Quicker way to power on/off external HD?

[DreadBelch]

Is there a quick, easy way to power on/off an external USB HD without either disconnecting the USB cord or turning off the computer?

I have the 4TB Seagate Backup Plus and the fan is really loud. Maybe there's some third party software that can quickly turn off the HD? Reaching down and disconnecting it every time is a PITA.

 

[Coup27]

I don't believe so.

 

[corkyg]

Yes. I have my external HDD powered through a switch box. I can turn it on and off simply by pressing a rocker switch on the box. Also works with eSATA.

 

[Charlie98]

Yes. I have my external HDD powered through a switch box. I can turn it on and off simply by pressing a rocker switch on the box. Also works with eSATA.

Corky, question for you: Is that doing any harm to the drive? I have a problem with one of my Seagates, it won't dismount when I try to disable it using the 'Safely Remove Hardware' utility, saying it's too busy right now even when I know it's not doing anything. Rather than pulling the USB cable and risking damaging the disk or something, I have to shut the computer down, pull the plug, and then restart.

 

[Elixer]

Corky, question for you: Is that doing any harm to the drive? I have a problem with one of my Seagates, it won't dismount when I try to disable it using the 'Safely Remove Hardware' utility, saying it's too busy right now even when I know it's not doing anything. Rather than pulling the USB cable and risking damaging the disk or something, I have to shut the computer down, pull the plug, and then restart.

You can force it to unmount via chkdsk /x e: (or whatever drive letter)
However, that will could take a long time depending on the files & sizes.
You can try this http://www.emptyloop.com/unlocker/ or you can kill explorer (your GUI shell!) then restart it back up... or...

Usually, it is windows indexing something on the drive in question.
I don't recommend you just yank the USB cable, though, sometimes, that isn't a issue, and I don't like to turn the drive off either.

 

[corkyg]

Corky, question for you: Is that doing any harm to the drive? I have a problem with one of my Seagates, it won't dismount when I try to disable it using the 'Safely Remove Hardware' utility, saying it's too busy right now even when I know it's not doing anything. Rather than pulling the USB cable and risking damaging the disk or something, I have to shut the computer down, pull the plug, and then restart.

No harm that I can see. Been doing it in all systems for years. This transcends XP, Vista, 7, and now 8 & 8.1. The trick is to use access disk policies under disk management, and set it to Optimize for Quick Removal. This disables write caching for that drive. That enables me just to turn off the power for that external drive. The only time I use Safely Remove is when it involves a device not so optimized.

 

[Charlie98]

No harm that I can see. Been doing it in all systems for years. This transcends XP, Vista, 7, and now 8 & 8.1. The trick is to use access disk policies under disk management, and set it to Optimize for Quick Removal. This disables write caching for that drive. That enables me just to turn off the power for that external drive. The only time I use Safely Remove is when it involves a device not so optimized.

Groovy... thanks for the tip, I'll check it. My other Seagate portable dismounts without problems, and the drive in question dismounts easily when it's on my HTPC (I use my portables to transfer ripped movie files from my desktop to HTPC) just not on the desktop. Makes sense. :thumbsup:

 

[PliotronX]

No harm that I can see. Been doing it in all systems for years. This transcends XP, Vista, 7, and now 8 & 8.1. The trick is to use access disk policies under disk management, and set it to Optimize for Quick Removal. This disables write caching for that drive. That enables me just to turn off the power for that external drive. The only time I use Safely Remove is when it involves a device not so optimized.

We have quite a sample size at work and you may want to rethink that strategy. My coworker has a penitent for removing without dismounting and it has corrupted directory structures with HDDs and in worst cases fried thumb drives beyond recovery (usually cheap generic brands). I always make an effort to hit safely remove or eject from explorer. No failures this route yet..

 

[Red Squirrel]

When you force power off a drive, doesn't that cause the head to crash on the platter because it did not get a chance to park properly? I always figured that was the reason hard drives often fail if the PC shut down hard due to not being on a UPS, or due to PSU failure. Question is, what process tells the drive to park the head, is it when you dismount the file system?

 

[BonzaiDuck]

When you force power off a drive, doesn't that cause the head to crash on the platter because it did not get a chance to park properly? I always figured that was the reason hard drives often fail if the PC shut down hard due to not being on a UPS, or due to PSU failure. Question is, what process tells the drive to park the head, is it when you dismount the file system?

Well, that's why I click the "safely remove" icon. I've seen failures on USB HDDs for leaving them running all the time -- or so you might suspect. I'm always careful how I power down an HDD.

I have a household hot-swap system, with the same type of bay in all five computers. What I discover is this: For AHCI-configured systems, you get a "safely remove" icon and menu. On RAID-mode setups, you need a utility like Hot!Swap. The thing I don't like about Hot!Swap is that it doesn't seem to have a path for de-installation. Once it's been installed on your system, I see no way of getting rid of it.

 

[C101]

When you force power off a drive, doesn't that cause the head to crash on the platter because it did not get a chance to park properly? I always figured that was the reason hard drives often fail if the PC shut down hard due to not being on a UPS, or due to PSU failure. Question is, what process tells the drive to park the head, is it when you dismount the file system?

When the hard drive detects a power loss on either the 5V line or the 12V line (5V only for laptop drives), the drive immediately performs a retract. How the retract procedure is handled will depend on the manufacturer of the drive, but for most drives, a simple disconnection should leave enough time for the drive to finish writing data to the current sector, before performing the retract. The retract function then quickly gets the heads off the disk before the platters spin down enough to cause the heads to make contact with the disk.

The normal parking(retract) process will be either firmware or hardware initiated. Hardware initiated will likely be due to a fault occurring, while the firmware process can use either ata commands from the OS to tell the drive to go into idle and unload the heads, or the drive itself can contain firmware that will unload the heads during periods of no read or write activity.

 

[Charlie98]

When you force power off a drive, doesn't that cause the head to crash on the platter because it did not get a chance to park properly? I always figured that was the reason hard drives often fail if the PC shut down hard due to not being on a UPS, or due to PSU failure. Question is, what process tells the drive to park the head, is it when you dismount the file system?

That's my main concern and why I go through the nonsense to remove my one portable drive. I used to just yank the USB cable out... but then I got 8 bad sectors. Whether that was related to that or not, I don't do it anymore (and, oddly enough, I haven't gotten any more bad sectors, and it's been over 2 years since.)

 

[corkyg]

When you force power off a drive, doesn't that cause the head to crash on the platter because it did not get a chance to park properly? I always figured that was the reason hard drives often fail if the PC shut down hard due to not being on a UPS, or due to PSU failure. Question is, what process tells the drive to park the head, is it when you dismount the file system?

No. Whwn I cut the power after a use, the heads park properly. See C101's reply above. As for what others in an office do is not my concern. I am the sole user of all my systems. Power Off switch works for me.

 

[Red Squirrel]

When the hard drive detects a power loss on either the 5V line or the 12V line (5V only for laptop drives), the drive immediately performs a retract. How the retract procedure is handled will depend on the manufacturer of the drive, but for most drives, a simple disconnection should leave enough time for the drive to finish writing data to the current sector, before performing the retract. The retract function then quickly gets the heads off the disk before the platters spin down enough to cause the heads to make contact with the disk.

The normal parking(retract) process will be either firmware or hardware initiated. Hardware initiated will likely be due to a fault occurring, while the firmware process can use either ata commands from the OS to tell the drive to go into idle and unload the heads, or the drive itself can contain firmware that will unload the heads during periods of no read or write activity.

Interesting, so I guess there is some capacitors that give it enough juice to do this?

 

[C101]

Interesting, so I guess there is some capacitors that give it enough juice to do this?

Capacitors give the drive the initial supply for the first few milliseconds after the power is removed, but the main source of power to allow the drive to conduct it's retract and shutdown procedures is generated from the still spinning platters. When the drive begins it's emergency power off procedures, it uses the spinning array of platters (now a sort of flywheel) as a generator.

Depending on the drive (number of platters, mass of platters, rpm), the drive could be internally powered for a number of seconds. The retract procedure is still initiated immediately, but the processor could execute code for a short amount of time after power-loss.

In enterprise (and some consumer) SSDs, you often see large banks of capacitors, as the SSD doesn't have a flywheel to keep it alive while it preserves any data that may be in it's cache when power is lost. A good example of this is this Kingston enterprise SSD.

In this SSD, all the black components with the white stripe and the (+) sign are large capacitors. You can even see that the drive was initially designed with more capacitance than the production design, shown by all the empty solder pads to the right of the row of capacitors.

 

[Red Squirrel]

Wow, that's really neat, I never even considered the platters being used as a fly wheel. They do have a decent amount of mass so they can definitely generate enough power to keep the circuit board going for well longer than needed to park the head.