How Hot Should a Hard-Drive Be?

Trying to debug this PC that ups and shuts-down...

Thought it was CPU fan... nope.

Only thing I can see is that HD temperature gradually rises to ~110°F.

Is that normal or does it portend failing bearings? ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson
Loading thread data ...

Still might be the heat sink. Sometime the plastic clamps deform.

cooler is always better. 110F isn't very hot. How are you measuring temperature? Thermometer or readout from some software? If you put an extra fan on the HD, does it still shut down?

Reply to
mike

=A0 =A0...Jim Thompson

Could be bad memory. Did you run a memory scan? Try this one.

formatting link

I had bad memory, then bought some replacement memory from crucial.com.

You might also need some more thermal paste between the CPU and heatsink. Arctic Silver is pretty good.

Rule out a software problem (virus?) by downloading a Linux Live CD (the stuff on ubuntu.com isn't too bad, although I dislike everything after version 10.04).

Have fun!

Michael

Reply to
Michael

Kind of a vague description of symptoms :< Does it "shut down" in an orderly fashion (i.e., because the system management features of the BIOS detect a "condition" that should signal a shutdown)? Or, does it unceremoniously *crash* while you are doing something?

Does it completely SHUT DOWN (i.e., power EVERYTHING off)? Or, does it just "lock up"?

Does it do this always when you are doing something specific?

With a "cold" machine, does the failure tend to occur after the same general time interval (for a given workload)?

Dust/loose heatsink. Bad caps (esp around the CPU... there are tens of amps shuttling around modern CPU's)

That's not warm (unless its a really crappy drive). Many disks (not true of the newer Green disks) are "pretty uncomfortable" to leave a hand on (even palm side!) for more than a short while.

Failing disks often (though not invariably!) show errors before any sort of drastic failure. Unfortunately, things like SMART don't often catch these errors. Or, report errors that do not predict failures. Depending on the OS you are running, there may be logs (if enabled!) that will give you a clue as to retries, etc. reported by the drivers -- from which you can glean some performance metrics.

Look elsewhere for your problem. Pull the disk and see how the machine behaves running off a live CD (if you want to be pedantic, put a comparable load on the drive's power connector -- perhaps a *dummy* disk -- just so you can stress the power supply comparably)

Reply to
Don Y

IF you were to boot say.. a Knoppix Live DVD, you could then run the disk utility or gparted and see what the SMART report is for temperature profile.

Likely HD failure up-n-coming, so I would back up now and buy soon as well.

Reply to
SoothSayer

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

110F sounds ok. My motherboard sensors show 100-133f. Harddisks run around 38c.

I'd look else where, maybe memory or a corrupt file. My last problem was similar, turned out to be the power button. Gold flakes (or a weak spring) in the switch would hold the button ON and the PC would shutoff during and after it was booted up. Tossed a good power supply because of it ;(

Cheers

Reply to
Martin Riddle

Any idiot can "list" every possible item for it to be.

You have provided nothing. Your skill hovers firmly at *none*.

Reply to
SoothSayer

Must be all those "Fake E-mails" you have gotten lately..

What was that now, "Only A FAKE" gets "Fake E-mails" ?

Maybe you should use your wits while you still have them.

My PC CPU operates at 68F normally, 110F for the CPU isn't hot.

Why don't you investigate the common problems with PC's like bad caps in the power supply or MB or get that dusty DMM and scope out and actually use some equipment to perform a voltage and noise test on your supply in the PC. Of course you could be having issues with one of your plugged in devices inside or out.

Yes, I know I am the last one you want to hear from, but guess what? I most likely am correct. Ofcourse, you'll never admit to that as most cranky old farts don't.

Jamie

Reply to
Jamie

43°C (I can't think in °F for electronics) is OK for a hard drive if it's the HD case or electronics temp. DRAM issues are the more likely issue. DRAM problems usually result in a BSOD or a locked up machine. Memory crashes will also cause reboot or power off, depending on how you have the bios set up. Memory is cheap.
Reply to
qrk

You know me. I keep stuff forever. This box is ~8 years old.

Maybe it's time to swap it out ?:-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

No, there is where you're wrong. I provided answers, he obviously didn't have. It does not mean they are the correct ones but they are ones that will cause that exact problem.

What I posted was from actual experience, how about you? Was your post just now, from experience? DO you like foaming at the mouth before you really know what is going on? Or are you one those that seizes the opportunity to open mouth and insert foot? Only to run the other way because you have no defense in your favor.

You may want to simply sit aside and leave it alone.

"The young are slow blind"

Jamie

Reply to
Jamie

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

The Core i5 & i7 stuff is fast. And its not that expensive. An i5 laptop is about $600 to start. Systems a bit more. Well worth the upgrade.

Cheers

Reply to
Martin Riddle

Hi Mark,

This is an 8 year-old box. What do you recommend as a replacement? (General use... my "COMMS" machine.) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

Doesn't sound extreme. Isn't this one of those boxes from the PC guy from the street corner shop? Maybe the electrolytics on the motherboard are finished.

And you call _me_ a cheapskate? SCNR ...

Probably. How about getting one from a real pro manufacturer this time? :-)

--
Regards, Joerg

http://www.analogconsultants.com/
Reply to
Joerg

Maybe a better question: "How hard should a hot drive be?"

Reply to
John S

Hard enough to keep her from getting bored with you within 5 seconds flat.

Reply to
MassiveProng

Bearing lube failure usually results in nonstarting; after the lubricants a= re warm, friction and losses go down. Bearing excess play usually results in excess retries (repetititive disk sounds are the key to diagnosing this). Neither causes computer shutdown.

Most likely, one or more power supplies is going out-of-bounds and shutting off at the power supply, just like blowing a fuse. Dying output = filter capacitors (very probable on an eight-year-old machine, that was in the middle of the low-ESD-capacitor-lifetime fiasco) on the logic board, or= in the power supply, is my bet. Look for bulges or brown stains.

Reply to
whit3rd

Almost anything you can buy now will be better than that! Even 2-3 y.o. second user ex-corporate gear for $100 or so!!!

If you are determined to get to the bottom of it download the latest memcheck and run it from a bootable CD/USB stick. My money is on one or more of the memory or CPU PSU capacitors giving up the ghost - you may be able to spot them by visual inspection. Typically they take on a rakish angle and the end caps swell as they get to the end of their tether, although they still measure OK on the bench they no longer work properly at high pulse currents and so the box goes crazy.

formatting link

May need a few hours to find a fault and keep an eye on the temperature as this is something of a stress test. The Mickeysoft memory test is pathetic by comparison and will pass units that are doomed to fail.

--
Regards,
Martin Brown
Reply to
Martin Brown

8 years! Take a look at the motherboard and power supply. Bet you'll find some bulging caps.
Reply to
JW

Mine is 8 yrs old as well, but I have changed out the PS and the vid card. I'd bet the Asus Mobo is fine for another half decade. The speed of the machine, however, is not... any more...

It will make a fine print server, and NAS and maybe HT video streamer/player.

My ION minis are almost faster. My i5 mini form factor machine is faster as well. And my ipad is too.

Maybe if I spend $4k on my next machine, it will be my last.

I already have several components for it. The PS. The HD (one of them) Still need a vid card and a mobo and ram and a cpu. Likely a dual Xeon if I can put it together.

He could slowly piece together the parts and not feel the hurt as much too. Then the final build is always a fun exercise too.

Reply to
SoothSayer

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.