Audio-gd Digital Interface - Page 169

Perhaps this question has been asked already, but when using the DI with a Mac OS, what is the proper volume setting? Is it 50%, or is there a way to pass through the volume to not allow the OS to control it? I normally have my volume set to 100% but when I set it down to 50% and controlled the volume with my amp some of the harshness I heard was gone.

Originally Posted by kr0gg 

 

>> If you connect external PSU(9-15VDC) to USB-ISO PWR-JACK, then the PSU
>> will supply power to USB HOST connector and on board isolator. [..]

 

as i understand, that means that connecting an external PSU substitutes the power sucked from PC.

Yup, but it goes through a BD9001F switched regulator...your choice which one is more evil ^^

Oleg was supposed to release a linear regulated dongle(read his comment at the bottom of the page), but that's still in the works AFAIK.

Quote:

Originally Posted by realmassy 
Yes

I tried it yesterday, with a 6V, 500ma PSU, and honestly I couldn't spot any difference.

I wouldn't expect any differences out of a switching-mode power supply when that is what is powering your USB ports and they add massive amounts of noise and ripple to the power. Also, not sure why you're commenting on sonic changes when you don't own a DI and that's what the thread is about; ergo most people are using the class A linear power supply that Audio-GD sells.

I think the point leeperry was making is that the problem is the power supply on the isolator itself, not which power supply is feeding the input to the isolator.  This probably also explains why I didn't find improvement feeding DI power supply through the OMILEX USB isolator.  Either way I hunted down a 12V 400 mA linear wall wart to power the isolator -  but the sound is still less focussed, detailed and articulation of transients is reduced a little (for example on vocal tracks small transients are almost impossible to hear with the isolator).  Maybe with a fully linear power supply things would improve, but I am personally surprised how many CMP project ppls use these isolators.

I have pretty much the opposite experience using the isolator. More detailed, more focused, broader and more encompassing soundstage, better dynamics and transient response, etc. I was testing various configurations of the DI earlier and reaffirmed my belief that the AGD power supply sounds best into the olimex, in fact this is the only time I could hear a definite improvement with the power supply. I don't think the switching isolator is anything to worry about, but something that could be improved upon like that inline linear psu he linked.

I mean this could be the case but in the ideal world a system would be better if it didn't need an isolator - all it can do is remove RF interference from the data lines, and at the same time AFAIK it must increase the rise time of the USB signal.  So you are swapping random jitter for deterministic jitter.  I am similarly suspicious of the BNC atennuators I have ordered - which in theory reduce signal reflections, but also definitely increase the rise time of the SPDIF signal.  Personally I think it would be better to tackle the problem at the source rather than add another piece of equipment to try and fix the original problem (eg. maybe run external power supply into motherboard and CPU, although again this might be a case of solving a problem that is not actually there in the first place.)  Or I might just give up and get a mac mini as a music server, or build a CMP system to complicate things further...

Quote:

Originally Posted by somestranger26 

I wouldn't expect any differences out of a switching-mode power supply when that is what is powering your USB ports and they add massive amounts of noise and ripple to the power. Also, not sure why you're commenting on sonic changes when you don't own a DI and that's what the thread is about; ergo most people are using the class A linear power supply that Audio-GD sells.

I wasn't posting about the DI, but about the OLIMEX isolator, which was talked about a few posts ago. Can't I give my opinion on it?

By the way, the PSU I tried is a linear regulated, not a switching one.

Can someone who has DI ver. A, motherboard soundcard and is using desktop PC tell me if there's a difference between usb and coax input?

Quote:

Originally Posted by realmassy 

I wasn't posting about the DI, but about the OLIMEX isolator, which was talked about a few posts ago. Can't I give my opinion on it?

We're talking about Olimex + DI. The combination of gear certainly matters. Several people have been posting that they like the PSU connected to Olimex using the Digital Interface. That you can't tell a difference using a different interface isn't really relevant to this thread. Also, most of the people using random PSUs mean crappy switching mode ones so I made a bad assumption.

Speaking of PSUs, has anyone tried using battery packs as a PSU with the DI? The cheapest, simplest option might be a 6 AA (NiMH) rechargeable battery pack. This should produce around 7V or just over - which is what Kingwa recommends.

Maybe something like this?

Sorry if this has already been addressed.

The digital Interface power supply is 5 volts and 6 x 1.5 batteries in series is 9 volts (significantly over spec).

Yes battery power would be interesting to explore though. I have considered LIPOs and will certainly try them on my Audiophellio 1's when they arrive.

Regards

Macrog

Audio-gd states that any PSU between 5V-10V will work and recommends 7V. 5V just happens to be the voltage of a standard USB power rail.

6 Alkaline AA batteries in series will output ~9V when fully charged. However, the rechargeable NiMH (Sanyo Eneloops, etc) batteries put out a bit less voltage (around 1.2 instead of 1.5V), giving around 7.5V or a bit less for 6 in series - right on the money according to Audio-gd. I'm not sure that LiPOs are necessary here; they might be overkill. The DI seems to require only 300 mA (or maybe a bit more) given the specs on the Audio-gd page. this means you'll probably get a good 5-6 hours or so on a fully charged, relatively fresh set of  NiMH AAs. The increased current output capability and energy density of LiPOs seems wasted here. Like LiPOs (and unlike Alkaline and lead-acid), NiMH has a relatively flat discharge curve (maintain a more constant voltage over discharge cycle).
 

Quote:

Originally Posted by macrog 

The digital Interface power supply is 5 volts and 6 x 1.5 batteries in series is 9 volts (significantly over spec).

 

Yes battery power would be interesting to explore though. I have considered LIPOs and will certainly try them on my Audiophellio 1's when they arrive.

 

Regards

 

Macrog

My error regarding digital interface PSU sorry. Yes you are right 7v 400ma

Regards

Macrog

I am sorry if this has been answered before, but I would like to ask whether there is a way to eliminate a short playback delay (about half a second or so) at the beginning of new tracks that Digital Interface sometimes causes.

Here's my system: PC (foobar) - DI - Ref. 7 - AT-HA5000 - ATH-W5000

Quote:

Originally Posted by macrog 

The digital Interface power supply is 5 volts and 6 x 1.5 batteries in series is 9 volts (significantly over spec).

 

Yes battery power would be interesting to explore though. I have considered LIPOs and will certainly try them on my Audiophellio 1's when they arrive.

 

Regards

 

Macrog

LiPO's can be very dangerous (fire) when charging and/or when they get wet. LiFE's are generally thought to be much safer but take note that the voltage per cell is a little different than LiPO.

LiPO's are also easily damaged if they are allowed to drain down too far before recharging.

Just throwing this out there....