I've posted to the Asus group, but as there is little traffic there, I might
not get a response.
ASUS m2v has a power connector with 12 pairs of receptors, as I guess most
of the newer boards do. After 2 or three weeks, computer failed to start.
Got a replacement PSU from vendor, it worked for about same amount of time,
then (apparently) *it* failed. A backup power supply with 10 pairs in its
connector is working now.
I attached the 12 type PSU to a backup 'puter with a 10 pin power connector,
and the "failed" unit worked, no prob.
What are the two extra pairs supposed to supply; what am I missing?
If not much is lost, nevertheless I'm concerned that given this one flaw
it's only a matter of time until more defects in mobo manifest, or it has an
irrecoverable failure. Your opinions, experience & insights would be much
appreciated.
--
RobF
Address anti-spammed
RobF wrote:
> I've posted to the Asus group, but as there is little traffic there, I might
> not get a response.
>
> ASUS m2v has a power connector with 12 pairs of receptors, as I guess most
> of the newer boards do. After 2 or three weeks, computer failed to start.
> Got a replacement PSU from vendor, it worked for about same amount of time,
> then (apparently) *it* failed. A backup power supply with 10 pairs in its
> connector is working now.
>
> I attached the 12 type PSU to a backup 'puter with a 10 pin power connector,
> and the "failed" unit worked, no prob.
>
> What are the two extra pairs supposed to supply; what am I missing?
> If not much is lost, nevertheless I'm concerned that given this one flaw
> it's only a matter of time until more defects in mobo manifest, or it has an
> irrecoverable failure. Your opinions, experience & insights would be much
> appreciated.
You can compare the 20 pin and 24 pin ATX specs here.
The four extra pins are (redundant) 3.3V, 5V, 12V, and GND. The only
time the 24 pin PSU is absolutely necessary, is if you have an SLI board,
and a couple video cards drawing 4A from the 12V rail from the video slots.
Otherwise, in many other situations, such as a motherboard with only one
video card slot, a 20 pin PSU plugged into the 24 pin socket is sufficient.
Motherboard power demands on 3.3V and 5V, are not great enough to strain
the existing multiple 3.3V and 5V pins and wires. And the separate 2x2
ATX12V connector for the processor, means the processor power does not
come through the main connector.
To understand your mystery, you'd need a clamp on DC ammeter, and measure the
actual current flow. (This would be a check for an overload, enough of an
overload to damage the supply.) An ordinary multimeter can be used, but
you'd need to buy and hack up an ATX power extender cable, to make current
measurements. The clamp-on type meter, is a non-contact method, and means
you don't need to butcher any wires.
You'd also want to measure the output voltages on the main connector, while
the thing was running - just in case there is a partial short between
rails, and one rail is pumping current into another rail.
This is the clamp-on AC/DC ammeter I use. There are probably better ones
than this by now. This design is a few years old. The jaws clamp around
one or more wires (like all the orange wires, at the same time, or the
red ones). The current flow of the combined wires, is summed together,
as the meter measures the magnetic field around the wires.
Many thanks, Paul. Apparently, I needn't worry about future glitches from
Asus, and will simply bend away the 11th and twelfth sockets (they're
hinged, on my honeycomb) and continue with what I've got.
--
RobF
Address anti-spammed
"Paul" <nospam@needed.com> wrote in message news:f3f1tm$ol3$1@aioe.org...
> RobF wrote:
>> I've posted to the Asus group, but as there is little traffic there, I
>> might not get a response.
>>
>> ASUS m2v has a power connector with 12 pairs of receptors, as I guess
>> most of the newer boards do. After 2 or three weeks, computer failed to
>> start. Got a replacement PSU from vendor, it worked for about same amount
>> of time, then (apparently) *it* failed. A backup power supply with 10
>> pairs in its connector is working now.
>>
>> I attached the 12 type PSU to a backup 'puter with a 10 pin power
>> connector, and the "failed" unit worked, no prob.
>>
>> What are the two extra pairs supposed to supply; what am I missing?
>> If not much is lost, nevertheless I'm concerned that given this one flaw
>> it's only a matter of time until more defects in mobo manifest, or it has
>> an irrecoverable failure. Your opinions, experience & insights would be
>> much appreciated.
>
>
> You can compare the 20 pin and 24 pin ATX specs here.
>
> http://www.formfactors.org/developer...X12V_1_3dg.pdf
> http://www.formfactors.org/developer...public_br2.pdf
>
> The four extra pins are (redundant) 3.3V, 5V, 12V, and GND. The only
> time the 24 pin PSU is absolutely necessary, is if you have an SLI board,
> and a couple video cards drawing 4A from the 12V rail from the video
> slots.
> Otherwise, in many other situations, such as a motherboard with only one
> video card slot, a 20 pin PSU plugged into the 24 pin socket is
> sufficient.
> Motherboard power demands on 3.3V and 5V, are not great enough to strain
> the existing multiple 3.3V and 5V pins and wires. And the separate 2x2
> ATX12V connector for the processor, means the processor power does not
> come through the main connector.
>
> To understand your mystery, you'd need a clamp on DC ammeter, and measure
> the
> actual current flow. (This would be a check for an overload, enough of an
> overload to damage the supply.) An ordinary multimeter can be used, but
> you'd need to buy and hack up an ATX power extender cable, to make current
> measurements. The clamp-on type meter, is a non-contact method, and means
> you don't need to butcher any wires.
>
> You'd also want to measure the output voltages on the main connector,
> while
> the thing was running - just in case there is a partial short between
> rails, and one rail is pumping current into another rail.
>
> This is the clamp-on AC/DC ammeter I use. There are probably better ones
> than this by now. This design is a few years old. The jaws clamp around
> one or more wires (like all the orange wires, at the same time, or the
> red ones). The current flow of the combined wires, is summed together,
> as the meter measures the magnetic field around the wires.
>
> http://exphil.com/images/products/Extech/380947.jpg
>
> Paul
ATX Power 
Linear Mode
