Mounting hard drives vertically

W

Waffles730

Limp Gawd
Joined
May 31, 2005
Messages
136
Hey all, I want to mount a couple of my hard drives vertically. I plan on suspending them to get rid of a bit of the noise. My question is if mounting vertically instead of the traditional horizontal position would harm the drives. I know SCSI disks in servers are frequently mounted vertically and normally that would convince me of it being fine, but I did a quick google and found people mentioning it might matter which side the drive was tipped to.
I would be mouting 2 Raptors and a Maxtor 80 and WD 160 gig IDE drives that way. Any thoughts or links you could point me to?
Thanks!
-Dan
 
It won't hurt them to be mounted vertically, in fact many OEM's mount them this way normally. I've mounted mine normal, vertical, at slants, etc. and never any issue.
 
From my experience vertical and horizontal are just fine. Diagonal is not recommended and is know to cause issues.
 
WHYYYYY back in the day it matter only if the drive was formated in that position ie
if it was formated with it vertical the it would need to stay like that but thats long since passed

NOW i would make sure thay cant move if the case is moved/tiped ect.
 
A HDD can be mounted in any position, as long as it's either vertical or horizontal :)

Suspending them, however, will seriously degrade performance.
 
Elledan said:
Suspending them, however, will seriously degrade performance.
Click to expand...
Do you have proof of said degradation?

Theoretically, your claim is sound, but I want hard data.
 
Elios said:
WHYYYYY back in the day it matter only if the drive was formated in that position ie if it was formated with it vertical the it would need to stay like that but thats long since passed.
Click to expand...
I still have some of those drives in operation. Dang saggy drive heads. :rolleyes:
 
Elledan said:
A HDD can be mounted in any position, as long as it's either vertical or horizontal :)

Suspending them, however, will seriously degrade performance.
Click to expand...


o_O;;

i seriously dont see ANY possible way that suspending drives to reduce noise(in most cases, significantly i might add) would degrade the drives physical head speeds as to degrade performace...if there are some reputable companies IE WD, Maxtor, etc etc that have white papers on it, i'd love to read them, until that fact I'm calling shens...
 
I have two that are vertical-on-side and two that are suspened-on-the-horizontal. Been running them for 4 years that way. No problems.
 
Yeah, Dell mounts their hard drives vertically, if I remember right, and they have failures more from the crappy hard drives they use, than because of the hdd position. :rolleyes:

In fact, the x-qpack case from aspire mounts the hdd's vertically. :)
 
xonik said:
Do you have proof of said degradation?

Theoretically, your claim is sound, but I want hard data.
Click to expand...
I've been trying to find the site which listed benchmarks on suspended vs rigidly mounted drives, but without luck :(

I did find these bits, though:
A potential concern is that the drive will be slower. The gist of it is that the drive may rely on rigid mounting in a chassis for accurate actuator positioning. A drive suspended in rubber allows movement of the drive when the actuator is moved. Drive performance can be degraded by up to 20%. I believe the drives tested were actually suspended swing like on rubber straps but I'm not sure.
Click to expand...
http://www.7volts.com/quiet.htm

This thread: http://forums.spodesabode.com/viewtopic.php?p=6010&highlight=

And from the Seagate 7200.8 Product Manual:
2.9.6 Vibration
All vibration specifications assume that the drive is mounted securely with the input vibration applied at the
drive mounting screws. Vibration may be applied in the X, Y or Z axis.

2.9.6.1 Operating vibration
The following table lists the maximum vibration levels that the drive may experience while meeting the performance standards specified in this document.
5-350 Hz -> 0.25 G acceleration

2.9.6.2 Nonoperating vibration
The following table lists the maximum nonoperating vibration that the drive may experience without incurring
physical damage or degradation in performance when subsequently put into operation.
5-22 Hz -> 0.25 G
22-350 Hz -> 5.0 Gs
Click to expand...

Because there's no way a suspended HDD will not vibrate if there's sufficient slack in the suspending contraption, the actuator will have a more difficult job positioning the head(s), hence performance will suffer.

Also, without the metal-to-metal coupling the drive will run much warmer, which will shorten its lifespan.

Truth must said, I have also suspended the drives in my main system, but I used a method which prevents HDD/case metal-to-metal coupling, cools the HDD (an aluminium profile mounted to each side) and has very little slack. I used little more than some aluminium profiles, rubber and some nuts and bolts.
 
The drives should be ok. You may want to do a full format of the disks in the intended orientation to allow the drive to write out the allocation units and get acclimated to the new orientation. The impact of natural gravity on HDD operation is very small, as the parts are ither fixed or levitated by the air currents caused by the moving parts.
 
Elledan said:
I've been trying to find the site which listed benchmarks on suspended vs rigidly mounted drives, but without luck :(

I did find these bits, though:

[...]

Because there's no way a suspended HDD will not vibrate if there's sufficient slack in the suspending contraption, the actuator will have a more difficult job positioning the head(s), hence performance will suffer.
Click to expand...
Right, like I said, your theory is sound but I still don't see any data to support a "serious degradation" in performance.
 
DougLite said:
The drives should be ok. You may want to do a full format of the disks in the intended orientation to allow the drive to write out the allocation units and get acclimated to the new orientation. The impact of natural gravity on HDD operation is very small, as the parts are ither fixed or levitated by the air currents caused by the moving parts.
Click to expand...
Is that really necessary? I don't really understand why formatting them in a different position would hlep anything. But I don't know much detail about the operations of hard drives either.

@ Ellenden about the aluminum profiles mounted to the sides. That is exactly what I planned on doing and cut out some little heatsink type things from the left over aluminum when I cut out my window. They're just U shaped brackets and I sanded the side that will touch the drive flat with my belt sander down to 320 grit so it has good contact. I plan on using rather stiff elastic cord as well to mount them, so it is rigid but at the same time will absorb enough vibration to quiet them.
 
xonik said:
Right, like I said, your theory is sound but I still don't see any data to support a "serious degradation" in performance.
Click to expand...
All I need is someone to run some HD benchmarks with both configurations :)

What I find very suspicious, however, is that in all of the reviews and DIY-articles of HD suspending contraptions I've seen so far, the performance impact is never mentioned.
 
Current technology hard drives do not matter what position you mount them in, but I would reccomend securely mounting them there. Suspended drives might affect the read heads accuracy, but noticeable amounts, I would wager not. Just doesnt sound like a good idea to dangle possible sensitive data...
 
Elledan said:
All I need is someone to run some HD benchmarks with both configurations :)

What I find very suspicious, however, is that in all of the reviews and DIY-articles of HD suspending contraptions I've seen so far, the performance impact is never mentioned.
Click to expand...
Recommend me some good programs with links and I'll do the benchmarks. It would be interesting to see if there is a performace hit.

Mickey21, I'll have them suspended but they're not just going to float around by a string. I'll have them secure enough so they don't swing back and forth when the case is moved.

I'll post pics and all when I finally get it done after I move back into my dorm this weekend.
 
Waffles730 said:
Is that really necessary? I don't really understand why formatting them in a different position would hlep anything.
Click to expand...
Older hard drives had a bit of slop in the read/write heads. If you formated it flat, then tipped it on its side, the drive heads would sag a bit and wouldn't be lined up exactly with the tracks. The theory was that you installed it, then formatted it in that position so the tracks would be in the right place. May or may not be true, but old hard drives had a mind of their own and any bit of extra voodoo you could come up with to make them run smoother was appreciated.
 
Zamboni said:
Older hard drives had a bit of slop in the read/write heads. If you formated it flat, then tipped it on its side, the drive heads would sag a bit and wouldn't be lined up exactly with the tracks. The theory was that you installed it, then formatted it in that position so the tracks would be in the right place. May or may not be true, but old hard drives had a mind of their own and any bit of extra voodoo you could come up with to make them run smoother was appreciated.
Click to expand...
Interesting. Mine are fairly new, so I'll try it without formatting first, and if there is a performance issue, I'll see what formatting would do to it.

I'll give HDTach a try when I move into my dorm this weekend and then do the mod some time next week hopefully. I'll let y'all know how it comes out :)
 
Well I did end up mounting the drives vertically the weekend I moved in here at RIT, just haven't gotten around to posting the pics and results. So here they are!
Original setup for the 2 Raptors in RAID 0:
Picture001.jpg

My mount constructed before installation:
Picture002.jpg

Views installed:
Picture003.jpg
Picture004.jpg
Picture006.jpg

Before benchmarks:
Short test
before-suspension-short-test.jpg

Long test
before-suspension-long-test.jpg

After benchmarks:
Short test
after-suspension-short-test.jpg

Long test
after-suspension-long-test.jpg


My construction for this was 4 heatsink type things I made out of the aluminum from my side panel window. I bent them U-shaped the best I could, drilled holes for the drive mount screws as well as for the elastics, and then sanded the surface that would make contact with the drives to make it as flat as possible for the most contact. For mounting to the case, I used a series of elastics secured to the top and bottom heatsinks. This creates a fairly rigid mount for the drives so swinging during transportation isn't a problem, but at the same time leaves the needed buffer for sound reduction.

As for noise, these are Raptors we're taking about. I guess I got my hopes up they would be near silent. They aren't, but the whine of theirs was cut down, as well as about half the seek noise. Much less noticable when I'm trying to get to sleep now :)

The benchmarks I ran were on the same install of Windows XP Pro x64, no format between tests. Average read was a little bit lower, burst read was a little bit higher, but the difference is pretty small so I consider it negligible. The CPU usage was 6% after though, compared to 2% and 4% before.

I think overall vertical mounting has helped me reduce noise without decreasing performance. They've been running a few weeks now, and its going great.
 
My manuals show how they can be mounted and basically udside down and rightside up horizontal and vertical is OK.

I would imagine the inertia that the heads have to counteract is much greater than gravity, and what is added up, is subtracted down so it should be a wash if anything.
 
You must log in or register to reply here.
Back
Top