Jump to content
IGNORED

External firewire buffer (tap test)


Recommended Posts

 

I went into my closet where my external firewire enclosure is stored. It's connect via Firewire 400 on a 10ft cable.

 

All of my music files are stored on (2) 1TB drives in this enclosure. They are WD "Green" version SATA drives.

 

I was putting my flashlight away and decided to lay it (gently) on the enclosure.

 

As I did this playback on my main zone, ASIO to EMU patchmix to DAC1, became garbled. The strnagest thing to me is that one a gently tap on the enclourse, playback doesn't lockup or skip but actually becomes skiddish like it was an analog signal; like you were jiggling a loose rca cable running from your radio to receiver.

 

I haven't had time to do any more tests or even think of any. This behavior is very surprising to me.

 

DC

 

Windows 10 x64 (no major tweaks)>JRMC v20>Adnaco S3B (Anker battery)>PPA USB>Auralic Vega (XLR output)>Tortuga Audio LDR v2 (custopm LPSU)>Decware EL34 (VCAPS, bias and UFO tranny mod)>Zu Union Cubes (Juptier Cap mod) - Cabling: Lectraline speaker, Antipodes Komako, Decware, and Huffman ICs

Link to comment

Wow, I've never heard of this or anything like it!

 

I can see a bunch of other things happening but not garbled sound. About the only thing I that come to mind is the possibility of something magnetic near the drive. I know that is a wild guess and most likely would not result in garbled output, but you really have me dumbfounded.

 

 

I would really appreciate if you can do a little more testing!

 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment

I tried tapping lightly on the enclosure at the start and end of a song with no difference. I thought maybe some kind of buffer was running out at the end of the song because it seemed to skip when I seeked to the end of track and tapped. It seems more like shock protection kicking in now, the illusion of garbled sound is possibly caused by the data stream being blocked rapidly by this??? STILL, the buffer should be a bit more beefy irregardless.

 

I thought about changing the write caching options on the drives.

 

the other test is to change the playback buffer in MC (default is .5 seconds); setting it too high can make changes (fades) less responsive.

 

DC

 

Windows 10 x64 (no major tweaks)>JRMC v20>Adnaco S3B (Anker battery)>PPA USB>Auralic Vega (XLR output)>Tortuga Audio LDR v2 (custopm LPSU)>Decware EL34 (VCAPS, bias and UFO tranny mod)>Zu Union Cubes (Juptier Cap mod) - Cabling: Lectraline speaker, Antipodes Komako, Decware, and Huffman ICs

Link to comment

Gang,

 

It is called a computer system for a reason. Everything you do can effect things that may not even appear to be related.

 

First off are all your power derived from the same wall outlet? All of your components including the drives and if the DAC1 is not isolated on the SPDIF then that and the rest of your stereo has too be on the same outlet. Not the same run, the same outlet or outlet pair.

 

Second 10f is too long for Firewire I think. I have had some issues in the past with longer Firewire cables.

 

What playback software? I take it you are using the EMU ASIO driver. It's pretty good but it's can be a bit flakey when presented with an error.

 

Thanks

Gordon

 

Link to comment

 

First off are all your power derived from the same wall outlet? All of your

components including the drives and if the DAC1 is not isolated on the SPDIF

then that and the rest of your stereo has too be on the same outlet. Not the

same run, the same outlet or outlet pair.

 

Are we talking about some kind of ground loop here? Not sure this matters because an external enclosure is not an audio component per se.

 

Wouldn't the enclosure end up being grounded to the HTPC chassis via the firewire cable itself?

 

I do have everything on the same dedicated 15A outlet, and even with power conditioning, tube amps, computers, etc. pull barley 5amps. No feedback loops or anything like so I think this is a simple case of data flow interruption due to physical shock. My worry was that it seemed to occur with very little stimulation i.e. LOUD bass....what if you have the floorstanders cranked way up? or DJ a gig with this enclosure sitting on a table near the speakers.

 

I think 15ft is the technical limit but so far so good with 10, and I definitely don't plan on sitting near my closet and tapping the drive while I try to enjoy music.

 

DC

 

Windows 10 x64 (no major tweaks)>JRMC v20>Adnaco S3B (Anker battery)>PPA USB>Auralic Vega (XLR output)>Tortuga Audio LDR v2 (custopm LPSU)>Decware EL34 (VCAPS, bias and UFO tranny mod)>Zu Union Cubes (Juptier Cap mod) - Cabling: Lectraline speaker, Antipodes Komako, Decware, and Huffman ICs

Link to comment

DC,

 

Yes even the drives are part of your system. It's possible that a ground loop is causing a drive fault or error when hit. Why not tell me exactly how you have this stuff powered including everything. I opened up my DAC1 and it is not isolated with transformers on any of the SPDIF other than the TOSLINK.

 

USB is capable of 5M, but that spec is reserved for stuff like keyboards and mice. Just because it says 5M it doesn't mean it's going to work or better off that the maker of your computer implemented the specification correctly.

 

We have found on allot of Firewire/USB ports that are WIRED off the mother board do not work nearly as well as the ones connected to the mother board. The good thing about Firewire is that the mother board is not sharing the usb with on boards stuff.

 

Yea at CES we had a ton of stuff plugged into the powerplant and it drew less than 1A for the entire system. Dual mono tube amps, tube preamp, tube USB dac and other stuff.

 

Thanks

Gordon

 

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...