audio shield anomaly

Status
Not open for further replies.
ETMoody3

ETMoody3

Well-known member
Hey there. I have been building a project around the teensy 3.1 and the audio shield. In the process of increasing the complexity of the system the audio shield stopped sending any audio to the line out jack but still sends generated tones out of the headphone jack. Additionally, line input seems to have stopped working. However...when the unit is not powered the signal from line in to line out is passed through. Powering the system cuts off the signal. It is noteworthy that this happens regardless of code or teensy used as I have a second uC handy and have tried various example programs on both processors in addition to my own code. I'm honestly unsure how this may have happened as I have checked my circuit for problems and found none. My first conclusion is that I have damaged the chip on the audio shield. Posting here due to puzzling pass-thru function. Something still works... thoughts? The behavior is the best description of the apparent problem I can offer. I am already planning to replace the unit, any information on this will help me decide what happens to the old one.
 
Last edited:
Yes. I'm sorry I didn't include any code here because I tried the shield with all of the examples and many variations of code and the output behavior was consistent as described. I even tried "toggling" that function(method?) just to see if it had any effect...
 
If I did this right I attached the last iteration of code I attempted. This was working...and didnt sound too godawful...:)
 

Attachments

  • fxboxone.ino
    6.6 KB · Views: 143
I think the only problem would be shorting VGND and GND (it says on the silkscreen not to do that).

teensy3_audio_back.jpg
 
Yeah. I didn't do that... I only attached wires to the line in/ line out areas. It was working quite nicely for a while. Too many possibilities including handling. For what it is worth I observed very good audio from the line out, and acceptable levels of heterodyning with all of the redundant signal mixing I was doing there just testing out potential. I'm a musician and have pretty good gear to interface this board with. I'm impressed.
 
Before we started shipping these, I tried shorting VGND and GND, and also the 2 output lines to GND and VGND, just to see what would happen. The chip does limit the current and things do recover, and the MK20 didn't seem to crash or sustain any damage (the supposed current limit is a little over the MK20's spec).

Of course, DO NOT actually try shorting VGND and GND, since it stresses the chips quite a lot. But I can tell you I did try it here on the prototype and everything survived.

As for the original question, I have no idea why the line in & out aren't working.
 
ETMoody3 said:
However...when the unit is not powered the signal from line in to line out is passed through. Powering the system cuts off the signal.
Click to expand...
I don't think the behavior of the chip is specified when it is powered off. I don't think the passthrough necessarily means much- there will be a signal path from input to output through the ESD protection diodes, for example, when the chip is powered off so that Vdd = GND.
 
Well, just to be thorough I just tried it again and I get a high frequency sqeal from left and 60 cycle hum from right. Fubar. Ordering two replacements ;)

Which is to say that if I were hiring engineers like y'all it would be costing a bit more than 30 bux. I should be donating, but suffice it to say I have a new favorite microcontroller dev board. Thanks beyond words for this work on this audio library. When I get good enough handle on what I need you can expect some musician-relevant requests.
 
Last edited:
Status
Not open for further replies.
Back
Top