Announcement

Collapse
No announcement yet.

Measuring the power response - How??

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Measuring the power response - How??

    Is it possible to measure the power response of a speaker with OmniMic?? IF so... how?!?

    Many thanks!
    Rich

  • #2
    Re: Measuring the power response - How??

    Originally posted by rmeinke View Post
    Is it possible to measure the power response of a speaker with OmniMic?? IF so... how?!?

    Many thanks!
    Rich
    Power response is an average total radiated response, so I'd take a measurement on-axis, 15, 30, 45, 60, 75, 90 degrees and average the results of all those. Realistically though, since you have to take all those measurements, you can generate a polar response graph with Omnimic which provides a lot more information than power response alone.
    I'm not deaf, I'm just not listening!

    Comment


    • #3
      Re: Measuring the power response - How??

      https://www.youtube.com/watch?v=j77VKw9Kx6U

      Rich,

      I like to use John Atkinson's method that he described at the 29:30 to 33:00 timeline point in the above referenced loudspeaker measurement video. Also called "spatially averaged in-room response." My understanding is that you set up an imaginary "grid" or "window" about 2 meters or so from your speaker. The window would be about 36" by 24" and you would proceed to take 20 measurements with the microphone positioned a various spots in the window. As OmniMic's swept sine wave is playing with "advanced functions" checked, I click "new average" then move the mic to another position in the window, then click "new average" and then move the mic to another position in the window. I keep doing this until I have averaged 20 mic locations.

      This is how I calculated the power response in post #209 of my signature Linehopper loudspeaker thread.

      Hope this helps,


      Bill
      SideTowers: http://techtalk.parts-express.com/fo...corundum-build
      Totally Flat: http://techtalk.parts-express.com/fo...5-totally-flat
      Plumber's Delight: http://techtalk.parts-express.com/fo...notech-winners
      Linehopper: http://techtalk.parts-express.com/sh...Esoteric-build

      Comment


      • #4
        Re: Measuring the power response - How??

        Originally posted by 4thtry View Post
        https://www.youtube.com/watch?v=j77VKw9Kx6U

        Rich,

        I like to use John Atkinson's method that he described at the 29:30 to 33:00 timeline point in the above referenced loudspeaker measurement video. Also called "spatially averaged in-room response." My understanding is that you set up an imaginary "grid" or "window" about 2 meters or so from your speaker. The window would be about 36" by 24" and you would proceed to take 20 measurements with the microphone positioned a various spots in the window. As OmniMic's swept sine wave is playing with "advanced functions" checked, I click "new average" then move the mic to another position in the window, then click "new average" and then move the mic to another position in the window. I keep doing this until I have averaged 20 mic locations.

        This is how I calculated the power response in post #209 of my signature Linehopper loudspeaker thread.

        Hope this helps,


        Bill
        Thanks Bill, this makes perfect sense to me. I'll have to read your post and watch John's video. I'm measuring DML panels and they are 24 x 30 and have good off-axis perfromance and wonder if my window has to be larger based on what you have stated for the window.

        The reason why I'm interested in measuring this is that in a technical paper on DMLs it was stated that the power response is such that tonally they have a gradually raising high frequncy response. After EQing the panels flat this weekend agree that they do indeed sound like the HF is hot. Given this, I want to see what is going on in the power response to verify and make sense of what I'm hearing.

        Sure, I can EQ to taste and that's the only thing that really matters to me for listening to music with these panels, but the computer scientist in me wants to understand the technical nature of these panels in this respect and what "flat" really should be on a DML panel.

        Thanks for the info Bill!

        Comment


        • #5
          Re: Measuring the power response - How??

          Originally posted by dcibel View Post
          Power response is an average total radiated response, so I'd take a measurement on-axis, 15, 30, 45, 60, 75, 90 degrees and average the results of all those. Realistically though, since you have to take all those measurements, you can generate a polar response graph with Omnimic which provides a lot more information than power response alone.
          Thanks dcibel, this makes sense to me. From what you know, do you consider this to be a somewhat standard method? I did imagine that a standard test method would be close to that you explain.

          Though I am very interested in the like to John A's method in the link that Bill provided as I have always expected J.A. opinion in Stereophile for many, many years and just wonder if it applies to a DML panel speak.

          Thanks again for the info man.

          Comment


          • #6
            Re: Measuring the power response - How??

            Originally posted by rmeinke View Post
            Thanks dcibel, this makes sense to me. From what you know, do you consider this to be a somewhat standard method? I did imagine that a standard test method would be close to that you explain.

            Though I am very interested in the like to John A's method in the link that Bill provided as I have always expected J.A. opinion in Stereophile for many, many years and just wonder if it applies to a DML panel speak.

            Thanks again for the info man.
            Go with the method shown above if you want the most "standard" method. However there is not much in the way of standardization in acoustic measurements. The key is to take whatever measurements you deem necessary to adequately evaluate what you want to. You may average however many measurements from however many points you like as long as you feel the result is of some benefit to you. That said, it is still my opinion that generating the polar graphs with the Omnimic will tell you more about your speaker than a simple average of multiple angles. The only time I've ever taken an average of multiple points was for room EQ. I took an average of multiple seating locations, and with the aid of the Omnimic software generated a parametric EQ to balance out the room resonances.
            I'm not deaf, I'm just not listening!

            Comment


            • #7
              Re: Measuring the power response - How??

              Hey dcibel, this is excellent. I really do want to get, in large part, the DML panel and the room so maybe a larger window than Bill and JA's. With a DML being dipolar there is more energy going into the room and that is what I need to assess. So I will use Bill's method of using "new average" after each sweep at a different location.

              I'm still surprised that there isn't a commonly accepted method to obtain a power response. It would be nearly impossible to compare power responses broadly then. I do remember JA's comments from many years ago that stated something to the effect that the power response in the room often told more about the tonal balance than the freq. resp. so I always assumed that there was a common way the measure it. More academic in nature but wanted to comments anyway. I have my marching orders and know what I want to achieve in my measurements.

              Should be interesting.

              Comment


              • #8
                Re: Measuring the power response - How??

                http://www.roger-russell.com/revrmpg.htm
                "It suggests that there is something that is happening in the real system that is not quite captured in the models."

                Comment

                Working...
                X