I'm servicing a PC that got zapped by a power surge following a neighborhood blackout. It killed the HDD, which is currently at a data recovery center, but that's a subject for a different thread. Anyway, while testing the rest of the system to test the functionality of other components, I noticed something odd... When I provide power to the PSU, either by plugging it into the surge protector or, switching on the PSU switch, the system seems to come to life (the CPU and chassis fans start spinning) for about a second then goes off again. The system can then be booted up normally from the chassis power button. But this is the first time I've ever seen an ATX system react (aside from flashing lights on the keyboard or mobo) to the PSU being swtched on or plugged in. Is this "normal" behavior for some ATX PSU's, or is it something to be concerned about? Could it be connected to the power surge that finished off the HDD?
Failing to power up the first time is abnormal behavior. However without details, no one can do anything but wildly speculate. For example, you don't know if that first power cycle is from something in the power supply controller or something else in the power supply. Is it powering up and down because power supply voltage starts excessively or does not achieve proper voltage? Or is voltage monitoring circuit measuring voltages too soon? You don't provide numbers, so the replies will be nothing more than speculation.
Surges typically don't harm disk drives. If a surge did, then the unit that should keep surges from happening AND contain other circuits to quash surges did not do its job. That is the power supply. But again, without further details such as what was damaged on that disk drive, then we can only speculate.
One thing common to all replies. Excessive voltage, or voltage monitor failure, or slow voltage startup, or disk drive hardware failure - each would have a common problem within that power supply. Without further data (at barest minimum, using the 3.5 digit multimeter), then we can only speculate. Even the voltage response on purple, gray and green wire from power supply to motherboard during power up would report significant information.
A neighborhood blackout does not create surges and should never cause any hardware failure > I'm servicing a PC that got zapped by a power surge following a
read it again, Tom. he was talking about the initial burst when the supply is connected to the AC line. Some supplies do this, till the + 5 VDC standby supply is stable, then it shuts the main supply down.
The hard drive is powered directly from the main power supply, so it is possible to be damaged. I've had numerous items damaged from poorly designed switchers, including the voltage regulator IC in monitor power supplies. The switcher draws more input current at lower line voltages. This increases the ripple to the switching regulator, and can let spikes through to the load. Exceed the breakdown voltage and you kill the ICs or short out tantalum capacitors.
A brownout can cause the supply to send the wrong voltages, with a lot of ripple, as explained above
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
Power supply makes damage from AC mains not possible. This assumes a power supply is properly constructed. However if that internal protection provided by a power supply is compromised, then other hardware would also be damaged.
Not only does a power supply make such voltage transients irrelevant. Supply also has other functions so that a power supply will not cause computer hardware damage. But again, to sell power supplies at discount prices, these required functions can be missing. Missing functions inside a 'dumped' power supply could permit disk drive hardware damage.
A brownout cannot cause wrong voltages with much ripple. Either the supply must provide sufficient voltages or the power supply must shutdown. And ripple 'spikes' that are destructive are also made irrelevant by power supply functions (again assumes the power supply did not cost less than $60 full retail).
Lower voltages with excessive ripple do not cause hardware damage. Low voltage would cause a non-destructive (to hardware) crash. After all, what does a power supply do when it powers down? The supply sees a brownout as line voltages slowly diminish over many milliseconds. If brownouts cause hardware damage, then powerdown also causes damage - in all electronics.
How low must AC main voltage drop before a computer (properly constructed) sees a brownout? Incandescent bulbs would be glowing at less than 40% intensity. Even when incandescent bulbs only glow at
40%, the computer must work just f> read it again, Tom. he was talking about the initial burst when the
You assume much! Many PC manufactures use crappy PSU to keep costs down and I've seen much hardware damage because of it. Yes in the perfect PSU, ones that are constructed properly you don't have issues. However, many PC dealers both large and small use crap to keep costs down! I would hazard to guess that a very great many PCs (probably at least 50%) are out there with PSUs that are not going to properly protect hardware!
You have repaired how many computer power supplies by replacing individual components inside? Therefore you know why damage happens? I have. Those 'crappy PSUs' are mostly found in clone computers where computer assemblers buy only on price and watts - don't demand numerical specifications. Mainstream manufacturers do provide better superior supplies.
I did not post by assuming. I repair things such as power supplies only to learn why failures happen. And too often, I find others claiming reasons for failure when they don't even know how to use an oscilloscope nor understand how electricity works, nor understand how a power supply works at the component level.
Meanwhile, let's look at the output from the computer grade UPS. 120 volts is really two 200 volt square waves with up to 270 volt spike between those square waves. Why is that UPS acceptable? Because computer power supplies even make crappier electricity from a computer grade UPS irrelevant.
I did not assume anything. Been doing this stuff for too long. Have found too many component failures falsely blamed on myths only because that other only knows how to swap boxes and boards.
A first test to identify an inferior power supply: operate an AM (longwave) radio adjacent to the computer. If computer causes interference, then power supply quality is suspect.
Man you need to get a sense of humor. My comment was meant as a joke. In the post that I replied to you made the following comments:
"Power supply makes damage from AC mains NOT POSSIBLE. This ASSUMES a power supply is properly constructed."
"again assumes the power supply did not cost less than $60 full retail"
I found it entertaining that you would post, "Power supply make damage from AC mains NOT POSSIBLE." But then in the next breath you basically said that it does happen. (" This assumes a power supply is properly constructed....However if that internal protection provided by a power supply is compromised, then other hardware would also be damaged.") And later you specified that it happens on PSUs "under $60."
Then in my reply I went on to agree with you, that hardware failure happens on cheap PSU's. And for my agreeing with you I got your "better than you" post and yet another assumption. In your reply you assume I have no experience with repairing PSU when indeed not only do I have experience with computer PSUs but also high voltage PSUs that are used inside Toshiba copy machines. (PS. I have also repaired a PSU in a touchscreen CRT monitor by replacing components within the PSU.)
Mainstream manufactures may demand specific specs but they can still get POS PSU's. I have seen personally more than a couple Dell PSU's take out hardware. My boss had a DELL PSU take out the motherboard the day she got the PC. I called dell and they overnighted a new PSU and Motherboard which I installed. I never did open that PSU because I had to ship it back to Dell.
Please man, get a sense of humor, I meant no disrespect, it was meant as a joke. And please stop assuming I'm some dumb-ass 14 year old going around with my pocket protector working on my aunties PC. I have two degrees and I work in the industry repairing and installing computers, and networks, and am a tech for Toshiba.
He is a troll, so just ignore him. I just picked up 14 power supplies from a computer store on the other side of town. They all failed in one day from electrical problems in one small area of town. It also fried 8 motherboards and a number of hard drives (Which I didn't get). Severe voltage droop, or extended brownout for up to a minute happen fairly often. I've lost two Air Conditioning compressors over the last couple years, when the line dropped to the point all of my the fluorescent lamps went out, but the compressor was still trying to run. Both times the run winding opened inside the sealed compressor. If they had been in a larger A/C the breaker would have tripped, but a compressor rated as a nominal 750 watt load on a 20 amp circuit just destroyed the motors.
Tom has no idea just how bad the power supply design is, and the big computer names are some of the worst. I've seen several that had an electrolytic explode and go out through the fan during a brownout. He also ignores the fact that a lot of people keep a computer till it can no longer be repaired. One that had a cap explode fried the motherboard, RAM, CPU, hard drive and CDROM drive. The only part that was still good was the front panel power switch, and it did not happen during a lightning strike.
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
From what I have seen, Michael was a good technican. He knows how to fix things; has much less appreciation of larger design issues or of why things fail. He admits to not having engineering training.
However, even a technician knows the difference between electronic devices - that are not harmed by low voltage - and motors - that are harmed by low voltages. Why would he cite an air conditioner compressor - except to confuse?
Properly designed electronics are not damaged by low voltages. Motors, however, draw excessive current during low voltage and therefore are harmed by brownouts. If voltage drops so low that flourescant lamps extinguish, then motor most certainly would be at risk. And any technician would know that. Meanwhile electronics would not be at risk - as an electronics technician should know.
How does an electrolytic fail during lower voltage? We all know of the many electrolytic capacitor manufacturers with defective electrolyte. Therefore its equivalent series resistance would be excessive. Failing capacitor would then fail on higher currents during low voltage. Capacitor failed in a condition that all electroncs must operate in without failure - as even defined by industry standards including CBEMA. But electronics with defective capacitors could fail on low voltage. Low voltage did not cause damage. Electrolytics capacitors were degrading - and would be failing soon if not for an extended low (but normal) voltage.
Michael - when are you going to return to being a good technician you once were? You admit to having only a technician's education. If you had understood the engineeering, then you would have never tried to cite a motor as proof that low voltage damages electronics. You should know better. Why did you do it - so that you could post insults? Such Rush Limbaugh logic does not become you - or are you also on drugs?
BTW, Michael, I am still using some 486 computers. But again you make assumptions without first learning facts. I also repair numerous old machines only to learn why failures happen. Low voltage is not destructive to electronics - despite myths that say otherwise.
Another little hint, Michael. One of my first engineering jobs was power supply design and manufacture.
A damaged air c> He is a troll, so just ignore him. I just picked up 14 power
I was an engineering tech, just before I ended up on 100% disability.
As an example to how often there are brownout problems. Or is that too complex a concept?
I've seen a lot of brand new computer power supplies that don't work at 90 volts. Some barely work at 105 volts. If you believe that every computer power supply meets your mythical spec, you have my sympathy.
The output capacitors can, and do, from time to time. The output of the switcher has more noise and ripple, which causes higher ripple current through the output capacitors. If they were choke input, the problem would likely go away.
You may have a degree, but the failure during a brownout is when the on time in the PWM is high enough, and at a high enough current to short the switching transistors on the input side of the supply. Even you can understand Ohms law, can't you? If the capacitors are smaller than needed, or they are losing capacitance from drying out, the added ripple causes even higher current peaks to the switching transistors.
It is truly amazing the low grade, non UL approved power supplies used to build computers these days. Its a wonder they work at all, and a lot of them will not put out the rated VA, right out of the box. Let them age, and you're just waiting to destroy something else in the computer.
Yes I am on drugs, thank you very much. I would be dead without them.
They are: Metformin Glipizide Simvastatin Triamterene Lisinopril Naproxen
See anything illegal in there? Mind altering?
What drugs are you on?
So you stop at 486? How about the original five slot PC? Have you converted any of the 256 KB IBM XT motherboards to 640 KB? Or a Commodore Pet? Maybe their IEEE-488 interfaced floppy drives with a pair of CPUs? I recently lost my Kim-1. I've repaired hundreds of Commodore 64 computers, about 75 VIC 20s. Oh, yes. I did the testing to select the power supply for the Microdyne/L3-com RCB 2000 which sold for $80,000 when it was released to the high end Telemetry market.
I've also worked on high voltage, high power DC power supplies at over 200 KW in Comark TV transmitters.
Good for you. Have you ever built a TV station from scratch?
You're one to talk about slinging insults. What do you call accusing me of being on drugs? You also sometimes snip relevant comments when you're losing an argument.
Don't you remember? I tested out of electronics tech school while in the US Army. I was told that it was impossible to pass the test, then got the highest score on record. I was awarded the MOS of Broadcast Engineer, which was the highest rated electronics job at the time. I worked on 2 MW pulsed RADAR, TV transmitters, CATV systems, weather instruments, and whatever else came along that had to be done right, on time, and usually with no budget. I left the service when they dropped my MOS to hire civilians.
So much for learning anything in tech school. The US government decided it was a waste of three years and the money, and put me right to work.
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
A 120 volt power supply that does not operate at 105 volts is completely defective. If numerous supplies fail at 90 volts, then a human purchasing those power supplies is suspect defective. Nothing new here. These requirements were even standard 30 years ago. Why are power supplies that fail under 105 volts suddenly acceptable when even industry standards in 1970s said otherwise? Did we forget how to make minimally acceptable electronics? Or are bean counters, masquerading as computer experts, now designing our computers?
No way around a standard that existed even 30 years ago. The CBEMA even provided charts that say low voltage must not cause hardware damage. If low voltage damages electronics, then electronics were defective. And yet some see such failures; then assume that is acceptable.
Even when lights dim to less than 40% intensity, a computer must not only work. It must even startup during the low voltage. This requirement is even in Intel specifications for power supplies. Just another engineering source that demands low voltage not be destructive. They did not see a failed supply and then assume failure is acceptable. They did an engineering analysis and learned such failures
- even 30+ years ago - were not acceptable. That has long been reality. Low voltage cannot damage working electronics. But then some even think a power supply that fails below 105 volts is acceptable. No wonder they would assume (without facts) that low voltage causes damage.
Nothing posted in this reply is for Michael's benefit. This reply provides the lurker with industry standards and facts - so that insults and junk science reasoning are ignored.
CBEMA in the 1970s demanded that computers start up and work just fine at 90 volts. In direct contradiction to Michael is this direct quote from an Intel specification:
That means a "brand new computer power supply" was OK when it did not operate as its own label claimed? Of course not. Michael deems personal observation more accurate than more than 30 years oft industry standards. He insists a defective supply means that industry standards are wrong. Apparently Michael works with numerous defective equipment.
Michael also claims (obviously in error) that power supply damage during low voltage is normal. Again his observations alone are sufficient to declare a classic junk science fact. Michael denies industry specs from even 30+ years ago.
What does that 1970s CBEMA specification demand? On that standards chart (also called ANSI/IEEE Std 446), in a region where voltage is well below 90 VAC, is this phrase:
Yes, but another industry standard that says Michael is wrong. Not only must a supply work at 90 volts to meet industry standard specs and labeled values. That supply also must be undamaged during lower voltages - also an industry standard requirement. If low voltage damages a supply, then low voltage during a power off transition would also cause damage. Another fact that says low voltage cannot cause power supply damage.
Posted insults were your first indication that Michael had misrepresented speculation as fact. Even his last posted numbers disagrees with his own 105 volt claim. Low voltage does not damage electronics. And yes, many defective power supplies are dumped into the market because many don't even bother to first learn well proven industry standards - as Michael demonstrates. Michael saw failure during low voltage. Using junk science reasoning, he declared low voltage as destructive to power supplies. Such reasoning creates a ripe market for defective computer supplies.
Tell Intel that all the power supplies on the market don't meet their requirements. The "junk science" is yours, in that you refuse to accept that standards have, and do change over decades. You quote a "30 year" figure. The only personal computers of that era used bulky linear power supplies and the S-100 (MITS Altair), Exocisor (Motorola) or similar industrial busses with 1 or 2 MHz processors on an eight bit buss. You are insulting, yet you use the term on others. Can you say "HYPOCRITE"? I know you're familiar with "TROLL".
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
But again Michael somehow knows only from personal observation. Some technicians never eliminate that bad habit. Those with technical grasp first learn facts and concepts. Meanwhile, back in the early 1970s, we were using switching power supplies. No wonder these standards of 30+ years ago are unknown to Michael Terrel who must deny even using insults.
Michael says that low voltage damaged his air conditioner motor. That is proof (to him) that low voltage also damages electronics. Again, he knows only from observation - classic junk science reasoning. He has seen electrolytic capacitors explode during a brownout. Again, observation alone proves low voltage damages electronics. He has seen a motherboard damaged when a power supply electrolytic exploded. Again, industry standards for power supplies make this not possible. But Michael need not know facts. He saw. Therefore that must be an acceptable failure.
Michael appears to use and find acceptable discounted power supplies that are missing essential and industry required functions. His observations are proof that all power supplies are built in violation even of 30+ year old standards? In his world - yes. But that is classic junk science reasoning.
A power supply whose label says it works even at 100 volts "barely work at 105 volts". That is proof that power supplies need not work at low voltage? Yes, in a world according to Michael. In reality, a computer power supply must even work when incandescent bulbs are at less than 40% intensity.
The lurker is cautioned about these types who routinely know - need not first learn concepts, facts, and industry standards. Low voltage must never damage electronics - as was standard even 30+ years ago. As even defined on the industry chart (and ANSI/IEEE Std 446) with a phrase in large letters:
Low voltage damages Michael's power supplies? A large and profitable industry makes defective supplies for clone computer assemblers. Supplies that are missing functions found standard even 30 years ago. Michael sees defective power supplies fail and then assumes such failures are normal. They are not. But then 'bean counter' type computer experts are alive; posting even on the internet. Instead, learn from those who were trained as an engineer, learned concepts, underlying principles, technical facts, industry standards, and apparently have many more decades of experience. If Michael had such knowledge, then he knew we were using switching power supplies in the early 1970s. Then he knew of industry standards that are that old and that well established.
Low voltage does not damage power supplies. And power supplies must also include functions so that motherboard and hard drive will not be damaged.
Tom, just keep kicking the crap out of your dead horse. Keep citing out context snips from out of date references from groups that no longer exist.
You still haven't answered what drugs you're on, after you accusing me of being on drugs. I posted the list of what I have to take every day. What are you afraid of?
Keep snipping anything you don't agree with, but whatever you do, don't you ever learn anything up to date information.
Everyone knows that you'll keep trolling, till they have to pry the keyboard out of your cold, dead hands.
--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
Nothing is obvious after a trip through the blender you use to distort the facts.
Keep spinning your lies. Anyone that has read the entire thread knows what kind of person you really are. You never answer questions, you take things out of context, and you snip away any links that refute your lies.
I'll bet that you're not man enough to look at the back of your computer and tell us what the AC input voltage ranges are, because it will prove you wrong. I have over 100 PCs right not, and not one of them says it will work at 90 VAC.
I also see that you aren't hanging around the electronics newsgroups after they showed you for what you are.
Well I thought I had him run off aus.electronics for a while but he showed up again like the bad stench he/she/it resembles - possibly something to do with thefact that I had not posted there for a couple of months ....
Yep "Mister" I second that you generally cannot believe a claim that w_tom makes as he generally tells Lies and makes outrageous and impossible claims
Hi Michael I was wondering why that half witted Moron w_tom was calling me Michael on another Newsgroup (aus.electronics) now I know he is confusing me with you I dont feel Insulted then ......
He's had it out for me for a long time because I got him all fired up, then left him to stew while the other regulars tore him to pieces on one of the sci.electronics newsgroups.
BTW, I used to read and post on aus.electronics from time to time, but I got sick of "toaster boi". He was repeating himself as much as w_tom. I figured that I would find a group that didn't smell of cats and burnt toasters.
--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
Member of DAV #85.
Michael A. Terrell
Central Florida
snipped - Lies from w_tom ----------------------------------------------------
Ha Ha Ha Ha is that who designed those crappy Switchers that I saw in the
80's - the ones that blew up if you looked at them Sideways I can see them being an effort that w_tom would be proud of ;-)
Note here how w_tom deliberately misquotes Michael - Michael pointed out that the PCs in the 1970's used Linear power Supplies so w_tom twists Michaels words to say that there were no Switchers in the 1970's
where does w_toms knowledge reside ? That is difficult to say as he does not seem to have any Genuine knowledge but just makes up 'facts' to suit himself - he assumes Reality follows fantasy must be when the hallucinogens kick in.
Somehow I doubt that w_tom even has a clue what an SMPS is ....
---- Snipped more crap from w_tom
Common ???? I think w_tom is about a decade early here - they were far from common in the 1970's and were only getting popular in the 80's - no wonder Mister can tell how wrong w_tom is proving to be.
Damn straight! This is a telling fact that w_tom thinks they were common in the 1970's and even I know that they were not ..... ( I was a child of the
70's)
I know yet another challenge for w_tom ( apart from the regular challenge of remembering to breath ) - Name 5 Personal computers of the 1970's that used SMPS.
Maybe that is because Mister can tell the stench of excrement when he sniffs it .....
Nothing 'w' you have been shown up once again for the ignorant liar that you are.....
Its always a pleasure and anyone who is putting the boot into w_tom cannot be all bad ;-)
I guess the Fact that he is calling me Michael though is evidence that he is on some pretty hard core Hallucinogenic drugs when we both live on opposite sides of the world .... (well near enough)
Ahh look I just despise any Idiot who advises folk to do stupid things - Like depend on MOVs for "100% Lightning protection" if does not end up causing someone's death by advising them to keep using their computer during a Thunderstorm it will only be due to the efforts of folk like you and me. I kind of got side-tracked rolling out what has become Australia's third largest ADSL Network (probably a piddly little thing by US standards as we only have a tenth of the population) so I have not had as much of a chance to post of late and I think when w_tom did his regular Google search for Lightning, surge etc he must have noticed that I had not been as active of late so he thought he would get away with writing crap on aus.electronics .
Anyway I just thought I would put the boot in to w_tom while I was over here ;-)
Yes when I found these Posts on Google I thought your name sounded familiar ....
Best Regards (to Michael that is not w_tom - loser extraordinaire)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.