Teensy Audio Shield Rev D2 dead?

B

boorch

Member
Hello,

When I was working on a sketch, experimenting with sgtl5000's compression values, suddenly all signs of life from either the input or output went dead. My Teensy 4.0 + Audio Shield is setup according to the Audio Tutorial.

Things that I've done so far to pinpoint the problem (with no real success):
- Ran Audio > Tutorial > Hardware test: All pots and buttons work properly with stable values printed.
- Ran various Hardware tests such as PassThroughStereo and PeakMeterStereo: No signs of life on the inputs/outputs.
- Ran Audio > Tutorial > PlayMusic: No sound (on both HP and LO)
- Modified PlayMusic script with the contents of PeakMeterStereo so I can monitor if any WAV is played off of SDCard at all (by wiring up a cord between playSd and peaks): i can monitor peaks of the WAV file, so AudioShield does not seem to be fully broken. No sound on any outputs tho.
- Went through all solder joints on Teensy AND Audio Shield with my Voltmeter, no shorts or loose solder.
- An interesting find: If I connect my Digitakt to Line-ins and crank the volume to the max, i can very very faintly hear what's coming from line-in through headphones. This is not dependent on the sketch loaded. Happens even with LED Blink sketch.
- Power cycled dozens of times (in hopes of resetting sgtl5000): Nothing changed.

I'm lost. Anyone have other ideas to try?

Cheers!
 
Last edited:
UPDATE: I randomly tried running the example Audio > Recorder sketch, I was able to successfully record from line in and play from headphone out (I didn't test lineout but I assume it would be working as well).

Following this, I tried running various sketches from PassThroughStereo to AudioEffects and all seems to work fine.

I can't come up with a logical explanation how it's been "revived" but I wonder if somehow something "reset" the sgtl5000. I'm up for any ideas/advice regarding what would be causing this because it seems super random and I'm sure I'll stumble upon this again.
 
After working flawlessly for hours, I tried to ground the mic input because I'm only using Line In and I've seen this advised to reduce line in<->mic crosstalk and get rid of the hum/noise (it's very prominent since I'm working on a stereo compressor sketch). Now it's dead again.
 
Last edited:
boorch said:
After working flawlessly for hours, I tried to ground the mic input because I'm only using Line In and I've seen this advised to reduce line in<->mic crosstalk and get rid of the hum/noise (it's very prominent since I'm working on a stereo compressor sketch). Now it's dead again.
Click to expand...

The mic input on the audio board has a 2.2K resistor to the microphone bias output of the SGTL500 codec. Perhaps the codec chip is objecting to excess current draw on that microphone bias output. The small electret microphones normally used with that input act like capacitors and pass very little or no DC current. Instead of a DC connection to ground, you might try a large capacitor across the input.
 
boorch said:
I can't come up with a logical explanation how it's been "revived"
....
Now it's dead again.
Click to expand...

Any chance one of the pins between Teensy and the Audio Shield isn't soldered well? Maybe the metal surfaces are just sitting next to each other and touch or stay separated by a microscopic gap depending on how you've recently touched the boards?


but I wonder if somehow something "reset" the sgtl5000.
Click to expand...

SGTL5000 doesn't have flash or other non-volatile memory, so power cycling should always fully return it to the original state.
 
You must log in or register to reply here.
Back
Top