Forum Home
Press F1
Thread ID: 126726	2012-09-15 03:31:00	Computer wont boot after power surge	antares (16890)	Press F1

Post ID	Timestamp	Content	User
1301088	2012-09-19 09:37:00	Nope. See how I use the word maybe. :)	Trev (427)
1301089	2012-09-20 06:10:00	Oh yeah...:)	ChazTheGeek (16619)
1301090	2012-09-20 12:16:00	One popular urban myth is that low voltage or sudden power loss can harm electronics. Never true even decades before the IBM PC existed. Also provided were numerous hard facts (at a layman's level) that further demonstrated why power off is not hardware destructive. But too many see a blackout. Then see damage. Then 'assume' a blackout did damage. Classic junk science reasoning is knowledge from observation. At last count I had repaired in excess of 700 switching power supplies of the same general design as computer power supplies. They employ pulse-width modulation to switch the rectified high voltage into the transformer primary, and feedback to regulate the voltage out of the secondary. Any event that effectively simulates or causes a double or triple switch-on pulse, or causes a spike that exceeds the voltage or current rating of the output switching device will cause a catastophic failure. These conclusions were based on observation, with due regard to underlying concepts and fundamental electrical and electronic principles. Similarly, running at low input voltage increases the on-time of the output device ans its current loading. Not surprisingly, many supplies die during brown-outs or brief supply interruptions. Another example of junk science reasoning: power cycling a light bulb causes premature bulb failure. If one knows otherwise, then the doubter posts facts and numbers that say 'why' he doubts. The life of any thermionic filament, whether valve or light bulb is determined primarily by the number of on-off cycles it can withstand. Ever wondered why light bulbs almost fail at switch-on? Have you not heard the 'ping'? They are heating devices but their filament resistance is low when cold and increases when hot, thus the inrush surge current at switch on is very high. It follows then, that since the filament gradually loses mass by evaporation during use, at 'end of life' stage it weakens and finally it cannot withstand the surge of inrush current and ruptures. There are many examples of filament bulbs that have run for many thousands of hours past their normal lifetime, simply because they were never switched off. In 1984 I was in one of Panasonics's torch battery factories in Japan and their party piece at the end of the tour was to produce a torch that had some of their original battery production in it, dating back decades. They would switch it on for a few seconds, and it would put out some light, and this supposedly proved the longevity of their product. In fact all it proved was the quality of their manufacture, because the lifetime in milliamp-hours would quite naturally be extreme if the current drain was just a few milliamps a week. I wouldn't try this with chinese batteries though, I've had some leak and go flat before they got out of the packet. In science: observation alone can very often be quite sufficient to create a hypothesis, but further investigation may be required to develop a proof. I think Darwin would take issue with your viewpoint. Cheers Billy 8-{) "I refuse to prove that I exist," says God, "for proof denies faith, and without faith I am nothing." "But," say Man, "the Babel fish is a dead giveaway, isn't it? It could not have evolved by chance. It proves you exist, and so therefore, by your own arguments, you don't. QED." "Oh dear," says God, "I hadn't thought of that" and promply vanishes in a puff of logic." -- Douglas Adams, from "A Hitchiker's Guide to the Galaxy"	Billy T (70)
1301091	2012-09-20 14:47:00	The life of any thermionic filament, whether valve or light bulb is determined primarily by the number of on-off cycles it can withstand. Ever wondered why light bulbs almost fail at switch-on? Defined was junk science: knowledge from observation. Knowledge that was not defined by numbers. Blaming power cycles is a classic example. Same reasoning also proved spontaneous reproduction and that bread mold breeds maggots. Light bulb industry has long known that power cycling does not determine bulb life expectancy. Industry even has formulas - numbers - that define life expectancy only in terms of voltage (temperature) and hours of operation. Why does science know something completely different from one who is using observation? Science even defines facts with numbers. The always required perspective. It was taught in junior high science. A hypothesis always requires a conclusion based in well proven principles. Knowledge means a valid hypothesis combined with extensive experimental evidence (ie controlled conditions). None of that exists with the power cycling speculation. Anyone could first learn this stuff from the IES Handbook - an industry standard. Unfortunately too many want to know without first learning whats and how. Without first doing their homework. Without first learning the numbers. Classic junk science. Anyone could see this. An orange traffic light flashes (power cycles) all night long. So it must be the first to burn out? Of course not. Green bulbs have longer hours of operation; fail sooner. Anyone can read numbers for life expectancy defined on a bulb's box. But no. Instead wild speculation based only in observation becomes knowledge. Too many foolishly make conclusions this way. One may feel it must be true. Therefore will even ignore observations that contradict those feelings. That is sufficient for knowledge? Of course not. Feelings are a common source of wild speculation. Knowledge always requires learning principles provided by hard facts and numbers. And no emotion. Anything less is classic junk science reasoning. Another perfect example of junk science - low voltage causes electronics damage. If true, then one can explain which part is damaged. Nobody does. Because the conclusion is only from observation. At minimum, one must first learn details; how a switching power supply operates. Necessary to have a hypothesis. Same reasoning also 'assumed' sudden power loss can destroy a hard drive. Classic junk science created by even ignoring what component failed on a drive. Too many cannot be bothered to learn this stuff. Are 'expert' only from observation and hearsay - because that is easy. That mistake is quickly identified in posts deviod of hard facts and numbers. If you know power cycling causes bulb failure, well, where are your numbers? No numbers indicates what?	westom (16792)
1301092	2012-09-20 20:35:00	Oh stop feeding the troll already. Westom, you ask for numbers and evidence but provide none yourself. You accuse others of being rude while ignoring their arguments and continually calling their arguments Junk science, again without proof. You write great walls of text to make minor points that could be expressed in a sentence. And to top it all off you are just plain wrong, inrush current stresses electrical and electronic components and is the reason as pointed out why many devices fail during power up. This is not junk science it's established fact and anyone who's studied electronics at even a basic level knows it (you want numbers, I have 27 years of experience dealing with faults in electronic equipment in the telecommunications industry) . Power cycling does in fact lower the life of almost every kind of electrical device (and especially filament bulbs) but guess what, so does simply being powered on which explains why the green light might fail before the orange - not because power cycling doesn't shorten the life, but because being powered on shortens it more. I suspect you don't even know what inrush current is. You either fail to understand or read the arguments made here, or you are deliberately ignoring them and posting this stuff either for entertainment or to advance your own point of view, either way it's long since become pointless.	dugimodo (138)
1301093	2012-09-21 04:13:00	Westom, you ask for numbers and evidence but provide none yourself . You accuse others of being rude while ignoring their arguments and continually calling their arguments Junk science, Well, if your belief is based in science, then that denial came with numbers or facts that said why . In reply was junk science: a blanket denial with no numbers . Knowledge from hearsay . Provided was where to read numbers for light bulb life expectancy . Defined were what those numbers are: voltage (temperature) and hours of operation . Posted but again are the numbers you accuse me of not providing . Those who actually study electronics know that some devices have power cycling limitations . For example (more numbers), the worst spec I saw for a disk drive (an IBM) was 40,000 power cycles . That is power cycling a disk drive seven times every day for . . . 15 years . Well, yes, every device does has a power cycle limitation . The drive is long obsolete and retired before power cycling does failure . So yes, some devices will eventually fail due to power cycling . And nobody cares . BTW, most drives should power cycle seven times daily for . . . 39 years . Perspective: power cycling causes damage - that would never be seen . Manufacturing defects are a more common reason for failure . Transistors have a power cycle limitation . How many thousand times does a power transistor (inside a supply) power cycle every second? A number so large as to be irrelevant; no longer be listed in datasheets . Many techs routinely use observation as if knowledge . See a failure . Only replace a box . Do not define what part failed . And therefore blame power cycling or fictional surges (earlier posts were classic examples) . Most failures are manufacturing defects . Or, in the OP's case, maybe a software failure . Diagnostics are executed to identify a defect before fixing it . But some techs still want to blame "the usual suspects" (even though we all knew who shot Major Strasser) . Bulb damage from power cycling is speculation justified by urban myth; promoted by hearsay and soundbytes . But again, if you know otherwise, then please repost your denials; this time with some actual facts and numbers .	westom (16792)
1301094	2012-09-21 05:26:00	this time with some actual facts and numbers Fact --- Power Supply in a PC fails - Customer only wants the PC going again. Change power Supply and test PC, charge Customer for new power Supply and labour = Happy Customer. OR Tech figures out what happened to power supply and cause failure - "could" spend hours doing so - charges customer for time customers doesn't give a sh1t about the cause - then charge for Power Supply and labour on top = customer would tell you where to put it.	wainuitech (129)
1301095	2012-09-21 06:09:00	Fact --- Power Supply in a PC fails - Customer only wants the PC going again. Change power Supply and test PC, charge Customer for new power Supply and labour = Happy Customer. OR Tech figures out what happened to power supply and cause failure - "could" spend hours doing so - charges customer for time customers doesn't give a sh1t about the cause - then charge for Power Supply and labour on top = customer would tell you where to put it. +1	KarameaDave (15222)
1301096	2012-09-21 06:51:00	Many techs routinely use observation as if knowledge . See a failure . Only replace a box . Do not define what part failed . And therefore blame power cycling or fictional surges (earlier posts were classic examples) . Most failures are manufacturing defects . Or, in the OP's case, maybe a software failure . Diagnostics are executed to identify a defect before fixing it . But some techs still want to blame "the usual suspects" Bulb damage from power cycling is speculation justified by urban myth; promoted by hearsay and soundbytes . But again, if you know otherwise, then please repost your denials; this time with some actual facts and numbers . Hmmm, Let's see now . This seems to be building into a classic case of 'if I can't be right, I'll be wrong louder and longer than anybody else . ' :D Firstly, and speaking for myself only, I have never been a 'replace the box' technician and apart from low-value smd-based devices that are not worth the time and effort, I repair to component level anything for which I have or can get parts, even items that I have to cut apart to get inside . My most recent repair was a set of rather cheap disco lights, the sort with a built-in mic and flash in time to the music . They had an unexpected drink and suffered a nasty electronic failure, but are on their way back to the owner tonight (freebie job for a friend) . Anyway, I'm not getting at you, but there is a shortage of factual information on which to assess the validity of your opinions, so let's start by asking how many switch-mode power supply failures you have diagnosed and repaired, computer or otherwise . That will do for starters . Then perhaps you could go on to explain how many designs you have seen where low input voltage on a smps has allowed 'the mark to exceed the space' to the point where the current and temperature limits of the junction rose to destruction level . Then tell us what you did about it . For statrters, I'll put up my credentials for scrutiny: One of my responsibilities was to analyse common failures and report back to manufacturers via my employer, with recommendations for changes . Exactly what revisions were implemented was entirely up to the manufacturer, but sometimes it was simply a change of output device manufacturer, a change of the device specification, or revisions to the control circuitry in relation to over-current conditions . The most popular protection system was to use cyclic interruptions to the drive to prevent overload failure (popularly known as hiccuping, but others would scream instead) . I can't imagine why my employer went to all that trouble of course, if they had been privy back then to your superior knowledge they could have just dismissed the failures as a myth and made bigger profits from repeat repairs . Testing was done non-destructively using jigs that allowed controlled run-up and run-down of the supply voltage with real-time current monitoring and an automatic current limiter to prevent junction breakdown, and frankly, I can't believe we wasted so much time on a non-existent failure mode . I also dispute your views on incandescent lamp failure . It is indeed arguable that light bulbs are not 'damaged' per-se by power cycling, though the difference is entirely semantic; they simply go through an aging process which brings them to an EOL (end of life) condition . EOL would be defined primarily by the number of hours of operation, the stability of the supply, and the number of on-off cycles; with temperature (enclosed operation) and vibration as a potential fourth and fifth factors . Did you ever wonder why 'long-life' incandescent lamps have sturdier filaments and more filament support? In the days of monochrome television, sets with series filaments failed through filament failure much more often that those with parallel filaments, so much so that thermistors were used to reduce the switch-on current surge, but they still blew filaments at regular intervals (they were developed originally from light bulb designs, and it was the blackening of the inside of the bulb by boiled-off filament material that was in part responsible for discovery of electron flow and the development of radio valves) but sets using parallel filament had the switch on surge restricted by the characteristics of the supply transformer, and they rarely suffered filament failure, but I digress . Manufacturers often quote their anticipated product life in 1000's of hours of 'normal usage' and I wonder also what that could mean, and what relationship it bears to your particular interpretations . Incandescent lamp life is, in measurable part, determined by on-off cycles, and to suggest otherwise is arrant nonsense . An excellent parallel is the extreme mileage between overhauls achieved by motor vehicles such as trucks and taxis which have significantly fewer cold starts per 1000km than Granny's shopping basket, but no doubt you have a theory on that too . The MTBF for electronics, especially power supplies, and also the humble incandescent lamp is determined by component quality, operating conditions, and usage (including on-off cycling), there is no getting around it . Cheers Billy 8-{) P . S . Did I mention I was a National Examiner in electronics for 10 years? No doubt you have qualifications and experience too, so maybe you should share them with us . That might put some weight behind your opinions .	Billy T (70)
1301097	2012-09-21 07:35:00	Oh stop feeding the troll already. Westom, you ask for numbers and evidence but provide none yourself. You accuse others of being rude while ignoring their arguments and continually calling their arguments Junk science, again without proof. You write great walls of text to make minor points that could be expressed in a sentence. And to top it all off you are just plain wrong, inrush current stresses electrical and electronic components and is the reason as pointed out why many devices fail during power up. This is not junk science it's established fact and anyone who's studied electronics at even a basic level knows it (you want numbers, I have 27 years of experience dealing with faults in electronic equipment in the telecommunications industry) . Power cycling does in fact lower the life of almost every kind of electrical device (and especially filament bulbs) but guess what, so does simply being powered on which explains why the green light might fail before the orange - not because power cycling doesn't shorten the life, but because being powered on shortens it more. I suspect you don't even know what inrush current is. You either fail to understand or read the arguments made here, or you are deliberately ignoring them and posting this stuff either for entertainment or to advance your own point of view, either way it's long since become pointless. +1, Westom pay some respect to members who have been on this forum longer than you.	ChazTheGeek (16619)
1 2 3 4 5 6