Testing a new PSU

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

I just purchased a new 500 watt PSU, but it's not a name-brand like
Antec, Enermax, or Sparkle. It's a Viomax/Acheive, which I heard is
among the best of the low end "generic" power supplies that perform
nearly as well as the bigger brands, just lower quality control in
manufacturing perhaps. My question is what's the best way to
test/stress the new PSU aside from hardware metering devices to make
sure it's getting adequate power on its various rails? Is there a
guide on how to properly test PSUs, and what software to use? Thanks.

-worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

motherboard monitor or some other such program should let you see what the
bios thinks your PS is putting out... no way to really "stress" it that i
know of... just load it up and see the numbers.

What MoBo are you running ? What chipset ?

NuTs


..
"Wormwood" <wormwood69@lycos.com> wrote in message
news:ggt370p9jhif9l4ntl69nhgk7qfk9be3mu@4ax.com...
> I just purchased a new 500 watt PSU, but it's not a name-brand like
> Antec, Enermax, or Sparkle. It's a Viomax/Acheive, which I heard is
> among the best of the low end "generic" power supplies that perform
> nearly as well as the bigger brands, just lower quality control in
> manufacturing perhaps. My question is what's the best way to
> test/stress the new PSU aside from hardware metering devices to make
> sure it's getting adequate power on its various rails? Is there a
> guide on how to properly test PSUs, and what software to use? Thanks.
>
> -worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

On Mon, 5 Apr 2004 19:27:59 -0500, "NuT CrAcKeR"
<nutcracker@internationalhacker.org> wrote:

>motherboard monitor or some other such program should let you see what the
>bios thinks your PS is putting out... no way to really "stress" it that i
>know of... just load it up and see the numbers.
>
>What MoBo are you running ? What chipset ?
>
>NuTs

NF7-S ver. 2, Nvida2 chipset.

-worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

I havent had an nForce board yet, so I dont know what nifty things there are
avilable for it. Via boards have a hardware monitor that often comes one the
CD that ships with the board. They have a tendency to ship chipset specific
builds... so that can be tedious.


"Wormwood" <wormwood69@lycos.com> wrote in message
news:901470lheam9qr5jsi94ksosf6icbmi1jb@4ax.com...
> On Mon, 5 Apr 2004 19:27:59 -0500, "NuT CrAcKeR"
> <nutcracker@internationalhacker.org> wrote:
>
> >motherboard monitor or some other such program should let you see what
the
> >bios thinks your PS is putting out... no way to really "stress" it that i
> >know of... just load it up and see the numbers.
> >
> >What MoBo are you running ? What chipset ?
> >
> >NuTs
>
> NF7-S ver. 2, Nvida2 chipset.
>
> -worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

On Mon, 05 Apr 2004 20:18:03 -0400, Wormwood <wormwood69@lycos.com> wrote:
> I just purchased a new 500 watt PSU, but it's not a name-brand like
> Antec, Enermax, or Sparkle. It's a Viomax/Acheive, which I heard is
> among the best of the low end "generic" power supplies that perform
> nearly as well as the bigger brands, just lower quality control in
> manufacturing perhaps. My question is what's the best way to
> test/stress the new PSU aside from hardware metering devices to make
> sure it's getting adequate power on its various rails?

The only way to test any type of power supply (computer or otherwise) is
to load it up to its rated ampere limits for its rated period of time. The
ATX specification may state what that period of time should be, or the
manufacturer may publish what its specs actually are. Unless
you have some proper electronic test equipment and some ammeters I doubt
you could do this effectively.

What I would do instead is to run your computer to its max limit with
that PS for a long period of time during the summer (when your room is
warmest). For instance, use various stress testing software such as
Prime95, which will run your CPU at 100%. But you also really need to
ensure that all of your peripherals are also running - eg try burning a
CD, run a hard drive test program PLUS Prime95, plus something that will
stress your video card, all at the same time!

I'm not sure Windows could handle this but Linux probably could. Have
Fun.

Larry Gagnon


--
********************************
to reply via email remove "fake"

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Running MBM 3.0, my voltages are the following (at nearly full load):

+12v line - flucuates between 11.92v & 11.86v

-12v line - 12.19v

+5.0v line - 5.05v

-5.0v line - 5.24v

+3.3v - 3.34v

I am running two 80 GB hard drives, 2 CD-Rewriters, ATI 9800 Pro,
NF7-S MB, 1 stick of 512 DDR PC3200, 4 USB devices, and 2 case fans,
and I am overclocking a Barton XP 2500 at 211 X 10.5 = 2.2 G.

So how about it? Do these look like heathly voltages that my new
Viomax 500-watt PSU is putting out? Anything look troublesome? System
appears to be stable. I should also mention that my S3
refusing-to-wake from Standby issues appear to be no more. Comments
welcome.

-worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Perhaps I'm missing something but why would one wish to stress test a PSU?
If it works it's probably OK. Does your housekeeper stress test the washing
machine or the gardener the lawnmower?

 

[Spajky]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

On Tue, 06 Apr 2004 04:15:05 -0000, Larry Gagnon
<lagagnon@fakeuniserve.com> wrote:

>What I would do instead is to run your computer to its max limit with
>that PS for a long period of time during the summer (when your room is
>warmest).

(simulating hot Summer) see my site under comp/testStability ...

>For instance, use various stress testing software such as
>Prime95, which will run your CPU at 100%.

http://users.ev1.net/~redelm/cpuburn4.zip
run with a high priority in Dos/cmd box
with Runprio program ( -P switch burnmmx.exe)
[can be found somewhere here http://www.benchtest.com/ ]
will stress CPU more !!! (check Vcore values drop in MBM or bettter
MBprobe!!!)

-- Regards, SPAJKY ®
& visit my site @ http://www.spajky.vze.com
"Tualatin OC-ed / BX-Slot1 / inaudible setup!"
E-mail AntiSpam: remove ##

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

On Tue, 06 Apr 2004 01:38:32 -0400, Wormwood <wormwood69@lycos.com>
wrote:
> Running MBM 3.0, my voltages are the following (at nearly full load):
>
> +12v line - flucuates between 11.92v & 11.86v
>
> -12v line - 12.19v
>
> +5.0v line - 5.05v
>
> -5.0v line - 5.24v
>
> +3.3v - 3.34v
>
> I am running two 80 GB hard drives, 2 CD-Rewriters, ATI 9800 Pro,
> NF7-S MB, 1 stick of 512 DDR PC3200, 4 USB devices, and 2 case fans,
> and I am overclocking a Barton XP 2500 at 211 X 10.5 = 2.2 G.
>
> So how about it? Do these look like heathly voltages that my new
> Viomax 500-watt PSU is putting out? Anything look troublesome? System
> appears to be stable. I should also mention that my S3
> refusing-to-wake from Standby issues appear to be no more. Comments
> welcome.
>
> -worm
Worm: I think you can see for yourself that your voltages are well
within recommended limits. If you are running all peripherals at full
load for a considerable period of time (difficult to do that, but you
have to ask the question: will there EVER be a time I will run
absolutely full load on all my peripherals? I doubt it). Obviously your
power supply seems fine at the moment. Stop worrying yourself and just
enjoy it!

Larry Gagnon

--
********************************
to reply via email remove "fake"

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

On Tue, 06 Apr 2004 15:43:20 -0000, Larry Gagnon
<lagagnon@fakeuniserve.com> wrote:

>Stop worrying yourself and just
>enjoy it!

Well, I am slightly, kind of worried, somewhere in the back of my
brain. After all, only paid $21 (USD) for the PSU, shipped, and
you hear all sorts of horror stories about so called generic power
supplies. One thing that did concern me though was that when I got
the PSU yesterday & took it out of the shipping box, I heard something
rattling inside the metal casing. On closer inspection I noticed a
small hard chunk of what appewared to be melted white plastic laying
loose on one of the fan blades. I thought about gettin' the old screw
driver out & taking the fan grill of to get at the offending "foreign"
object when it dropped off the fan blade & fell back into the casing.
To open the PSU casing itself would most likely void the warranty so
for now I'll just let it, hoping it doesn't cause mischief with a
running fan blade, or inside on the PCB. It's little things like this
that made me insecure. Heh.

Next up for me is further overclocking experiments. Right now my
Barton XP2500 is @ 3200 speeds and perfectly stable w/ Prime95 for 6
hours, but with the new PSU, a new Vantec Areoflow cooler still
sitting in its packaging & a new "solar flare" Chieftec Dragon case
($45, shipped) on the way, I'm going to have some fun! Thanks
everyone for your jelp.

-worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

You have assumed that voltage and current output are the
only functions inside a power supply. Other essential
functions are 'forgotten' to sell at well below $70. For
example, should supply suffer a failure, then nothing else
inside computer can be damaged by that supply. But since so
many computer assemblers don't even have basic knowledge, then
they don't even check specs for overvoltage protection - an
essential function. When the failed supply destroys
motherboard and disk, then they *assume* this is normal rather
than blame the bean counter mentality who only understood one
spec - price.

You cannot test for some essential functions. However if
the power supply is properly constructed, then short out all
DC power connectors, turn on power supply, and supply must not
be damaged. In fact, industry specs go so far as to define
how big the shorting wire must be - because a shorted supply
cannot be damaged - assuming essential functions exist.
Shorting is one test you can perform to learn if supply
contains same functions that were even defacto standard 30
years ago.

Put an AM radio adjacent to power supply. If power supply
contains essential functions, then AM radio suffers no
interference. Interference would suggest other missing
functions including internal transient protection. That's
correct. Any protection effective at the computer should
already be inside that power supply. Another essential
function that cannot be tested.

Motherboard voltage monitor does not provide valid voltage
readings. Monitor is not for measuring voltage. Monitor is
for detecting voltage changes. A 3.5 digit multimeter is
necessary to discover what voltages really are.


Wormwood wrote:
> Well, I am slightly, kind of worried, somewhere in the back of my
> brain. After all, only paid $21 (USD) for the PSU, shipped, and
> you hear all sorts of horror stories about so called generic power
> supplies. One thing that did concern me though was that when I got
> the PSU yesterday & took it out of the shipping box, I heard something
> rattling inside the metal casing. On closer inspection I noticed a
> small hard chunk of what appewared to be melted white plastic laying
> loose on one of the fan blades. I thought about gettin' the old screw
> driver out & taking the fan grill of to get at the offending "foreign"
> object when it dropped off the fan blade & fell back into the casing.
> To open the PSU casing itself would most likely void the warranty so
> for now I'll just let it, hoping it doesn't cause mischief with a
> running fan blade, or inside on the PCB. It's little things like this
> that made me insecure. Heh.
>
> Next up for me is further overclocking experiments. Right now my
> Barton XP2500 is @ 3200 speeds and perfectly stable w/ Prime95 for 6
> hours, but with the new PSU, a new Vantec Areoflow cooler still
> sitting in its packaging & a new "solar flare" Chieftec Dragon case
> ($45, shipped) on the way, I'm going to have some fun! Thanks
> everyone for your jelp.
>
> -worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

"h2so4" <elly.fant@blueyonder.co.uk> wrote in message
news:zMtcc.28006$Id.23969@news-binary.blueyonder.co.uk...
> Perhaps I'm missing something but why would one wish to stress test a PSU?
> If it works it's probably OK. Does your housekeeper stress test the
washing
> machine or the gardener the lawnmower?

You lawnmower probably don't have thousands of dollars worth hardware
relying on it! If your lawnmower fails, you won't get your lawn trimmed, big
deal. A blow in psu could roast everything from mb, cpu, ram and gfxcard to
peripherals.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

>If it works it's probably OK. Does your housekeeper stress test the washing
>machine or the gardener the lawnmower?

Ahhhhhhh no, but my wife keeps sending me out for batteries. She must be stress
testing something in our bedroom.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

sounds like you should be stess testing your tool instead of her stress
testing her toys.

NuTs

"PRIVATE1964" <private1964@aol.com> wrote in message
news:20040406203745.18008.00000984@mb-m06.aol.com...
> >If it works it's probably OK. Does your housekeeper stress test the
washing
> >machine or the gardener the lawnmower?
>
> Ahhhhhhh no, but my wife keeps sending me out for batteries. She must be
stress
> testing something in our bedroom.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

First, to sell a power supply at that $20 price, then power
supply is missing essential functions. As Deming taught,
quality cannot be tested. Quality is designed inside. No
test exists for those missing and essential functions. So
those functions can get forgotten.

In the meantime, some basic functions are best tested inside
the computer, with computer under full load (ie. XP
multitasking so that all peripherals are working
simultaneously). Essential tool is a ubiquitous 3.5 digit
multimeter. Procedure is defined in "I think my power supply
is dead" in alt.comp.hardware on 5 Feb 2004 or at
http://tinyurl.com/yvbw9 .

Motherboard voltmeter is not acceptable for voltage
measurements. Its function is to monitor for voltage
changes. As noted by that above procedure, your voltmeter
readings must be within upper 3/4 of limits in table. This is
minimal acceptable testing. As jojo so accurately notes,
cheap supplies almost 'invent' numbers because they are
selling to the technically naive who would claim to be expert.

Seek recent posts by Kony where identical supply (same
power), sold under different trade names, claims different
power ratings. There is good reason why minimally acceptable
supplies sell for $70 (which also does not mean all $70
supplies are minimally acceptable). Essential functions cost
money.

To sell to technically naive 'experts', many essential power
supply functions are forgotten. Forgotten functions because
some 'experts' only understand one spec - price. 'Experts'
who make inferior supplies so profitable.

Wormwood wrote:
> I just purchased a new 500 watt PSU, but it's not a name-brand like
> Antec, Enermax, or Sparkle. It's a Viomax/Acheive, which I heard is
> among the best of the low end "generic" power supplies that perform
> nearly as well as the bigger brands, just lower quality control in
> manufacturing perhaps. My question is what's the best way to
> test/stress the new PSU aside from hardware metering devices to make
> sure it's getting adequate power on its various rails? Is there a
> guide on how to properly test PSUs, and what software to use? Thanks.
>
> -worm

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Why would you risk your entire setup with a cheap A$$ power supply.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Agree. In old days of Pentium2 and 3, PSU were cheap, but they were solid.
Not like today. So it might take some time and money for most people to
realise that fact, old days were better

"Mecho" <mecho@aol.com> wrote in message
news:20040408065045.10106.00000002@mb-m17.aol.com...
> Why would you risk your entire setup with a cheap A$$ power supply.