| View previous topic :: View next topic |
| Author |
Message |
alangarf
Joined: 24 Jul 2009
Posts: 3
|
| DSP Analog board AIC23B CODEC only outputting random noise |
 Posted: Sun Jul 26, 2009 10:53 pm |
 |
|
Hi all,
I'm hoping someone can help me. I've got a newly purchased CCS DSP Analog development board that seems to have a defunct AIC23B CODEC. The examples that came along with it don't compile due to version difference in the compiler (and a few obvious errors) and so I've had to re-work most of them to make them work.
But... once I started trying to use the CODEC as soon as the DACs in the AIC23B are enabled via SPI I'm greeted with a bucket load of random noise. Even when the dsPIC isn't clocking any data to the CODEC. Is this normal? I've also tried to turn on the AIC23B's audio bypass and indeed the audio is bypassed, but the roar of noise is also still there in the output as well.
Anyone have any ideas? I think the CODEC is dead...
Also does anyone have a good working example? The documentation is a little thin and the DCI functions in PCD seem mysterious and broken.
Help would be greatly appreciated!
Many thanks,
Alan. |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
|
alangarf
Joined: 24 Jul 2009
Posts: 3
|
|
| Posted: Mon Jul 27, 2009 5:27 pm |
|
|
Have done, but as yet no response. It seems like it's a hardware problem in my case because the noise continues even if I turn on the analog bypass which would make any "digital" section of the CODEC disabled and unable to inject any noise due to incorrect driving by the software in the dsPIC. It's a weird one!
Thanks,
Alan. |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Mon Jul 27, 2009 5:42 pm |
|
|
CCS doesn't post their development board manuals online, so I can't
look at the schematic. I've used the TLV320AIC23B on another (non-PIC)
board. I have a 3.5mm stereo audio connector that feeds the input
circuit to the Line In pins on the TLV chip. The connector has a spring-
loaded "shunt" inside it. When there's nothing plugged into it, the inputs
are grounded. This should mean there is no noise (coming from that
path) to be amplified. |
|
|
alangarf
Joined: 24 Jul 2009
Posts: 3
|
|
| Posted: Mon Jul 27, 2009 5:55 pm |
|
|
| PCM programmer wrote: | I have a 3.5mm stereo audio connector that feeds the input
circuit to the Line In pins on the TLV chip. The connector has a spring-
loaded "shunt" inside it. When there's nothing plugged into it, the inputs
are grounded. This should mean there is no noise (coming from that
path) to be amplified. |
It's similar, it's got a 3.5mm socket for the Mic, and Line-In, and a socket for Line-Out and an amplified Speaker-Out. The Speaker-out is amplified via a couple of LM386's. The inputs don't look like they have a shunt and the schematic doesn't show as much either.
When I feed a signal in the Line-In and have headphones connected to the Speaker-Out, when I enable the TLV via the SPI control interface the TLV starts outputting random noise. I've stuck a CRO across the SDIN and SDOUT, and I can see SDIN is stable from the dsPIC and the SDOUT is going crazy outputting really scattered IO nothing you'd expect from a serial interface. When the TLV is in master mode it doesn't output a BCLK or frame signal either.
Yet the SPI interface seems to work perfectly....
If I instruct the TLV via SPI to enable it's analog bypass the line-in should be directed directly to the line-out and speaker-out and the ADCs and DACs should be removed from the signal path. The line-in signal does indeed end up on the line-out and speaker-out but so does the noise as well.
It's looking like a dead TLV to me. Unless I'm completely silly and missed something trivial! 
Thanks,
Alan. |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
