• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 209 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA1650: Ground / earth / chassis . ref connections

David Koster
Prodigy 30 points
Part Number: INA1650

Tool/software:

We're working on a pro audio device with differential inputs and chose the INA1650 as our line receivers. I followed the example circuit on page 22 of the datasheet called "8.2.1 Line Receiver for Differential Audio Signals in a Split-Supply System"

Now that the boards have come in I've been getting a lot of remarks on the design that I myself don't have any answers to:

  • Shouldn't the 100kΩ resistors have been connected to the system ground, instead of the chassis?
  • Wouldn't a lower input impedance (instead of the 100kΩ per leg) be better, noise wise, and regarding static pops etc. when plugging in cables?
  • Shouldn't R3 and R4 be connected to the system ground, instead of the chassis? I believe this is the case for the evaluation board, though it doesn't have a separate chassis. 

I'm trying to run some tests to see what is 'best', but am struggling to find concise answers to these. I hope someone with more experience would be able to help me out!
Much appreciated, 
David

  • 0
    TI__Mastermind 25210 points

    Hi David, 

    David Koster said:
    Wouldn't a lower input impedance (instead of the 100kΩ per leg) be better, noise wise, and regarding static pops etc. when plugging in cables?

    In the small signal AC analysis you would kill all DC voltages and both sides of the resistors would have zero potential across them and therefore the resistors would fall out of the equation. So lower wouldn't necessarily be better. The 100k resistors are bleeders that discharge any DC that builds up on the AC coupling caps. You honestly never know what the DC output voltages are of whatever is getting plugged in. The 100k resistors can limit how fast current discharges. 

    David Koster said:
    Shouldn't the 100kΩ resistors have been connected to the system ground, instead of the chassis?

    It is best practice as shown below to connect pin 1 to chassis ground. For this reason that is why you see in our design we have it connected to chassis ground.  By going to chassis ground you can be grounded right at the connector. See the image below that follows the AES48 standard specified by Audio Engineering Society. There is an in depth discussion that can be found here:

    https://electronics.stackexchange.com/questions/460889/the-pin-1-problem-in-audio-star-and-mesh-ground-and-contradictory-grouding-pra

    I hope this information helps. 

    Best Regards, 
    Chris Featherstone

  • 0
    TI__Mastermind 25210 points

    Hi David, 

    David Koster said:
    Shouldn't the 100kΩ resistors have been connected to the system ground, instead of the chassis?

    To follow up on my last response. Pin 1 needs to be attached to chassis ground. If the bleeder resistors go to circuit ground that is fine. The chassis and circuit ground do need to be connected together at some point so that they are at the same potential. 

    Best Regards, 

    Chris Featherstone

  • 0 in reply to Chris Featherstone
    Prodigy 30 points

    Hey Chris,

    Thank you for both the responses! We have the Pin 1 to chassis, and the chassis and circuit ground are connected together only inside of our amplifier module. 

    So, if I understand correctly:

    • The 100kΩ resistors (R1, R2, R4, R5) are only there to bleed the DC voltage from the input coupling caps after an input device is unplugged.
    • And thus, since our chassis and circuit ground are connected at out amplifier module, it does not matter if we hook them up to chassis or circuit ground.
    • Does this also go for the 1MΩ resistors in the image? R3 and.. also R4? Those would then also be connected to system ground instead of chassis?

    I made the assumption these resistors were to deliberately lower the input impedance of the input circuit. Would it for any reason make sense to want to do that? 

    Much appreciated!

    With kind regards,

    David 

  • 0 in reply to David Koster
    TI__Mastermind 25210 points

    Hi David,

    A colleague of mine shared this application note from RANE Commercial Audio Products below. There is a lot of great information in the app note on how to connect the Chassis and Signal Grounds. 

    Grounding_&_Shielding_of_Audio_Devices.pdf

    David Koster said:
    The 100kΩ resistors (R1, R2, R4, R5) are only there to bleed the DC voltage from the input coupling caps after an input device is unplugged.

    This is correct. 

    David Koster said:
    And thus, since our chassis and circuit ground are connected at out amplifier module, it does not matter if we hook them up to chassis or circuit ground.

    I believe this logically makes sense since our goal is to have the chassis ground and signal ground at the same potential. 

    David Koster said:
    Does this also go for the 1MΩ resistors in the image? R3 and.. also R4? Those would then also be connected to system ground instead of chassis?

    Sorry I don't see the 1M resistors in the image. Can you point them out to me?

    David Koster said:
    I made the assumption these resistors were to deliberately lower the input impedance of the input circuit. Would it for any reason make sense to want to do that? 

    If the input impedance is too low you may attenuate the signal coming from the input. For this reason we try to keep the input impedance high. 

    I hope this information is helpful. Let me know if I can be of further assistance. 

    Best Regards, 

    Chris Featherstone

  • 0 in reply to Chris Featherstone
    Prodigy 30 points

    Hey Chris,

    Great application note! 

    Here are the 1M resistors pointed out:

    Those would move together with the 100kΩ resistors then right? To either signal GND or chassis? 

    Looking forward to hear,

    With kind regards,

    David 

  • 0 in reply to David Koster
    TI__Mastermind 25210 points

    Hi David, 

    Having the Rcom = R3 = R4 at 1M improves the CMRR. I think it makes more sense to connect the Rcom to signal ground. But either way the system and chassis grounds will need to be at the same potential at some point so either will technically work. Here is a snippet discussing that. 

    Best Regards, 
    Chris Featherstone