Thermaltake Volcano CPU Fan and my work-around

[Lawrence]

I haven't found anything similar to my following post yet.... looking for
educated opinions....

Here's what happened -
I shut off my PC for the first time in months and months the other day to
accommodate some furniture re-arrangements.
The machine was off all night. The next morning, I reconnected things, and
turned it on. Everything came up just fine, I checked my email, then left
the computer and went about putting furniture pieces back in their
places.... After a few moments, I noticed an odor, like hot dust... which
slowly turned into a stinky plasticy smell....

As the only thing powered up, was my computer, I checked things out (using
PC Alert: temp, cpu fan RPMs, etc.) and things looked ok. But that stinky
smell never means anything good when electronics are involved, so I shut
things down, and took off my chassis cover to inspect the innards of the PC.
No loose wires, no melted plastic, nothing. Just that (by now) pungent
stink.

With the cover off, I turned on the machine and watched the PSU, GPU, and
CPU fans all spin up. Windows XP booted fine again, but after a moment or
so, I noticed the Thermaltake Volcano 7+ fan slow, then stop, then just sort
of twitter. PC Alert kicked in, and reported the CPU fan RPMs at 0, and
sounded the audio alarm. (Nice to know that the software and sensor actually
works! - Although it sounded like a car alarm, and had I not been right at
my PC, I would have thought it was just that, and probably not been worried
that my CPU was at risk of frying. I think I am gonna change the wave file
that plays... -FYI)

Needless to say, a lot of cussing, and troubleshooting ensued...
Here is my 'solution' to get things going without replacing any hardware
(yet):

There were two leads from the Thermaltake 1) has just the sensor wire, and
was plugged into a 3-pinned JP site on my motherboard, 2) had just 12+ volt
and 1 ground (in a 3-holed housing), and adapted to one of the many power
supply leads (I tried moving it around to several different leads, but the
fan would always shut off after a few moments).

Then, with the computer on, I simply unplugged the power lead to the fan,
and plugged it back in... the fan would work for 10 or 15 seconds then stop
and twitter.... unplug then plug back in... start, stop, and twitter...
consistently.

I carefully read my manual, and cruised MSI's site for any info regarding
the (thankfully) redundant option of connecting the CPU fan directly to the
motherboard, and found no indication that there were any issues with doing
this. I cruised Thermaltake's website and found it to frustratingly bereft
of any decent troubleshooting steps, or signs and symptoms of problems. It
was no help whatsoever.

So, going with the notion that there was something screwy with the what
controls output to the power supply leads for internal devices, I carefully
removed the single sensor wire from its housing, and inserted it into the
empty one on the 2-wire fan-power housing, then plugged it into the CPU fan
JP site on the motherboard.

It worked! Great! I have been putting the machine through its paces for
the last couple of days without ANY failures: Unreal Tournament, Unreal II,
Nvidia Earthviewer, Sisoft Benchmarks, and FutureMark's Benchmark. All with
smooth sailing.


I am supposing that I should consider replacing my power supply....
[it is ample on paper, but it is supporting the following:]
AMD 1700+ OCed to 1900+
MSI Motherboard (ms-6380)
1 Gig DDR RAM
2 Hard drives
DVD Drive
CD-RW Drive
A seldom used floppy drive
Win TV Card
Geforce MX 440
NIC card for DSL
56k Fax/Modem
6 USB ports of which I use 3
2 9-pin serial ports of which I use one
1 game port of which I use one for MIDI


In general, it has been my experience that the Power Supply is more
susceptible to damage than internal components, as it is on the 'front line'
because it is connected to the power source. Of course, that is assuming
that you have cooling, and a decent case, and stable hardware inside to
begin with.

Sorry for a long-winded post.... But I am curious what others think about
my situation. The only times I have ever had problems with computers of
this magnitude, it has been when I turned them on from a cold state- not
while they are already running... that's the main reason I keep mine on all
the time. You typically only see lightbulbs blow when you turn them on,
too.... very rarely will one fail after being on for hours.... cold metal
wire infused with structure-buckling electrons..... and that's when
something with no moving parts can break.

Thanks in advance for ideas, warnings, and suggestions.
-Lawrence

 

[Paul]

Lawrence,
If I am understanding you correctly, you are plugging the fan directly into the
motherboard for the 12V power supply? If this is the case, have you double
checked to see what the power rating of that connection is? My motherboard
states it can only support a fan drawing up to 350mA. The Volcano 7+ can draw
up to 1+ Amps.

If you are plugging the fan directly into the 4-pin drive power connection, how
can you stand the load noise? I have my Volcano 7+ plugged into the resistor
divider (known as the speed control) and am running it in slow speed to keep the
noise level down.

As far as the 10 to 15 second and then spin down goes, the plug itself my have a
bad or corroded contact. I had a situation with a 4-pin drive splitter that
would intermittently cause a loss of continuity and the drive would spin down
and be unavailable until I rebooted. It took me a while to find it because the
action of taken the side cover off the tower would giggle the connector and
power would be reapplied.

Lastly, this type of problem would probably have been more suited for the MSI
mainboard NG since this the fan does not pertain to XP in any way.

Paul
P.S. I am snip your original message out from my reply for brevity
-- snip --

 

[Lawrence]

Hi Paul, et al.

I'll keep this brief, since I posed it in the wrong NG....

I have plugged the fan into the CPU fan (3-pin) jumper labeled on the mobo -
as opposed to using the adapter to the 4-pin drive connector. Yes. I am
using the motherboard's designated site for power and sensor.

The previous connection - per Thermaltake's instructions *was* via an
adapter, through the 4-pin drive connector, with the lone sensor wire going
to the CPU fan sensor pin.

As regards possible contact corrosion, I am going to log off now, and look
for that.... but I think I gave things a pretty thorough going over while
trying to identify the problem originally, prior to trying what I eventually
did.

If I find anything new, I'll post one last note here in this NG, regarding
it.

Thanks for the info, though.
-L

 

[Lawrence]

Hmmm... no corrosion... thoroughly searched....
I haven't found the Amp specs for the CPU Jumper in MSI land yet.
I am happy that things are stable, and working.... but I certainly don't
want a time-bomb.... It's clear that something went wrong somewhere.

Right now, the primary suspect is the Thermaltake... but I have no
Voltmeter, and nothing *looks* fried -- although I confess that I have NOT
removed the fan from the CPU.... I don't have any Thermal Grease to put back
on it.... soooooo.

Anyway, I agree with your cautionary tone. Just because you can stick a
fork in an outlet as long as you don't touch it, doesn't mean that you
should.....

I guess I'll take the box into my local vendor, and have them check things
out.....

Still, so far, no problems, no temperature spikes, and smoke... etc.

Thanks!

I'll try posing the question in a more MSI-related place.
-L

 

[Lawrence]

Well crud....

I found the info. what I was looking for.... and although I am not happy
about it.... it makes more sense, now.

I found postings in *alt.comp.periphs.mainboard.msi-microstar* that roughly
pertained to me.

I do not have a Volcano 7+... I have a Volcano 7 (which has no manual speed
switch - which explains why I couldn't find a switch anywhere.) If I am
lucky, the CPU fan jumper I am using will last until tomorrow afternoon,
until I can get a new Volcano fan.

The rated consumption for my fan is .18 - .45amps, so hopefully I can keep
it on the low end of that range until the fan is replaced. I now think the
Temperature Control Sensor is what failed, based on what I've read.
http://www.thermaltake.com/heatsink/v7.htm

Funny... it hasn't peeped at all, and I have been hammering it for the last
few days, with the most intense games I've got. And it's been fine. Normal
temps... maybe I've just been lucky so far.

Thanks for the correspondance!
-Lawrence

 

[Lawrence]

Replaced the flakey Volcano 7, with a new Volcano 7+.

 

[Paul]

I did not catch the fact that you had the standard Volcano 7. I thought you
mentioned it was a 7+ that is why I was responding. Now that you clarified
that, I would tend to agree with your assessment about the temp. sensor.

Are you running the 7+ with the speed controller? What speed do you have it set
at if you are using it? What temps. do you see?

Paul

 

[Lawrence]

It was my mistake regarding the "7+", initially... I thought that's what I
had.

[posted for those who happen to not know]
The Volcano 7 has a larger fan, an aluminum heat-sink, and automated RPM
control (the sensor for which is what seems to have died)
The Volcano 7+ has a slightly smaller fan, but a copper heat sink, and a
manual 3-setting switch. There are FAR MORE convolutions on the copper heat
sink, and each one is much much thinner than on the aluminum one (which
seems crude by comparison). I bet the actual cooling surface area is at
least 5 times greater on it than on the aluminum one. It must be far better
at dissipating heat, and the slightly smaller fan is more than ample to just
blow that heat away.

I found the High-speed setting on the new fan to make things a bit too loud,
though not unbearable.
The Low-speed setting was virtually silent, but as the Medium setting wasn't
much louder -- I chose it for safety's sake.

It's very warm here in Seattle right now.
With the switch set to Medium, it's running at ~4750 RPMs.
The temp in my apt. is 31C, the case-air temp is 42C, and the CPU temp is
48C.

I bought that Volcano 7 13 months ago.... although it still works, its
power connection is screwy, and it doesn't like to be connected the way it
was designed to be. Wouldn't you know it.... just past the 1 Year mark, and
it fails. Maybe the sensor was smarter than I thought.

Good luck all,
-Lawrence

 

[Lawrence]

I forgot to mention Overclocking...

Mine is a locked XP 1700+ running at 143FSB (1600MHz, instead of 1467), so
it shows up as 1900+, in WCPUID (available at http://www.h-oda.com/, for
those interested in testing/spec-ing their own CPU)

-L

Thanks for the feedback,

I have keep mine on low speed. The CPU diode is at 46C, the mobo is at 30C
ambient is probably around 25C. The fan is at 3550RPM.

Under full, continuous loading, the CPU never gets above 51C. The CPU is a 1800+
with no OC'ing.

Paul

It was my mistake regarding the "7+", initially... I thought that's what I
had.

[posted for those who happen to not know]
The Volcano 7 has a larger fan, an aluminum heat-sink, and automated RPM
control (the sensor for which is what seems to have died)
The Volcano 7+ has a slightly smaller fan, but a copper heat sink, and a
manual 3-setting switch. There are FAR MORE convolutions on the copper heat
sink, and each one is much much thinner than on the aluminum one (which
seems crude by comparison). I bet the actual cooling surface area is at
least 5 times greater on it than on the aluminum one. It must be far better
at dissipating heat, and the slightly smaller fan is more than ample to just
blow that heat away.

I found the High-speed setting on the new fan to make things a bit too loud,
though not unbearable.
The Low-speed setting was virtually silent, but as the Medium setting wasn't
much louder -- I chose it for safety's sake.

It's very warm here in Seattle right now.
With the switch set to Medium, it's running at ~4750 RPMs.
The temp in my apt. is 31C, the case-air temp is 42C, and the CPU temp is
48C.

I bought that Volcano 7 13 months ago.... although it still works, its
power connection is screwy, and it doesn't like to be connected the way it
was designed to be. Wouldn't you know it.... just past the 1 Year mark, and
it fails. Maybe the sensor was smarter than I thought.

Good luck all,
-Lawrence

I did not catch the fact that you had the standard Volcano 7. I

thought
you

mentioned it was a 7+ that is why I was responding. Now that you clarified
that, I would tend to agree with your assessment about the temp. sensor.

Are you running the 7+ with the speed controller? What speed do you

have
it set

at if you are using it? What temps. do you see?

Paul

Replaced the flakey Volcano 7, with a new Volcano 7+.

 

[Lawrence]

dangit.... I meant 145 Front Side Bus... not 143... not that it really
matters... (but if you're trying to calculate what the heck I meant....)

I forgot to mention Overclocking...

Mine is a locked XP 1700+ running at 143FSB (1600MHz, instead of 1467), so
it shows up as 1900+, in WCPUID (available at http://www.h-oda.com/, for
those interested in testing/spec-ing their own CPU)

-L

Thanks for the feedback,

I have keep mine on low speed. The CPU diode is at 46C, the mobo is at 30C
ambient is probably around 25C. The fan is at 3550RPM.

Under full, continuous loading, the CPU never gets above 51C. The CPU is

a

1800+
with no OC'ing.

Paul

what

I

had.

[posted for those who happen to not know]
The Volcano 7 has a larger fan, an aluminum heat-sink, and automated RPM
control (the sensor for which is what seems to have died)
The Volcano 7+ has a slightly smaller fan, but a copper heat sink, and a
manual 3-setting switch. There are FAR MORE convolutions on the

copper

heat

to

just

way

it

mark,

and

 

[Murray McNeill]

I am very surprised that the header on the motherboard had enough power
to run the fan on a Volcano 7. The reason for the split connectors was
to give the fan the full power of a hard drive connector, yet still give
the motherboard the RPM signal it needs. I have read of people blowing
the CPU fan header on their systems, and being forced to run it the way
yours was originally wired.

I haven't found anything similar to my following post yet.... looking for
educated opinions....

Here's what happened -
I shut off my PC for the first time in months and months the other day to
accommodate some furniture re-arrangements.
The machine was off all night. The next morning, I reconnected things, and
turned it on. Everything came up just fine, I checked my email, then left
the computer and went about putting furniture pieces back in their
places.... After a few moments, I noticed an odor, like hot dust... which
slowly turned into a stinky plasticy smell....

As the only thing powered up, was my computer, I checked things out (using
PC Alert: temp, cpu fan RPMs, etc.) and things looked ok. But that stinky
smell never means anything good when electronics are involved, so I shut
things down, and took off my chassis cover to inspect the innards of the PC.
No loose wires, no melted plastic, nothing. Just that (by now) pungent
stink.

With the cover off, I turned on the machine and watched the PSU, GPU, and
CPU fans all spin up. Windows XP booted fine again, but after a moment or
so, I noticed the Thermaltake Volcano 7+ fan slow, then stop, then just sort
of twitter. PC Alert kicked in, and reported the CPU fan RPMs at 0, and
sounded the audio alarm. (Nice to know that the software and sensor actually
works! - Although it sounded like a car alarm, and had I not been right at
my PC, I would have thought it was just that, and probably not been worried
that my CPU was at risk of frying. I think I am gonna change the wave file
that plays... -FYI)

Needless to say, a lot of cussing, and troubleshooting ensued...
Here is my 'solution' to get things going without replacing any hardware
(yet):

There were two leads from the Thermaltake 1) has just the sensor wire, and
was plugged into a 3-pinned JP site on my motherboard, 2) had just 12+ volt
and 1 ground (in a 3-holed housing), and adapted to one of the many power
supply leads (I tried moving it around to several different leads, but the
fan would always shut off after a few moments).

Then, with the computer on, I simply unplugged the power lead to the fan,
and plugged it back in... the fan would work for 10 or 15 seconds then stop
and twitter.... unplug then plug back in... start, stop, and twitter...
consistently.

I carefully read my manual, and cruised MSI's site for any info regarding
the (thankfully) redundant option of connecting the CPU fan directly to the
motherboard, and found no indication that there were any issues with doing
this. I cruised Thermaltake's website and found it to frustratingly bereft
of any decent troubleshooting steps, or signs and symptoms of problems. It
was no help whatsoever.

So, going with the notion that there was something screwy with the what
controls output to the power supply leads for internal devices, I carefully
removed the single sensor wire from its housing, and inserted it into the
empty one on the 2-wire fan-power housing, then plugged it into the CPU fan
JP site on the motherboard.

It worked! Great! I have been putting the machine through its paces for
the last couple of days without ANY failures: Unreal Tournament, Unreal II,
Nvidia Earthviewer, Sisoft Benchmarks, and FutureMark's Benchmark. All with
smooth sailing.


I am supposing that I should consider replacing my power supply....
[it is ample on paper, but it is supporting the following:]
AMD 1700+ OCed to 1900+
MSI Motherboard (ms-6380)
1 Gig DDR RAM
2 Hard drives
DVD Drive
CD-RW Drive
A seldom used floppy drive
Win TV Card
Geforce MX 440
NIC card for DSL
56k Fax/Modem
6 USB ports of which I use 3
2 9-pin serial ports of which I use one
1 game port of which I use one for MIDI


In general, it has been my experience that the Power Supply is more
susceptible to damage than internal components, as it is on the 'front line'
because it is connected to the power source. Of course, that is assuming
that you have cooling, and a decent case, and stable hardware inside to
begin with.

Sorry for a long-winded post.... But I am curious what others think about
my situation. The only times I have ever had problems with computers of
this magnitude, it has been when I turned them on from a cold state- not
while they are already running... that's the main reason I keep mine on all
the time. You typically only see lightbulbs blow when you turn them on,
too.... very rarely will one fail after being on for hours.... cold metal
wire infused with structure-buckling electrons..... and that's when
something with no moving parts can break.

Thanks in advance for ideas, warnings, and suggestions.
-Lawrence

 

[Lawrence]

Yes.... I have read concerns similar to yours since having replaced the fan
with a sporting new 7+...

But it purred along just fine plugged directly into one of the mobo headers.
At slower speed: 3068 rpms.... but not a peep out of it. And in warm
weather, too. But I only ran it that way for a little over 24 hours.

I just hated the nagging thought of an *eventual* *untimely* burnout of the
header at a later date... plus I don't like it when I know there's something
wrong inside the case.

It's an MSI board. 14 months old. K7T266 Pro2 version 2.0
It has always done whatever I have asked it.

Incidentally, the old number 7 has been gutted, and is now a 'personal fan'
on my desk. Works fine.

Thanks for the feedback,
-Lawrence in Seattle

I am very surprised that the header on the motherboard had enough power
to run the fan on a Volcano 7. The reason for the split connectors was
to give the fan the full power of a hard drive connector, yet still give
the motherboard the RPM signal it needs. I have read of people blowing
the CPU fan header on their systems, and being forced to run it the way
yours was originally wired.

I haven't found anything similar to my following post yet.... looking for
educated opinions....

Here's what happened -
I shut off my PC for the first time in months and months the other day to
accommodate some furniture re-arrangements.
The machine was off all night. The next morning, I reconnected things, and
turned it on. Everything came up just fine, I checked my email, then left
the computer and went about putting furniture pieces back in their
places.... After a few moments, I noticed an odor, like hot dust... which
slowly turned into a stinky plasticy smell....

As the only thing powered up, was my computer, I checked things out (using
PC Alert: temp, cpu fan RPMs, etc.) and things looked ok. But that stinky
smell never means anything good when electronics are involved, so I shut
things down, and took off my chassis cover to inspect the innards of the PC.
No loose wires, no melted plastic, nothing. Just that (by now) pungent
stink.

With the cover off, I turned on the machine and watched the PSU, GPU, and
CPU fans all spin up. Windows XP booted fine again, but after a moment or
so, I noticed the Thermaltake Volcano 7+ fan slow, then stop, then just sort
of twitter. PC Alert kicked in, and reported the CPU fan RPMs at 0, and
sounded the audio alarm. (Nice to know that the software and sensor actually
works! - Although it sounded like a car alarm, and had I not been right at
my PC, I would have thought it was just that, and probably not been worried
that my CPU was at risk of frying. I think I am gonna change the wave file
that plays... -FYI)

Needless to say, a lot of cussing, and troubleshooting ensued...
Here is my 'solution' to get things going without replacing any hardware
(yet):

There were two leads from the Thermaltake 1) has just the sensor wire, and
was plugged into a 3-pinned JP site on my motherboard, 2) had just 12+ volt
and 1 ground (in a 3-holed housing), and adapted to one of the many power
supply leads (I tried moving it around to several different leads, but the
fan would always shut off after a few moments).

Then, with the computer on, I simply unplugged the power lead to the fan,
and plugged it back in... the fan would work for 10 or 15 seconds then stop
and twitter.... unplug then plug back in... start, stop, and twitter...
consistently.

I carefully read my manual, and cruised MSI's site for any info regarding
the (thankfully) redundant option of connecting the CPU fan directly to the
motherboard, and found no indication that there were any issues with doing
this. I cruised Thermaltake's website and found it to frustratingly bereft
of any decent troubleshooting steps, or signs and symptoms of problems. It
was no help whatsoever.

So, going with the notion that there was something screwy with the what
controls output to the power supply leads for internal devices, I carefully
removed the single sensor wire from its housing, and inserted it into the
empty one on the 2-wire fan-power housing, then plugged it into the CPU fan
JP site on the motherboard.

It worked! Great! I have been putting the machine through its paces for
the last couple of days without ANY failures: Unreal Tournament, Unreal II,
Nvidia Earthviewer, Sisoft Benchmarks, and FutureMark's Benchmark. All with
smooth sailing.


I am supposing that I should consider replacing my power supply....
[it is ample on paper, but it is supporting the following:]
AMD 1700+ OCed to 1900+
MSI Motherboard (ms-6380)
1 Gig DDR RAM
2 Hard drives
DVD Drive
CD-RW Drive
A seldom used floppy drive
Win TV Card
Geforce MX 440
NIC card for DSL
56k Fax/Modem
6 USB ports of which I use 3
2 9-pin serial ports of which I use one
1 game port of which I use one for MIDI


In general, it has been my experience that the Power Supply is more
susceptible to damage than internal components, as it is on the 'front line'
because it is connected to the power source. Of course, that is assuming
that you have cooling, and a decent case, and stable hardware inside to
begin with.

Sorry for a long-winded post.... But I am curious what others think about
my situation. The only times I have ever had problems with computers of
this magnitude, it has been when I turned them on from a cold state- not
while they are already running... that's the main reason I keep mine on all
the time. You typically only see lightbulbs blow when you turn them on,
too.... very rarely will one fail after being on for hours.... cold metal
wire infused with structure-buckling electrons..... and that's when
something with no moving parts can break.

Thanks in advance for ideas, warnings, and suggestions.
-Lawrence