Windows not booting after power failure, suspect HDD.

[-Slacker-]

So after the power failure yesterday, when I switched my machine on I got this weird screen, like "3rd master hard drive error" or something" and it wouldn't boot to windows.

I also noticed that I couldn't open the boot device screen (F11) anymore, only the setup screen (F2), which was fine too I guess because it does the same thing ... still found it strange though.

I tried unplugging/plugging the power and sata connectors from the HDD - same problem, then I plugged in different connectors from the same (and only) cable group from the psu - didn't work either.

Now, the only thing left that I can try out without some inspiration, anshort of RMAing the damn thing is to reinstall windows, but I don;t really want to do that yet, mainly because there's almost zero guarantee it will solve anything, since the BIOS won;t even detect the HDD...

Any ideas?

 

[Blain]

* What model PS and how old is it?
* Are you running any power protection ahead of the PC... surge... UPS?

* I'd unplug the PS and clear the CMOS memory via the jumper
* I'd test the PS
* I'd try a known good PS
* I'd connect the HD to a different machine and find out if data can be accessed

 

[-Slacker-]

* What model PS and how old is it?

It's a brand new 550w Chieftec APS A-135 S

* Are you running any power protection ahead of the PC... surge... UPS?

It's times like these I wish I were ... but no. Bare naked psu.

* I'd unplug the PS and clear the CMOS memory via the jumper

You mean take the jumper out and put it back in or...?

* I'd test the PS

I don't have anything to test it with ... other than my own setup, which seems to be working integrally except for the HDD

* I'd try a known good PS

I thought about that shortly, but the psu I bought was recommended as a budget psu by hardwaresecrets (or xbitlabs, can't remember from the top of my head). It wasn't a random purchase, but if I had enough money I would have definitely gone for a SeaSonic or something like that.

* I'd connect the HD to a different machine and find out if data can be accessed

That's not an option right now .. I only have another much older desktop that doesn't have a sata mobo...

 

[westom]

It's a brand new 550w Chieftec APS A-135 S ...
I don't have anything to test it with ... other than my own setup, which seems to be working integrally except for the HDD.

Those who better know hardware can only stay silent without critical facts and numbers. Those who know better will not post 'try this or try that'. But you must provide some critical facts.

For example, suggested was a power problem. The naked supply connects to AC mains exactly as one through a power strip or typical UPS. Ignore that nonsense to move on to real problems.

You need to buy (for about $18) or borrow a 3.5 digit multimeter. Sold in Radio Shack, K-mart, Wal-mart, Sears, or most any store that sells hammers - typically for less than the price of a hammer.

On the 20 VDC scale, measure the purple wire where it connects the power supply to motherboard. Simply push the probe into that nylon connector to touch metal and read the voltage. The numbers should be something around 5 volts - always - even when a computer is powered off but connected to the wall. Report the actual number in 3 digits.

Do same for any one red, yellow, and orange wire after the power switch is pressed. Then the next answer will report on the Chieftec (and other critical components) in the very next reply - definitively. That means the answer with be so accurate that you move on to other suspects and never look back - never suspect the Cheiftec again.

As I understand, it does boot, but does not read data from the disk drive. What else do you have to boot from? CD-rom? USB memory stick? Once voltages have been established, then move on to the hard drive by loading the disk drive manufacturer's comprehensive diagnostics. Your failure is why better manufacturers provide diagnostics. Only better computer manufacturers provide diagnostics for every component with a provided CD or from the web site. If yours is not, then you must obtain those diagnostics from each relevant component manufacturer.

Identify the disk drive manufacturer and model numbers. Then download the diagnostic from their web site or from the Ultimate Boot CD web site.
http://www.ultimatebootcd.com/index.html
A diagnostic boots from a CD-rom or from that memory stick. And without complications created by booting Windows.

Informed techs never go about wildly swapping parts. First get facts and numbers using the meter and diagnostics. Then things you do not know can be answered by those who really know this stuff. Without those facts and numbers, then those who actually know this stuff can only remain silent.

 

[Blain]

A $14.99 ATX PS tester (plus shipping), may make a for a better investment than a simple multimeter, in your situation.

 

[westom]

A $14.99 ATX PS tester (plus shipping), may make a for a better investment than a simple multimeter, in your situation.

Power supply testers are virutally useless for numerous reasons.

First you must disconnect wires. Good diagnostic procedure says you disconnect nothing until after facts are collected.

Second, it provides no loading. Therefore a completely defective power supply can be reported good by the tester.

Third, it provides no numbers. Numbers (to three significant digits) contain significant additional facts. Power supply testers provide no numbers. Therefore additional information from the better informed is not possible.

Fourth, a PS tester will not perform the many other homeowner tasks that a multimeter does - such as determine whether the car battery is dead or the alternator is defective. A homeowner using a multimeter becomes better informed - and fixes other household appliances faster. A multimeter can even locate other computer problems.

A PS tester that costs so much money does none of those. Can even report a defective supply as good. PS testers are strongly discouraged. Meters are sold even to K-mart shoppers because the meter is so ubiquitious, so useful, and so simple.

 

[Blain]

Thank you for your diatribe.

 

[westom]

Thank you for your diatribe.

Its only a diatribe if you are emotional. It is hard facts if you know this stuff. Please stop trying to make this nasty. PS testers are often recommended when a magic box will answer all questions; when thinking is not desireable. You may not like it. But that is the hard reality.

Apparently you do not understand how insulting that PS tester recommendation was. Its like telling the general that he could have won the war if he had only given his soldiers bullets.

 

[RebateMonger]

Apparently you do not understand how insulting that PS tester recommendation was.

I don't understand why you are insulted.

I own a couple of DVMs, but I use a power supply tester as linked by Blain when I need to check a power supply. They are fast, don't require three hands, and the voltages are provided to two or three significant digits, depending on which voltge is being measured. They also measure the time delay to the "Power Good" signal, which can indicate a problem if too fast or too slow.

They DO provide a load, which is necessary for power supplies to work properly.

They also provide a safe way to test voltages on a "suspicious" power supply. To test a power supply, you need to hook it up to a motherboard or to a power supply tester. Hooking a questionable power supply to a motherboard is not a good idea.

I have nothing against a DVM, but I don't have the time or patience to individually probe all the connectors and I need something that can check a power supply without risking the destruction of a motherboard.

 

[westom]

I own a couple of DVMs, but I use a power supply tester ...
They DO provide a load, which is necessary for power supplies to work properly.

The load exists. Is obviously near zero. Even tester heat makes that obvious. Woefully insufficient to report anything useful. Even that response time for the Power Good signal is useless due to insufficient load.

If a Power Good signal takes too long, then what happens. Nothing. Does not in any way obstruct booting. You should know that and also know why. The tester is a complete waste of money. And is popular among techs who never learned how electricity works. Who then take insult when they post bad and insulting recommendations. A PS tester is an obvious indication that electrical basics are completely unknown. Should have been obvious from spec numbers - that tester is virtually zero load.

Reason for a meter should be obvious if the OP posts voltage numbers.

 

[RebateMonger]

O.P.:

Your symptoms certainly sound like a failed hard disk. Standard procedure is to hook the disk to that second PC and test it. That can be done with your other PC if you use a SATA-to-USB adapter cable, a SATA-to-USB enclosure, or a PCI SATA hard disk controller installed in the second PC.

You can also try the disk maker's disk diagnostics boot disk. If your disk maker doesn't offer such a download, you can use other disk maker's diagnostics. Chances are it won't see the disk, either, but you might as well run the test.

 

[Emulex]

yup just had an X25-m do the same thing. should have advanced swap overnight new replacement on monday.. sad

 

[-Slacker-]

Oh wow 11 replies already...

Well, sorry for the late comeback ... I bumped into even more problems since I last posted (not the least of which not having internet on my machines...) ... Anyway...

...So, yeah, the HDD is definitely not working anymore, and this, combined with the fact that it happened just after a power failure would confirm the suspicions of the posters in this thread - that it's the power supply.

Interestingly enough (well ... not interestingly ... more like goddamn annoying) I found another sata hdd from another old machine laying around and I decided to test it - it worked about 2-3 hours (I managed to reinstall windows on it and all), but during this time it kept getting incredibly hot, like, boiling hot, and then, all out the sudden it stopped working. However, it was a POS HDD to begin with and the reason why I stopped using it was probably the heating issue, so I'm not sure if the PSU is t blame on this one...

 

[Emulex]

you don't have a ups like a be350 with a usb cord to shutdown? or maybe a cyberpower pure-sine-wave?(splurge?)

you know the way they kill people and animals is to do 3 or 4 surges to burn off the carcass off a power line (branches too) - this rapid succession on/off easily will kill a drive.

so lesson learned : good battery backup cyberpower PURESINE can handle the pickiest of power supplies (high efficiency energy star ones reject PWM/square wave).

 

[-Slacker-]

^I don't have an UPS, but, if I remember correctly, the PSU model I have has poor compatibility with UPS's...

 

[RebateMonger]

^I don't have an UPS, but, if I remember correctly, the PSU model I have has poor compatibility with UPS's...

The new, hgih-efficiency power supplies can have very high startup-currents, which, in theory, can trip the UPS protection circuitry if the UPS isn't rated for enough current. I don't know if this is a "real" problem or not. I don't recall a single person posting here complaining of this problem.

Hard disks use 12V for their motor and 5V for their PCB circuitry. I suppose that an overvoltage in either portion of a power supply could cause overheating. Or, it might just be a bad bearing in the old hard drive.

 

[westom]

^I don't have an UPS, but, if I remember correctly, the PSU model I have has poor compatibility with UPS's...

When one has near zero electrical knowledge - is educated by popular computer user myths - then a UPS get recommended.

What does a power supply do? Makes those AC mains variations completely irrelevant. And many other functions. What happens when AC voltage increases so great that incandescent bulbs burn out almost 3 times faster? That is also ideal perfect voltage to every power supply because of what power supplies must do.

What happens when AC voltage drops to so that incandescent bulbs are at 40% intensity? This voltage can be destructive to refrigerators, air conditioner, and vacuum cleaner. And ideal perfect electricity because of what all power supplies must do.

Apparently you did not do what the electrically trained (many generations) said. You know that was the superior answer because it resulted in numbers. Yes, numbers are the first indication of those who know this stuff verses other who only parrot what they were told to believe.

Sounds like you might have destroyed another drive. Maybe due to the supply. Maybe not. But nobody can say until you measure (with a multimeter) and post those numbers (to 3 digits).

Are output voltages good or bad? Those numbers will be good or bad with or without a UPS. UPS recommendation says another has no idea how a power supply works or even what a UPS does. That UPS recommendation says that poster has zero electrical knowledge. His erroneous claims can be confirmed with numbers from that multimeter.

Failure could be due to a defective drive, due to a failure in the drive interface electronic, or failure via power lines. One can keep swapping parts on wild speculation. Or one can first identify or exonerate the many suspects - without any damage, much faster, and with answers that are only definitive. That starts with numbers and a multimeter.

 

[westom]

The new, hgih-efficiency power supplies can have very high startup-currents, ... I don't know if this is a "real" problem or not. I don't recall a single person posting here complaining of this problem.

Another popular myth. Better electronics contain circuits such as an inrush current limiter. Why? Electronic are even more robust when the voltage (and therefore incoming current) is kept low during startup. Electric motor appliances hate and can be harmed by low voltage on startup. Therefore the electrically naive assume computers are harmed by low voltage. In reality, better computers will draw less power on startup due to circuits such as the inrush current limiter. And due to so many functions that remain disabled until enabled later during the power on process.

UPS is typically made a cheap as possible. Often so cheap that a new battery costs almost as much as an entire new UPS. Therefore low voltage (or other electrical anomalies that are not low voltage) can cause a UPS to switch to battery power. Then users automatically assume the computer has a massive power on surge. If he makes a claim without numbers, then suspect lies based in junk science. Subjective claims and conclusions based only in observation are how junk science conclusions get believed by a majority.

Which power supply might have a larger inrush current? A crappiest and cheapest because it is missing functions long considered essential in all power supplies.

 

[Emulex]

try the cyberpower "PURESINE" ups - it will work - or you can take it back. the new high efficiency power supplies require a pure sine wave output on battery.

most cheapies are modified square wave. real simple.

 

[-Slacker-]

On the 20 VDC scale, measure the purple wire where it connects the power supply to motherboard. Simply push the probe into that nylon connector to touch metal and read the voltage. The numbers should be something around 5 volts - always - even when a computer is powered off but connected to the wall. Report the actual number in 3 digits.

Ok, I finally found myself a multimeter, and now I don't know how to enable a 20 VDC scale (checked on the manufacturer's website and didn't find anything useful)

The multimeter model is this one (fluke 8060 A)

http://www.teknetelectronics.com/Search.asp?p_ID=85&pDo=DETAIL

Anybody know what buttons on that thing I need to mash to set it on a 20 VDC scale?

 

[westom]

Anybody know what buttons on that thing I need to mash to set it on a 20 VDC scale?

Pictures are too poor to answer with certainty. But push the V button (A is amps and Omega for ohms). Push the 20 button above (or related side mount switch). If working, then display may show a small 20 and a VDC. Touch probes to a nine volt (or other) battery terminals. It will report battery voltage (which then says how good the battery is). Now you know the meter is measuring DC voltage.

Those side switches may be how you select the scale. For example, what looks like a 20 volt pushbutton may only be referencing the side mount switch that must be pressed in to select that 20 scale.

Make sure the meter is measuring VDC – not VAC.