Weird hard drive squeal

Silencing hard drives, optical drives and other storage devices

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Weird hard drive squeal

Post by BaconTastesGood » Sat Mar 01, 2003 10:46 pm

One of my systems is quiet enough now that I hear this really, REALLY weird chirping from the hard drive (I think it's the hard drive at least). This is when the system is completely idle, no activity. It's a chirp and squeal, then it does it again, a little thump, then it's quiet. Then a few seconds or even a few minutes later, it repeats it. Almost like some periodic check of something.

I have no idea what could be causing this -- any ideas?

Ralf Hutter
SPCR Reviewer
Posts: 8636
Joined: Sat Nov 23, 2002 6:33 am
Location: Sunny SoCal

Post by Ralf Hutter » Sun Mar 02, 2003 5:41 am

Backup your data.

Backup your data.

Backup your data.

SpyderCat
Posts: 208
Joined: Sun Feb 23, 2003 12:22 pm
Location: The Netherlands

Re: Weird hard drive squeal

Post by SpyderCat » Sun Mar 02, 2003 6:16 am

BaconTastesGood wrote:One of my systems is quiet enough now that I hear this really, REALLY weird chirping from the hard drive (I think it's the hard drive at least). This is when the system is completely idle, no activity. It's a chirp and squeal, then it does it again, a little thump, then it's quiet. Then a few seconds or even a few minutes later, it repeats it. Almost like some periodic check of something.

I have no idea what could be causing this -- any ideas?
You just beat me in posting on this subject.
Is your drive a IBM Deskstar GXP perhaps?
My drive makes more or less the same noises, about once every 10 hours of operation, and mostly when idleling.
My guess was that it's some kind of "de-magnitizing routine".
I attach what SiSoft Sandra says about this "cool decoupled drive" running at 37*C/98.6*F 8).

Questions:

Sandra says: "Acoustic Management Enabled : No"
How do I enable it, and does it help?

Could any of the more knowledgeable people comment on:
My guess was that it's some kind of "de-magnitizing routine"?

Regards, Han.


Results from SiSoftware Sandra:

General Capabilities
Channel : Master
Type : ATA
Removable Device : No
Model : IC35L120AVVA07-0
Revision : VA6OA52A
Serial Number : VNC602A6C5R8EA
ATA/ATAPI Approved Version : 5.21
Cache Buffer Type : Multi-Sector with Read Cache
Cache Buffer Size : 1863kB (3727 sectors)

Drive Geometry
CHS Geometry : 16383 x 16 x 63
CHS Total Sectors : 16514064
LBA Total Sectors : 241254720
Number of ECC Bytes : 52
Disk Capacity : 115.0GB

Translation Mode Geometry
CHS Geometry : 16383 x 16 x 63
CHS Total Sectors : 16514064

Properties
IORDY Support : Yes
IORDY Disableable : Yes
LBA Support : Yes
DMA Support : Yes

Features
SMART Support : Yes
Security Support : Yes
Power Management Support : Yes
Advanced Power Management Support : Yes
Power-up in Standby : Yes
Packet Command Interface : No
Removable Media : No
Look-Ahead Buffer : Yes
Write-Back Cache : Yes
Host Protect Area : Yes
Microcode Update : No
Acoustic Management : Yes
48-bit LBA : No
Device Config Overlay : Yes

Active Features
SMART Enabled : Yes
Security Enabled : No
Power Management Enabled : Yes
Advanced Power Management Enabled : No
Power-up in Standby Enabled : No
Look-Ahead Buffer Enabled : Yes
Write-Back Cache Enabled : Yes
Host Protect Area Enabled : Yes
Acoustic Management Enabled : No

SMART Information
Version : 1.01
ATA Commands Support : Yes
ATAPI Commands Support : Yes
SMART Commands Support : Yes

SMART Data
Raw Read Error Rate (01) : 0 (Th 60; Min 100; Max 100)
Throughput Performance (02) : 0 (Th 50; Min 100; Max 100)
Spin-up Time (03) : 298 (Th 24; Min 115; Max 115)
Start/Stop Count (04) : 309 (Th 0; Min 100; Max 100)
Re-Allocated Sector Count (05) : 0 (Th 5; Min 100; Max 100)
Seek Error Rate (07) : 0 (Th 67; Min 100; Max 100)
Seek Time Performance (08 ) : 0 (Th 20; Min 100; Max 100)
Power-On Time Count (09) : 1770 (Th 0; Min 100; Max 100)
Spin Retry Count (0A) : 0 (Th 60; Min 100; Max 100)
Power Cycle Count (0C) : 275 (Th 0; Min 100; Max 100)
Power-Off Park Count (C0) : 367 (Th 50; Min 100; Max 100)
Load/Unload Cycle Count (C1) : 367 (Th 50; Min 100; Max 100)
Drive Temperature (C2) : 37 (Th 0; Min 148; Max 148)
Re-Allocated Data Count (C4) : 0 (Th 0; Min 100; Max 100)
Pending Sector Count (C5) : 0 (Th 0; Min 100; Max 100)
Un-Correctable Sector Count (C6) : 0 (Th 0; Min 100; Max 100)
CRC Error Count (C7) : 0 (Th 0; Min 200; Max 200)

Environment Monitor(s)
HD / Aux Temperature : 37.0°C / 98.6°F

Transfer Modes Support
Queue Depth : 32
Maximum Block Transfer : 16
Transfer Width : 16-bit
Max PIO Mode : PIO-4
Max DMA MW Mode : MWDMA-2
Max UDMA Mode : UDMA-5

Transfer Modes Active
Current Block Transfer : 16
Current Active Mode : UDMA-5

Herb W.
Posts: 113
Joined: Sun Sep 15, 2002 8:35 am
Location: Toronto

Post by Herb W. » Sun Mar 02, 2003 9:17 am

I wouldn't hit the panic button yet - yes, it's probably just a calibration routine. IBM has included something like this in their newer drives to recalibrate after temperature changes. This was discussed in a thread on the 180GXP...

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Sun Mar 02, 2003 9:54 am

This is an 80GB Cuda. It's not exhibiting any problems, works fine, but the noise is basically constant -- it's a tiny squeal/thump every couple seconds on average. I never would have heard it if I hadn't made the rest of the system quiet.

What's real weird is that I just noticed that WinXP calls it a SCSI IDE hard drive (?!).

Justin_R
Posts: 319
Joined: Sun Aug 11, 2002 3:26 pm
Location: Philadelphia, PA, USA

Post by Justin_R » Sun Mar 02, 2003 1:22 pm

I'll second the "back up the data" vote. It never hurts.

There was an issue a while back with Barracuda's running some kind of "SMART diagnostic" or something which they are supposed to complete just once. What would happen was, the diagnostic would start, then get interrupted by a request from the OS, service the OS request, and then restart the diagnostic. The fix was to leave the hard drive powered up but disconnected from any IDE controller overnight.

You may also want to check the physical mounting of the drive, although that seems unlikely.

If your warranty is still good and the noise doesn't go away after a week, I'd say to RMA it.

And finally, just to check, are you sure that it's the hard drive? Might it be a floppy, or a ZIP drive? You seemed unsure in your original post.

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Sun Mar 02, 2003 1:24 pm

Justin,

That sounds like it me doing that. It's not the floppy or CD-ROM drive, it's definitely the hard drive, unless there are mechanical parts on my mobo I don't know about =) (I just listened closely so that I wasn't unsure anymore)

Any reference to that SMART diagnostic bug?

Maybe I could boot up the system, go into BIOS, and leave it like that overnight instead of disconnecting it? I can't get into my case very easily (rack mounted).

Ralf Hutter
SPCR Reviewer
Posts: 8636
Joined: Sat Nov 23, 2002 6:33 am
Location: Sunny SoCal

Post by Ralf Hutter » Mon Mar 03, 2003 6:12 am

I have four 80GB cuda IV's and have never had the noise that you're talking about, nor have I had that SMART diagnostic issue. I've also built at least 30-40 systems for other people using the 80GB cuda IV and have never had issues with those drives either.

I'll repeat my "back up" recommendation once more. Do this before you do ANY other troubleshooting or diagnostics. I'd call Seagate and RMA the drive. That noise is not normal, whether it's constant or not.

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Mon Mar 03, 2003 10:10 am

My data is already backed up, no worries there, but for some reason I don't think it's a hard drive failure.

Here are the weird things that concern me:

- if the machine is on but NOT in Windows, e.g. in the BIOS, it does not make that noise

- WinXP thinks it's a SCSI device, what is up with that?!

Justin_R
Posts: 319
Joined: Sun Aug 11, 2002 3:26 pm
Location: Philadelphia, PA, USA

Post by Justin_R » Mon Mar 03, 2003 12:40 pm

Ralf-

Check this thread and the link there to the Storage Review thread. But if the sound occurs only with Windows loaded, I doubt this is the problem. (Although it shouldn't hurt to try the fix.)

Bacon-

If the hard drive only makes the noise in Windows, I suppose it must be some Windows software doing it. Check the settings on any kind of background-running software you have that might access the hard drive: anti-virus programs, indexing services, (hopefully not) spyware. Dan's Data addresses a somewhat similar problem here.

As for the "SCSI" thing, is your HD attached to one of the chipset IDE controllers on your motherboard, or to something else (IDE controller PCI card, motherboard RAID controller)? I know Win2K will identify the latter as a SCSI device sometimes.

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Mon Mar 03, 2003 12:49 pm

Justin,

Thanks for the links. I have no idea what's causing this, as it's now become intermittent. It MIGHT have something to do with the motherboard though.

It turns out that the motherboard is exporting its IDE interfaces as SCSI -- my IDE DVD-ROM is also reporting as SCSI. Must be some idiosyncracy of nForce2.

I tried leaving it on overnight, still get the same squeaks. I'll try the Seagate diagnostics, and messing with the sleep stuff.

thanks!

powergyoza
Patron of SPCR
Posts: 543
Joined: Mon Oct 21, 2002 1:01 am
Location: Vancouver, BC, Canada
Contact:

Post by powergyoza » Mon Mar 03, 2003 2:07 pm

BaconTastesGood wrote:It turns out that the motherboard is exporting its IDE interfaces as SCSI -- my IDE DVD-ROM is also reporting as SCSI. Must be some idiosyncracy of nForce2.
It's not a idiosyncracy of the nForce. Windows NT/2k/XP is built in a certain way where it assumes everything is SCSI. The translation from IDE -> SCSI is taken care of by the IDE/ATAPI miniport driver, AFAIK.

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

Re: Weird hard drive squeal

Post by mg1394 » Wed Mar 19, 2003 8:09 pm

BaconTastesGood wrote:One of my systems is quiet enough now that I hear this really, REALLY weird chirping from the hard drive (I think it's the hard drive at least). This is when the system is completely idle, no activity. It's a chirp and squeal, then it does it again, a little thump, then it's quiet. Then a few seconds or even a few minutes later, it repeats it. Almost like some periodic check of something.

I have no idea what could be causing this -- any ideas?
Well, I have exactly the same problem with 4 different computers with B-IV drives (80MB.)

I've noticed it for months, and cannot find a cause. All the drives are new, all act the same. Computers have both Intel and ASUS motherboards, all P4's, with very quiet P/S (Q-Tech and PC P&C) and Pabst fans.

All systems run W2k sp2.

The systems are actually quieter when the disks are working, seeking and copying files. The "squeals" only occur when the system is idle, it stops after some time, but periodically returns. I have done a number of reinstalls from scratch - no difference. They do all use the Intel Accelerator code.

Ralf, I understand your comments, but this isn't a single drive problem. I guess I can't rule out a large bad batch of drives, but this is a real problem that is not a drive failing.

For over four months I've tried the Seagate site and seen no mention of the problem and found nothing in the archives -

To repeat - these drives are quiet when working - yes, I can hear spin noise (with a stethoscope), and the quiet click of seeks when happening. This "squealing" occurs when the system is idle and it comes and goes with time (5 seconds on, off for awhile, again, etc. - but not a fixed pattern at all.)

If there are others with knowledge of this or suggestions, I appreciate your help.

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Wed Mar 19, 2003 8:21 pm

I've tried the Seagate SMART diags, the FeatureTool and enabled acoustic quiet mode, etc. I've left my computer on overnight, etc. No difference. And it doesn't seem to be correlated to much of anything, with the possible exception that it goes away when the computer is NOT idle.

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

Post by mg1394 » Thu Mar 20, 2003 10:25 am

Well, we (BTG and I) must have the strangest Seagate Barracuda IV drives ever built - and I have 6 that all do the same thing (two spares along with the above 4). Quite amazing that this only exists with us and not most other users.

I haven't seen the quiet mode program for Seagate - is there a working link to it somewhere??

Yes, there is no squeal ever when the computer is not idle - I set up a script to copy 4GB files from one drive to another continuously and monitored this for 3 hours.

When idle, the squeal sequence happens for about 1-2 minutes, then goes silent for 20-40 minutes, then repeats (this from last night).

I am next going to load Linux on one of the machines and see what the effect of that is. For no good reason, I suspect the OS is involved in at least triggering this oddity.

rpc180
*Lifetime Patron*
Posts: 309
Joined: Sun Feb 09, 2003 8:01 pm
Location: Washington, DC

Post by rpc180 » Thu Mar 20, 2003 4:00 pm

could it be the hard disk calibration that windows does every few minutes? I get idling read/writes, but no squealing. I also would not hear it if the system wasn't silent. I heard somewhere that WIndows XP does this in order to maintain file integretity or something. Some people say it is a setting in the registery that can set how long between cycles it occurs. I last heard about it through an Windows XP/Registry optimization online guide, but I didn't pay too much attention to it until now.

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

Post by mg1394 » Fri Mar 21, 2003 11:25 am

Could be something like that, although why it causes a squeal is still a mystery. Note my machines are all W2k, not XP, but they may be the same. I'll look for pointers to register settings. Thanks.

Dru
Patron of SPCR
Posts: 154
Joined: Sun Aug 11, 2002 3:26 pm
Location: Gilbert, Arizona, USA
Contact:

Post by Dru » Sat Mar 22, 2003 10:33 am

I have a 40GB Cuda IV exhibiting the same problem. The firmware is 3.04 noted on the drive's label.

I have 2 40GB Cuda IVs running in Raid. These two don't have this problem.

The drive performs without any problems but the squealing does bring an eerie feeling. I've had this drive for 2 years now. I "think" the sound has creeped up but have no proof. I may have not noticed it before? I have 1 year left on the warranty. I wonder if Seagate would honor a complaint and replace the drive?

rpc180
*Lifetime Patron*
Posts: 309
Joined: Sun Feb 09, 2003 8:01 pm
Location: Washington, DC

Post by rpc180 » Sun Mar 23, 2003 12:04 am

I found this on the Seagate Website while researching this problem:



Re: I hear the "click-click" of the head all the time even when the machine is totally idle.

Seagate Technician's Response:

When the drive is totally idle, that is with no command activity being sent from the host and none pending, it will do seeks based on a timer. Usually this happens every 15 to 30 seconds, depending on the model of drive you have.

This is called STIR (seek to improve reliability). This is designed to prevent the heads from remaining in any given position for extended periods of time. Since the heads are flying a very short distance above the media, and are flying on a cushion of air, the compression of the air beneath the head does impart a small amount of heat to the media because of the friction involved. If the heads remained over a cylinder for extended periods of time, the heat build-up on the media in a small localized area could degrade the media in that particular locality. So, periodically during idle, the heads will be moved to another cylinder and the servo recalibrated for all heads. This is perfectly normal.

If you experiencing more than the standard "click-click" on a timed basis, it would indicate the system is probably doing something in the background.

Original Email Message:

The drive continuously seeks. I hear the "click-click" of the head all the time even when the machine is totally idle. Can you explain this for me?

Dru
Patron of SPCR
Posts: 154
Joined: Sun Aug 11, 2002 3:26 pm
Location: Gilbert, Arizona, USA
Contact:

Post by Dru » Sun Mar 23, 2003 5:45 am

I've noticed my 7200.7 does that click-click noise, which is fine. However, that squeal noise is totally different. So, this does not pertain to the squeal noise.

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

Post by mg1394 » Wed Mar 26, 2003 9:31 pm

Hmmmm, so we have at least three people with one or more Seagate B-IV drives that squeal, both 40 and 80GB models, and of the thousands of other users (and probably hundreds on this forum) no one else has observed the problem or any idea of a fix?

I also ran the acoustic management program on two of the computers and it made no difference at all in the squeal.

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

PROBLEM SOLVED!!

Post by mg1394 » Sat May 17, 2003 2:32 pm

:D :D :D

My problem of squealing Seagate Barracuda IV drives is finally solved. I ran across a comment on the Yahoo Silent-PC forum and tracked down a mysterous utility from Seagate called SMSCAN.EXE (DOS program). It seems that B-IV drives built prior to October 2001 (early in their life) had a problem with the self-test code (runs when drive is released and idle) and some made squealing and other noises. This utility fixes the noise problem. What an Odyssey!

The program can be found in the Yahoo Group archives.

BaconTastesGood
Posts: 137
Joined: Mon Nov 04, 2002 3:55 pm

Post by BaconTastesGood » Sat May 17, 2003 5:38 pm

Do you need to boot into DOS mode or anything like that special, or will it run correctly under XP's CMD window?

mg1394
Posts: 167
Joined: Sun Aug 11, 2002 3:26 pm
Location: Silicon Valley

Post by mg1394 » Sat May 17, 2003 8:00 pm

It's a pure DOS program and needs to be run from a DOS floppy or equivalent. Good luck - we've been waiting for quite awhile. :)

Post Reply