[f] AMB on Facebook

γ3 On the Move

gamma3, gamma24 (plug-in for the γ3)

γ3 On the Move

Postby amb » March 12th, 2016, 6:58 am

Hmm. I suppose it's possible, but that's a very strange failure mode.
User avatar
amb
Site Admin
 
Posts: 10432
Likes: 176 posts
Liked in: 495 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 12th, 2016, 5:26 pm

Another thing,
if "Digital Outputs-Loop Out", should that sampling frequency be the native base frequency on JP2 y24 44.100 on red book CD? LCDuino display says 44.100, output from the coax is 44100 I am pretty sure.
All I can see is 96kHz or 192kHz up-sampled frequency on JP2.
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: γ3 On the Move

Postby amb » March 12th, 2016, 5:40 pm

The LCDuino-1 display will always show the native sample rate. If you set "Digital Outputs - Loop-out" in the menu, then the digital outputs will preserve the sample rate of the input stream, but only for inputs 1-4. If your input is set to 0 (USB), the digital output will always be upsampled. This is because the SRC4392 has no "loop-out" path from its serial (I2S) ports to the DIT. The DIR has such paths to the DIT.

Also, you will always see an upsampled rate at γ24's JP2, because that's the I2S bus to the DAC(s), not to the digital outputs.
User avatar
amb
Site Admin
 
Posts: 10432
Likes: 176 posts
Liked in: 495 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 12th, 2016, 5:45 pm

OK, when playing CD from CDP coax out , input #2 in DAC, digital output in loop out mode, the LRCKA is 96kHz and BCKA is 6.14403MHz .
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: γ3 On the Move

Postby amb » March 12th, 2016, 5:49 pm

This is correct. I just added one more sentence to my post above that explains why.
User avatar
amb
Site Admin
 
Posts: 10432
Likes: 176 posts
Liked in: 495 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 12th, 2016, 6:02 pm

OK that's good. All clocks checked and looking pretty good and solid.
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 13th, 2016, 3:23 pm

Another observation: when playing from PC (USB only connected), if I select other input and get the popping noise (popping glitches are like 1.4Hz), the DC offset on XLR outputs goes to 45mVDC and RCA outputs are 22mVDC(glitches spiking higher above that DC voltage). If muted, the offset drops to normal value 2.6mVDC on RCA if any input is selected and music played. Is the muting done on DAC chips? Because I can still see the data flowing on SDOUTA on JP3. There is still some data on SDOUTA present, even though the music is not played through any input. And like I wrote before, when the coax locks on, everything is normal. Could the PLL be an problem? It looks like it can not lock on to the other selected input if no data flows to it.
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 13th, 2016, 3:31 pm

I think if the coaxial input would be given basic 44.100kHz when selected, the problem may disappear. Even though nothing connected to it, just to lock it initially on.
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

Re: γ3 On the Move

Postby amb » March 13th, 2016, 4:13 pm

The volume control and mute are both done on the DAC chips.

It does seem that you have a lock-related problem, but it is not required to have a lock condition at all times. An unconnected input should not cause popping noises. In fact the two bits in the register 0E I mentioned earlier specifically tells the SRC4392 to mute when a loss-of-lock condition exists. You could try changing the the LOL bit to 0 but keep the RXAMLL bit at 1 to see if that makes a difference.
User avatar
amb
Site Admin
 
Posts: 10432
Likes: 176 posts
Liked in: 495 posts
Joined: March 29th, 2010, 9:49 pm
Location: Sunnyvale, CA
Country: United States (us)

Re: γ3 On the Move

Postby cobretti » March 14th, 2016, 11:41 am

OK, I will give it a shot but anyway, I have a feeling that SRC chip itself is somehow damaged. I will order a new one and some chipquick to desolder it and see. If the firmware is good, which it must be if works OK for others, then there is nothing else left but that chip.
User avatar
cobretti
 
Posts: 604
Likes: 18 posts
Liked in: 24 posts
Joined: July 9th, 2011, 5:41 am
Location: Weymouth, Taxachusetts
Country: United States (us)

PreviousNext

Return to γ3 high resolution DAC and γ24 DAC core module

Who is online

Users browsing this forum: No registered users and 0 guests

Quick Links

AMB audio DIY resources
AMB audio shop