Announcement

Collapse
No announcement yet.

ARTA Help

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

  • ARTA Help

    For those that might have used ARTA, can you explain why I may be running into the following issue:

    I'm trying to take gated measurements of a woofer. I performed 4 measurements with the mic on axis with the woofer.

    For all 4 measurements, no settings were changed: the mic location was kept the same, the speaker location was kept the same, the distance was kept the same, same gating, same software settings, etc.....same everything.

    However, my low-frequency response changes with every new measurement. Anyone know why this would be the case?

    Click image for larger version

Name:	Arta Low Frequency Issues-edited.jpg
Views:	1
Size:	170.4 KB
ID:	1399215

  • #2
    I mean you're aware that data is insignificant anyway as it's after your gate? Basically a random reaction to minor differences in the impulses?

    Comment


    • #3
      How many ms IS your gate?
      I don't seem to be able to take "averaged" readings (of several samples) w/out some wonkiness?

      Comment


      • #4
        Thanks for responding Civit. I figured this was the region where near-field measurements are spliced to far-field measurements (i.e.300hz - 600hz). The variance appears to be 1-3 db in the 300hz range about 1db in the 600hz range. My concern is that if I don't splice at the right merge level, it may impact the tonal balance of the speakers.

        Furthermore - I don t get this inconsistency when using Holm Impulse or REW.

        Comment


        • #5
          Originally posted by Chris Roemer View Post
          How many ms IS your gate?
          I don't seem to be able to take "averaged" readings (of several samples) w/out some wonkiness?
          Chris,

          I believe gate time was about 4.5ms.

          Mic was about 800mm from baffle, 1150mm from the nearest boundary (i.e., ceiling).

          Comment


          • #6
            If its of any help, I did observe the following:

            When making consecutive measurements, sometimes the impulse response starts a level zero, and some times slightly above/below it. It seems to correlate with how the FR is then derived (with more wave-ness when the IM response is not starting at level zero). See attached images - I'm really not explaining it well enough.

            Click image for larger version

Name:	Difference in IM.jpg
Views:	2
Size:	159.5 KB
ID:	1399317Click image for larger version

Name:	Resulting FR Issues.jpg
Views:	1
Size:	210.1 KB
ID:	1399316
            Attached Files

            Comment


            • #7
              Try closing the gate once just before that little "blip" (just before your "8" mark) - possibly < 3.0msec?

              Comment


              • #8
                I'd open the gate by starting a smidge closer to 4.0 and as Chris mentioned closing the top end gate to the flat spot. I assume you are using a 3 sweep average as recommended in Charlie Laub's white paper? Overall I agree with Civit that you are splicing in this range and the near field with baffle added should be able to make a good match up to 900 Hz or so.
                John H

                Synergy Horn, SLS-85, BMR-3L, Mini-TL, BR-2, Titan OB, B452, Udique, Vultus, Latus1, Seriatim, Aperivox,Pencil Tower

                Comment


                • #9
                  jhollander, could you please provide a link to Charlie Laub's white paper? Thanks. Rick

                  Comment


                  • #10
                    Measuring with ARTA http://audio.claub.net/tutorials/FR%...ing%20ARTA.pdf

                    Here's Charlie's software page http://audio.claub.net/software.html


                    John H

                    Synergy Horn, SLS-85, BMR-3L, Mini-TL, BR-2, Titan OB, B452, Udique, Vultus, Latus1, Seriatim, Aperivox,Pencil Tower

                    Comment


                    • #11
                      Thanks Jhollander,

                      I'll try and adjust gating. I posted on Diyaudio as well - and couple members suggested that the issue can be related to "DC offset" observed in images below.

                      While I can merge closer to 900hz, if the room permits, I'd like to come down to 300 - 500hz region. Seems like that should be doable in a 8' ceiling setting. I'm going to try what you, Chris, and Civit suggest - but I think I still need to resolve the DC offset issue :-(

                      Comment


                      • #12
                        Originally posted by jhollander View Post
                        I'd open the gate by starting a smidge closer to 4.0 and as Chris mentioned closing the top end gate to the flat spot.
                        Jholander,

                        Doing the above definitely helped with the waviness experienced on the low end (as well as in higher frequencies).

                        Just so I understand, is it more accurate to have shorter gate time and then merge near-field low frequency response (adjusted for simulated diffraction) with fair-field response at a high frequency (i.e. 900Hz), or is it better to have longer gate time and merge with near-field low frequency response at a lower frequency (i.e. 300 - 400hz)?

                        I noticed that by starting the gate time closer to 4, and closing it right where Chris mentioned, even my high frequency response cleaned up.

                        Comment


                        • #13
                          Well that is a bit of a trick question. The proper response is it’s best to have a measurement that is free of reflections. For a closed box I know I can splice the near field (with baffle added) to the far field and get a good response from 900 Hz down. So, in that application I’m reducing my gate to ensure the far field measurement is reflection free. That said you should be able to get down to 400 Hz in room. Over lap of the far field and near field can be good as you can see if your baffle assumptions are correct.

                          There are times when I’m measuring a dipole where I know the near field is not accurate then I move the measurement set up outside and try to get a wider gate to a lower frequency.

                          Note, for the ARTA near field measurement, make sure you gate the response to 20 or 35 ms depending on the capability of your system. I had a problem where I got very strange responses with an open gate time.
                          John H

                          Synergy Horn, SLS-85, BMR-3L, Mini-TL, BR-2, Titan OB, B452, Udique, Vultus, Latus1, Seriatim, Aperivox,Pencil Tower

                          Comment

                          Working...
                          X