Vista - hardware problems or virus

On Aug 14, 10:34 am, K8 notsogrand
<K8notsogr...@discussions.microsoft.com> wrote:
> I'm not trying to be rude or anything.
> But your saying even if the computer is up and working fine. I can still
> have a bad PSU

Yes. For example, a power supply can have failing electrolytic
capacitors. Therefore its ripple voltage is excessive and its voltage
regulation is poor. The computer will work just fine except when load
is maximum AND will fail months later as the electrolytics fail. The
power supply is already defective. Computer still boots OK. Meter
can identify that failure during the max load (multitasking) test -
especially when the failure only creates a rare computer crash.

Another defective power supply works just fine in another computer
but will not work in this one. It may be undersized - but just large
enough to run that other computer. Its signaling voltages may be
defective, just enough to be seen OK by another computer, but not
properly read by this computer's power supply controller.

Two example of how a power supply can work and yet still be
defective. In the second example, one would blame the power supply
controller rather than a defective power supply. Also an example of
why shotgunning can lead to total confusion.

This could explane my problems i'm having. (Thanks)

Could it be that the old PSU had damaged somthing (vidio card, hard drive,)?
I better get my hands on a meter to see if the new PSU is even working right.


I can't see it being undersized since the new one is 550w ATX. (the old one
was 500w)Or did I even hook it up right?
The new one has one 20 pin and two 4 pin connecters for the motherboard.
but I only room for one of the 4 pins. One has two yellow and two black
wires (this is the one I hooked up) and the second 4 pin has Red, black,
ornage and yellow.

Now the motherboard manual says that the 4pin should be two +12v and two
ground.

Is there a way to test the max load with out a meter?



"w_tom" wrote:

> On Aug 14, 10:34 am, K8 notsogrand
> <K8notsogr...@discussions.microsoft.com> wrote:
> > I'm not trying to be rude or anything.
> > But your saying even if the computer is up and working fine. I can still
> > have a bad PSU
>
> Yes. For example, a power supply can have failing electrolytic
> capacitors. Therefore its ripple voltage is excessive and its voltage
> regulation is poor. The computer will work just fine except when load
> is maximum AND will fail months later as the electrolytics fail. The
> power supply is already defective. Computer still boots OK. Meter
> can identify that failure during the max load (multitasking) test -
> especially when the failure only creates a rare computer crash.
>
> Another defective power supply works just fine in another computer
> but will not work in this one. It may be undersized - but just large
> enough to run that other computer. Its signaling voltages may be
> defective, just enough to be seen OK by another computer, but not
> properly read by this computer's power supply controller.
>
> Two example of how a power supply can work and yet still be
> defective. In the second example, one would blame the power supply
> controller rather than a defective power supply. Also an example of
> why shotgunning can lead to total confusion.
>
>

If a power supply creates damage to any computer part, then another
function was missing inside that power supply. To damage a video card
or hard drive, then voltage must be excessive. But even power
supplies 30 years ago contained separate circuits so that overvoltage
would not happen. It is required in all supplies today. However, with
so many A+ Certified computer techs having no electrical knowledge,
the market is chock full of power supplies missing essential functions
and selling only on watts and price. To dump supplies missing
essential functions, a supply manufacturer need only forget to provide
that long list of numeric specs. What did your supply manufacturer
do?

Take the 500 watt number. That could be a 350 watt supply measured
differently. They did not lie. They only (intentionally?) deceived.
Again, they can do what in a market where computer assemblers have
little electrical knowledge - don't even know how to read spec
numbers. Then a supply may not even do 350 watts. Put the computer
into maximum power consumption mode (multitask to all peripherals) and
take voltages on orange, red, yellow, and purple wires. If power
supply is sufficient, then voltages will remain above 3.23, 4.87, and
11.7 volts. Best way to confirm size of a supply - measure it under
full load.

No conductor has same electricity at both ends. CPUs can go from
consuming less than 1 amp to demanding tens of amps - in only
microseconds. Therefore PC traces through motherboard are too long.
We put 12 volts at the CPU with that 4 (black and yellow) wire next to
CPU's power supply. To provide sufficient power fast enough, a CPU
has a power supply adjacent and a dedicated 12 volt power source.
That four wire connector would connect somewhere next to CPU.

We test power supplies even better outside the computer. That is
hundreds of dollars in test equipment with dynamic loads, cooling for
those loads, and measurements even for response times. We also use an
oscilloscope. Either you do that, or you do something hundreds of
times less expensive. Using the meter is the only useful test you
have without major expenses. It must be 3.5 digits because an analog
meter is not sufficiently accurate. If your motherboard has a voltage
monitor, then the meter is necessary to calibrate that monitor.

On Aug 15, 1:46 pm, K8 notsogrand
<K8notsogr...@discussions.microsoft.com> wrote:
> This could explane my problems i'm having. (Thanks)
>
> Could it be that the old PSU had damaged somthing (vidio card, hard drive,)?
> I better get my hands on a meter to see if the new PSU is even working right.
>
> I can't see it being undersized since the new one is 550w ATX. (the old one
> was 500w)Or did I even hook it up right?
> The new one has one 20 pin and two 4 pin connecters for the motherboard.
> but I only room for one of the 4 pins. One has two yellow and two black
> wires (this is the one I hooked up) and the second 4 pin has Red, black,
> ornage and yellow.
>
> Now the motherboard manual says that the 4pin should be two +12v and two
> ground.
>
> Is there a way to test the max load with out a meter?

Actually for a power supply to cause problems the voltage can be
low or not clean enough of a wave length (tough generally computers
are fairly forgiving of either of these issues). High voltage or a high
forced amperage can also cause issues, especially spikes.

"w_tom" <w_tom1@usa.net> wrote in message news:1187287730.604173.204080@19g2000hsx.googlegroups.com...
> If a power supply creates damage to any computer part, then another
> function was missing inside that power supply. To damage a video card
> or hard drive, then voltage must be excessive. But even power
> supplies 30 years ago contained separate circuits so that overvoltage
> would not happen. It is required in all supplies today. However, with
> so many A+ Certified computer techs having no electrical knowledge,
> the market is chock full of power supplies missing essential functions
> and selling only on watts and price. To dump supplies missing
> essential functions, a supply manufacturer need only forget to provide
> that long list of numeric specs. What did your supply manufacturer
> do?
>
> Take the 500 watt number. That could be a 350 watt supply measured
> differently. They did not lie. They only (intentionally?) deceived.
> Again, they can do what in a market where computer assemblers have
> little electrical knowledge - don't even know how to read spec
> numbers. Then a supply may not even do 350 watts. Put the computer
> into maximum power consumption mode (multitask to all peripherals) and
> take voltages on orange, red, yellow, and purple wires. If power
> supply is sufficient, then voltages will remain above 3.23, 4.87, and
> 11.7 volts. Best way to confirm size of a supply - measure it under
> full load.
>
> No conductor has same electricity at both ends. CPUs can go from
> consuming less than 1 amp to demanding tens of amps - in only
> microseconds. Therefore PC traces through motherboard are too long.
> We put 12 volts at the CPU with that 4 (black and yellow) wire next to
> CPU's power supply. To provide sufficient power fast enough, a CPU
> has a power supply adjacent and a dedicated 12 volt power source.
> That four wire connector would connect somewhere next to CPU.
>
> We test power supplies even better outside the computer. That is
> hundreds of dollars in test equipment with dynamic loads, cooling for
> those loads, and measurements even for response times. We also use an
> oscilloscope. Either you do that, or you do something hundreds of
> times less expensive. Using the meter is the only useful test you
> have without major expenses. It must be 3.5 digits because an analog
> meter is not sufficiently accurate. If your motherboard has a voltage
> monitor, then the meter is necessary to calibrate that monitor.
>
> On Aug 15, 1:46 pm, K8 notsogrand
> <K8notsogr...@discussions.microsoft.com> wrote:
>> This could explane my problems i'm having. (Thanks)
>>
>> Could it be that the old PSU had damaged somthing (vidio card, hard drive,)?
>> I better get my hands on a meter to see if the new PSU is even working right.
>>
>> I can't see it being undersized since the new one is 550w ATX. (the old one
>> was 500w)Or did I even hook it up right?
>> The new one has one 20 pin and two 4 pin connecters for the motherboard.
>> but I only room for one of the 4 pins. One has two yellow and two black
>> wires (this is the one I hooked up) and the second 4 pin has Red, black,
>> ornage and yellow.
>>
>> Now the motherboard manual says that the 4pin should be two +12v and two
>> ground.
>>
>> Is there a way to test the max load with out a meter?
>

On Aug 16, 2:25 pm, "Spirit" <no...@notthere.net> wrote:
> Actually for a power supply to cause problems the voltage can be
> low or not clean enough of a wave length (tough generally computers
> are fairly forgiving of either of these issues). High voltage or a high
> forced amperage can also cause issues, especially spikes.

Power supplies must maintain perfectly in-spec power even when AC
mains voltage drops so low that light bulbs are only at 40%
intensity. Even Intel specs for power supplies make that demand.
That is another function of power supplies - to make AC power problems
irrelevant. But again, when computer assemblers only buy on dollars
and watts, then dumped into the market are many supplies missing those
required functions.

Let's look at the output of a computer grade UPS when in battery
backup mode. That 120 volt modified sine wave is ... two 200 volts
square waves with up to a 270 volt spike between those square waves.
Is that clean electricity? Of course not. Why does that one UPS not
cause computer problems? Again, power supplies make such power
problems irrelevant. That is required.

That UPS outputs power so 'dirty' as to damage some small electric
motors. Same electricity is perfectally acceptable to properly
constructed computers. Just another function that must be inside that
power supply.

However we don't let such power problems remain. For example, if
wall receptacles are installed with wires pushed into the back, then
intermittents can occur that exceed power supply functions. Whereas
'push in the back' wiring is acceptable for lights, it is completely
unacceptable for computers. A quick inspection by removing receptacle
cover plates will confirm good wiring wrapped around and secured by a
side mounted screw. Using the screw to clamp wires is important so
that power supply can make other electrical problems irrelevant.

But when a supply is failing, the computer may still boot. Only way
to find that failure is using a tool so ubiquitous as to be sold even
in K-mart, Sears, Radio Shack, Wal-mart, hardware stores, Lowes, etc.
That 3.5 digit multimeter costs only $20 - less in Wal-mart and when
on sale. Two mnutes with the meter means previous posts were
unnecessary beause meter numbers are so definitive - not subjective..

Thank's for your reply
But I think I'm going to take in and get it looked at.
Becouse if the new supply is not working right, theres not much I can do
about it.
Besides buy a new one.



"w_tom" wrote:

> On Aug 16, 2:25 pm, "Spirit" <no...@notthere.net> wrote:
> > Actually for a power supply to cause problems the voltage can be
> > low or not clean enough of a wave length (tough generally computers
> > are fairly forgiving of either of these issues). High voltage or a high
> > forced amperage can also cause issues, especially spikes.
>
> Power supplies must maintain perfectly in-spec power even when AC
> mains voltage drops so low that light bulbs are only at 40%
> intensity. Even Intel specs for power supplies make that demand.
> That is another function of power supplies - to make AC power problems
> irrelevant. But again, when computer assemblers only buy on dollars
> and watts, then dumped into the market are many supplies missing those
> required functions.
>
> Let's look at the output of a computer grade UPS when in battery
> backup mode. That 120 volt modified sine wave is ... two 200 volts
> square waves with up to a 270 volt spike between those square waves.
> Is that clean electricity? Of course not. Why does that one UPS not
> cause computer problems? Again, power supplies make such power
> problems irrelevant. That is required.
>
> That UPS outputs power so 'dirty' as to damage some small electric
> motors. Same electricity is perfectally acceptable to properly
> constructed computers. Just another function that must be inside that
> power supply.
>
> However we don't let such power problems remain. For example, if
> wall receptacles are installed with wires pushed into the back, then
> intermittents can occur that exceed power supply functions. Whereas
> 'push in the back' wiring is acceptable for lights, it is completely
> unacceptable for computers. A quick inspection by removing receptacle
> cover plates will confirm good wiring wrapped around and secured by a
> side mounted screw. Using the screw to clamp wires is important so
> that power supply can make other electrical problems irrelevant.
>
> But when a supply is failing, the computer may still boot. Only way
> to find that failure is using a tool so ubiquitous as to be sold even
> in K-mart, Sears, Radio Shack, Wal-mart, hardware stores, Lowes, etc.
> That 3.5 digit multimeter costs only $20 - less in Wal-mart and when
> on sale. Two mnutes with the meter means previous posts were
> unnecessary beause meter numbers are so definitive - not subjective..
>
>

On Aug 19, 4:08 pm, K8 notsogrand
<K8notsogr...@discussions.microsoft.com> wrote:
> Thank's for your reply
> But I think I'm going to take in and get it looked at.
> Becouse if the new supply is not working right, theres not much I can do
> about it.
> Besides buy a new one.

There were a few things to fix the system without 'buying a new'
something. But then the reason we fix things is to learn. My bet all
along was an intermiittent problem on the video card. But without
facts (the numbers), then this entire discusson was wasted time;
nothing was learned.

Well the problem seems to be fixed. (for now)
It turned out to be a bug in the OS. One or two driver's in Vista were not
booting up at restart. So I put in Vista's DVD rom and asked it to repair my
computer. (and so far so good) Its restarting just fine and everything seems
to be working. But with this thing...

"w_tom" wrote:

> On Aug 19, 4:08 pm, K8 notsogrand
> <K8notsogr...@discussions.microsoft.com> wrote:
> > Thank's for your reply
> > But I think I'm going to take in and get it looked at.
> > Becouse if the new supply is not working right, theres not much I can do
> > about it.
> > Besides buy a new one.
>
> There were a few things to fix the system without 'buying a new'
> something. But then the reason we fix things is to learn. My bet all
> along was an intermiittent problem on the video card. But without
> facts (the numbers), then this entire discusson was wasted time;
> nothing was learned.
>
>