I've blown a few power supplies on my home PC. The PC is connecting to a
working surge protector. It was suggested to me that I could have an
over-voltage problem. I thought a surge protector handled that. How would
I know if I have an over-voltage problem? does this make any sense?
Surge protectors handle surges, not higher than normal voltage. Get a meter
and see exactly what you do have. Do this a few times a day. When I
started replacing light bulbs too frequently, I did that and found we had up
to 140 volts for a time in the morning. The power company denied it, but
once I made the call, it never happened again. They also replaced some
equipment at the nearby sub station.
hows the dust level in your home. power supplies attract dust, which
prevents proper cooling and causes power supply failures. plus a
underlying main computer problem can cause power supply failures. you
might up the wattage of your supply espically if you have added cards
to your computer.
do check your line voltage, but that may not be it
Agree with all the above. However, one would think that there is
quite a bit of margin in a PC supply. I would think you'd have to see
a pretty hefty over-voltage to burn it out. Is the supply adequate
for the total load of the PC, ie is this just a typical PC or one
fully loaded with add-in boards? Fans/ventilation working?
I think pretty much all PC power supplies (and most other power
supplies) these days are auto ranging switching type power supplies
which will accept an input voltage range from about 100-240V without any
issues. If you have an overvoltage situation on your power line it's
very unlikely that it will be high enough to be any issue for the PC
supply before a lot of other electrical items in your house are melting
down, particularly all your light bulbs.
I did have a moderate overvoltage situation at my house once, about
136V. I noticed it when my UPS switched to battery while power was
obviously still on. Checking the UPS I found it reporting 136V on the
input and switched to battery protectively. I confirmed the reading with
a separate meter as 135.6V (the UPS only reports three digits), and
called the utility.
The CSR I talked to at the utility had no idea what I was talking about,
but promised to pass the info to a tech. Much to my surprise I got a
call back from a tech in the area within 10 min and after explaining
what I found to him, he was parked in my driveway another 10 min later.
By the time I got out to the truck he had confirmed the reading and was
on the radio to another tech heading to the regulator bank up the road.
A few min later I had a nice 126V reading and it's been fine ever since.
A lot better response from the utility than I would have expected.
Both of those are way overkill for most machines. I've got a good
quality 330W in one machine, a 220W in the other. Both have been fine
for many years.
You only need a huge power supply if you are driving a high-end graphics
card or many hard drives.
Indeed. For comparison, consider this server from Apple:
Two 3 GHz dual-core Xeon processors
32 GB of RAM
3 300 GB 15000 RPM SAS drives
It's got a 650W supply. Or how about this one from Pogo Linux?
Four 2.8 GHz dual-core Opteron processors
64 GB of RAM
5 SCSI hard disks
12 case fans
(I'm not bothering to list things like CD drives, ethernet, etc, on any
That's got 850W. He must have one hell of a server if he really needs
My home server has used 27 KWh over the last 225 hours, so that works
out to 120W (it runs 24/7).
For a home non-server machine, 500W *may* be reasonable. A high-end
graphics card might use 120W. So, figure two of those, in some kind of
SLI setup or something like that. Now throw in a pair of fast dual-core
processors, lots of RAM, high end sound card, a couple big fast disks,
and you could maybe need 500W.
I'm involved with a local access TV
studio that had a similar problem. We blew
the power supply in a color laser
printer twice, a power supply in a PC and a
surge protector. In one case the UPS
actually switched on a high voltage ...
many don't. Anyway, it turned out the
be a bad neutral from a 3 phase
I actually measured 180 volts at one of
This is in a commercial building and the
transformer was mounted up in the ceiling
area of the TV studio .... humming all
the time. Anyway the electrical department
determined that the transformer was bad,
so they replaced it putting it in an
room where it should have been
originally. I'm really not so sure that
it was the
transformer ... it could have been a bad
connection somewhere. But anyway,
new transform = no humming and the
problem hasn't come back. As others have
said, measure the voltage at the outlet.
One wonders if you are in the military. You seem to think all
powersupplies 'must' conform to certain standards. No they do not have
to. They can be just as cheap and junky as their manufacturer desires them
Specifications for PC power supplies are quite specific. However
some clone power supplies are, instead, marketed to computer
assemblers who often don't even know how electricity works. These
power supplies may be missing essential functions required by those
How to suspect a power supply as 'cheap and junky'? Such supplies
will be missing the full page of numerical specifications. If is does
not claim to meet those well defined industry standards, then they can
sell it at discounted prices. Many who never learned technology will
buy a supply only based upon watts and dollars.
Since A+ Certified techs need not even know how electricity works to
be certified, then this is a perfect market for supplies that are
missing essential functions.
For example, all outputs from a PC supply can be shorted together
without power supply damaged. Industry specs even define wire size
for that test. But this same required function may be missing in
'cheap and junky' power supplies. Those who did not learn technology
might blame the short instead of a human buyer for damage that must
Yes standards do exist for PC power supplies. If a supply permits a
PC to interfere with an adjacent AM radio, then the power supply does
not conform to another industry standard. If a power supply is
'cheap and junky', then blame falls upon the human because those
Only such body I am aware of is UL or United Laboratories. I dont think
all power supplies get the UL stamp of approval?
Anyway, the point is if its specified, it needs to be tested. I dont know
what body is validating that these supplies are meeting the stated
specifications. Do you?
Were you around when car amplifiers put out 100W per channel? and sound
more terrible than another one that puts out just 15W/ channel?
UL does not care if a power supply even works. Their mandate is
human safety; that a power supply does not harm humans. If supply
destroys transistors - UL does not care. As long as destruction does
not harm humans, then supply can obtain UL approval.
Another standard is FCC Part 15. Supply need not meet FCC
requirements when not stated in written specs. A clone computer
assembler is responsible for compouter meeting those Part 15
requirements. Since many computer assemblers do not even know what
Part 15 is, then a supply manufacturer can 'forget' to include
essential filters; leaving computer assembler liable.
Intel demands a long list of requirements. Does the power supply
conform to those requirements? Again, responsibility lies with the
computer assembler. He must demand a long list of written
manufacturer specs. But many clone computer manufacturers know the
technical competence of their customers. If functions claimed on a
written spec sheets were missing, only then is a supply manufacturer
liable. Best to not provide any numerical specifications since so
many computer assemblers only buy on watts and price. Eyes glaze over
when many computer assemblers look at electrical parameters. Even A+
Certified technicians neither know what these functions are nor demand
those numerical specifications. North American clone computer
industry is ripe for dumping.
Explained are simplest tasks, tools, and numbers to answer the OPs
question. 200 volts could harm his computer and would be completely
ignored by a surge protector. This is simple stuff that every
responsible computer tech should know. Even A+ Certification does not
require such electrical knowledge. Therefore the OP is told a surge
protector would have eliminated overvoltage. No wonder a majority of
America's engineers are immigrants and foreigners. Too many computer
experts who don't even know how electricity works.
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.