Jump to content
IGNORED

Hard Drives Make a Difference in Sound


Recommended Posts

Looking at the specs for an Intel SSD and a Seagate Barracuda 4TB drive, the electrical power consumption of both are radically different. Not looking at how much power is consumed, but the rate of rise of that power.

 

The Intel 520 series draws 600mW at idle and 820mW. The change in power levels is 220mW in milliseconds as the drive reads and writes. There's no data to tell if the 5V or 12V rail draws the most, it's not important, since it's the rate of change of the power (more importantly the current) that the power supply has provide. The return of course is reflected in the ground plane. The faster the rate of rise of the ground current and magnitude combined with wires leads to noise.

 

The Seagate Pipeline 3.5in spinning drives can draw 3500mW - this is an average.... so the rate of current maybe slower, however the quantity is ten times that of the SSD. By inference, so does the noise and explains why an SSD 'sounds better' than a spinning drive.

To the extremes, an SD Card draws very little power again, below the SSD, that explains Superdad's findings of better performance.

 

You could introduce a linear supply with shielded 5V and 12V cables for the power section only of a PC's drives, since the actual power consumption of the drives are not that large, 15W would be plenty for audio use. The SATA cables themselves can be as is, always wiser to keep them short. To avoid differences in the ground plane, the Hard drive's power supply 0V needs to be connected to the 0V on the motherboard, fairly simple to do, there's usually spare molex connectors for CD drives that can be hacked.

 

The biggest trap is make sure the single connection to earth in the PC stays with the main PSU for the computer. There should be enough rejection in the Hard drive supply for a great deal of noise, the shielded cables help a lot to avoid emissions within the PC's housing.

 

Maybe the ideal is to have an SD card exclusively for the OS and stream music from NAS over Ethernet --> PC ---> DAC, and do away with the hard drives totally in really small drives like a Mac Mini. This way, the noise emanating from the hard drives is gone, problem solved! For OSX it does work, although I read somewhere that Windows has a bit of a problem installing on removable drives, but like most software, can be hacked, overcoming this issue.

AS Profile Equipment List        Say NO to MQA

Link to comment
I don’t agree my Intel 530 SSD sounds better than my Seagate hard drive. I have not tested an Intel 520 but I get the idea. The SSD treble kind of attacks you and over powers the music which leads to listener fatigue. It is not a relaxed sound immerging from the speakers floating in the air like it should be. I figure the sound is being affected by the internal electronics in the SSD. The other thing is my 1050 watt power supply is very large so it probably is less affected by power draw than a MAC.

 

I have tried serving files from my server using Seagate drives. The sound is good just like having a local Seagate drive except I don’t get top end air. I was thinking maybe use the network for older recording which don’t have top end air. It would be kind of a 2 level system. The best quality music would be local and the lesser music files would be on the network.

 

I don’t think Microsoft is going to allow TFTP remote booting of their servers either.

 

I would be interested in a linear power supply but I don’t think one exists which would work with my system. I am more of a computer guy than an electronics guy so I am not positive.

 

I have bought some shielded SATA cables to try to see if anything changes. I am looking for a good short one which I have not found on eBay. I will look at Fry’s when I go there.

 

I just chose the Intel 530 as an example and not specific to your build. Principally the noise is still there, no matter how large the computer's power supply is, since the rate is the same. A large PSU can ride through demands from the system, however there comes a point where the load is small, so the regulation circuits have nothing to measure and voltage tolerance start to exceed.

 

If you have another PSU, can be a small one, and join the two blacks at one point, and drive the hard drives only from this new supply. Not all that keen to use shielded SATA cables, since the frequencies these things run at needs something properly designed to avoid rounding the edges with weird capacitance. That's why I suggested to shield the power cables to the drives.

 

The differences in hard drives is coupling different noise levels to the DAC/Preamp, the idea is to lure the noise to the source (the PSU) rather than the DAC. To be really effective, the noise needs to be identified and a suitable filter trap built. That's easier said than done though, just ask John Swenson.

AS Profile Equipment List        Say NO to MQA

Link to comment
  • 2 months later...
In case others have misread what I said in my reply 262, I have accentuated a key word.

 

Which jitter could cause amplitude variations?

 

- Period Jitter ? So the timing wanders off spec. The timing would have to be so bad as to cause reflections in say a transmission line. In a computer - no transmission lines, and the music would be unlistenable.

 

- Determinisitic jitter would be more the culprit, since noise/EMC is the great instigator of this type. However this type of jitter doesn't increase in amplitude, tends to keep this under control.

 

Juan's problem is in the new Mac, since all other equipment is the same. Unless the OS is performing some trickery that is underhanded compared with the MacPro of old and new. OR there is more noise than signal, the music would reveal this in quieter passages.

 

The diagnosis requires a bit more thought, rather than plucking out concepts from hearsay that aggravate people.

AS Profile Equipment List        Say NO to MQA

Link to comment
You appear to be easily aggravated these days. I don't snipe each time you get on your hobbyhorse and recommend extensive mains supply modifications.

 

Let's just say if there's a cause for a thread becoming an all out bitch fight, sandyk is usually involved. Quite frankly am rather tired of the fighting on these threads, just venting my aggravation towards you in a VERY controlled manner. Come to think of it, why am I even bothering to write or read here, when arguments stir around the vortex of a cesspool with no end in sight.

 

This was an interesting place, one time.

 

Goodbye CA.

AS Profile Equipment List        Say NO to MQA

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...