Teensy 4.1 + Genuine Audio Shield Rev D - SGTL5000 enable) fails, no devices found

Hi everyone,

I'm hoping someone can help me diagnose an issue with a Teensy 4.1 and a genuine PJRC Audio Adapter Rev D.

Hardware:

Teensy 4.1
Genuine PJRC Audio Adapter Rev D (DigiKey part 1568-15845-ND)
Arduino IDE 2.x
Teensy package 1.62
Audio shield connected via jumper wires (not stacked directly onto the Teensy)

Wiring:

Audio Shield → Teensy

3.3V (main header) → 3V
GND (main header) → GND
Additional GND (near pin 0) → Teensy GND

MCLK → 23
BCLK → 21
LRCLK → 20
DIN → 7
SDA → 18
SCL → 19

What works:

Teensy uploads sketches correctly (Blink works).
Teensy Audio library is installed correctly.
Audio examples compile.
USB Serial works.
Audio test sketch runs.
LED blink confirms the sketch is executing.
The problem
Using the standard SGTL5000 test code:

"if (audioShield.enable()) {
Serial.println("SGTL5000 enabled OK");
} else {
Serial.println("SGTL5000 enable FAILED");
}"

I always get:

"Starting audio test...
SGTL5000 enable FAILED
Tone should be playing now"

No audio is produced.


I have also run a standard I²C scanner.
Result:

=================================
I2C Scanner
=================================
No I2C devices found.


I also tried scanning Wire, Wire1 and Wire2 with the same result.

Other tests done:

3.3V measured at the Audio Shield with a multimeter.
Continuity confirmed from Teensy to Audio Shield for:
3.3V
GND
SDA
SCL
...SDA and SCL are not shorted together.
Tried swapping SDA/SCL (just to rule it out).
Added external 10k pull-up resistors from SDA and SCL to 3.3V.
Tried powering from both the auxiliary 3.3V/GND pads and the main header 3.3V/GND pads.
Added an additional ground connection between the shield GND near pin 0 and the Teensy GND.
Inspected solder joints and checked continuity after soldering.
Despite all of this, the scanner still reports no I²C devices....


At this point I'm unsure whether I've overlooked something fundamental with the wiring,
there's something specific about using the Rev D shield with jumper wires rather than stacking,
or whether the Audio Shield itself could be faulty.
I've attached photos of the wiring from and back.
Any suggestions would be greatly appreciated.

Thanks!
 

Attachments

  • 1000047887_forum.jpg
    1000047887_forum.jpg
    172 KB · Views: 19
  • 1000047886_forum.jpg
    1000047886_forum.jpg
    195.6 KB · Views: 19
Looks like you've wired it as if it was a Rev C. See the table here.
The two I2C pins might work better with 3k3 resistors.

Pete
 
Thanks Pete — I should clarify: DIN only looks like it goes to 22 in the photo. It actually runs in between the pins and underneath the Teensy into pin 7. Current wiring is MCLK 23, BCLK 21, LRCLK 20, DIN 7, SDA 18, SCL 19, 3.3V to 3V, and GND to GND. I’ll try 3k3 pull-ups on SDA/SCL next. Gonna have to order and though, hence only trying the 10k so far. I kinda convinced myself that the board already had pull ups tbh, but maybe it works different with the flying wires rather than the stack?
 
You might check which ground you're connected to...only one of the ground pins is actually used/connected on the Audio Adapter.

Hope that helps...

Mark J Culross
KD5RXT
 
"No audio is produced."
How are you connecting the output to hear the audio? The photo's don't show connection to headphones and there are no soldered wires to the line outputs (that I can see).

BTW, the headphone connector is only meant for connection to headphones, not an audio cable to an amplifier nor powered speakers. Be sure to read the "Caution" text on bottom of pcb about VGND.
 
If you happen to have six 10k resistors, you could put three in parallel on each of SDA and SCL.

Pete
 
If you happen to have six 10k resistors, you could put three in parallel on each of SDA and SCL.

Pete
Hi Pete, figured as much but for the sake of a few quid I've ordered some anyway, they'll arrive today so will post an update then. Thanks
 
Last edited:
"No audio is produced."
How are you connecting the output to hear the audio? The photo's don't show connection to headphones and there are no soldered wires to the line outputs (that I can see).

BTW, the headphone connector is only meant for connection to headphones, not an audio cable to an amplifier nor powered speakers. Be sure to read the "Caution" text on bottom of pcb about VGND.
I've been using headphones, I have just unplugged them at this stage. Interesting that it won't work for speakers though, I did initially try some desktop speakers but soon switched to testing with headphones as figured they'd be more reliable. Thanks
 
The audio shield already has 2.2K pullup resistors on SDA and SCL (pins 18 and 19). Adding more pullup resistors on SDA and SCL won't fix this.

Check your connections for LRCLK and DIN (which connect to Teensy pins 20 and 7). You can't get audio output without those 2 signals. DIN is the actual audio data, and LRCLK is needed to tell which bits are the left vs right sample and where each begins within the bit stream.

In your photo, no wires are soldered to those locations!

1783239447755.png


Here's a copy of the website info (scroll to "Signals to Teensy"), rotated and flipped so the orientation is same as the photo:

1783239457775.png

Many pins are options.

Pins 10-13 are not needed for audio output. They're only used for the SD card.

Pin 8 (DOUT) is only needed if you want the audio shield to be send audio to Teensy.

Pin 15 (Vol) is only used if you add a trim pot.

Pin 6 (MEMCS) is only used if you add a memory chip.
 
Last edited:
Because you're not using the SD card, for the sake of testing I'd recommend running File > Examples > Audio > Synthesis > Guitar.

This program plays a continuous sound which only uses I2S output. A non-stop output is easier than using beeps or other tests that require you to do something each time you want to know if you're new connections work.
 
Hi Paul, thanks so much for this.

I think here lies the issue, that pinout is completely different to the pinout I've been using as a reference that I got from Google (see attached).

I'll try again using the correct pin out.. hopefully that'll do it
 

Attachments

  • Screenshot_20260705-095429.png
    Screenshot_20260705-095429.png
    635.1 KB · Views: 11
Having looked at the link I now see it, I have indeed wired it up as if it were a RevC but to the RevD positions on the Teensy... Seems so obvious now, lesson learnt, always check the source rather than just trusting Google!
 
Back
Top