Commuting with a laptop that is powered on

Question

How safe is it to leave my laptop running a build or other time-consuming task when I am traveling? I mostly travel on a two-wheeler, with my laptop in a backpack, strapped vertically to my back (left edge towards Earth, right edge towards sky or vice versa).

I have heard that damage can occur if the hard disk's spinning platter comes in contact with the drive head, but isn't that only a problem with much older hard disks? My laptop was purchased in 2010.

I found two related questions, but I am particular about whether the laptop is safe when positioned vertically.

Solution 1

The orientation of hard drive was historically considered important.

Seeing that some versions of iPod's have the drive turned around all the time, one would think drive orientation change wouldn't matter.

However from personal experience with desktop drives, the "field" knowledge from multiple PC repair techs I have known over the years, as well as that expressed by other comments in various forums, I've personally taken a cautious view that changing orientation during use could be a real problem.

Specifically the common theme I've heard and experienced is that if you originally formatted the drive when it was horizontal, then using it vertically may make it more prone to read/write errors. Similarly if you originally formatted it while vertical, then it would be problematic when used horizontally. I have used that tidbit of info as an additional precaution while recovering an old drive that started clicking. However in laptop drives it may not be applicable since manufacturers expect the drive to be moved around a lot more than desktop drives. YMMV.

Your laptop drive was likely originally formatted when horizontal, so positioning vertically, especially while running builds because of large amount of reads and writes could be a problem, if the above holds for laptop drives.

Hard drive reader heads aren't exactly an equal length lever on both sides of the fulcrum (pivot). One side of the head is long and thin while the other side is short and heavy, crudely shown below:

[][]=(o)------------e

See a more accurate view of the geometry of the drives, here. This might be interesting as well, knowing that the core drive mechanical structure hasn't really changed in multiple decades.

The design is engineered to have the same weight on both sides of the lever to minimize mechanical energy expended in starting and stopping the motion. Two things happen here..

1. When rotating the head on another axis besides it's own pivot axis, the head is subject to gyro forces and as a result subject to mechanical stress during motion.

2. When pivot is vertical (i.e. head motion plane is horizontal) then the pivot has no stress moving it sideways. However when the pivot is horizontal (i.e. the head motion plane is vertical or the drive is vertical) then due to the weight of the head mechanism, the pivot will (due to its weight) put pressure on the bearings it is positioned on due to gravitational pull. No matter how precise the engineering, there will be some measurable effect on the bearings that will only come into play in vertical drives. This I guess, will be enough to cause the head to read the disk surface slightly offset.

Item #1 (gyro effect) could be a significant issue on multiple axes especially when the laptop is in a backpack and moving/bouncing up and down and rocking back and forth on the owner's back as the owner is walking. Though human motion is way slower compared to drive platters or readers' motion, so it may not be as significant as might be expected.

Item #2 could be an issue as well albeit less significant and much more automatically corrected with normal calibration that the modern drives already do.

In either case don't move your spinning drive like a pompom :-)

Regarding heat build up: Others have mentioned the heat accumulation inside the bag. I want to reiterate from personal experience that this can be a HUGE issue, and I'd really advise that you avoid putting a working laptop inside a bag for any longer than 1-2 minutes. Be careful with the sleep button, too. Sleep can fail if an error dialog pops up and you don't realize it. If that happens, your laptop can overheat, CPU protection may kick in, and you could lose your work, or worse.

If you really have to keep your laptop in the bag, see if you're able to get a vertical airpath with at least one air entry in the bag near the bottom and at least one air exit near the top. And make sure your laptop has vents that align with the bag's air entry points, only then will it be safe to leave your laptop in there for any longer than 5 minutes.

Reference:

https://superuser.com/a/50422/212609

Does vertical position affect the lifespan or integrity of a hard drive?

https://serverfault.com/a/14245

http://hardforum.com/showthread.php?t=1512482

http://hardforum.com/showpost.php?s=8249ef649e3616f32823f34b635aeeac&p=1035610869&postcount=2

http://www.tomshardware.com/forum/241026-32-vertical-horizontal-mounted-harddrive#5980965

http://en.wikipedia.org/wiki/Wikipedia:Reference_desk/Archives/Computing/2012_October_3

http://answers.yahoo.com/question/index?qid=20110321205213AAhwSDV

http://www.extremetech.com/extreme/142911-ibm-3390-the-worlds-largest-and-most-expensive-hard-drive-teardown

http://en.wikipedia.org/wiki/File:Harddrive-engineerguy.ogv

Solution 2

In general that's not a good idea.

Hard drives with active protection systems will park the head on sudden shock (and these are more common than they used to be) and SSDs are nearly insensitive to shock. There seem to be more than one implementation of this, so you'll need to check with your hard drive model to be sure.

That said, your system will likely cook in a closed bag - especially when building files. Its also probably not healthy for your cooling system, and your overheating system may shut down midway through.

Solution 3

I think that the spinning hdd will take damage with high-G movements, as the reading needle can accidently scratch the disk plates (the strong magnets prevent this from low-G movements).

Apart from this, the battery time and ventilation should be your main concern. When I do this myself, I only do it for shorter travels (max. 1 hour) and place my laptop in the bag with the ventilation opening upwards and the backpack slightly open to enable it to cool just a little.

If it gets to warm, my laptop goes to sleep, so no real damage happens though...

Solution 4

With all the precautions in the world, eventually something will distract you or displace your machine while you are on the move. Speed bump. Phone call. Shady looking passerby. Suddenly, you notice that it's time to get on or off a bus or train. It doesn't take much.

Accordingly, it doesn't take much to fry a laptop's processor. If you are already doing something processor intensive and shove it into a backpack, weird things can happen.

You can start a shutdown or hibernate, stuff it into your bag, and then happen to pull the thing back out 30 minutes later later and find that a modal dialog interrupted your shutdown. Your bag will be warm, and your rig will be red-hot. They're not designed for it.

In summary, if you ever stow your laptop, it's best that you know for sure that it's fully OFF, or fully hibernated in the least.

Solution 5

Another aspect is the battery connection. I have seen interruptions caused by impacts to the laptop. In that case your SW will very likely crash, unless the duration of the power failure is extremely short.

View more solutions

Comments

[174140]

With a general purpose laptop is definitely not a good idea.

[Ivan Viktorovic]

If you want to do that i would be sure that there is sufficent air to avoid overheating and replace the disk with a SSD. SSD is a flash based memory with no spinning parts just like a external USB stick.

[kobaltz]

Not to mention that you're probably not going to be getting good ventilation.

[deed02392]

I had a Dell E6410 which I kid you not loved turning itself on from OFF in my bag. It came out scalding hot. I no longer pack and carry it around as I did back then so I don't know if the problem persists but years on it's fine. I'd suggest laptops are less susceptible to overheating than we think, even if it feels concerning to find it hot to the touch.

[Krease]

Leaving it running a task while travelling? I wouldn't even consider this unless you're using an SSD hard disk. Regular hard disks are very susceptible to shock when they're not running; even more so while they're spinning and reading/writing data.

[oliver]

I would suggest switching to an SSD if you want to move the laptop while it's powered on.

[prakash]

@deed02392: I've got a four year old HP laptop with a dead video card that attests to the risks of overheating.

[James Snell]

A lot of the shock protection systems are also intended to park during 'freefall' events (so when being dropped) and would not protect the system at all from the kind of shock you'd get in a pannier negotiating a typical road/cycle route.

[Martin Beckett]

I lost one laptop to overheating by leaving it to re-encode a videofile while in a laptop bag. The passively cooled GPU literally burned.

[tvdo]

@JamesSnell Some do, though, which is why you need to check. HP's one (if enabled) will park heads with the slightest movement when the lid is closed. But continuous start/stops isn't good for the drive either.

[Ian Howson]

Historically, it used to matter. Modern drives (less than 20 years old) use a closed-loop control system to ensure that the heads are positioned accurately. There's no 'tolerances' or 'movement' or 'gravity' at play; the control system ensures that the heads are in the right place regardless of orientation or motion.

[BeowulfNode42]

@IanHowson Newer drives also have significantly smaller targets to find to place the data in or read from. With any mechanical system there is no such thing as perfect fit, no slack, or no play in the mechanical parts. The systems used to monitor where the drive heads are currently located are also subject to some inaccuracies. HDD manufacturers use advances in positioning accuracy to allow for the data to occupy a physically smaller location resulting in much the same robustness as older drives but with increased speed and capacity. I use tolerances to describe the complete system accuracy.

[Thalys]

Compilation + insulation may end in deflagation? ;p

[Rook]

Just so there is no confusion, "hibernated" is fully off.

[MSalters]

@BeowulfNode42: You're missing the point. The small sizes are exactly why modern drives use the closed-loop system. It's now simply impossible to hit the right sector by guessing. Modern systems guess, check where they ended up, estimate how far off they are, and guess closer. That's why it's called a closed loop. Reorientation doesn't matter: the closed loop system design just means that the first estimation differs, which pretty much fixes every first-order effect. IOW, it's hard to exactly hit track #103925 from zero, but it's a lot easier when you're already on #103924.

[jpaugh]

The orientation thing in this post is mostly superstitious. Another answer in the linked article (as well as the above comments) confirms that it doesn't matter.

[LMSingh]

@jpaugh none of your references confirm nothing and I would never take a chance with my data based on your logic. Your comment is highly irresponsible in telling people not to be cautious with valuable data on their drives. All of your references are equally partially constructed opinions based on incomplete logic. Provide a definitive statement by a known expert on hard disk design to supersede my cautionary statement based on years of experience. I gave both sides of the argument, you gave one sided callous statement. FYI, I have repaired more drives than most people use in their lifetime.