Couple of hardware questions

[LouisG]

I'm running Xp Pro Sp3 and lately my computer has been acting up and i'm
not sure whether its software or hardware related ,, i'm leaning to
hardware but i was wondering if anyone else thinks this.

A few weeks ago i noticed that after a reboot the sound would work for a
while then it would just die, i went out and bought a soundcard thinking
that maybe it was a problem with the onboard sound, it's still happening
with the soundcard.

Also the front and side usb ports on the case would stop working, same
thing reboot they worked for a time then died and wouldn't work until a
reboot.

With both of these problems there was nothing showing in the device
manager, running dxdiag came back with Directsound issues. Removed and
reinstalled drivers to no difference.

All through this the computer would sometimes just freeze up requring a
cold boot to get it back.

Yesterday had major problems connecting the internet, though every other
computer on the network could connect, the main computer could not, the
same one that all the above problems are occuring on.

There have been no major software additions and only the soundcard added.

I took out a stick of ram the other day to see what type/size it was and
it was extremely hot, this along with the sound and usb problems lead me
to think that it's only a hardware issue, but before going out and
building a new machine, i though maybe i could get some suggestions here.

Thanks

 

[Paul]

I'm running Xp Pro Sp3 and lately my computer has been acting up and i'm
not sure whether its software or hardware related ,, i'm leaning to
hardware but i was wondering if anyone else thinks this.

A few weeks ago i noticed that after a reboot the sound would work for a
while then it would just die, i went out and bought a soundcard thinking
that maybe it was a problem with the onboard sound, it's still happening
with the soundcard.

Also the front and side usb ports on the case would stop working, same
thing reboot they worked for a time then died and wouldn't work until a
reboot.

With both of these problems there was nothing showing in the device
manager, running dxdiag came back with Directsound issues. Removed and
reinstalled drivers to no difference.

All through this the computer would sometimes just freeze up requring a
cold boot to get it back.

Yesterday had major problems connecting the internet, though every other
computer on the network could connect, the main computer could not, the
same one that all the above problems are occuring on.

There have been no major software additions and only the soundcard added.

I took out a stick of ram the other day to see what type/size it was and
it was extremely hot, this along with the sound and usb problems lead me
to think that it's only a hardware issue, but before going out and
building a new machine, i though maybe i could get some suggestions here.

Thanks

The RAM being extremely hot, isn't a good thing.

The motherboard has a number of DC regulators, for creating
voltages not available on the power supply. One of those could
be bad.

Knowing the make and model number of motherboard would help.
Or, if it's a pre-built computer, perhaps the make and model
number of the whole computer. What I'd be checking for, is
existing known issues with that model.

Onboard regulators come in two types - linear or switching.
When the linear ones fail, some components get so hot, the
solder underneath them melts. Good fun. The switching
ones, can have protection (overcurrent/overvoltage),
in which case the motherboard likely won't POST any more.
So whatever is going on, nothing has checked it and found
it to be too far out of spec.

Occasionally, the BIOS plays a part in this. Some motherboards,
the BIOS has been known to mis-program a regulator, for
the voltages that can be changed under firmware control.
In which case, all the symptoms would have existed from
the first day of usage.

Paul

 

[LouisG]

The RAM being extremely hot, isn't a good thing.

The motherboard has a number of DC regulators, for creating
voltages not available on the power supply. One of those could
be bad.

Knowing the make and model number of motherboard would help.
Or, if it's a pre-built computer, perhaps the make and model
number of the whole computer. What I'd be checking for, is
existing known issues with that model.

Onboard regulators come in two types - linear or switching.
When the linear ones fail, some components get so hot, the
solder underneath them melts. Good fun. The switching
ones, can have protection (overcurrent/overvoltage),
in which case the motherboard likely won't POST any more.
So whatever is going on, nothing has checked it and found
it to be too far out of spec.

Occasionally, the BIOS plays a part in this. Some motherboards,
the BIOS has been known to mis-program a regulator, for
the voltages that can be changed under firmware control.
In which case, all the symptoms would have existed from
the first day of usage.

Paul

Thanks for the reply Paul ,, the mobo is a Asus P5QL-VM DO

 

[LouisG]

The RAM being extremely hot, isn't a good thing.

The motherboard has a number of DC regulators, for creating
voltages not available on the power supply. One of those could
be bad.

Knowing the make and model number of motherboard would help.
Or, if it's a pre-built computer, perhaps the make and model
number of the whole computer. What I'd be checking for, is
existing known issues with that model.

Onboard regulators come in two types - linear or switching.
When the linear ones fail, some components get so hot, the
solder underneath them melts. Good fun. The switching
ones, can have protection (overcurrent/overvoltage),
in which case the motherboard likely won't POST any more.
So whatever is going on, nothing has checked it and found
it to be too far out of spec.

Occasionally, the BIOS plays a part in this. Some motherboards,
the BIOS has been known to mis-program a regulator, for
the voltages that can be changed under firmware control.
In which case, all the symptoms would have existed from
the first day of usage.

Paul

Just to add to the mobo # Paul , this computer has the BIOS updated to
the latest , the system is about 3-4 years old and has just started all
this within the last 2 months, but more things seem to be cropping up.

Are there any tests i could run on it to see if somethings going bad? Or
should i just forget about it and look at another computer?

 

[Paul]

Thanks for the reply Paul ,, the mobo is a Asus P5QL-VM DO

OK, I have the PDF version of the manual here.

VDimm available voltages are 1.87, 2.00, 2.13, 2.26.
The nominal value is 1.8V for DDR2. You would check in
the BIOS screen, to see if it has been bumped up. It would
take a two bit code to control that value.

The VDimm regulator, is in the upper left of this photo. It's
a switcher, judging by the toroidal coil.

http://images10.newegg.com/NeweggImage/productimage/13-131-401-10.jpg

The manual also mentions Northbridge (NB) and Southbridge (SB)
voltage controls. Some of your other chips, could be tied into
those voltage rails.

The sound, that's a strange one. The sound card you installed,
could be running off 5V (from the power supply). The onboard
sound (VIA chip), likely uses a little linear located on the
motherboard, as the audio chip needs "clean" power. Those
two kinds of sound, don't have much of a relationship with
one another, in terms of failure points. The digital signals
going to the VIA onboard sound, come from the Southbridge, so
an incorrect SB voltage setting might upset it.

The USB ports could be related to the Southbridge (SB) voltage
setting (the USB core logic runs off the SB core voltage). USB
also requires a 5V source (for D+ D- signals on the cable), and
then the problem could be with the ATX power supply.

I really don't see all these things, pointing at just one
source of a problem.

Start by checking the BIOS first, Section 2.4 of the manual, and
see if any voltages have been jacked up. If the hardware was
defective, the settings might not actually control the thing.
(I.e. A hardware failure can cause the voltage to be some value
other than the BIOS specified value.)

The hardware monitor, doesn't measure anything useful, so
staring at that screen in the BIOS will only give you a headache.
It would have been very useful, if the PC architecture
specified a 16 channel ADC for hardware monitoring, so more
of these goofy voltages could be hooked up and monitored.
(So the user could quickly glance at the hardware monitor
screen, and instantly know everything was in spec.)
But I suppose that's too much to ask.

I've only seen a couple motherboards, with "voltage points"
(little metal legs sticking up, to hint where to measure).
When measurement points aren't visibly present, you
really need a schematic to know where to probe and check
stuff. (I'm embarrassed to report, I tried to trace down
the regulator chain on my P4C800-E Deluxe, and just
couldn't make sense of it. The voltages on the legs of
the MOSFETs seemed to be totally unrelated to what I
thought they should be. A schematic would have helped,
but those don't grow on trees. I could tell my board
used linears for NB and SB voltages, but the related
circuits had the wrong voltages present. I could never
get a decent reading in there. And the board worked
and there was nothing wrong with it. Just I couldn't
figure it out.)

So, try checking the BIOS screen first. It might not
be the source of the problem, but it is worth checking.

You can also try clearing CMOS. Just in case a BIOS value
you cannot see in the interface, is corrupted. Clearing the
CMOS will return the system to defaults. If you have any
custom settings you've entered, you'd have to enter those
settings again. When you clear the CMOS, the computer
should be completely powered off - and unplugging it is
the safest way to do that. You can damage the motherboard,
if you use the clear CMOS jumper or grounding points, with
the system powered. (See section 1.6 of the manual,
for instructions on the CLRTC jumper for clearing CMOS.)

Then, after clearing CMOS, you'd check RAM temperature again
and see if it was running a bit cooler.

Paul

 

[LouisG]

Louis, could be a heating issue. Have you tried using a can of
compressed air to clean out accumulated dust inside?

George

Yes George , that was my first action actually when this all started , in
fact you just reminded me of why i took that canned air to the computer ,
i had to replace the case power supply due to the fact that one of the
two fans had got really clogged and after i cleaned it out it would'nt
keep turning after a reboot , i actually had to spray some air into it to
get it going again and even then it was slow , though the other fan in
the power supply was working fine.

 

[VanguardLH]

A few weeks ago i noticed that after a reboot the sound would work for a
while then it would just die, i went out and bought a soundcard thinking
that maybe it was a problem with the onboard sound, it's still happening
with the soundcard.

Do you use powered speakers that have an auto-mute option? That is,
some powered speakers will mute themselves if they've been idle for
awhile (i.e., below some minimal threshold of volume). My old Altec
ADA-305 have this option and it's always a nuisance when they auto-mute.
Instead of waiting to be below the volume threshold for some time, they
just auto-mute after a fixed time. Either I have to play something loud
to whack it way past the min threshold or hit a button on the speaker to
kick it in the ass to start making sound again (even if it's below the
min threshold, like I'm listening to soft music or in a stealth game and
trying to listen for footsteps afar).

Just in case the problem is with the external powered speakers, have you
tried plugging in headphones (that you test are working okay by plugging
into, say, your stereo system, MP3 players, or whatever)?

Also the front and side usb ports on the case would stop working, same
thing reboot they worked for a time then died and wouldn't work until a
reboot.

What's the size of the power supply? Did you buy a pre-built or build
the computer yourself? From its initial delivery or build (and before
adding the new soundcard), did you add more hardware? How old is the
PSU (power supply unit)?

I took out a stick of ram the other day to see what type/size it was and
it was extremely hot, ...

That doesn't say how hot was the memory module. If it's DDR3 memory
then expect it to *feel* hot. If you're overclocking then it'll be even
hotter. Say the operating temperature range for the memory -40C to
+85C. 85C is 185F. Would that feel comfortable in your fingers? No.
While most folks would want their memory running at lower than its peak
operating temperature, that is still its continuous *operating*
temperature range.

A one-second exposure to 160F water causes 3rd degree burns. 30 seconds
exposure to 130F causes 3rd degree burns. The Consumer Product Safety
Commission recommends water heaters be set to a maximum temperature of
120F. So have you been able to take a 100% hot-water shower (just the
hot water tap turned on) and it got to max temperature? Ow ow ow.

 

[VanguardLH]

Just to add to the mobo # Paul , this computer has the BIOS updated to
the latest , the system is about 3-4 years old and has just started all
this within the last 2 months, but more things seem to be cropping up.

Are there any tests i could run on it to see if somethings going bad? Or
should i just forget about it and look at another computer?

Ever run memtest to check the memory?

http://memtest.org/

 

[Paul]

Ever run memtest to check the memory?

http://memtest.org/

"I took out a stick of ram the other day to see what type/size it was and
it was extremely hot"

I see no reason to be testing anything, until the computer is
returned to "nominal" conditions. Something is wrong with some
voltage on that machine. The symptoms don't particularly lead me
to conclude just one thing is wrong (symptoms are pointing
at more than one voltage). But RAM being "hot" is not a good sign,
and should be fixed first. Then, once the RAM is closer to the
proper temperature, you can try a memory test and see how well
it's working. Either the visible settings in the BIOS are wrong,
and should be corrected. Or, invisible settings (things stored
in CMOS RAM) are corrupted, and some of those are used to control
some of the onboard regulators. Clearing CMOS, may correct that.

There have been cases, where a user flash updates the BIOS, and
the BIOS has a programming error, where it uses the wrong
voltage for VCore on a processor. There have been cases,
where users report "my processor is too hot", another user
says "me too", and upon checking, it appears the BIOS is
using a voltage setting for another family of processor.
I don't think that's happening here, because too many things
are screwed up. So it doesn't match the symptoms of that kind
of problem.

If you flash the *wrong* BIOS into a board, the
GPIO signals that control the regulators can end up being
wrongly programmed, because the BIOS is intended for another
motherboard, using different GPIO signals, for different
purposes. Then, virtually all your custom logic on the board,
is bonkers (everything ends up set wrong). Based on what Louis
has reported so far, I don't see any hints of that kind of
problem. If he'd reported problems getting the BIOS to flash,
I'd be more suspicious of a wrong file being involved. The BIOS
flasher programs, do have rudimentary checks they can do, but
if you use the right (older) version of flasher, you can
override those checks.

Paul

 

[LouisG]

"I took out a stick of ram the other day to see what type/size it was
and it was extremely hot"

I see no reason to be testing anything, until the computer is
returned to "nominal" conditions. Something is wrong with some
voltage on that machine. The symptoms don't particularly lead me
to conclude just one thing is wrong (symptoms are pointing
at more than one voltage). But RAM being "hot" is not a good sign,
and should be fixed first. Then, once the RAM is closer to the
proper temperature, you can try a memory test and see how well
it's working. Either the visible settings in the BIOS are wrong,
and should be corrected. Or, invisible settings (things stored
in CMOS RAM) are corrupted, and some of those are used to control
some of the onboard regulators. Clearing CMOS, may correct that.

There have been cases, where a user flash updates the BIOS, and
the BIOS has a programming error, where it uses the wrong
voltage for VCore on a processor. There have been cases,
where users report "my processor is too hot", another user
says "me too", and upon checking, it appears the BIOS is
using a voltage setting for another family of processor.
I don't think that's happening here, because too many things
are screwed up. So it doesn't match the symptoms of that kind
of problem.

If you flash the *wrong* BIOS into a board, the
GPIO signals that control the regulators can end up being
wrongly programmed, because the BIOS is intended for another
motherboard, using different GPIO signals, for different
purposes. Then, virtually all your custom logic on the board,
is bonkers (everything ends up set wrong). Based on what Louis
has reported so far, I don't see any hints of that kind of
problem. If he'd reported problems getting the BIOS to flash,
I'd be more suspicious of a wrong file being involved. The BIOS
flasher programs, do have rudimentary checks they can do, but
if you use the right (older) version of flasher, you can
override those checks.

Paul

The Bios had been flashed , but that was a year or more ago.
After your last post regarding clearing the Bios , i went ahead and did
that leaving all defaults the way they were after resetting. I don't
usually play around in the Bios anyway as a lot of the settings are out of
my realm.
But the sound and usb aren't working again , the computer seems pretty
stable though without any freeze ups.
It just seems to "drop" these for no reason , though if you go into the
device mgr everything is okay, or so it says.

 

[Char Jackson]

"I took out a stick of ram the other day to see what type/size it was and
it was extremely hot"

I see no reason to be testing anything, until the computer is
returned to "nominal" conditions. Something is wrong with some
voltage on that machine. The symptoms don't particularly lead me
to conclude just one thing is wrong (symptoms are pointing
at more than one voltage). But RAM being "hot" is not a good sign,
and should be fixed first.

I don't see anything that would lead me to believe there's a voltage
problem with the RAM. It's completely normal for it to be far too hot
to hold comfortably in your hand.

 

[VanguardLH]

"I took out a stick of ram the other day to see what type/size it was and
it was extremely hot"

I see no reason to be testing anything, until the computer is
returned to "nominal" conditions.

A memory module running at 85C (which is WITHIN the operational
temperature range of a DDR3 module) will obviously feel hot. 1-second
exposure to water at 74C causes scalding. "Hot", even "extremely hot"
doesn't say HOW hot. It just means, without a real value, that it was
[very] uncomfortable to hold. I discussed temperature in my other post.

85C is 158F. You don't think that will feel scalding hot to your
fingers? While I chose 85C, some DDR3 are rated for an OPERATING
temperature range of up to 105C, or 221F. When was the last time you
dipped your hand into boiling water and it didn't feel "extremely hot"?

Besides, the computer was presumably powered down and it was some time
afterward that the OP removed the memory module. That means there was
no cooling air moving over the memory and it would be at a raised
temperature rather than at its cooled temperature within operating
range.

It was hot. So HOW hot was it? We don't know.

 

[Chris S.]

I don't see anything that would lead me to believe there's a voltage
problem with the RAM. It's completely normal for it to be far too hot
to hold comfortably in your hand.

Huh?

Just reached over and touched my RAM sticks. (This MB and system are
in "test/setup" mode on a tabletop). To my touch, almost room temperature.
"far too hot to touch" it NOT normal.

Chris

 

[Char Jackson]

Huh?

Just reached over and touched my RAM sticks. (This MB and system are
in "test/setup" mode on a tabletop). To my touch, almost room temperature.
"far too hot to touch" it NOT normal.

"far too hot to touch" it NOT normal"

....for YOUR system, perhaps. I've seen systems where the RAM runs at
about room temperature, but not lately.

 

[VanguardLH]

"far too hot to touch" it NOT normal"

...for YOUR system, perhaps. I've seen systems where the RAM runs at
about room temperature, but not lately.

Chris didn't mention if he has DDR, DDR2, DDR3, at what clock rate the
memory module(s) were rated, what voltage they were running, or if he is
overclocking or not (probably not). Can't tell what Chris has so we
cannot know what would be the *operating* temperature range of his
memory. In addition, we don't know what app(s) Chris was running at the
time (i.e., how saturated was the memory bus regarding max bandwidth for
his modules). His computer might've been sitting rather idle just
displaying the Windows desktop versus running something like Prime95 or
playing a video game where the screen content changes a LOT requiring
high bandwidth. We also don't know what cooling Chris has in his case.
The OP might just have the backside fans (sometimes just the PSU fan(s)
but sometimes with an optional backpanel fan). Chris might have a front
panel fan along with cabling that is out of the way or otherwise better
air flow over the memory modules. He should check his memory temps when
only the PSU fan is providing air movement while playing a fast-paced
video game or running a continuous memory benchmark. However, we also
don't know what the OP has for cooling or what apps he was running. So
far, we're comparing an unknown to another unknown.

 

[Paul]

A memory module running at 85C (which is WITHIN the operational
temperature range of a DDR3 module) will obviously feel hot.

Again, this may not be the only aspect of temperature.

First of all, if the ambient was high (computer case has a sock
stuffed in the vent, and there is no air movement), and the
air temperature is 50C inside the computer, the RAM module,
with its delta_T, could end up at an unnaturally high temperature.
In such a combination of circumstances, I don't necessarily see immediate
damage as the outcome. The voltage level is normal, and the case
temperature is higher than it should be. If you leave it like that
for five to ten years, yes, it may eventually result in a
shortened life. But not in an overnight accident (your sock falls
into the intake vent).

When the computer case is adequately cooled, and yet, the RAM
module has a high temperature, it implies a high voltage is being
used. Say you had DDR3, with nominal 1.5V VDimm. And the motherboard
applies 1.9V to it. OK, the module gets warm. That's a *hint*
something is wrong. Now, the question is, how long does that
particular brand of RAM, withstand 1.9V applied. In the history
of RAM, there were RAM designs that were "high voltage compatible" (Winbond)
and ones that were not. Over on xtremesystems forums, you could
find idiots applying too much voltage to RAM modules, and they'd
report "it failed in my overnight test, and was dead in the morning".
For some brands of RAM, that's how we learn, how much is too much.

So this is not a "temperature induced failure" we're worrying about,
it's an electrical breakdown failure inside the RAM. The operating
temperature of the module, is hinting to me, that the voltage is
set wrong. And it's a voltage induced failure I'm worried about.

Just like, the Northwood processors, you could set VCore a little
higher, and if you selected just the right value, the processor
would die *instantly* on the next POST. These are things you learn,
from data mining the enthusiast sites. I learned, for example, that
I should not apply more than 1.4V long term, to my current processor,
if I expected it to have a long life. And that was learned... by
killing some processors, on an enthusiast site. Soon the word spread,
and if a noob came along and said "I got 4GHz at 1.5V", the others
would pipe up and tell him to turn it down to 1.4V or less.

So my advice here, is to see whether the high temperature makes
sense or not. If you're an enthusiast, you bought a "RAM booster"
(that was a plugin module, intended for abusing RAM sticks), and
you had a high operating temp, than both you and I would realize
what you were doing. Suiciding your RAM, for a high bench. Fine.
No problemo. But if you're an ordinary user, your RAM is hot,
that's a hint things are not adjusted the way you think
they're adjusted. It would be satisfying to understand why.
If you did that on purpose, well, that's your business.

Since other subsystems are "winking out", it implies something
funny is going on. Misadjusted chipset supply voltages can do
that (Vsb).

Paul

 

[VanguardLH]

Again, this may not be the only aspect of temperature.

First of all, if the ambient was high (computer case has a sock
stuffed in the vent, and there is no air movement), and the
air temperature is 50C inside the computer, the RAM module,
with its delta_T, could end up at an unnaturally high temperature.
In such a combination of circumstances, I don't necessarily see immediate
damage as the outcome. The voltage level is normal, and the case
temperature is higher than it should be. If you leave it like that
for five to ten years, yes, it may eventually result in a
shortened life. But not in an overnight accident (your sock falls
into the intake vent).

When the computer case is adequately cooled, and yet, the RAM
module has a high temperature, it implies a high voltage is being
used. Say you had DDR3, with nominal 1.5V VDimm. And the motherboard
applies 1.9V to it. OK, the module gets warm. That's a *hint*
something is wrong. Now, the question is, how long does that
particular brand of RAM, withstand 1.9V applied. In the history
of RAM, there were RAM designs that were "high voltage compatible" (Winbond)
and ones that were not. Over on xtremesystems forums, you could
find idiots applying too much voltage to RAM modules, and they'd
report "it failed in my overnight test, and was dead in the morning".
For some brands of RAM, that's how we learn, how much is too much.

So this is not a "temperature induced failure" we're worrying about,
it's an electrical breakdown failure inside the RAM. The operating
temperature of the module, is hinting to me, that the voltage is
set wrong. And it's a voltage induced failure I'm worried about.

Just like, the Northwood processors, you could set VCore a little
higher, and if you selected just the right value, the processor
would die *instantly* on the next POST. These are things you learn,
from data mining the enthusiast sites. I learned, for example, that
I should not apply more than 1.4V long term, to my current processor,
if I expected it to have a long life. And that was learned... by
killing some processors, on an enthusiast site. Soon the word spread,
and if a noob came along and said "I got 4GHz at 1.5V", the others
would pipe up and tell him to turn it down to 1.4V or less.

So my advice here, is to see whether the high temperature makes
sense or not. If you're an enthusiast, you bought a "RAM booster"
(that was a plugin module, intended for abusing RAM sticks), and
you had a high operating temp, than both you and I would realize
what you were doing. Suiciding your RAM, for a high bench. Fine.
No problemo. But if you're an ordinary user, your RAM is hot,
that's a hint things are not adjusted the way you think
they're adjusted. It would be satisfying to understand why.
If you did that on purpose, well, that's your business.

Since other subsystems are "winking out", it implies something
funny is going on. Misadjusted chipset supply voltages can do
that (Vsb).

Paul

All of which still points to *we* still don't know how hot was
"extremely hot". Hot to the touch is not necessarily outside of the
operating temperature range.

With all the points you made about why the memory was overheated or
over-volted, I have to wonder what condition the CPU is in, or for how
long the hard disks have been overheated. At this point, it seems the
OP needs to reduce the hardware complement to its minimal config and
start testing there, along with dusting it out and making sure any
cabling, especially flat cables, aren't blocking the airflow, and if the
case is constricted then adding another fan. Yet fried components are
going to heal so replacements are likely.

The OP said he dusted out his computer but hasn't mention since if the
memory modules still feel hot or not. Hopefully he also ensured
adequate airflow through the case. We'll have to wait until the OP
tells us what happened regarding temps.

 

[LouisG]

You seem to have a lot going on - I would start by running SiSoft Sandra
or similar which stress tests all your hardware and flags up anything
odd, there may be other utilities for free that do the same job but i
havnt been down this path for some time.
Others have suggested well but I dont think it is economic in time and
effort to correct a m/b at component level.
If you do go for a replacement m/b then getting a new one with XP
drivers now may be more difficult! If you can find an exact replacement
for pennies on ebay etc. then you can just swap out the the whole thing
and restart as before!
Cant remember but in the thread did your new powersupply start the cycle
of problems, apart from the old one's fan sticking? You can always with
the case open - point a good powerful cold fan blower on the m/b as a
test to cool things down and see what the effect is on your problems.

Actually , i noticed the fan problem when i went into the case to replace
the soundcard and clean out the dust bunnies from the case.

 

[LouisG]

All of which still points to *we* still don't know how hot was
"extremely hot". Hot to the touch is not necessarily outside of the
operating temperature range.

With all the points you made about why the memory was overheated or
over-volted, I have to wonder what condition the CPU is in, or for how
long the hard disks have been overheated. At this point, it seems the
OP needs to reduce the hardware complement to its minimal config and
start testing there, along with dusting it out and making sure any
cabling, especially flat cables, aren't blocking the airflow, and if
the case is constricted then adding another fan. Yet fried components
are going to heal so replacements are likely.

The OP said he dusted out his computer but hasn't mention since if the
memory modules still feel hot or not. Hopefully he also ensured
adequate airflow through the case. We'll have to wait until the OP
tells us what happened regarding temps.

The ram modules were hot , but not to hot to hold ,, i'm guessing by
whats been posted that they seem to be at the normal operation temps.
The case had a power supply that had a fan over the cpu and one out the
back of the case. The cpu has its own fan , an upgraded on from what came
with it , sort of a radiator apparatus with a fan attached to it , plus
there is a fan mounted in the front lower half of the case blowing over
the hd's and another mounted in the back of the case sucking air out of
the case.
I think i've got the cooling process of the computer covered. I still
think by the way its acting that there must be something wrong with the
mobo to have just the sound and usb problems.
To clarify , the usb ports mounted on the back of the system board work
fine , its the ones that are running off the board to the front of the
case and also a add on that fits in the floppy bay that has usb and sd
card readers. I forget the name of the company that makes it.

 

[glee]

The ram modules were hot , but not to hot to hold ,, i'm guessing by
whats been posted that they seem to be at the normal operation temps.
The case had a power supply that had a fan over the cpu and one out
the
back of the case. The cpu has its own fan , an upgraded on from what
came
with it , sort of a radiator apparatus with a fan attached to it ,
plus
there is a fan mounted in the front lower half of the case blowing
over
the hd's and another mounted in the back of the case sucking air out
of
the case.
I think i've got the cooling process of the computer covered. I still
think by the way its acting that there must be something wrong with
the
mobo to have just the sound and usb problems.
To clarify , the usb ports mounted on the back of the system board
work
fine , its the ones that are running off the board to the front of the
case and also a add on that fits in the floppy bay that has usb and sd
card readers. I forget the name of the company that makes it.

The USB ports on the back are integrated directly with the motherboard.
The front USB ports, as well as the adapter in the floppy bay, are
connected with a cable to separate headers on the motherboard. Those
headers could be faulty, or the cables connecting them could be wired in
the wrong sequence.... the header pin-outs for USB are not universal.

You say the sound is intermittent also whether using onboard audio or an
add-in sound card. Are you connecting speakers to the speaker port on
the rear of the computer, or are you connecting to a speaker port on the
front, when you use onboard sound?

Can I assume that when you install a sound card, you are connecting
speakers to the lime green speaker port on the back of the sound card?