APx555: Should I or should I not?

Discussion in 'Tales from the Bully Pulpit' started by purr1n, May 17, 2021.

  1. Tachikoma

    Tachikoma Facebook Friend

    Contributor
    Joined:
    Oct 4, 2015
    Likes Received:
    198
    Trophy Points:
    43
    Maybe you could cheat by injecting spikes into the signal that can be used as references to sync the recordings?
     
  2. Earbuds Hub

    Earbuds Hub Rando

    Contributor
    Joined:
    Wednesday
    Likes Received:
    2
    Trophy Points:
    3
    Location:
    San Jose
    Home Page:
    Made a small donation.
     
    • Like / Agree Like / Agree x 2
    • List
  3. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    15,874
    Trophy Points:
    113
    Location:
    On planet
    Consider the number of combinations necessary to answer this question thoroughly:

    1. output impedances for various DACs

    2. input impedances for various Amps

    3. Type of Amp input
    a) potentiometer or stepped attenuator
    b) transformer
    c) direct input to a gain stage with level controlled by feedback

    4. For resistor based inputs what is the next stage after the potentiometer output
    a) grid of a tube
    b) fet input
    c) bjt input
    d) transformer
    e) op-amp with further subcategories of fet of bjt inputs

    5. What position is the volume control in when unit is off and system tested
    a) at max volume the input impedance will be dependent on 4 a through e above
    b) at min volume the input impedance is the maximum value of the potentiometer
    c) at any other volume the input impedance is some combination of a and b.

    While the above is a comprehensive list, it may not completely describe the universe of possibilities.

    Several examples:
    DAC 600R output impedance
    Amp 100K pot input with output of pot to tube grid
    In this case the amp turned off at worst will be a 100K load to the DAC and probably not cause any degradation of consequence.

    DAC 900R output impedance
    Amp 10K pot input with output of pot to a bjt gain stage that represents a near short circuit when amp is unpowered
    At min volume there is only a 10K load on a 900R output impedance. No significant distortion.
    At max volume there is nearly a short circuit on the 900R output impedance. Very significant distortion will occur.

    This is why it is difficult to answer the question in a generalized way and explains @schiit response to @rhythmdevils .
    Typical Real Engineer(tm) will assume worst case with a short circuit presented to the DAC output when the Amp is unpowered.
    The DAC output impedance protects the DAC from damage or destruction in such case but sound quality to a second Amp is not guaranteed to be any good.
     
    • Like / Agree Like / Agree x 5
    • Epic Epic x 3
    • List
    Last edited: Jun 10, 2021 at 9:39 AM
  4. bboris77

    bboris77 Friend

    Friend
    Joined:
    Dec 12, 2015
    Likes Received:
    727
    Trophy Points:
    93
    Agreed...way too many variables to make any kind of general conclusion. The only way would be to measure specific scenarios involving the most popular DACs and headphone/speaker amplifiers.

    A side question, would inserting an RCA switch in place of the Y-splitter resolve all of these potential issues? I am asking because most switchers do not act as a direct connection between the DAC and the AMP - the ones with shared ground have a significant degree of crosstalk between various inputs, meaning that they are not electrically isolated from each other.
     
    • Like / Agree Like / Agree x 1
    • List
  5. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Friend MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    11,887
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    There is this software which attempts to correct for timing drift, as well as automatically match, align, and correct for other factors: https://deltaw.org/

    Haven't used it myself, but have wanted to for a while. No idea how accurate or insightful it really is, but it's a cool idea nonetheless. Thankfully, the developer has disclaimers all over the place about it still being a work in progress.
     
  6. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    15,874
    Trophy Points:
    113
    Location:
    On planet
    Here is a measurement of a Goldpoint SA2 used as an output switch. The attenuator has been set to maximum such that it only represents an additional 25K load to the DAC output. Putting a 10K load on the unselected output here is a sweep of leakage vs. frequency:
    20210610 SA2 unselected output leakage with 10K load.png
    At 1 KHz the leakage is 77 dB down.
     
    • Like / Agree Like / Agree x 1
    • List
  7. dasman66

    dasman66 Self proclaimed lazy ass - friend

    Friend
    Joined:
    Mar 13, 2018
    Likes Received:
    1,078
    Trophy Points:
    93
    Location:
    NW Pennsylvania
    For us non-EE's... can we assume this represents a good result?
     
    • Like / Agree Like / Agree x 1
    • List
  8. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    15,874
    Trophy Points:
    113
    Location:
    On planet
    It is a decent, predictable behavior. If this were a preamp input with two active sources, one playing 0 dBu sines at 100, 1000, and 10000 Hz and the other input had music at 0 dBu, the leakage from the source playing sines would contaminate the music source by -97 dB at 100 Hz, -77 dB at 1000 Hz and -57 dB at 10KHz.

    Goldpoint SA2 leakage between inputs / outputs
    00 SA2 unselected output leakage with 10K load + selected output - annotated.png
    SA2 is very predictable

    Little Bear MC102 leakage between inputs / outputs
    02 MC102 unselected output leakage with 10K load + selected output - less clutter.png
    The MC102 is a bit erratic between 100 Hz and 3 KHz.

    Eddie Current Studio B leakage between inputs
    03 Studio B unselected input leakage + selected input - less clutter.png
    Studio B is much better behaved
     
    • Like / Agree Like / Agree x 4
    • List

Share This Page