Twisted Pear Audio Support
»
Product Support
»
Digital
»
Buffalo DAC
»
Stable I2S connection on BII only after a warm-up ?
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
Hi,
I've a problem with the I2S connection between a DSP and a BuffaloII.
I've found other thread with some suggestion, but my case seems to be different.
The connection between the two device (the Buffalo II and the DSP ) are very short (2cm).
When I first turn on the Buffalo II it loose the lock (green led turn off) for about 1 to 2 second every 5 to 10 second.
This continue for about five minute, and after that the lock became stable.
It seems to me that Buffalo need to be "warmed" up before becoming stable, but this is quite strange....
If I use the SPDIF input, then it work fine from the very beginning.
The setup is as follow:
Buffalo II with placid ps
IVY-III with placid bp ps
|
|
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 2/10/2009(UTC)
Posts: 299
Location: Italy
|
Could it be your DSP that need to be warmed up to deliver a good I2S signal?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
That behavior is "normal".
It is unrelated to the I2S wires
It is related to the DPLL setting used for I2S (but unless you are using an external controlling, you don't have control on the DPLL setting).
My suggestion is to use spdif.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
NicMac wrote:Could it be your DSP that need to be warmed up to deliver a good I2S signal? No, I've already made many test, and it's unrelated to the DSP
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
glt wrote:That behavior is "normal".
It is unrelated to the I2S wires
It is related to the DPLL setting used for I2S (but unless you are using an external controlling, you don't have control on the DPLL setting).
My suggestion is to use spdif. Using SPDIF is not so easy for me my DSP is used as an electronic xover and have only one SPDIF out that can transport only two channel. This is why I need I2S in order to connect two Buffalo II The second Buffalo II is on the way (should be sent this week), but I can't use it if this is the normal behaviour of DPLL setting as you say
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
OK, Then you'll need to build an external controller or wait for the TPA product.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
glt wrote:.....or wait for the TPA product. Any hint on this ? Edited by user Wednesday, March 2, 2011 4:20:19 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
I an send you some alternate firmware if you can program the chip.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
Russ White wrote:I an send you some alternate firmware if you can program the chip. Not sure I have the necessary tool to do that. By the way, Brian will send me the second BII by this week. If you can add two chip with a different firmware I'll pay for them.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/12/2011(UTC)
Posts: 1
Location: Japan
|
Russ White wrote:I an send you some alternate firmware if you can program the chip. Hi, I tried I2S connection 80MHz / 100MHz version BII DAC. Older one(80MHz) is stable, but New one(100MHz) is unstable. So, I tried swapping firmware, 100MHz BII DAC + 80MHz's firmware is stable, and 80MHz BII DAC + 100MHz's firmware is unstable. Did you change the firmware? Can you send alternate firmware?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
Dear Russ,
Since I still haven't solved this problem, today I've try to put a small heatsink on top to the crystal. (100mhz version)
I've done it when the Buffalo was warm and stable.
As soon as I put the heatsink the lock led start flashing every few second.
I know that this is probably a quite stupid experiment, but It seems clear to me that there is something related to temperature that must settle down at startup.
Is something that can be solved with a different firmware?
Thank you in advance for your help
Michele
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Possibly, but you would have to open the bandwidth way up which would be worse for jitter rejection.
What you are describing should only occur if your source has enough jitter to test the DPLL.
Why not just run your DAC warm? That would be better than increasing the DPLL bandwidth. All XOs I have ever seen will settle into their frequency with temperature. Though this is usually just a very slight change from startup condition. The ESS DPLL is quite sensitive to such changes though.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
I agree with you not to open the bandwidth, but what is confusing me is that using SPDIF the lock led is "on" from the very beginning, while using I2S I must wait several minute in order to have a stable lock (and sound....) I ever believed that SPDIF have more jitter than I2S. Am I wrong? Edited by user Thursday, March 17, 2011 8:12:09 AM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
Here is my guess on what is going on: http://hifiduino.wordpre...-on-sabre32-i2s-locking/Also, if you feed higher sample rate through spdif, you will start seeing drops with dpll set at lowest. The "solution" is to set dpll to "best" for all cases. "best" has never been defined in detail.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
The real practical problem here is that running in "best" does in effect open up the bandwidth. But I don't disagree that it is a good solution. Edited by user Thursday, March 17, 2011 3:24:00 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 7/21/2010(UTC)
Posts: 75
Location: Colorado
Thanks: 1 times
|
glt wrote:Here is my guess on what is going on: http://hifiduino.wordpre...-on-sabre32-i2s-locking/Also, if you feed higher sample rate through spdif, you will start seeing drops with dpll set at lowest. The "solution" is to set dpll to "best" for all cases. "best" has never been defined in detail. Hmmm? I am playing 24/192 and 24/176.4 into the B-II with no dropouts whatsoever. The source is a Wavelength Audio Wavelink. I guess this result speaks well for the output jitter levels of the Wavelink, as I am using stock DPLL bandwidth settings from TPA.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
glt wrote:Here is my guess on what is going on: http://hifiduino.wordpre...-on-sabre32-i2s-locking/Also, if you feed higher sample rate through spdif, you will start seeing drops with dpll set at lowest. The "solution" is to set dpll to "best" for all cases. "best" has never been defined in detail. Thank you for the link. I'll try to assembly an arduino controller and use your code to see what happen I've just one more doubt. Is it possible to control two Buffulo II with the same arduino controller? Ps: thanks for sharing your work. Really interesting Edited by user Sunday, March 20, 2011 1:06:31 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 3/1/2011(UTC)
Posts: 23
Location: Italia
|
Indigo wrote:glt wrote:Here is my guess on what is going on: http://hifiduino.wordpre...-on-sabre32-i2s-locking/Also, if you feed higher sample rate through spdif, you will start seeing drops with dpll set at lowest. The "solution" is to set dpll to "best" for all cases. "best" has never been defined in detail. Thank you for the link. I'll try to assembly an arduino controller and use your code to see what happen I've just one more doubt. Is it possible to control two Buffulo II with the same arduino controller? Ps: thanks for sharing your work. Really interesting I finished to put togheter an Arduino controller with HifiDuinoCode vB06e. It is working flawlessly now, and can adjust the bandwidth very easily Now I don't have any unlock, even at very first startup, when the XO is cold... Now, I've only to find out how to control two Buffalo II at the same time with only one Arduino. Any help will be very welcome
|
|
|
|
|
|
Twisted Pear Audio Support
»
Product Support
»
Digital
»
Buffalo DAC
»
Stable I2S connection on BII only after a warm-up ?
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.
