Power Supply Efficiency

[trebuchet03]

I'm setting up a new computer... And I took a peek at my power supply....

100-127V - 8A

So that's fully loaded power....

So 110V *8A = 880Watts

The supply is rated to deliver 300W max...

That's to say, fully loaded my power supply has an efficiency of 34% Yes, fully loaded is one of the least efficient zones - but still, 34%?

I found a similar power supply (same model and power reqs/specs with a couple different connectors) that claims 70% minimum efficiency
ATX12V-G300BT (Bestec ATX-300-12EB3 Power Supply)

Any other explanation? What's your box say?

[TestDrive]

Look for articles about multiple Power Supplies compared at Tom's Hardware: Hardware News, Tests and Reviews

No time now - got to leave in a minute.
Will look up the specific article and write more on this tomorrow.

[NeilBlanchard]

Hi,

A much better source of info about power supplies is one of my other haunts: Silent PC Review.

silentpcreview.com | Everything about Silent / Quiet Computers

Typical older units were in the 65-75% efficiency range (though some are lower), and now 80-85% are fairly common. A few units are approaching 90% now; and it also depends a lot on your actual consumption. Most computers are 100-200 watts of DC consumption, so add ~25-30% to that for the AC use.

[TestDrive]

Originally Posted by NeilBlanchard

A much better source of info about power supplies is one of my other haunts: Silent PC Review.

silentpcreview.com | Everything about Silent / Quiet Computers

Wasn't aware of that site. Good find, looks like their tests are as through as Tom's Hardware.

Here a link that takes you straight to the Power Supply Articles & Reviews at Tom's Hardware.

And here is a link that takes you straight to the Power Supply Articles & Reviews at Silent PC Review.

Silent PC's article about Power Supply Fundamentals looks particularly good.

The specific article at Tom's Hardware that I had in mind is
Inadequate and Deceptive Product Labeling: Comparison of 21 Power Supplies

In spite of the high end-user retail prices, our lab technicians were surprised by the test results. No fewer than 6 power supplies struggled under full load: 3 of the candidates simply went up in smoke, while the other 3 shut down prematurely. The "black sheep" among all the test participants was Maxtron, with the TOP-520P4. According to the manufacturer's specifications, this power supply is represented to be capable of withstanding 520 watts - divided up over the corresponding voltage ranges. However, in our test, it ceased working at 446 watts. Completely dead!

Lawyers, politicians, statisticians and Power Supply Manufacturers - they all tend to bend the truth and a significant number of each one tell out and out lies. Take manufacturer's specification numbers with a great, big block of salt. Shop for specific units that have performed well in tests conducted at one of the above sites. Or shop for brands that consistently perform well in multiple tests conducted at one of the above sites.

[Daox]

Take a look at The 80 PLUS Program | Home for efficiency info on different maker's power supplies. I've been looking into this a lot recently. My power supply is probably 8 years old or so, it claims 70% efficiency at full load. I look forward to finding a new 85+% for whenever I do a new build.

Its pretty horrible when you look at the numbers. 70% efficiency and my computer at idle consumes 150W. That means I'm wasting 45 watts as a best case scenario!

[roflwaffle]

Originally Posted by trebuchet03

What's your box say?

Supposedly it's ~80% efficient at full load. Not bad for a $5 PS.

[NiHaoMike]

The 8A is the fuse rating. It has nothing to do with actual load. Also, you'll need a True-RMS meter to accurately measure the current as it is very nonlinear.

[The Atomic Ass]

Originally Posted by Daox

Its pretty horrible when you look at the numbers. 70% efficiency and my computer at idle consumes 150W. That means I'm wasting 45 watts as a best case scenario!

Have you verified that number by a Kill-A-Watt or other meter? I HIGHLY doubt you'll be drawing 150w at idle, unless your monitor is factored in. With my Mac Mini, (smaller than your computer I imagine, but keep reading), 2 external hard drives, external DVD-writer on idle, USB hub with a cellphone charging off it, and a broadband modem, all while running file-sharing to the tune of 10-15mbps, (A lot of disk access), was only drawing about 65 watts average, at the wall, through a UPS. I only get up to about 150w with my monitor running.

Trebuchet: The 8A is maximum burst current draw. As in when all the motors in all the hard drives, dvd drives, and the processor are all drawing maximum load, while the PS capacitors are charging up, which is what happens in a start-up condition, when you have enough equipment tacked on to run the PS at full load afterwards.

[NeilBlanchard]

Hi,

Mac Mini's and the new iMac's are exceptionally efficient and use very little power. 150watts is a bit higher than typical, but with a gaming video card, it is not hard to get the idle in that range.

Check out Silent PC Review for all sorts of good info on power supplies and efficiencies.

[Daox]

Yep, thats what the ol kill-a-watt tells me.

I just reformatted this weekend and ended up ripping out my sound card and extra network card. I'm now using onboard sound and lan. This dropped me down to 141W at idle.

Specs are:

Athlon xp 2500 barton core
Radeon X800 GTO
Seagate Cheetah ST336607LW 37gb scsi hd
Western Digital Caviar SE16 WD2500KS 250gb sata hd
Enermax EG365P-VE power supply


I'd be interested in knowing how much juice the scsi card and HD eat up.