Audio-GD DI-20

[sajunky]

The di20 needs over 700h of burn-in. Just be patient. Give it 2 additional weeks. At least. 2 weeks is far from enough.

It doesn't seem a temporary burnin regression, it is something about building up static electricity or S/PDIF receiver works on the border of losing synchronisation.

@ddpete, I would check grounding of all devices. A weak link seems to be S/PDIF connection between DI-20 and a DAC. Is it the same when playing 44.1 and 48kHz files? Different oscilators are used for two families, a small deviation of base frequency can cause such symptoms. Mojo Audio guys are audio perfectionist, they could use very tight synchronisation loop, I would also ask them wheter it is a known problem. If you have a diferent DAC, it would be the easiest way to identify a culprint. You can also try a different S/PDIF cable, it can be faulty.

 

[higginsd]

Any informations about a new firmware version?

 

[ddpete]

The di20 needs over 700h of burn-in. Just be patient. Give it 2 additional weeks. At least. 2 weeks is far from enough.

Hi,

Thank you for the input. I have already 'burned-in' my DI-20 for about 2oo hours, so I thought things would improve ... but they haven't. I'll keep burning away and shoot for another 2 weeks as you suggest. Thanks again.

 

[ddpete]

It doesn't seem a temporary burnin regression, it is something about building up static electricity or S/PDIF receiver works on the border of losing synchronisation.

@ddpete, I would check grounding of all devices. A weak link seems to be S/PDIF connection between DI-20 and a DAC. Is it the same when playing 44.1 and 48kHz files? Different oscilators are used for two families, a small deviation of base frequency can cause such symptoms. Mojo Audio guys are audio perfectionist, they could use very tight synchronisation loop, I would also ask them wheter it is a known problem. If you have a diferent DAC, it would be the easiest way to identify a culprint. You can also try a different S/PDIF cable, it can be faulty.

Thank you for your reply. Maybe it would be a grounding issue, but, as I use a sensitive tube amp, I thought I would've had other grounding issues before. Anyway, I use a 1.5m Black Cat S/PDIF going from the DI-20 to my Mojo Audio DAC. This cable sounds great when going straight from my transport to the DAC. But I'll swap it with a Kimber Illuminations D60 that I have and see if it makes a difference. Also, I do have another DAC so I'll replace the Mojo with that one and see if it makes a difference. I use JRiver on my Linux server. Do you think I might have something set up wrong on JRiver? Thanks again.

 

[PeterCraig]

I had a similar sounding problem years ago with a different DDC then the DI20. The issue occurred only when I changed sample rates and the dac lost the "lock" on the signal and the sound quality degraded. But turning everything off and on solved this issue. Until I again played music with a different sample rate.

The solution in this case was a new DDC.

 

[ddpete]

Also, as far as sound quality, rebooting the computer once in a while is a good idea.

Thanks for another reply to my inquiry. I have rebooted my server a couple of times .. but it did not change anything. It did change things the first time I did that, but the 2nd and 3rd time didn't make any change. The sound quality is not really a high fidelity issue. The sound that comes out is fuzzy with a lot of background hiss/fuzz. The music can still play simple single instrument pieces but the fuzziness is obvious (even to my son who is not a 'listener') and when the music starts to get a little more complicated and more instruments come in or the tempo changes quickly, the sound of the instruments deteriorates to very fuzzy sound as well. I'm using JRiver, but I dont know if maybe I have something set wrong on that. Thanks again.

 

[ddpete]

I had a similar sounding problem years ago with a different DDC then the DI20. The issue occurred only when I changed sample rates and the dac lost the "lock" on the signal and the sound quality degraded. But turning everything off and on solved this issue. Until I again played music with a different sample rate.

The solution in this case was a new DDC.

Thank you for your input on this. I'm going to try another DAC that I have and see if it changes. I hope it's not the case as I really like my Mojo DAC and that is the reason I went with the DI-20 to begin with.

 

[FredA]

Thank you for your input on this. I'm going to try another DAC that I have and see if it changes. I hope it's not the case as I really like my Mojo DAC and that is the reason I went with the DI-20 to begin with.

The rca out is more likely to work pefectly than the bnc, which is a current output, by the way.

 

[sajunky]

The sound quality is not really a high fidelity issue. The sound that comes out is fuzzy with a lot of background hiss/fuzz. The music can still play simple single instrument pieces but the fuzziness is obvious (even to my son who is not a 'listener') and when the music starts to get a little more complicated and more instruments come in or the tempo changes quickly, the sound of the instruments deteriorates to very fuzzy sound as well. I'm using JRiver, but I dont know if maybe I have something set wrong on that.

Changing tempo confirms that it is PLL synchronisation issue. It happens when a source clock frequency differ to much from a nominal frequency a receiver is tuned for or error rate is to high. When a sample rate changes from 44.1kHz to 48kHz or in reverse, a different oscilator is involved and problem probably will go away. In this situation if by example problem is at 44.1kHz, upsampling to a multiple (88.2 176.4...) is expected to make it worse. But resampling to 48kHz should fix a problem. I don't say it is a permanent solution, but worth checking...

It has probably nothing with JRiver, but here you can resample using SoX (I think) from 44.1kHz to 48 or in reverse to see which base frequency gives a problem.

The easiest way to determine a cause is to swap devices/cables, please report a progress.

 

[ddpete]

Changing tempo confirms that it is PLL synchronisation issue. It happens when a source clock frequency differ to much from a nominal frequency a receiver is tuned for or error rate is to high. When a sample rate changes from 44.1kHz to 48kHz or in reverse, a different oscilator is involved and problem probably will go away. In this situation if by example problem is at 44.1kHz, upsampling to a multiple (88.2 176.4...) is expected to make it worse. But resampling to 48kHz should fix a problem. I don't say it is a permanent solution, but worth checking...

It has probably nothing with JRiver, but here you can resample using SoX (I think) from 44.1kHz to 48 or in reverse to see which base frequency gives a problem.

The easiest way to determine a cause is to swap devices/cables, please report a progress.

Thank you for your response. I had a fleeting thought last week that this may be due to some synchronization issue. I will certainly try to use JRiver to resample the output and see what happens. I am not very familiar with it (JRiver), but I'll figure it out. I'm also going to try swapping cables and DACs. In the mean time, the burn-in process continues. I really do appreciate and value all the suggestions I have received concerning this issue. I have an insane work schedule and wont be able to try any of these things until Tues. I will try them on Tuesday and report back at that time. Thanks again.

 

[ddpete]

The rca out is more likely to work pefectly than the bnc, which is a current output, by the way.

Thank you for that tidbit of information. I'll stay away form the BNC output connector on the DI-20 until I get everything running perfectly.

 

[PeterCraig]

Looking forward to more of this story. The ultimate test bench is to be able to switch out and try different gear.

 

[sajunky]

Looking forward to more of this story. The ultimate test bench is to be able to switch out and try different gear.

There is no ultimate test in this story. If one device works slightly outside a nominal frequency limits and other the same but in the same direction, they will still work together. It doesn't mean that those two are right, as a device that stays perfectly inside prescribed limits may not work. Measurement is the best, statistical next, unfortunately none of them is available for home user. All we can do is to (1) confirm a cause, (2) trying one of possible action.

 

[ddpete]

Thank you for your response. I had a fleeting thought last week that this may be due to some synchronization issue. I will certainly try to use JRiver to resample the output and see what happens. I am not very familiar with it (JRiver), but I'll figure it out. I'm also going to try swapping cables and DACs. In the mean time, the burn-in process continues. I really do appreciate and value all the suggestions I have received concerning this issue. I have an insane work schedule and wont be able to try any of these things until Tues. I will try them on Tuesday and report back at that time. Thanks again.

I guess another thing that troubles me about this whole thing is that, prior to obtaining the DI-20, I was using a M2Tech HiFace2. It simply plugged into one of the usb ports on my server, JRiver saw it and I selected the HiFace, plugged the S/PDIF cable into the HiFace and my dac and launched the music. It worked with everything with no problem and it sounded good. I just didn't like plugging and unplugging my S/PDIF cable when I wanted to use the CD transport. Now I feel like I've got a high-maintenance device that is taking the fun out of audio. Basically, I want to just sit and listen .. I don't want to get up to turn equipment on and off when things don't sound right and, given the investment I have in my system, any sound which is not high fidelity makes me cringe. Plus constantly rebooting this equipment, In addition to being a big pain in the butt, will also wear out the equipment fast. So I'll try the things that have been suggested and give it a bit more burn-in time. But I've already spent a lot of time and personal bandwidth on this thing just to get it to work. Hopefully I'll find a solution soon. Else I'll need to start to consider alternatives (i.e. I still have the HiFace2)

 

[gimmeheadroom]

Haven't understood the issue but a few questions, feel free to ignore if you don't find them helpful.

1. Are you using USB input to the DI-20 and have you confirmed that USB is selected on the DI-20?
2. Are you using coaxial S/PDIF input to the DI-20 and have you confirmed coax is selected on the DI-20?
3. Have you tried switch between parallel and serial on the DI-20, and does it make any difference?
4. Have you checked the DI-20 settings to verify everything is as desired? Do you perhaps have external clock selected without a 10 MHz clock connected?
5. Have you verified the DI-20 appears reliably from the standpoint of the Linux box? I don't remember if Amanero is needed for Linux but I guess it is. If you press the button on the back it's possible to wipe Amanero and it would have to be reinstalled.