comp on 24/7

[darkorb]

hey guys my old computer/parents new one, is used for their work banking and all and its on 24/7. it is a Amd Athlon xp 1900+, 1.6ghz, 1gb ram, 40gb hard drive, asus a7n266-vm, and a geforce 4 mx420. it has a 300watt psu, and its temps are 35 mobo, 46 cpu. they never game or anything, just like word files, internet banking and all. the 300 watt psu, i read in everest that the 12 volt rail is at 11.37. i dont no if its accurate or anything, but its ok for it to be on 24/7? or shoujld i swap in my 400watt psu that i have lying around from my old hp. my dad hates it when i open up computers or anything, so changing psu/and adding a fan would be the last resort. how are the temops and the rails doing on the pc? thx alot guys!

just to let u no, there is only 1 fan in the case, which is the cpu fan running at 6k rpm lol, i tried lowering it with speedfan but its either on/off. and when i did add a exhaust fan, it only lowered temps by like 2 degrees so i dont think its worth it. there is an intake, but the holes in the front for the air to get through are relaly small, but still very little air gets in.

[nam-ng]

hey guys my old computer/parents new one, is used for their work banking and all and its on 24/7. it is a Amd Athlon xp 1900+, 1.6ghz, 1gb ram, 40gb hard drive, asus a7n266-vm, and a geforce 4 mx420. it has a 300watt psu, and its temps are 35 mobo, 46 cpu. they never game or anything, just like word files, internet banking and all. the 300 watt psu, i read in everest that the 12 volt rail is at 11.37. i dont no if its accurate or anything, but its ok for it to be on 24/7? or shoujld i swap in my 400watt psu that i have lying around from my old hp. my dad hates it when i open up computers or anything, so changing psu/and adding a fan would be the last resort. how are the temops and the rails doing on the pc? thx alot guys!

just to let u no, there is only 1 fan in the case, which is the cpu fan running at 6k rpm lol, i tried lowering it with speedfan but its either on/off. and when i did add a exhaust fan, it only lowered temps by like 2 degrees so i dont think its worth it. there is an intake, but the holes in the front for the air to get through are relaly small, but still very little air gets in.
12v hardware had been around nearly forever, they normally operated within the range of valid lead-acid batttery output - 10.8V to 13.2V, a fully charged lead-acid battery stabilized out at 11.8V nominal. If you bought any common 12V hardware which could not do so, get rid of them, they were designed or built by retards.

The above is the norm for most all 12V voltage regulator devices for years and years but few exceptions.

What you're looking for is deviation from nominal voltage to determine the quality, if your 11.37V output essentially never deviate from 11.37V or little if any deviance detected at all under all loading condition then it is a good 12V supply.

The larger the deviance percentage from nominal under all loading condition the worst the 12V supply. A power supply with "wandering voltage level" from nominal is a bad power supply, a good perfect one simply doesn't deviate fom nominal voltage under any load condition, up or down.

[Enigmachine]

Ok I'm not an expert or anything, but one thing you should do is look at the efficiency of the power supply...

Normally the 300 W will waste less electricity as heat and so cost less in the long run, but these things keep improving so a dramatically more efficient 400 W PSU might cost less money in the long run.

Also, you want to make sure your HDs don't power off/go to sleep, because most hard drives fail when powered up (it causes more wear on the mechanical parts).

Anybody correct me if I'm wrong. :)

[nam-ng]

Normally the 300 W will waste less electricity as heat and so cost less in the long run, but these things keep improving so a dramatically more efficient 400 W PSU might cost less money in the long run.
With a brand new "dramatically more efficient 400 W PSU", he'll save enough money after 2 yrs of 24/7 for a Double-Whopper at the nearest Burger King, assuming the Double-Whopper would still be the same price as it is now.

[darkorb]

nam-ng, sry i dont undesrtand what u mean lol, english plz

bascially its ok right?

[nam-ng]

nam-ng, sry i dont undesrtand what u mean lol, english plz

bascially its ok right?
You've to verify for the below to know for sure...
What you're looking for is deviation from nominal voltage to determine the quality, if your 11.37V output essentially never deviate from 11.37V or little if any deviance detected at all under all loading condition then it is a good 12V supply.

[darkorb]

o ok, il check in a few hours

[GIBSON]

I understand what you mean about solid power supplies nam-ng, however isn't a voltage as close as possible to 12.00V (solid voltage of course) preferred over the lower or higher voltage. Or doesn't it make that much of a difference?

[nam-ng]

I understand what you mean about solid power supplies nam-ng, however isn't a voltage as close as possible to 12.00V (solid voltage of course) preferred over the lower or higher voltage. Or doesn't it make that much of a difference?
It doesn't make much of a difference as long as you're close enough.

Wandering voltage levels up or down under varied loads is indicative of insufficient power, the greater the wandering variation levels the greater the indication. When you have lots of power there is no wandering, or little if any at all.

[darkorb]

that psu used to be in another hp, which has a stronger cpu, and it was in a warmer case, and it worked fine.

i just checked it, its still 11.37

[werty316]

11.37V on the 12V rail is dangerously. Sounds like that computer is using a POS PSU. Might be a good idea to get a decent 400W or higher PSU but since it isn't a gaming rig and if it seems stable then I would worry.

[darkorb]

i would worry? so i should change it?

[werty316]

To prevent any damage done I would replace the PSU since 11.37V is way too low on the 12V rail. Usually you don't want the voltages to deviate by about 3-5% out of spec.

[darkorb]

i just chcked it and its at 11.43, i turned the comp off. lets see what it is tomorw morning

[nam-ng]

i just chcked it and its at 11.43, i turned the comp off. lets see what it is tomorw morning
It'd gone up, interesting... Just because 12V level is wandering from lack of power, it isn't wandering randomly.

Typically any A+ certified JOEs with 20yrs expertise can tell you to test by running CD/harddisk tests/benches for example to check for 12V level dropping lower down. They usually don't know the tests/benchmarks to make it goes up.

[GIBSON]

It'd gone up, interesting... Just because 12V level is wandering from lack of power, it isn't wandering randomly.

Typically any A+ certified JOEs with 20yrs expertise can tell you to test by running CD/harddisk tests/benches for example to check for 12V level dropping lower down. They usually don't know the tests/benchmarks to make it goes up.

Well, I wouldn't know any. Could you give an example. Or are you talking about decreasing the load? (like getting drives out etc.)

[nam-ng]

Well, I wouldn't know any. Could you give an example.
I can't find the thread examples I'd given before in this forum, they had been deleted.
Or are you talking about decreasing the load? (like getting drives out etc.)
Any A+ certified JOEs could do that, but it isn't practical nor useful.

The system mentioned by darkorb pre-dated 12V mobo requirement pioneered by Intel from old P4s excessive wasteful power needs, the old AMD Athlons never did Which is why most old AMD mobos didn't care about the 24 pins power connector or the extra 12V connector in the mobos on power-ups, only the newer AMD mobos do.

The only things which really used 12V on that machine are relays, CD, hard disk, and fans; all combined would used at worst 50W out of 12V. In those days there were excess 12V power available, which was why Intel used the excess 12V power to supplement P4s power demands.

Relays/Fans used minimal 12V power, CD used negligible 12V power at idle, how would anyone run benchmarks and voltage monitoring progams with the only hard drive also removed?

[GIBSON]

Hmm yes. Besides that the only thing I could personally think of would be to constantly load and unload the system and check for the voltages going up as well as down.
Would it be something along the lines of that? If not, please tell the actualy way it is tested/benchmarked. (This also makes me wondering, if it seems (well, to me atleast) to be so hard to get the voltage go up instead of down, then what is the sence of looking at it's peaks (the kind of peaks you don't see during regular testing)).

[nam-ng]

Hmm yes. Besides that the only thing I could personally think of would be to constantly load and unload the system and check for the voltages going up as well as down.
Would it be something along the lines of that? If not, please tell the actualy way it is tested/benchmarked.
No, not along that line at all. I don't normally tell people what to do, if they understood my answer they don't need me to tell them what to do.
(This also makes me wondering, if it seems (well, to me atleast) to be so hard to get the voltage go up instead of down, then what is the sence of looking at it's peaks (the kind of peaks you don't see during regular testing)).
"Wandering voltage levels up or down under varied loads is indicative of insufficient power". "down" is simpler to understand to any Joes when 12V is loaded, "up" from simpler unloaded 12V power.

"up" is slightly more difficult because "insufficient power" didn't only have to be with 12V, but 12V will go up just the same due to "insufficient power" from 5V and 3.3V.

There're signs of power-supplies being overloaded for one reason or another. One of the most common signs for marginal/in-sufficient 3.3V + 5V combined power is the inverse level up shifting of 12V instead of dropping.

When 3.3V and 5V get loaded down and dropping downward, 12V level will shift upward (-5V and -12V will be more negative!!!).

[GIBSON]

Oh, I understand what you mean now. Thanks for the info nam-ng! :)

[nam-ng]

Oh, I understand what you mean now. Thanks for the info nam-ng! :)
You're welcome... With a few questions you'd learned more in a short time than what most A+ certified PSU "experts" ever did in 20+ yrs.

I normally don't volunteer factual information no one had asked for as "experts" usually do (they needed to prove they had large e-penus despite the apparent deficient in brain power for 6th grade explanations or "simple garden-variety common sense").

Do you want another very common pitfall most 20+ yrs "experts" also didn't get per subject topic at hand?

[DragonMaster]

11.37V on the 12V rail is dangerously. Sounds like that computer is using a POS PSU. Might be a good idea to get a decent 400W or higher PSU but since it isn't a gaming rig and if it seems stable then I would worry.

The best way to know the real voltage would be to check the 12V rails with a calibrated multimeter. You can't really be sure that a motherboard will show you accurate readings, since it would be too expensive to do so : The only multimeters/voltmeters that stay accurate over the years cost over $100, while your mobo costs probably less than that. Why should you spend ½ of the mobo's price on the voltage readings?

Your PSU is probably OK. Your computer always ran fine? It should.
I have an Athlon XP 2000 computer that runs w/o crashing since 2003 ~6-12 hours a day with a generic PSU. The mobo is reading -14v as -12v, 5.5v for 5v and 11.3v for 12v, but I'm sure that with a DMM, the readings are going to be much better.
You could compare with the HP PSU just for fun.