Need a new PSU?

Notmyrealname

espresso wrote: »

To use your words, "Not wishing to be rude" but you are simply showing your total ignorance on this matter by posting that rubbish!
.

Indeed. Electrolytic capacitors are the only ones likely to fail however we're talking over a decade or more. And as they're made to within 20% tolerance and typically only used for smoothing, you'd not notice anyway.

robmar0se

espresso wrote: »

To use your words, "Not wishing to be rude" but you are simply showing your total ignorance on this matter by posting that rubbish!

The advice posted by A.Penny.Saved above is spot on. Bigger is not better if the extra capacity is not required or used.

notmyrealname wrote:

Indeed. Electrolytic capacitors are the only ones likely to fail however we're talking over a decade or more. And as they're made to within 20% tolerance and typically only used for smoothing, you'd not notice anyway.

Ignorance?

Here is a quote from Thermaltake website. I leave it to others to decide if they prefer to take posters' views or that from a PSU manufacturer, and a good one at that!!!

Thermaltake
Electrolytic capacitor aging. When used heavily or over an extended period of time (1+ years) a PSU will slowly lose some of its initial wattage capacity. We recommend you add 20% if you plan to keep your PSU for more than 1 year, or 25-30% for 24/7 usage and 1+ years.

espresso

robmar0se wrote: »

Ignorance?

Here is a quote from Thermaltake website. I leave it to others to decide if they prefer to take posters' views or that from a PSU manufacturer, and a good one at that!!!

Oh dear!

It is obvious from your posts that you are not a qualified electronic design engineer, however if your PSU uses correcly specified and toleranced capacitors rated for 3000 hours at 105 degrees C, these capacitor would be rated for 192000 hours at 45 degrees C.

How long do you expect a PC PSU to last?

A higher wattage PSU is pointless unless more power is actually required - a higher wattage PSU will not extend the capacitors life and will not aid the PSU's efficiency!

You obviously do not understand basic PC power supply specifications if you believe that bigger is better. Modern motherboards are designed to take the bulk of the power from the +12V rail.

Look at these specs. for three PSU's from a well know manufacturer:



You will see that the maximum load current for all rails except the +12V rail is the same - only the +12V maximum load current increases in line with the total output rating.

USB devices take mimimal power from the +5V rail, so why would you specify a higher wattage PSU to cater for some extra USB periferals which use very little?

Ignorance is indeed a dangerous thing, especially when you know very little about electronic circuit design.

smos585

espresso wrote: »

Oh dear!

It is obvious from your posts that you are not a qualified electronic design engineer, however if your PSU uses correcly specified and toleranced capacitors rated for 3000 hours at 105 degrees C, these capacitor would be rated for 192000 hours at 45 degrees C.

Well....I have to say that from my own experience I cannot agree less with Expresso.

As Robmar0se says, its upto others to make up their own mind, but posts such as this are dangerous, as they tend to mislead. Would I listen to Expresso "expert" opinion (after all that is all in fact it is, and under freedom of speach he may do so), or from manufacturers themselves?

You only have to google "Electrolytic capacitor aging" and I believe you may form your own opinion.

expresso wrote:

Ignorance is indeed a dangerous thing, especially when you know very little about electronic circuit design.

We can at least agree on that!

robmar0se

If one uses one of the PSU calculator websites (eg http://www.thermaltake.outervision.com/ ), you’ll probably find that if you calculate based on minimum usage that the suggested wattage on a new system is quite low, my guess well less than 500w (unless a high spec gaming rig),.

As Expresso has used OCZ (which incidentally I also use), it is difficult to find one less than 500w in their current range, so probably in many cases there would be headroom anyway.

However I would suggest anyone building their own rig, take the advice of the manufacturers and include the allowance for "electrolytic capacitor aging”, rather than the contrary advice of anonymous experts on this forum, although I’m sure they are posting in good faith.

mikey72

The point of ageing is good, and expressos accelerated ageing with temperature is valid. PSU are designed to achieve a maximum operating temperature at maximum power dissipation, running below maximum power will achieve below maximum temperature, and so less ageing. Buy bigger, it'll run cooler and last longer. More efficient PSU's dissipate less heat as well, so again less premature ageing.

Notmyrealname

robmar0se wrote: »

Ignorance?

Here is a quote from Thermaltake website. I leave it to others to decide if they prefer to take posters' views or that from a PSU manufacturer, and a good one at that!!!

Not ignorance at all. I'm a fully qualified electronics engineer. I've worked with kit decades old.

I'm speaking from experience, not regurgitating something from the marketing department on the website of a company with a vested interest in selling you something you don't need at a higher price.

Capacitors do age and do alter their values with age however in the design of a switch mode power supply, it isn't something you're going to notice and by the time it gets to the point you would, you'd have changed computer or upgraded anyway.

mikey72

Notmyrealname wrote: »

Not ignorance at all. I'm a fully qualified electronics engineer. I've worked with kit decades old.

I'm speaking from experience, not regurgitating something from the marketing department on the website of a company with a vested interest in selling you something you don't need at a higher price.

Capacitors do age and do alter their values with age however in the design of a switch mode power supply, it isn't something you're going to notice and by the time it gets to the point you would, you'd have changed computer or upgraded anyway.

You're half right anyway. Tantalum capacitors tend to suffer catastrophic failure, and causes the switcher to fail, usually in a cloud of smoke, and a bad smell. You'd notice.

espresso

mikey72 wrote: »

The point of ageing is good, and expressos accelerated ageing with temperature is valid. PSU are designed to achieve a maximum operating temperature at maximum power dissipation, running below maximum power will achieve below maximum temperature, and so less ageing. Buy bigger, it'll run cooler and last longer. More efficient PSU's dissipate less heat as well, so again less premature ageing.

"You're half right anyway". This thread was about buying a new replacement PSU, until is was derailed by one patronising poster, who is not an electronic design engineer and has not added any useful input to this thread. Modern PSU's are much more efficient ranging typically between 85 - 95% efficiency and this is certified at 20%, 50% and 100% full load. Therefore even one rated at only 85% efficiency, has a maximum of 15% of the full rated output wattage that could generate heat and consequently they run much cooler than the old PSU's.

Properly designed PSU's are toleranced to take capacitor ageing into account and will run at their full output rating 24/7, as a safe overhead is designed in. MTBF is typically in excess of 100,000 hours, which is more than 11 years for 24/7 operation. I have checked the expelled air temp leaving my PSU at 35 degrees C and this is not only the small amount of heat generated by the efficiently designed PSU but heat generated by the CPU as well.

The one who boasts "Been in IT for over 40 years!" says "Ask any electrician" - you may as well ask any plumber as they would know just as much about electronic circuit design!