Forum Rule: Always post complete source code & details to reproduce any issue!
Results 1 to 13 of 13

Thread: Audio shield headphone out connected to amplified line-in

  1. #1
    Junior Member
    Join Date
    Mar 2019
    Posts
    11

    Audio shield headphone out connected to consumer stereo amplifier line-in

    Hi;
    I was working through some of the Tutorial material for the Audio shield and Teensy 3.2 over the weekend. Things were going great but I found that the next morning I was not able to get ANY output from the headphone connector using the same tutorial material, even the first Hardware Test tutorial that generates the beep. The shield is not completely dead as I can read the floating input values from the button and pot inputs and talk to the SD card. Tried a couple of different sets of headphones.

    After quite a bit of pondering I got to thinking about the silkscreen note on the board warning not to short the headphone VGND to the board GND (if I'm interpreting that correctly). While working with the board previously I had connected the headphone output to the line in of an AC powered stereo receiver to play output on the speakers. Seemed to work ok but since realized that probably connects the headphone GND back to the board GND via my computer's AC GND and USB cable. I was being lazy and didn't use the line-out on the shield, figuring headphone out should have been ok given the relatively high input impedance of the stereo AMP line-in, but had not thought of the ground connection implications.

    Is connecting the headphones to the line-in on an AC connected stereo amp something that could wipe out the headphone output HW in the SGTL5000 due to the GND path back through the wall outlets and computer's USB, or is that warning just about preventing ground loop noise?
    Thanks for your time.
    -AG6GU
    Last edited by AG6GU; 03-25-2019 at 09:35 PM. Reason: Typo in title

  2. #2
    Junior Member
    Join Date
    Mar 2019
    Posts
    11
    Grr! 8)
    Found this in the data sheet.
    "Do not connect HP_VGND to system ground, even when unused. This is a virtual ground (DC voltage) that should never connect to an actual “0 Volt ground”. Use the widest, shortest trace possible for the HP_VGND."

    I even have the audio isolation transformer but didn't use it a couple of times when making the connection to the AMP. Lesson learned.

  3. #3
    Junior Member
    Join Date
    Mar 2019
    Posts
    11
    Looks like maybe the line-out on shield board may still work since it seems to just reference system GND. Won't be able to try that until tomorrow. Sorry about not 'posting full source code' but I doubt anyone wants to try to reproduce this 8)

  4. #4
    Senior Member+ Theremingenieur's Avatar
    Join Date
    Feb 2014
    Location
    Colmar, France
    Posts
    2,505
    Several people have already gone that path. It seems that fully reading a documentation and data sheets has become unmodern... odds are good that the line out will still work if the SGTL has not heated up too much due to the 1.65V short to GND.

  5. #5
    Junior Member
    Join Date
    Mar 2019
    Posts
    11
    Well thanks for the reply and the insult. I did read the note but it had not occurred to me that I was making that GND connection when I plugged in the AMP.
    I look forward to not posting again 8)

  6. #6
    Senior Member+ Theremingenieur's Avatar
    Join Date
    Feb 2014
    Location
    Colmar, France
    Posts
    2,505
    I didn’t intend to insult you. If it came out this way, please accept my apologies! I know that my non-native English is not up to refined diplomacy, but I do my best to improve.

    The first part of my previous post was not meant as a personal attack but it was just the summary of what I saw happen in that context over the past years. The second part was meant to be encouraging. Again, sorry if that was not clear.

  7. #7
    Junior Member
    Join Date
    Mar 2019
    Posts
    11
    Quote Originally Posted by Theremingenieur View Post
    I didnít intend to insult you. If it came out this way, please accept my apologies! I know that my non-native English is not up to refined diplomacy, but I do my best to improve.

    The first part of my previous post was not meant as a personal attack but it was just the summary of what I saw happen in that context over the past years. The second part was meant to be encouraging. Again, sorry if that was not clear.
    Fair enough and thanks for the reply. I figured I had already admitted to doing something careless and stupid in my original posting so I was a little more sensitive to the wording of your reply than was warranted 8)

  8. #8
    Just to be clear, I should not connect my headphone 1/8-inch output jack on my audio board to an amplifier input, but use the line out pins, right?

  9. #9
    Senior Member+ Theremingenieur's Avatar
    Join Date
    Feb 2014
    Location
    Colmar, France
    Posts
    2,505
    Quote Originally Posted by ofishal View Post
    Just to be clear, I should not connect my headphone 1/8-inch output jack on my audio board to an amplifier input, but use the line out pins, right?
    Yes, because the ground of the headphone output is a virtual ground which has an offset of +1.65V. If you connect it to a “true” GND, these 1.65V will be shorted and the SGL5000 risks to be damaged.

  10. #10
    Thanks. This is good to know. I'm not quite there with my project, but eventually I need output to a powered speaker so I will use the output pins on the audio shield. Thanks again

  11. #11
    Junior Member
    Join Date
    Dec 2016
    Posts
    15
    For what it's worth, I have had good success isolating the headphone out left and right signal through two 220uF capacitors and using system ground for ground reference, leaving the virtual ground floating. This gives a nice clean sounding headphone signal into an amplifier with no fear of shorting the virtual ground. I think the datasheet mentions this connection method somewhere.

  12. #12
    Senior Member+ Theremingenieur's Avatar
    Join Date
    Feb 2014
    Location
    Colmar, France
    Posts
    2,505
    That’s also a possible “solution”, but why add two bulky 220uF capacitors if one can have a clean signal without these directly at the audio out pins?

  13. #13
    Junior Member
    Join Date
    Dec 2016
    Posts
    15
    It allows software control over speaker volume in a small synthesizer I am constructing using the headphone volume settings for the amplified internal speakers. This is while line out remains at a fixed level for recording. Having two semi-independent outputs seemed a good idea. The unit is plenty bulky already with keyboard, knobs, lights and display, lots of space for capacitors, two more make little difference to the dozen or so already present.

    It actually drives the headphone jack directly too, not recommended for line out levels I believe. This is why I used 220uF and not a smaller value which would perhaps be sufficient for driving higher impedances.

    I find the surround sound functions and bass boost work well to help compensate for small speakers placed too close together as well. Are the DSP effects available over line out?

    Edit - On review, I recall I decided to tap the headphone _before_ the 220uF and use the virtual ground on a panel-insulated headphone jack. Inserting phones loads the signal and makes the internal speakers drop in volume, but only by a small amount. None of this should affect the line out though, I hope, they run straight to the back panel jacks.
    Last edited by prickle; 09-07-2019 at 04:08 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •