SSD Noise (or Darn These Silent Computers)

DaveLessnau · Post by **DaveLessnau** » Thu Oct 29, 2009 10:39 am

I've got an Intel 160GB SSD in this computer. At first, I found it dead silent. But, I've now linked a sound to the SSD. When there's a lot of disk activity going on (for instance, during a virus scan), I hear faint, high-pitched, whining or scratching "bursts" of sound which vary with that disk activity. So, at first, I thought it was the SSD (it's definitely not coming from the speakers). Initially, I had just placed it on the bottom of my Antec Solo's drive cage (I wasn't happy with it in the standard Solo suspension, standard drive screws don't work with the Solo's rubber drive isolators, and the Solo's long drive screws don't fit the 2.5" drive). In an attempt to remove any coupling to anything that might vibrate, I used a type of adhesive gum and stuck the SSD to the bottom of the case, instead. The problem persists. Touching the drive while the noise is happening reveals no vibrations. Plus, moving my ear around the innards of the case leads me to believe that the sound isn't actually coming from the drive. It seems loudest in the areas of either the CPU or the power supply (neither of which made any such noise before I swapped my mechanical hard drive for the SSD). Has anyone else run across anything similar?

If it weren't for this computer being so darn quiet, there's no way I'd have ever heard it.

Darn these (near) silent computers and darn SPCR for addicting me to them

.

cmthomson · Post by **cmthomson** » Thu Oct 29, 2009 10:58 am

What you're hearing is almost certainly power converter noise. This usually emanates from the coils associated with DC/DC converters. This noise is a common complaint with high-end graphics cards, and is less common with power supplies or motherboards.

The reason you only hear it sometimes has to do with the specifics of the power being drawn while you're running that particular load. It is most likely due to the CPU alternating between loaded and idle while it is either waiting for the drive or scanning the data retrieved.

If you have the time and energy, you could experiment with very slightly modified CPU voltage or clock to avoid the particular load that is causing the issue.

JVM · Post by **JVM** » Thu Oct 29, 2009 3:07 pm

I have an Intel 160GB SSD G2 as main drive and did two virus scans without hearing anything peculiar.

green · Post by **green** » Fri Oct 30, 2009 5:46 am

see if symptoms match:
http://forum.corsair.com/v3/showthread.php?t=75330

DaveLessnau · Post by **DaveLessnau** » Fri Oct 30, 2009 8:08 am

At some point, I might try under-volting the system to see if that helps. But, I think I'll start with the C1E and EIST settings green recommended from his link. I'm almost positive I turned those on at some point. Thanks.

JVM · Post by **JVM** » Fri Oct 30, 2009 10:48 am

DaveLessnau wrote:At some point, I might try under-volting the system to see if that helps. But, I think I'll start with the C1E and EIST settings green recommended from his link. I'm almost positive I turned those on at some point. Thanks.

If I were you, I would first just disable CIE and see how it goes, unless there is some reason both need to be disabled.

DaveLessnau · Post by **DaveLessnau** » Fri Oct 30, 2009 11:13 am

Interesting. There sure appears to be a link between the various energy saving states and that noise. On this GA-EP45T-UD3P, there are four energy state options in the BIOS:

EIST
C1E
C2/C2E
C4/C4E

I had all of those on. If I turn them all off, the bursts of noise seem much less loud than with them on (but they're still present). I also noted that if C4/C4E is on, I get a set of error messages (one for each processor core) in Event Viewer every time I reboot (I'd been trying to track it down and this did it):

Idle power management features on processor 1 in group 0 are disabled due to a firmware problem. Check with the computer manufacturer for updated firmware.

and then:

Processor 1 in group 0 exposes the following:

1 idle state(s)
2 performance state(s)
8 throttle state(s)

After playing around with turning various of those settings on and off, I decided to just turn off the C4/C4E option and leave all the others on. That removes the errors I'd been seeing and seems to reduce the noise a bit (though it seems louder than if all those things were off). I figure the occasional barely audible bits of noise are acceptable when compared to turning off all my power saving states (particularly EIST and C1E).

Once I figure out the undervolting thing, I'll give that a try, too. It just seems odd that any of this would be tied in any way to the SSD.

JVM · Post by **JVM** » Fri Oct 30, 2009 11:22 am

Did this issue with noise start after the firmware update with TRIM?

DaveLessnau · Post by **DaveLessnau** » Fri Oct 30, 2009 1:11 pm

No. I started noticing the noise before I applied the TRIM-containing firmware update. I'm pretty sure it'd been there since I first installed the SSD: I just didn't notice it (it's only during heavy disk use and is quiet enough that, if the air conditioning or heat was on at those moments, it could get lost in the ambient noise). I suppose it's possible the noise was there even when I was using my mechanical drive, but the head seek noise during high disk use covered it. Or, maybe the speed of the SSD is fast enough that the CPU load/idle state switching (as cmthomson noted) passed a threshold and became audible.

mumford · Post by **mumford** » Fri Oct 30, 2009 1:12 pm

Scythe Quiet Drive? Not just for hard drives anymore.

JVM · Post by **JVM** » Fri Oct 30, 2009 2:26 pm

DaveLessnau wrote:No. I started noticing the noise before I applied the TRIM-containing firmware update. I'm pretty sure it'd been there since I first installed the SSD: I just didn't notice it (it's only during heavy disk use and is quiet enough that, if the air conditioning or heat was on at those moments, it could get lost in the ambient noise). I suppose it's possible the noise was there even when I was using my mechanical drive, but the head seek noise during high disk use covered it. Or, maybe the speed of the SSD is fast enough that the CPU load/idle state switching (as cmthomson noted) passed a threshold and became audible.

If that were the case, then why don't I hear it? And I haven't heard of anyone else having that problem.

I did run two virus scans, quick ones, and there was no such noise that I could hear.

ntavlas · Post by **ntavlas** » Fri Oct 30, 2009 4:45 pm

I`m pretty sure it is coil whine that was previously masked by the disk drive. Probably coming from the motherboard power circuitry though the psu isn`t ruled out. Anyway, it`s worth trying to undervolt a bit.

I also had errors showing up with the c4 state enabled as well as trouble resuming from standby. That`s on a similar-ish gigabyte p43 board.

DaveLessnau · Post by **DaveLessnau** » Sat Oct 31, 2009 6:32 am

Well, this morning, I temporarily replaced the SSD with my old WD6401AALS (WD 640GB Caviar Black) to test the noise. I left the drive out of the machine, set it on one pillow and stuck another over it to muffle the noise. When I booted up and ran a virus check, sure enough, I had the same noises coming from the area of the CPU or power supply. Removing the top pillow confirmed that the seek noises were pretty much in sync with those noises and were enough to disguise/mask them. So, it looks like the problem is coil whine and not specific to the SSD. Sorry.

cmthomson · Post by **cmthomson** » Sat Oct 31, 2009 3:39 pm

Don't be sorry; experiment with voltage/frequency and tell us what makes the noise go away!

RoGuE · Post by **RoGuE** » Sat Oct 31, 2009 3:52 pm

yep..I saw that one comin...

Oh well, good idea with the experiment at least...