power supply damaged even with ups ???????????

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View
here is my problem, i run my new pc with a ups 800 VA, power supply
480 watt, ’lc power scorpio’ but even with that after 3months it is
dead !!!!!

fortunately my pc is ok, i got replaced the psu, ok, but my concern is

whom foult it was ?

the ups that allowed some strange signal to pass to the psu or the psu
itself was so poor?     i took it 95 euro !!!!

the strange thing is that the psu  didn’t ’die’ during the time that
pc was on, but one morning that i tried to switch it on, it did never
switch on until i replaced the psu. until that moment i never had even
small problems with my pc.

my pc is p4 3ghz s775, mobo asus p5gd1 pro, vga ati x300 pci-e, ram
1024, 1hdd sata,1tv card,and 1printer, lg lcd display,

i put in ups only the pc and the monitor lcd.

thank u in advance !

Posted using the http://www.hardwareforumz.com interface, at author's request
Articles individually checked for conformance to usenet standards
Topic URL:
Visit Topic URL to contact author (reg. req'd).  Report abuse:

Re: power supply damaged even with ups ???????????

On 30 Dec 2005 05:39:11 GMT, graxtax

Quoted text here. Click to load it

Sounds generic, what was the system like that it was

Quoted text here. Click to load it

Probably the manufacturer of the power supply for spec'ing
it as a 480W unit when it may've been closer to 250-300W in
reality, but weak on some rails per any particular use.

Quoted text here. Click to load it

It's doubtful to be the UPS fault, after all your AC power
before the UPS shouldn't be THAT bad or your other
electronic products should be showing faults as well.

Quoted text here. Click to load it

Don't use generic power supplies... nobody else likes paying
more for good brands either but they do to avoid such

Re: power supply damaged even with ups ???????????

Quoted text here. Click to load it

Unless the UPS incorporates surge protection (and you properly connect your
hardware), it is only a UPS - *uninterruptible* power supply, which has
nothing to do with surge protection.  I have a huge 2kVA UPS with a 60-pound
isolation transformer that also protects the ground circuit.  Most
consumer-grade UPS'es have no surge protection.  You use those for *data*
protection (regarding up-time and prevention of catastrophic system crash
causing data corruption).  Consumer-grade UPS'es do NOT provide for hardware

The best would be to install surge protection at the entry point for power
into your house (i.e., a whole-home surge arrestor).  Unless it has a means
of monitoring itself, you won't know when it fails.  Every surge stresses
the protector until it eventually fails.  If you use an end-point surge
protection, like the type you plug into a wall outlet, use only ONE for all
your computer equipment (and also for any equipment you connect to your
computer).  Two surge protectors plugged into the same double wall outlet
probably each have a 6-foot cord.  That means there is 12 feet of cord
between them resulting in a high-voltage impedance than can cause a
differential of 400V, or more, between the two surge protectors during a
surge.  If you connect any other equipment to your computer (phone cable to
modem, stereo to audio jacks) then having each component on a different
surge protector can result in a differential between them that is also
experienced as a surge.  Put all the interconnected equipment upstream of
the SAME surge protector.  For a modem, use a surge protector with built-in
phone surge protection (so the surge release at the phone protector is very
close to the input side of the power surge protection.

Don't go stringing a bunch of surge protectors from every wall outlet and
then end up connecting equipment on one surge protector across to equipment
on another surge protector (unless that other surge protector is upstream of
the first one, in which case you don't need to waste the money on a surge
strip and just use a power strip).

Most of the consumer-grade surge protectors don't seem to afford much
protection plus you don't know when it has failed.  I'm still on the fence
regarding surge protectors (that shunt the surge) and surge suppressors
(that absorb the surge and dissipate it slowly); see http://snipurl.com/l622
and http://www.zerosurge.com/HTML/about.html .  It seems the suppressors
would not incur the problems of parallel protectors separated by some
distance where the shunted surge across the distance between the protectors
and the impedance therein ends up itself being a surge.  Rather than shunt
it and hope it goes elsewhere, suck it in and clamp it down.  But then if
you arrest it an the point-of-entry then it isn't inside the house to deal
with (typically with consumers using cheap power strips claiming to provide
surge protection).

Since w_tom is a regular here, hopefully he will chime in, too.  Despite any
disagreement between us, it is always good to get varied on-topic
viewpoints.  You might want to ping him for his opinion (get ready for him
rolling his eyes when asking why a UPS didn't stop a surge).  Check his
other posts regarding surges
(http://groups.google.com/groups?q=author%3Aw_tom+group%3Aalt.comp.hardware+surge&start=0&scoring=d ).

Regarding failure of the PSU, they die early (premature death) or they die
late.  When new, they may be weak or have a defect that causes failure which
is often exhibited within a month or two.  If they survive a couple months,
they'll last a long time, like years, and then die.  Also goes for other
components: a hard drive fails early or it lasts for years, a memory stick
dies immediately or soon or it lasts for years, and so on.  The warranty
really only gets used for premature death; otherwise, the product usually
lasts longer than the warranty (so when it dies later then you have no
warranty).  Since your UPS probably does nothing regarding surge protection
or suppresion, anything after the UPS can still get fried by a surge.
*Good* computer switching power supplies are pretty resistant to surges but
then, of course, I'm not experienced with the real cheapy units.  If I get
stuck with a case with a low-grade PSU, out it goes to get replaced.  It is
the blood supply of your electronic computer and is not a part where you
want to go cheap.  At 95 Euro (~$113 USD), it wasn't priced cheap.

"it is dead" really doesn't say anything intelligible about the condition of
PSU.  Could be it refuses to accept any input power, like from a blown fuse
or snapped breaker.  Could be it comes up but goes down in a short time
because of excessive heat.  Could be some or all taps have no voltage.
Could be there is a reduction in the output capacity of a tap or shared taps
which results in flaky equipment operation.  Could be something is sucking
out so many amps that voltage regulation is extremely poor.  Could be some
circuitry fried so the ripple is too huge.  Could be just the fan stopped
spinning but the supply works okay; however, it might check fan speed to
stop itself if the fan doesn't spin (fans are moving parts and fail more

I've never heard of LC Power (http://www.lc-power.com /) before now but then
maybe they have no distributors in my country (rather than list their
distributors, they make you go clicking around on a map hunting for
distributors, and none of the countries shown in their map is mine). When
you read the spec decal they show
(http://www.lc-power.com/htm/products_detail.php?lang=2&h_id=1&u_id=21&a_id=67 ),
the total power would be 739.7 watts.  They neglect to mention which are
common railed outputs that would limit each other for amperage output.  I
didn't see any downloads for a manual that might explain in there what you
can really get out from each tap and which taps are dependent.  From 740
down to 480 watts, there definitely is some dependency between taps.  They
don't mention if 480W is the maximum at failure or a suggested maximum but
that there is still room for overage (i.e., you can safely go above 480W).
"Our manufacturing facilities are located in Shenzhen, China".  Might be
their plant, or might be whomever they contract from where to buy the PSUs
on which they slap their brand sticker.  Maybe kony or someone else here has
some experience with this brand.

BTW: Could you ask your forum admin to reduce the size of the signature they
tack onto posts that they migrate from their forum to Usenet using their
gateway?  Just tell us the post came through their Usenet gateway, don't
bother telling us about its attempt to conform to standards, don't bother
with an abuse URL (the X-Complaints-To header already notes that, and
probably should be changed if not correct), and don't bother showing the
forum URLs to the original post because we in Usenet are not going to bother
web browsing over to your forum.

** Post replies to the newsgroup. Share with others. **
For e-mail, remove "NIX" and append "#VC811" to Subject.

Re: power supply damaged even with ups ???????????

  Surge protection adjacent to a power supply would already be
inside that power supply.  So why did that power supply fail?
Just too common is that a power supply failed; therefore it
must be a surge.

  Meanwhile best evidence is that dead supply.  Without
details - such as what has specifically failed - then a human
has no idea what caused that damage.

  I have yet to see a supply damaged by surge.  Seen numerous
supplies with component failure typical of manufacturing
defects.  Learning is why we repair supplies.  For example, a
recent supply accused of failing due to power on was, instead,
found to have its +5VSB power supply damaged.  Power on would
have no effect because this part of a power supply is always

  In another failure blamed on transients, failure was
actually due to a failed pullup resistor.  A problem so
difficult to find that I literally removed and individually
tested every other component until I finally discovered that
resistor.  Clearly something only done to learn - not to save
money.  But again, this part failed on its own.  The part may
have failed during normal operation.  But since its function
was only for power up restarts, then the failure was not
observed until the power supply was power cycled.  And yet
some would immediately blame surges from power cycling -
because they did not first obtain facts.

  (BTW, same kind of resistor failure is why electrolytic
capacitors are routinely shorted before touching with a hand.
In one supply, a 'bleed off' resistor failed so that one
electrolytic capacitor remain fully charged - hazardous to
human hands.  Again, a failure in normal operation that could
have cause a human to suffer surge damage.  The point being
that part do fail often not because of surges.)

  In another failure, a switching transistor failed.  In this
case, about 30 same power supplies were connected and running
from same electrical source.  But only this one power
transistor failed - again an internal or manufacturing defect
unique to that one supply.  Another example of a failure that
some elsewhere would have blamed on surges.  Again, the only
reason for repairing it was to learn why a failure happened.
And again, not due to AC line transients.

  If a power supply is damaged by transients, then damage
would more likely appear in places where a surge would be most
destructive such as in line filter, diode bridge, or
electrolytic capacitors.  Too often, a failed power supply has
no damage in its power input section. Therefore failure is not
due to external transients.  Furthermore, a transient requires
an output path before damage could happen.  One path it will
not use is through galvanic isolation required in all power
supplies and rated at typically thousands of volts.  Yes,
power supplies also have significant galvanic isolation; just
another part of internal protection.

  So what damaged a supply?  More often due to internal or
manufacturing defects.  Computer grade UPSes connect AC mains
directly to computer when not in battery backup mode.  So
where is that UPS protection anyway?  Protection that was
going to protect a computer is already inside its power
supply.  If power supply has failed, no one can say accurately
until a failed internal component has been identified - the
autopsy of the best evidence - the dead body.

  I can not remember the last time a power supply was
identified failed due to surges.  Have seen some where
components were suspiciously too marginal meaning that a
certain percentage would always fail in normal operation -
maybe even months or years later.  And yet many will blame a
surge for conditions more likely attributed to marginal
components or manufacturing defects.

  Of course a UPS may have "allowed some strange signal to
pass to the psu".  Computer grade UPSes don't even claim to
stop, block, or absorb such transients.  But long before one
can suggest what caused the failure, one must first identify
its failed component and building wide circuit that would have
caused that damage.

  Was the 480 watt supply really only 250 watts or not capable
of outputting sufficient power on one voltage?   These are
other possibilities when one does not first trace down a
specific reason for each power supply failure.  One who fixes
things by shotgunning would never learn these useful lessons
about why things fail.

Vanguard wrote:
Quoted text here. Click to load it
(http://groups.google.com/groups?q=author%3Aw_tom+group%3Aalt.comp.hardware+surge&start=0&scoring=d ).
Quoted text here. Click to load it
(http://www.lc-power.com/htm/products_detail.php?lang=2&h_id=1&u_id=21&a_id=67 ),
Quoted text here. Click to load it

Re: power supply damaged even with ups ???????????

graxtax wrote:
Quoted text here. Click to load it

Every UPS I've checked had an elaborate surge protector built into it,
and all but the worst PSUs have an EMI filter that can also block some
surges.  Unfortunately the "LC" in the name of your PSU may indicate
that it's made by Deer, which produces several brands including L&C,
Codegen, and Logic, the worst which have no EMI filters, and even the
best Deers aren't very good.  One expert, JonnyGuru.com, found that a
500W Deer wouldn't even start if the initial load was over 150W,
although it could put out more power if the load was increased

If you want very high quality at a moderate price, check out PSUs made
by Fortron-Source, which sells under brands like Fortron, Sparkle,
Trend, Hi-Q, PowerQ, Powertech (not Powertech), and sometimes
Casedge.(look for a missing screw on the side).  Most of these PSUs
have model numbers starting with "FSP".

Re: power supply damaged even with ups ???????????

Quoted text here. Click to load it

What is failing? The +5vsb is *always* on, and without cooling fans
circulating air, is prone to failure in poorly designed units. If
it *is* the +5vsb that is failing (or even if it isn't), try a better
brand of PSU.

Re: power supply damaged even with ups ???????????

Quoted text here. Click to load it

Not to mention that poweron is controlled by the mainboard. Could be failing
caps causing a strain on the PSU.

Finally, most PSU's put out dirtier power than you get from the outlet since
it's usually a square wave instead of a sine wave. You're protected from
power losses, but the switching is also pretty dirty.

Re: power supply damaged even with ups ???????????

Noozer wrote:

Quoted text here. Click to load it

Don't you mean most UPSes puts out dirtier power?  PC PSUs output DC
with less than 0.1V peak-peak AC noise.

Re: power supply damaged even with ups ???????????

Quoted text here. Click to load it

Yup... Fingers going faster than my brain again.

: )

Re: power supply damaged even with ups ???????????

Noozer wrote:

Quoted text here. Click to load it

Assuming you were writing about a *UPS's* square wave, a SMPS has a
300vdc buss on its input. This should withstand a square wave just
fine. Some cheaper "auto switching" circuits (90-264 vac) *are* quite
sensitive to input waveforms, however.

Site Timeline