How to Build a Cheap and Easy Measurement Rig

Discussion in 'Measurement Setups, Systems, and Standards' started by OJneg, Feb 28, 2016.

  1. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    I'm doing this mostly to absolve myself of all the people who are asking if I can build/sell them a measurement rig. I believe this project is the cheapest and easiest way to get reliable and precise measurements out of a headphone.

    BTW I reserve all rights to turn this into a commercial project and sell it for $199.

    What you need for this project:


    Assembly:

    Start with the soldering. You should put together the circuit based on this schematic.

    [​IMG]

    Example layout

    [​IMG]


    This is basically the same cascode circuit used in John Conover's writeup (Figure XXXVIII) on the WM61. Advantages are:
    • Stupid simple
    • Low part count
    • No coupling caps to limit LF response
    • Extra gain takes load off the preamp
    • Low distortion
    Disadvantages mainly include single-ended wiring, so don't run long cables to your input. I've never had a problem with a ~1m cable run into my pre FWIW.

    [​IMG]

    I place our transistor near the end so I can bend it off the board and our whole profile remains relatively flat.

    [​IMG]

    Securing the board can be easily done with tape. I do some crossbracing to keep it tight. The pre-soldered mic leads should be just about perfect to reach from the edge through the foam. If your foam is solid, just make a small slice with a thin blade in the center and slide the mic on through. Should look like this from the front:

    [​IMG]

    Here's what it'll look like in action with a creatology square:

    [​IMG]


    During usage you can lie it flat on a desk, or stack in sideways with some books or foam. Feel free to use different square materials for coupling, you'll get varying results therein. You can also use certain techniques to break the seal. It's still all a matter of looking at the results and interpreting based on what looks most indicative of actual performance.
     
    Last edited: Aug 15, 2016
  2. Serious

    Serious Inquisitive Frequency Response Plot

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,224
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    A big advantage that you get with the WM-61A is that you can also do measurements with the mic in your ear. These have proven quite useful for me to see the effect of various materials on the ring around the driver in the HD800. With a normal coupler the differences were usually pretty small and sometimes didn't quite match what I heard.

    I am using this as a phantom power supply. This way you can just connect it to the sound card in your pc.
    I get a lot of 2nd order distortion above around 90db, I'm probably going to change it to something more complicated later on.

    I feel what we need is a standard coupler that works well with most headphones. I also don't feel that the Marv v2 measurement rig (C.U.N.T.) works very well with the HD800 and it frequently needs felt to break the seal or else we get too much bass on the measurements.
     
    philipmorgan likes this.
  3. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    I know how to solder. I do not know how to read diagrams. Can someone make a diagram that is more similar to the meathead-level guitar pickup wiring diagrams I am used to? :p

    Also, does anyone know of good resources to make in-ear mics to measure while the headphones are on your head?
     
  4. Serious

    Serious Inquisitive Frequency Response Plot

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,224
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
  5. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    That's very good! Only thing is, I have a full recording interface with XLR ins and phantom power... So I just need to know how to hook it up to an XLR! I could probably guess my way to it though.
     
  6. DANIU

    DANIU Taking a week off for being a dumbass - Rando

    Joined:
    Apr 29, 2016
    Likes Received:
    0
    Dislikes Received:
    12
    Trophy Points:
    1
    I would suggest a 78L05 to provide the regulated power supply.

    Or a blue LED instead to lower noise from a Zener Diode.


    people use Zener Diodes as Noise source, FYI.


    ref:

    http://www.ti.com/lit/ds/symlink/lm78l05.pdf
     
    Last edited: Apr 29, 2016
  7. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    How would you propose to use a regulator in this circuit? Remember the phantom power is both the output and our supply.

    LED would be better for lower dynamic impedance, but a single one wouldn't do. After the junction drop, that would leave less than a 1V on the drain of the mic's internal JFET. I'd like to keep it closer to 5V for maximum linearity and gain. Regardless a chain of LEDs might be preferable although we might loose our dynamic impedance / noise advantage. Can't be bothered to do the math ATM.

    Datasheet of the 2SK3372 which is purported to be internal to the WM61:

    http://media.digikey.com/pdf/Data Sheets/Panasonic Semiconductors ICs PDFs/2SK3372.pdf
     
  8. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    I made an in-ear mic with an electret condenser capsule from an AVR "tuner mic". I simply wired it directly to an XLR, and then I plugged it into a proper interface and applied 48v phantom power. It totally gives signal, but the high end bias is STRONG. The signal starts to blend into the noise in the sub bass region. Could this be a specific wiring mistake, like maybe I need to add more components, or maybe the cable is just way too long? (Although I thought that would eat the highs more). It could easily just be the capsule being bad, but it IS made to do EQ calibration to an AVR, so I'd assume it should be able to hear low frequencies louder than the self-noise?
     
  9. zerodeefex

    zerodeefex Grumpiest admin

    Staff Member Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    7,679
    Dislikes Received:
    32
    Trophy Points:
    113
    @OJneg built me one of these and it is fantastic for learning about measurements. I haven't had time to dive deeply into the measurements world but this thing really works well with my little focusrite.
     
  10. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    Assuming the receiver mic is more or less equivalent (electret with internal JFET) to the WM61, I would guess that plugging it directly into a phantom line is overbiasing and perhaps causing clipping. There's a reason I put together this simple circuit. Mind sharing how exactly you wired it up?
     
  11. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    I simply kept the same wiring as was in the microphone already, which was + going to hot and - going to ground. I just replaced the mono jack with an XLR, wired like any other mic. The actual microphone had no internal components. Just wires going to the jack. If there were other components, they are on the receiving end of the AVR system instead of inside the mic.

    Overbiasing is a thing then? I had no idea. I guess maybe I should add something that can "eat" some of that juice and reduce the voltage?
     
  12. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    If someone could draw a literal diagram that shows the actual components, with the name of each part, I could do it in seconds. I never learned to read electrical symbols like that. I just know how to solder things.

    Edit: okay, after studying the symbols online and reading a little, I am starting to see how this works. I still don't know WHAT these all do, but I get what the layout should be! For the XLR connections, I see that pin 3 is on the top, and pin 1 is the lower right. But what goes to pin 2? The black wire?
     
    Last edited: May 1, 2016
    JBoogie likes this.
  13. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    2 pin is left floating. As I mentioned, it's single ended in this case

    What is important is that you connect in a way that the capsule is biased correctly.
     
  14. MF_Kitten

    MF_Kitten Almost "Made"

    Joined:
    Jan 1, 2016
    Likes Received:
    123
    Dislikes Received:
    4
    Trophy Points:
    43
    Are the components you used specific to that capsule, or can I just copy that?

    Edit: I just remembered how thesemic capsules function, and realized what you meant by biasing it. Still don't get what each component is doing, but I'm starting to see what's happening!
     
  15. JBoogie

    JBoogie Rando

    Joined:
    May 31, 2016
    Likes Received:
    11
    Dislikes Received:
    0
    Trophy Points:
    3
    In the first picture of the assembled board there looks to be 2 different resistors but the parts list only mentions 1, am I missing something?
     
  16. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    Combined in series to get equivalent value I think. It's not a mission critical value, you just want a mA or so through the zener.
     
  17. JBoogie

    JBoogie Rando

    Joined:
    May 31, 2016
    Likes Received:
    11
    Dislikes Received:
    0
    Trophy Points:
    3
    ahhh now i see, thanks
     
  18. spwath

    spwath Collegiate hijinks master

    Friend
    Joined:
    Dec 13, 2015
    Likes Received:
    2,778
    Dislikes Received:
    18
    Trophy Points:
    113
    Location:
    Hartford, CT
    Im confused about the diagram. Any help?
     
  19. OJneg

    OJneg The Most Insufferable

    Friend
    Joined:
    Sep 25, 2015
    Likes Received:
    3,253
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    Chicago
    Maybe read up a bit on how to hook up basic circuits. Doesn't get easier than this.
     
    bazelio likes this.
  20. spwath

    spwath Collegiate hijinks master

    Friend
    Joined:
    Dec 13, 2015
    Likes Received:
    2,778
    Dislikes Received:
    18
    Trophy Points:
    113
    Location:
    Hartford, CT
    I have done a bunch of circuit stuff, and am currently in the middle of building a nixie tube clock.

    Main question is, what is the big thing on the left,
    where does the mic go, and what is the number 1 atached too?

    EDIT:
    or is that thing on the left the mic?
    If so, what do the 3 wires that go nowhere on the bottom conect to?
     

Share This Page