Detailed examination of the cable would probably reveal the root cause.


0
 
LVL 12

Expert Comment

by:public
ID: 16883898
>I'm going to try and melt a wire of the same guage as the usb ones by shorting across a 9 volt battery to see what happens.

I do not mean to be disrespectful or arrogant, but if you want to be taken seriously please do not make statements like the one above.
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16886944
I read a story about a boy that left to a friends house, and when he returned his family had all perished from the Lockerbie incident.  How horrific.

Funny how people that saw a missile heading towards the flight TWA800 were quickly discredited.  

Anyhow, the power is only going to run through the motherboard to the front USB connector pins, and from there to the front ports on the case.  Since the most damaged area is the front USB connector, is it possible that whatever regulates the power to the front USB was defective, therefore sending an overvoltage, and causing the meltdown?  

I don't know enough about the electronics aspect of motherboards, but this is just a guess.
0
 
LVL 12

Expert Comment

by:public
ID: 16887826
>Since the most damaged area is the front USB connector, is it possible that whatever regulates the power to the front USB was defective, therefore

Nothing regulates usb power. It is connected directly to the 5V/40A power supply. That is why shorts are destructive.

>Funny how people that saw a missile heading towards the flight TWA800 were quickly discredited.

Many think that ignorance is bliss.....
Try reading some more.
0
 
LVL 11

Expert Comment

by:maramom
ID: 16887964
Which USB devices were connected at the time? Anything, perhaps, that may send power back through the ports?
0
 
LVL 11

Expert Comment

by:maramom
ID: 16888025
Just reread (lots to read!)...and saw that you didn't attach any devices to the front port.
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16889272
So if the main motherboard ATX connector supplies the 5V to the USB devices, is it possible that this was the first area the power passes through, therefore taking the brunt of the power supplies wrath?  I don't know if is a chain-like path that the power passes through from connector to connector on the motherboard, or if there are multiple paths that that it could take.  Again, I am not an electronics expert by any means...

A power supply failure at 5V explains why the SCSI drive wasn't damaged.  I took it out, put it in another computer and ran a FULL disk diagnostic with zero errors.  The CPU shows no signs of anything wrong.  I will test the CD rom drive, and memory soon.

The only variable I can see at this time is the power supply, due to the fact that no changes have been made, and I wasn't using the USB ports for anything.
0
 
LVL 12

Expert Comment

by:public
ID: 16890200
>So if the main motherboard ATX connector supplies the 5V to the USB devices, is it possible that this was the first area the power passes through, therefore taking the brunt of the power supplies wrath?

The motherboard has internal power planes that can handle very high current.
It is unlikely that the power supply fault was responsible.
Look at the usb cable more closely.
0
 
LVL 2

Author Comment

by:bleujaegel
ID: 16890710
It looks pretty fried.  It's hard to make heads or tails when you're not sure exactly what you're looking for in a charred mess.  I would think the cable would be most probable, since the motherboard definitely isn't able to move.
0
 
LVL 38

Expert Comment

by:BillDL
ID: 16900066
Thank you, bleujaegel.
0
 
LVL 12

Expert Comment

by:public
ID: 16909925
>It's hard to make heads or tails when you're not sure exactly what you're looking for in a charred mess.

You may find some indication of a molten metal weld in some spot.
0

Featured Post

How to run any project with ease

Manage projects of all sizes how you want. Great for personal to-do lists, project milestones, team priorities and launch plans.
- Combine task lists, docs, spreadsheets, and chat in one
- View and edit from mobile/offline
- Cut down on emails

Join & Write a Comment

I use more than 1 computer in my office for various reasons. Multiple keyboards and mice take up more than just extra space, they make working a little more complicated. Using one mouse and keyboard for all of my computers makes life easier. This co…
Skype is a P2P (Peer to Peer) instant messaging and VOIP (Voice over IP) service – as well as a whole lot more.
Excel styles will make formatting consistent and let you apply and change formatting faster. In this tutorial, you'll learn how to use Excel's built-in styles, how to modify styles, and how to create your own. You'll also learn how to use your custo…
This demo shows you how to set up the containerized NetScaler CPX with NetScaler Management and Analytics System in a non-routable Mesos/Marathon environment for use with Micro-Services applications.

746 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

12 Experts available now in Live!

Get 1:1 Help Now