Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Hi all, I am planning to use my BIII (8 channel mode, unbalanced) in voltage out mode for some testing (I might use an I/V stage later) and was wondering about the following things: - Do I need a passive filter on the outputs ? If so what would you suggest on values - does the input impendance of the amps come to play here ? - Should I implement some kind of isolation on the outputs in order to prevent damage to the dac, for instance when shoring the output wiring accidentally - would the resistance in series of the filter mentionned above have that function/ suffice ? - Should I expect a serious performance degradation usinhg a "simple" LT1764 regulator circuit ? Thanks in advance, Max Edited by user Wednesday, January 15, 2014 5:12:58 PM(UTC)
| Reason: Modifying subject for troubleshooting
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
Ideally you would have a low-pass filter. It is not strictly required, but it is a good idea. An active filter would be better.
You do not need to isolate the outputs, as they can tolerate shorts (essentially, this is current-mode).
Where are you using this regulator? As the pre-regulator, or in place of the local regs (Tridents)?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Dear All,
I asked a question over at the DiyAudio forum and Russ kindly invited me to post the problem here for easier troubleshooting.
First things first, here is the actual setup:
MiniDSP Mini-Sharc (including the digital input-board) and VolumeFP control board -> TPA III (8 channel I2S lines, 5 cm flat cable directly soldered to Minisharc using the connector on the TPAIII board)
As mentionned in the post above, in order to do some testing I simply hooked up a 1K resistor in series with the positive outputs and a 3300pF cap in parrallel of Ground and positive output as Low-Pass filter.
I have omitted a DC blocking cap as I do have one in the amps.
The Main board is powered by a LT1963A regulator using the application note schematics and putting out 5.23V (both load/idle) feeded from a 12V SLA battery.
The Main board has the reference V 1.01, the AVCC is ref. 1.0 and the tridents are V 3.0 with OPA2209A.
Everything is working as it should (Signal on all 8 channels) but a high pitch whining noise on the output (all channels).
We have tried several DIP-switch settings (as DPL settings, etc.) but nothings helps.
Right now all settings are OFF, only SP/DIF bypass is ON (as per instructions for the new firmware).
When unhooking the I2S lines the noise has a stable frequency and level, when connecting I2S on the fly is shows frequency variations and an overall lower frequency, not by much though.
I haven't looked into it with a scope, but could do so with some assistance.
If I need to tackle the boards with the soldering iron, no problem with SMD parts but for the DAC chip itself maybe.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Originally Posted by: Brian Donegan  Ideally you would have a low-pass filter. It is not strictly required, but it is a good idea. An active filter would be better.
You do not need to isolate the outputs, as they can tolerate shorts (essentially, this is current-mode).
Where are you using this regulator? As the pre-regulator, or in place of the local regs (Tridents)?
Thanks Brian for chiming in, I was posting when you did.... the LP is fitted (correctly I think). THe LT1963A is the pre-regulator for the board, the tridents are in place. Cheers, Max
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
If it's a pre-reg, you should not have any degradation, as the local Trident regulators will dominate the performance. I use LCDPS (LM317-based) with mine.
If you have the latest firmware, the default position for all the switches is ON (the way they ship from the factory, and the way they arrive from me). That may be your problem...
Aside from that, can we see a picture of your setup? Typically I2S wiring can be noise prone, unless done right.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Brian, here are some pics of the I2S wiring as requested, sorry for the noisy images. With the DIP switches set per factory default (i.e. all ON) I only get mono-sound due to the remap parameters. Any thoughts ? Max EDIT: As the noise is present with our without the I2S connections I ruled I2S connections out actually... Edited by user Tuesday, January 14, 2014 3:46:18 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 know you said your amp has caps, but just for a test did you try adding them to the output of the DAC?
Also I only see output wires on one side of the DAC... Is this why you see "mono" sound?
|
|
|
|
|
|
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)
|
Also in your config you could actually use the DAC with no firmware at all.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Thanks Russ,
If I put a cap in series with the dac output, i don't have any sound...and no noise....
Of course the output are correctly wired both sides. I just disconnect wires for the picture.
What do you mean by not using any firmware ?
|
|
|
|
|
|
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)
|
You can take out the controller. (the DIP8 in the socket) Since in your situation(8 channel I2S input) the DAC default register settings will work just fine.
In any case - the fact that the DAC is now silent with the caps in place is actually a good sign. And it means your amp actually probably can't handle DC on the inputs. So I would leave them for the remainder of the testing.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Russ,
thanks a lot for your hints.
Taking the Controller out doesn't make any difference! DO I understand you right that taking the controller out will disable all DIP switches as we are running on the DAC default register ?
Still the whining sound on the output.
When connecting a 4.7uF MKP in series with the output the noise goes away, but the output is muted as well (no sound).
Hints ?
Cheers,
Max
|
|
|
|
|
|
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 controller is what reads the switches, so they have no effect at all when it is not present. With no controller the DAC is operating in it's default mode, which is perfectly fine for I2S input.
It's time to start checking your I2S signal. :)
Leave the series caps, the high pitched noise you are hearing without them is likely your amp not being happy. The fact that you don't hear anything with the caps present just means that the DAC is mute, or possibly that the signal is carrying silence.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Problem solved! Actually nothing wrong with our implementation, just ripple on the AVCC as described by many users (including the solution) here: Sawtooth-ripple-on-AVCC-moduleWe have done a quick and dirty job with short trimmed through hole ceramics (100pF) on top of the PCB - result: working like a charm and dead silent! OPA 2209 and OPA2211 are on their way and will probably replace the LMP next week! Big thanks to DIYAudio-member Zaireeka for pointing us straight to the solution! Edited by user Wednesday, January 15, 2014 4:00:54 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)
|
Ah yes, you have the older AVCC module. I should have picked up on that. I thought that because you had the new tridents you also had the new AVCC module. Occasionally the old one could indeed exhibit the oscillation behavior you saw. Glad someone showed you how easy it is to resolve. All the best! Glad it is sorted. Edited by user Wednesday, January 15, 2014 6:35:25 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/27/2013(UTC)
Posts: 10
Thanks: 1 times
|
Originally Posted by: coolnose Dear All,
I asked a question over at the DiyAudio forum and Russ kindly invited me to post the problem here for easier troubleshooting.
First things first, here is the actual setup:
MiniDSP Mini-Sharc (including the digital input-board) and VolumeFP control board -> TPA III (8 channel I2S lines, 5 cm flat cable directly soldered to Minisharc using the connector on the TPAIII board)
As mentionned in the post above, in order to do some testing I simply hooked up a 1K resistor in series with the positive outputs and a 3300pF cap in parrallel of Ground and positive output as Low-Pass filter.
I have omitted a DC blocking cap as I do have one in the amps.
The Main board is powered by a LT1963A regulator using the application note schematics and putting out 5.23V (both load/idle) feeded from a 12V SLA battery.
The Main board has the reference V 1.01, the AVCC is ref. 1.0 and the tridents are V 3.0 with OPA2209A.
Everything is working as it should (Signal on all 8 channels) but a high pitch whining noise on the output (all channels).
We have tried several DIP-switch settings (as DPL settings, etc.) but nothings helps.
Right now all settings are OFF, only SP/DIF bypass is ON (as per instructions for the new firmware).
When unhooking the I2S lines the noise has a stable frequency and level, when connecting I2S on the fly is shows frequency variations and an overall lower frequency, not by much though.
I haven't looked into it with a scope, but could do so with some assistance.
If I need to tackle the boards with the soldering iron, no problem with SMD parts but for the DAC chip itself maybe. Russ, I mentionned it briefly, but at that time was not aware how crucial the info could be. I would have been more explicit in that case, especially if I had been aware of the issues with AVCC 1.0. Thanks all for your help! Cheers, Max
|
|
|
|
|
|
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)
|
Your welcome, once again I apologize I missed that detail in your post. Happy listening!
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
It is a tough case to catch. There are several hundred AVCC 1.0 in service, and only a handful exhibiting this problem. Glad you got it sorted.
|
|
|
|
|
|
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.
