Sign in to follow this  
Caracal

Question about 'wear and tear'

Recommended Posts

I've recently bought some external USB2 enclosures for my drives, with the intention of just plugging them in, as and when i need the data that's on them.

However i've just realised that one of the drives has data on, which i need constant access too (Outlook PSTs and various other bits), so i was just wondering... assuming the cooling is adequate, would keeping the drive switched on, and accessing the data on it, via USB2, put any more 'strain' on it, compared to if it was connected normally, internally, via IDE?

1 quick other thing as well! - According to my enclosure manuals (Icybox) the 'correct' way to attach the drives, is to power them on THEN plug the USB lead into the computer...... i was wondering, is there any difference (or could i cause any damage) if i simply left the USB leads connected, and simply powered them up to gain access to them? Sorry if that sounds dumb, im just REALLY paranoid about losing GBs of data :D

Thanks

Share this post


Link to post
Share on other sites

I've been using two extenal USB2 drives with a Linksys NSLU2 Network Storage Link for over two years now. The two drives, one a Samsung, the other a Seagate, have been on contunuously in cheap enclosures with no fan for the full two years. They both get mildly warm to the touch and I have had no problems.

I have had no problems with other USB2 drives plugging them in when already running, or plugging them in and then turning them on. Just be sure to unmount the disk before disconnecting when you are done.

Share this post


Link to post
Share on other sites

Probably the answer is "it depends" on the drive and controller.

Most USB enclosures never spin down their drives, so the drives may wear their bearings faster, and spinning all the time potentially increases exposure to damaging impacts.

Most USB enclosures are also unable to send a shutdown command to the drive. Some drives are rated for a certain number of stop/start cycles if sent proper commands to spin down or shut off, and a completely different number of stop cycles if the power is just yanked. If this second number is considerably less for your drive, then repeatedly shutting it off via the emergency shutdown mechanism may wear the disk out faster.

Of course if you have disabled power management for the internal disks and never turn the computer off, then there should be no difference.

Share this post


Link to post
Share on other sites

From what I gather, bearing wear doesn't occur with FDBs.

Is this right? So leaving the drives on wouldn't wear out the bearing, assuming it's a modern drive.

Share this post


Link to post
Share on other sites

I think that's kind of like assuming the differential in your car never wears because it has a design life of over a million miles (by which time something serious in the rest of the car would have failed anyway). The wear may indeed be virtually negligible, but usually the failure of a related component like a seal causes catastrophic failure long before the design life is achieved. That's why there is a market for used differentials in junkyards, though of course FDB are non-serviceable (and any wobble may have caused irreparable damage to the platters by then anyway). Certainly we've seen autopsy pictures of seized FDB here before, but it's probably pretty rare.

Another question is if heat-cycling the drive will cause more "wear" to solid state components (as in cold solder joints) than continuous operation wears seals or outgasses lubricants on the spindle and voice coil motors or platters themselves (moving parts fail, wear-leveling algorithms do exactly that).

While any modern drive has a good chance of becoming obsolete before dying, it is foolish to think of any drive, internal or external, as reliable enough to need no backup of "GBs of data."

Share this post


Link to post
Share on other sites

Thanks for the replies - that puts my mind at ease :D

Ohh, will Windows detect an external drive just as easily if its currently plugged in THEN switched on, as opposed to having the drive switched on THEN plugged in?

Share this post


Link to post
Share on other sites
I think that's kind of like assuming the differential in your car never wears because it has a design life of over a million miles (by which time something serious in the rest of the car would have failed anyway). The wear may indeed be virtually negligible, but usually the failure of a related component like a seal causes catastrophic failure long before the design life is achieved. That's why there is a market for used differentials in junkyards, though of course FDB are non-serviceable (and any wobble may have caused irreparable damage to the platters by then anyway). Certainly we've seen autopsy pictures of seized FDB here before, but it's probably pretty rare.

Another question is if heat-cycling the drive will cause more "wear" to solid state components (as in cold solder joints) than continuous operation wears seals or outgasses lubricants on the spindle and voice coil motors or platters themselves (moving parts fail, wear-leveling algorithms do exactly that).

While any modern drive has a good chance of becoming obsolete before dying, it is foolish to think of any drive, internal or external, as reliable enough to need no backup of "GBs of data."

OK I see. Of course, I was taking "wear" literally, in the sense of a ball bearing wearing out. With use, any moving part will tend to increase its chances of failing somehow. I wouldn't dispute that.

I also agree that backups are a necessity and didn't want to imply otherwise.

Share this post


Link to post
Share on other sites
Most USB enclosures never spin down their drives, so the drives may wear their bearings faster, and spinning all the time potentially increases exposure to damaging impacts.

If the drive actuator does not use ramp load, spinning down and spinning up the drive will wear out the heads faster than spinning all the time will wear out the bearings. Reason being the drives without ramp load park the heads in a textured landing area on the media when they spin down. While the drive is spinning the head floats above the media, causing little or no wear.

I think the bigger concern is heat. Many of the external boxes do not have adequate ventilation, and that will shorten drive life for sure.

Share this post


Link to post
Share on other sites

Drives have a minimum of 50,000 spin up/down cycles. To put this into perspective, spinning a drive up and down 10 times a day over 8 years, would bring the number of cycles to only 30,000.

All I'm doing here is pointing this out. ;)

Share this post


Link to post
Share on other sites

As I mentioned, start/stop cycles (the number of normal spindle start/stop cycles, usually about 50,000) is a completely different rating than power-off retract count (the number of emergency retract cycles). The problem is that most drive manufacturers do not actually specify what that rating is, or for that matter the design number of power-on head parks while the disc continues to spin (which may be in the millions). This makes it impossible to know how bad it really is or to make any logical decisions on using power management or not for best reliability ;)

Hitachi rates some of their drives at 20,000 power-off retracts, and have this to say about it:

Emergency unload

When hard disk drive power is interrupted while the heads are still loaded the micro code cannot operate and the normal 5-volt power is unavailable to unload the heads. In this case, normal unload is not possible. The heads are unloaded by routing the back EMF of the spinning motor to the voice coil. The actuator velocity is greater than the normal case and the unload process is inherently less controllable without a normal seek current profile. Emergency unload is intended to be invoked in rare situations. Because this operation is inherently uncontrolled, it is more mechanically stressful than a normal unload.

Sounds like they are using the spindle motor as a generator and shorting the current across the voice coil to force a quick park. With most external enclosures, every power off is this way.

I suggest not worrying about wear and tear but just treat these devices as consumables and expect they may fail at any time. Especially because the OP seems to be wondering how much these things may be trusted, and my answer is not at all, whether internal or external :blink:

Share this post


Link to post
Share on other sites

OK i'm not worried about excess spinning now :P

But back to my previous Q:

"ill Windows detect an external drive just as easily if its currently plugged in THEN switched on, as opposed to having the drive switched on THEN plugged in?

Share this post


Link to post
Share on other sites

"YES!" is usually the correct answer but there's exceptions to every rule. (Sometimes exceptions to a rule have their own exceptions... which in turn have their own exceptions and so on.)

It may depend on your enclosure. With any decent enclosure, it shouldn't matter which way you do it. With some very cheap crap enclosures, there may be some specific orden in which to use.

For example my Mentor enclosure uses external power brick to power the HDD and electronics. USB supplies +5V, and connecting the enclosure to USB without using power brick, power LED on enclosure starts emitting light. Also, if I do this, Windows will report that one of my USB devices is in error state. If I first connect USB and then power within first ~5 seconds or so, it'll detect. If I power the HDD first, then connect USB cable, it'll detect. But leaving USB connected for too long, it'll enter error state.

All other(*) enclosures (non-Mentor) work perfectly fine either way I do it. I usually leave the USB cable plugged unless I need to take the HDD with me. If I just want to stop it spinning, I'll cut the power (after first unmounting it from the OS).

* 2x Mapower "Warps" KC31 USB2

2x OneTouch (1st gen) USB2+FW (I have not tested FW so I can't say how it works)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this