Slashdot Mirror
Brand Names Take On Generics In PSU Showdown
The Raindog writes "The power supply is perhaps the most overlooked element of a modern PC, and yet it's the one component that can irreparably damage the rest of a system. The market is littered with generic PSUs that are often much cheaper than name-brand alternatives, but can you trust them? The Tech Report aims to find out in its latest power supply round-up, which compares the performance, efficiency, and noise levels of a collection of reputable PSUs with some budget, no-name competition. As it turns out, any money you save on a generic PSU purchase will likely cost you more in the long run."
And before you think that all your PSUs are failing because you bought them on the cheap, you should also check your UPS. I had 13 PSUs die at Suso and thought it was just horrible luck with power supplies, until I realized that the 5 year old UPS that those servers were on was having issues. Since I replaced it, haven't had any problems since. *knock* *knock*
I have owned several HP/Compaq machines and NEVER have lost a PSU. And all of the ones I have built myself with parts (Antec) have had a PSU fail multiple times...
Generic power supplies are an awful plague upon our fair world. Why, just last week I was using my homebrew PSU rack to power my uncle's hospice life support system. Now I'm going to his funeral this weekend. Tragic, just tragic...
512 MB RAM, 20 GB disk, 200 GB transfer, five datacenters. $19.95/month.
Actually, some friends of mine had a no-name PSU explode, literally. One of the electrolytic caps decided it had had enough. I replaced it with an Antec and no problems for as long as they used it.
Oh, and the Antec PSU in my personal machine died one day, possibly because it was plugged into an 11-year-old surge protector. I got a free replacement (minus shipping), and the replacement is a nicer unit and handles more power.
Say what you will, but their service is nothing to complain about.
The difference between spam and poop is that you don't have to dig through septic tanks looking for real food. -- Me
I'm genuinely curious, what parts are different or better in a more expensive PSU? Or is it just a combination of confirmation bias and shelling out too many bucks to say it's not worth it? Not really an accusation, I'm more curious about what parts in a PSU can be "better", considering all the parts are mass manufactured, anyway.
Comment removed based on user account deletion
...As it turns out, any money you save on a generic PSU purchase will likely cost you more in the long run..."
To such statements, I say "Ohh puhleeze!" I use generic power supplies for all my PCs, which I never switch off by the way. Apart from increased noise after about 3 years of constant humming, I have no complaints for a product that costs me about 18 dollars.
I heard Google uses the same stuff too.
There aren't very many factories that actually make PSUs. You'll find that a great many PSUs share designs. So you can have cheapies that look like brand name PSUs. Ok so what's the money difference? Parts quality. The company making generics says "Ya give me the cheapest caps, fan controls, etc. I need lowest cost no matter what." The good brands say "Give us higher temperature parts, better quality, etc." Just because they look the same, doesn't mean they are built ot the same standard.
The difference between good and crap in electronics can often be as simple as the parts used. However, good parts cost more money so you are going to pay more for the finished product.
Personally, I'm a Corsair fanboy. They seem to spec really high grade electronics in to their powersupplies and those things do a great job.
They should have checked each power supply for a UL marking, and an entry in the UL Certification Database. Things seem to be getting better, though; the power supplies tested did not blow up or catch fire at full load. That's a big improvement from a few years ago.
The basic UL requirements are 1) no overload problems at full load, 2) no explosion or fire under output overload/short conditions, and 3) no single component failure can cause a fire (i.e. there should be a fuse of some kind in there.) It's permitted for an overloaded unit to fail and never work again; that's not a safety issue. Some no-name power supplies had real problems meeting those basic conditions.
Right. Bose is one of the best brands of speakers, and neon lights make your computer go faster. Sorry, but Antec is all marketing and no quality. All the ricer overclockers want to believe that Antec PSUs make their computers +0.2% faster, but the truth is, Antec uses some of the lowest quality parts in their PSUs and if you check something like badcaps.net, you'll realize that Antec is one of the most commonly reported brands in the PSU category, and most widely complained about brand in the forums.
Why does every article on power supplies feel the need to remind us that power supplies are probably the most overlooked component? Judging by the number of online reviews and by the 560 (!) power supplies available at Newegg, I think it's safe to finally retire the "overlooked" cliche.
Really? Most reviews of their PSUs have them with pretty good ratings, and their PSUs are pretty power efficient too. I have their EarthWatts 500 PSU and it's really good, my GPU technically requires 550W but it's fine. I've never heard of any bad things about Antec before, the worst ones are the ones from Best Buy et al that have some random company you've never heard of.
All your base are belong to Wii.
Except, of course, for the Antec EarthWatts. And the TruePower series. Both of which are... made by Seasonic.
jonnyguru.com is probably the best site for power supply reviews. the guy does FULL disassembly, even removing rectifiers and diodes and stuff. Because of this he is able to determine what OEM manufactured every power supply he tests, and also gives you comments that, if you payed enough attention, would allow YOU to determine the layouts typical of each OEM. He's also the guy that started using an actual load tester to test power supplies, instead of testing them by placing them in systems. The ORIGINAL hard core power supply test site!
I KUT J00 M4NG!!!
Taking an old 300 watt atx power supply apart to make a variable 0-5v and 0-12v bench power supply for electronics projects will teach you what you will need to know about quality of components if you have a decent DMM, or better a USB scope to see the difference when you switch out caps and other components. An often overlooked problem with cheap power supplies is that bad soldering and cheap PCBs cause the capacitors to work harder to move the same amount of charge. This effects voltage as it relates to capacitance's formula C = Q/V and its integral (dammit give me MathML) sigh read Wikipedia's definition. Doesn't matter how good your caps are if the circuit was soldered by someone being paid slave wages working 14 hour shifts. The real cost of the plenitude of cheap electronics is going to be an interesting story. Besides the obvious horrors of the places of manufacture in Asia and the dumping grounds in Africa where similar autoimmune disorders, cancers and genetic abnormalities are rampant is the story of what happens between those two places. It is not just the disposable aspect that technology plays in our culture that fascinates me but how little knowledge of the design and manufacture of technology is present in the people that purchase it, sell it or even repair it. PC techs without any electronics understanding are the worse, "Who needs an anti-static work area when I can use any flat surface including a pizza box?" and than wonder why they have half a dozen motherboards laying around the house that won't work. /rant
If I have the time before Xmas I will post my pics and schematics on my blog. Did it for an EE project. Working on LCDs and circuitry to display voltage and amperage (currently have 4 lobotomized dmm pcb w/ attached backlit lcd doing a reading each.
An Education is the Font of All Liberty
As a PC repairman I have run into the Bestec brand a few times, even built a few machine with them, and I can tell you it ain't the brand, although they are cheap and I tend not to use them(they have lousy fans if they even bother to have a fan at all) except in the occasional "granny" machines that are only doing basic web stuff. I can tell you the problem with both the HP and the eMachine you describe, because I have ran into it in the shop, and the problem is this:
For awhile there both HP and eMachine would put in VASTLY underpowered PSUs in their machines, which would naturally strain the living hell out of the PSU and cause it to blow, taking out the PC with it in many cases. I have actually seen fairly modern machines from HP and eMachine(a few years maybe 3 old) that had a Bestec PSU with a MAX rating of 200 watts! for any modern machine that is simply unacceptable. And woe be to anyone that actually tried to upgrade anything in one of those machines without yanking the PSU first, because with that little headroom it was bye bye PC.
As I put in an earlier post with the cheap PSUs you need to have at LEAST 150 watts of headroom ABOVE what you machine uses under max load. For example when my PC used 300 watts max ( figured using a calc like this) I used a 450 watt cheapo(NOT Bestec because I require a fan in my PC) and when I upgraded my PC and its power usage went to 350 I changed it out for a 550 watt PSU. So don't really blame Bestec. I have used them in a few ultra low power HTPC and ultra quiet desktops and they still work fine. Blame HP and eMachine for building a PC that used 200 watts and putting a 200 watt PSU in it instead of spending an extra dollar and getting one with a little more juice. That IMHO is the big difference between the cheapos and the big brands, with the cheapos you really need a LOT more headroom IMHO than you do with the big brands. But of course YMMV and everyone seems to have preferences when it comes to PC parts.
ACs don't waste your time replying, your posts are never seen by me.
Funny.
I've been "doing" computers since the early 90s. I've never had much problem with power supplies. And I do mean *never*.
I took an ancient, generic 286 computer, and upgraded it through 386SX, 486 DX/2, Cx 6x86, and AMD Athlon motherboards before finally switching to ATX. It was a cheezy, god-only-knows-who-made it power supply that came from a 'not-quite-aluminum-foil' AT case.
And I've done plenty of computers since. I've *always* bought the cheapest, craptastic cases and power supplies, and generally had years of excellent service before dying. I'd say my average life expectancy for a power supply is over 5 years, and I say that because I generally give up on the computer before the power supply dies.
What makes a computer last a long, LONG time?
1) Under-clock the CPU. Really. 10-20% makes a significant difference in reliability because it runs cooler, but almost never makes enough difference to notice for real-world, day-to-day usability. As you approach the thermal limits of our CPU, the longevity drops off sharply. 10% makes a *huge* difference.
2) Dust out the computer every year or so. Dust is an insulator, which causes heat "hot spots" that play hell with components.
3) Replace the fans regularly. I keep servers running for years on years on years by replacing the fans every other year or so. Along with dusting out the server, they keep on ticking far longer than you ever thought possible. I've had systems last well over 10 years with decent reliability by doing this when performance simply wasn't an issue.
4) Don't turn it off! Computers that are turned on/off every day last a few years. Servers that are babysat, running 24x7 at a consistent temperature run damn near forever. This costs money, so run the numbers to see what uptimes vs power consumption really costs you.
I've never noticed power supplies (cheap, expensive) being much of an issue. I've seen craptastically cheap hardware run under heavy loads for a very long time without complaint, and I've seen plenty of expensive, "high end" hardware die well before it's expected life time. /shrug/
I have no problem with your religion until you decide it's reason to deprive others of the truth.
I think most problems would be with older equipment made in the days when both the US and the EU countries were trying to make inroads in the Chinese suppliers. For a time the certification bodies seemed to go a little crazy and let the Chinese get away with murder because they all wanted to be the primary Chinese certification body. One of the best incidents I remember was an auditor going around a Chinese plant with ISO 9002 certification. All the documentation was there, all the procedures written up. In English. And no-one in the entire factory spoke English. I doubt this is the case with electricals any more.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
"You can hide a multitude of sins behind one of those "Warranty void if broken" paper seals. ;)"
Slashdot has one of those.
Shai Schticks:"You don't make peace with friends, you make peace with enemies"
If you look, that supply is only "UR" component recognition, i.e. it is not UL approved as a standalone item. UR just means that, if a UL tester finds it inside a computer being tested as a whole, he does not test the PSU individually but treats it as a single component. Replying to my own post, below, I should clarify that this seems to be a loophole that I have encountered before. Nobody should sell a UR assembly to an end user, it should only be sold to an OEM to replace an identical item in a UL piece of equipment. I hope this clarifies things.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
"We've narrowed our focus on PSUs in the 350-500W range, which should be enough power for most budget and mid-range systems."
This isn't quite true. The more important question is the amps on the +12V Rails? Even better if yours has a monorail design were all the power-hungry parts can get what they need. Also sustained rating is important. Not peak. And last even the better brands can be/go bad. My PC Power & Cooling 750 silencer was recieved DOA. It happens even to the best...much like hard drives.
Shai Schticks:"You don't make peace with friends, you make peace with enemies"
The main things you'll see in a high end PSU:
1) Voltage stabilizing in case the power coming to the PSU is not very good
2) Quieter fans
3) Output voltage/watts and efficiency stay within reason at higher load
4) Some generic heat up quite a bit.
Don't forget power factor correction (PFC). Especially in a datacenter.
Research and development is costly, to find out which components are good. Additionally, each part in the supply chain adds their profit margin to the product, so adding value in the early levels of the supply chain (as in manufacturing) can easily add a significant amount of cost to the end price.
Except, of course, for the Antec EarthWatts. And the TruePower series. Both of which are... made by Seasonic.
Not entirely true.
The older EarthWatts were made by Seasonic, the newer ones are made by Delta.
Source: http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story&reid=110
I've never underclocked a CPU, don't replace fans unless they break or get noisy and turn my computer off an average of two times a day. Never, ever has a computer broken even though I'm ignoring 3 of your rules.
The thing is my above observation means absolutely nothing as the sample is far to small to be of any statistical significance. I expect the same is true for your experience with PSUs. If someone has done testing on a reasonable scale, in monitored condition then it would be of real use to people buying 100s of units who want to minimise TCO.
I tend to buy reasonably good PSUs but mainly because I'm after energy efficiency (which is also why I power down twice a day).
Here's a partial list:
I've seen companies use components rated at or just below their stated current rating in order to save money (using 8 amp transistors in a 10 amp supply, for example). They'll often work right out of the box, but, since there's no margin built in they will run hot and eventually fail. As to component quality, take it from someone who designs and manufactures precision instrumentation, I can tell you that there can be an enormous difference in quality from one manufacturer to another. "considering all the parts are mass manufactured, anyway" is not a valid argument.
After you had 2-3 fail why did you buy 3-4 more?
Not a Twitter sockpuppet... but I wish I was.
Among other things granted to the Chinese during the Clinton administration was the relaxation of having to meet FCC part 15 subpart B for electronics. Some Chinese power supplies put out tremendous amounts of RF all across the spectrum. Might check that too if you are having problems with radio reception. When I can get them I always buy Taiwan made power supplies because #1 they are better engineered and built, and #2 they do not radiate RF. They have to meet the RF emissions requirements. So far I've had a large failure rate of Chinese power supplies because of being under engineered and the poorly made components. Electronics are like everything else being made in China today. They cut corners and mismanage and put out crap clothes, electronics, baby food, pet food, toys, medicines, and every body just keeps letting them get by with it. Most of my Chinese made shirts don't come with pockets anymore, the thread count of the cloth is lower, the sewing thread is smaller diameter and of poorer quality. The seams are sewn right up against the edges so they rip out in the washing machine and buttons disappear after a few washings too. And if you take medicines ask your pharmacist what the country of origin is. It might surprise the hell out of you. Almost everything I take now (blood pressure and kidney meds) are made in India. That scares the bejesus out of me too but not nearly as much as Chinese made meds.
Too lazy to create a sig...
Sadly, PC Power & Cooling is no longer an independent company, but was bought by OCZ last year. I'm not sure how this might affect quality in the future, but I still buy PC P&C PSUs for now.
After ~20 years in the industry, through thousands (probably tens of thousands) of PCs - everything from no-name dsektops to high-end IBM blade servers, I think I've witnessed (or received direct reports of) 3 or 4 PSU failures ever.
Heck, if someone asked me to rank the components most likely to fail in a computer, the PSU would probably be sitting just above screw holes and mounting posts.
WTF are you people doing to your computers ? Is the power supply in Australia really that much better than the rest of the world ?
Yes, I agree that jonnyguru.com is the best review source for PSU stability (noise and transients are measured with oscilloscope and compared with ATX specs) and build quality (inside pics and commentary on components used). If you additionally care about the noise your PSU is making at various loads, silentpcreview.com has those measurements. I bought a couple of power supplies based on the review on to those two sites, and never had issues stability or noise wise.
Lots of problems with this statement:
Another problem is with your advice to buy cheap power supplies. Cheap power supplies are made out of less environmentally friendly materials, and are also less efficient. The lower efficiency is both an environmental issue and a money issue.
Find free books.
It probably depends on the quality of your powergrid, too.
I've heard the american grid, on average, delivers much more fragile and "dirty" electrons than, for example, the european grid.
I guess it makes a difference whether your PSU has to deal with spikes and brownouts on a daily/weekly basis or whether it's just humming along on nominal line voltage.