Horizontal Vs. Vertical mounting

Thetech

Senior member
Mar 12, 2005
571
0
0

I have an Antec P180 case, and I was thinking about moving the hard drives to the bottom of the case where they would be mounted vertically instead of horizontally.

Would having a drive mounted vertically increase the risk of failure?
I'm concerned about if the case gets bumped with the drives mounted vertically.

 

Denithor

Diamond Member
Apr 11, 2004
6,298
23
81
Dust? Inside a harddrive? I thought these devices were sealed & produced in rigorous clean-room environment?

Also, Blain - a bit disappointed, to tell the truth. I saw your name responding and honestly expected some kind of vaguely inappropriate joke about how he mounts his HDD. And with the drive orientation remark above - wow, just didn't take advantage, did you? Feeling ill?

If it's a new drive you're probably ok to mount it however you like. If you're worried about it being damaged due to a bump - if it takes a hit that will damage it, the damage will occur no matter how it's oriented in the case. I've only lost one drive to a bump in nearly 20 years of working on these things. That one was a freaking oops - bare drive slipped out of my fingers during file copy and fell 1 inch to desk surface. Wouldn't read or access at all immediately. Moral of the story: don't handle your drives while they're working.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
The drive would definitely work either way... the question is, is it more RELIABLE one way or another...in theory drives should work either way... but from CONJECTURE I observed higher failure in vertical... the question is... do I actually base any decision on it or dismiss it as (focused) random chance... I haven't made up my mind yet.
 

ShawnD1

Lifer
May 24, 2003
15,987
2
81
Originally posted by: taltamir
The drive would definitely work either way... the question is, is it more RELIABLE one way or another...in theory drives should work either way... but from CONJECTURE I observed higher failure in vertical... the question is... do I actually base any decision on it or dismiss it as (focused) random chance... I haven't made up my mind yet.

You're probably right. I remember reading an article about the Xbox 360's DVD drives scratching the discs and it turned out to be a lot more common when the DVD drive was vertical. There's just something about vertical spinning things that sucks.

It might have something to do with how the parts of the drive are moving. A DVD drive will have a laser while a hard drive has a magnet, and these parts move back and forth along the surface of the disc. If the disc is vertical, the laser or magnet is constantly moving with or against gravity. It might sound like something that doesn't make a difference, but we're talking about a little arm that moves fairly quick and it's doing this nonstop while the computer is running. If the drive is destined to fail, it seems reasonable to assume that it will fail much faster when it's fighting against gravity.
 

Blain

Lifer
Oct 9, 1999
23,643
3
81
Originally posted by: ShawnD1
If the drive is destined to fail, it seems reasonable to assume that it will fail much faster when it's fighting against gravity.
So you're saying the most reliable mounting orientation is to have the cable connectors at the top with the platters below, so that the heads are more free to swing back and forth?

 

ShawnD1

Lifer
May 24, 2003
15,987
2
81
Originally posted by: Blain
Originally posted by: ShawnD1
If the drive is destined to fail, it seems reasonable to assume that it will fail much faster when it's fighting against gravity.
So you're saying the most reliable mounting orientation is to have the cable connectors at the top with the platters below, so that the heads are more free to swing back and forth?
I think I've had too much to drink since I can't quite picture what you're describing. If my hypothesis is correct, the drive would be under the least stress if the drive is horizontal.

Another thing I would like to know is if having the drive upside down is bad for it. All of the SATA drives in my HTPC are upside down because it's the only way to make the SATA power connectors line up correctly. If I put the drives the normal way (electronics facing down), the SATA power cable would need to be longer. I should take a picture of this to show what I mean.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
Originally posted by: Blain
Originally posted by: ShawnD1
If the drive is destined to fail, it seems reasonable to assume that it will fail much faster when it's fighting against gravity.
So you're saying the most reliable mounting orientation is to have the cable connectors at the top with the platters below, so that the heads are more free to swing back and forth?

what he is saying makes perfect sense, don't ridicule him as a response.
 

Emulex

Diamond Member
Jan 28, 2001
9,759
1
71
i dunno i have 2 brand new servers that mount horizontal and 2 brand new servers that mount vertical. they are all hp and use the same drives (15K SAS). i'd have to go with the fact that it doesn't really matter but what fits best in the rack is what they are after.

fwiw most 2.5" are vertical in 2U for density to hold 16 drives. 3.5" don't quite fit in 2U vertically; have to move up to 3-5U.

i would guess that a big company like hp would maybe sorta kinda have put their cash into R&D to understand the dynamics of vibration;noise;cooling;orientation as a package.

you ever see that iometer test on youtube where the dude screams into the raid and the i/o slows down? crazy.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
back in the day when drive did their own low level format, you had to reformat a drive when changing it from horizontal to vertical.

Ha, I was about to reference that screaming dude and i noticed you did..
 

Blain

Lifer
Oct 9, 1999
23,643
3
81
Originally posted by: taltamir
Originally posted by: Blain
Originally posted by: ShawnD1
If the drive is destined to fail, it seems reasonable to assume that it will fail much faster when it's fighting against gravity.
So you're saying the most reliable mounting orientation is to have the cable connectors at the top with the platters below, so that the heads are more free to swing back and forth?

what he is saying makes perfect sense, don't ridicule him as a response.
LoL... You can't know my intent, so I'll let that slide.

I've thought about this issue before also. I was asking him if he theorized that drives would be best mounted with the cable connectors up Like This.

 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
btw, in addition to the arm going against gravity (I am more concerned about the part where gravity "helps"), there is also the issue of vibration on a platter spinning horizontally instead of vertically.
You wanna see something catastrophical? put your PC in horizontal position, put a CD in... start READING something off of the CD... and lift the the PC to vertical position WHILE the disk is reading... this is NOT gonna be pretty...
 

Denithor

Diamond Member
Apr 11, 2004
6,298
23
81
Originally posted by: ShawnD1
There's just something about vertical spinning things that sucks.

Really? So you don't like the flywheel and tires on your car, Ferris Wheels, fans (including most likely the CPU fan cooling the computer you are on right now) and numerous other items that spin in a vertical orientation?

I have worked on numerous systems over the years. A fairly high percentage of the OEM systems have the HDD mounted vertically on the front grate of the box (especially prevalent in the old P3 Pavilions). Somehow I doubt Dell/HP/etc would design a mounting system that would result in higher HDD failure.

The blade-style servers have an HDD installed in the same direction as the case - which is then mounted vertically in the rack.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
I have worked on numerous systems over the years. A fairly high percentage of the OEM systems have the HDD mounted vertically on the front grate of the box (especially prevalent in the old P3 Pavilions). Somehow I doubt Dell/HP/etc would design a mounting system that would result in higher HDD failure.
You have a lot of faith in them...
 

Denithor

Diamond Member
Apr 11, 2004
6,298
23
81
Not really, I just doubt they would accept a design that yielded significantly higher HDD failure rates (and therefore likely cost them more money in warranty service work). A poor design generally doesn't last long.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
I was once told that horizontal was best for drives, but I haven't seen data to back that up.


Originally posted by: taltamirYou wanna see something catastrophical? put your PC in horizontal position, put a CD in... start READING something off of the CD... and lift the the PC to vertical position WHILE the disk is reading... this is NOT gonna be pretty...

AFAIK, many cd drives have the motor mounted independantly of the tray, for alignment purposes I think. When you tilt the drive, the motor and disk tilt relative to the tray due to inertia, since it isn't locked in place. I don't believe hard disks have this "issue", though the heads have been known to scratch the surface of the disk when bumped.
 

bryanl

Golden Member
Oct 15, 2006
1,157
8
81
Originally posted by: Blain
If you change a drives mounting orientation, you should backup your data and probably reformat.

Scott Moulton (of My Hard Drive Died), says that drives develop a sort of memory (my word). And that dust particles can be knocked loose when changing orientation or moving the drive, resulting in data loss or drive failure.

> Scott addresses the issue almost 47 minutes into the interview

Mr. Moulton's advice is important for 20 megabyte hard disks with stepper motors instead of servo mechanisms.

Many, many years ago, the owner of a disk drive rebuilding business scoffed at some of the information printed in his Upgrading and Repairing PCs.

Vertical or horizontal doesn't matter, but vertical provides a small but significant improvement in natural convection cooling. Drives create dust when the heads skid into the platters, which they do when the drive is bumped or when the heads seek and suddenly lose lift. This is why drives contain an air filter, as this photo of a Toshiba 2.5" drive shows, in the upper left corner (white object). This is not a breather filter. It also contains dissicant to remove moisture.

http://img.directindustry.com/...k-drive-hdd-320775.jpg



 

Elixer

Lifer
May 7, 2002
10,371
762
126
Originally posted by: taltamir
btw, in addition to the arm going against gravity (I am more concerned about the part where gravity "helps"), there is also the issue of vibration on a platter spinning horizontally instead of vertically.
You wanna see something catastrophical? put your PC in horizontal position, put a CD in... start READING something off of the CD... and lift the the PC to vertical position WHILE the disk is reading... this is NOT gonna be pretty...

Now we know what you do in your free time! ;)

For what it is worth, I have seen HD racks that have the HDs in both orientations, and I can't say that horizontal rack has had more issues than the vertical rack.

Having said that, there is a big sticker on one of the DAT drive that says "Only use in horizontal orientation", and when I asked why it had that, they said it had to do with dust.
 

Emulex

Diamond Member
Jan 28, 2001
9,759
1
71
DAT tapes? you ever seen what they do? they need gravity in order to lift the tape out and around the drum.

you ever seen a see-through vcr and how it works?