Announcement

Collapse
No announcement yet.

Crossover design - importance of impulse and step response?

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

  • Crossover design - importance of impulse and step response?

    Well, the title of the topic is basically the question - how important is the predicted impulse and step response, when designing a crossover?

    ​I'm working on another x-over design at the moment, and XSim predicts the following for the impulse and step response. What are these graphs telling me about the x-over and how the end result will perform?

    Brian Steele
    www.diysubwoofers.org

  • #2
    Good question, I don't understand either of these things or how to interpret the graphs. So whoever answers please speak slowly and use small words for me :D

    Comment


    • #3
      Brian - also glad you asked this question!

      I think I saw a comment recently "I could tell from the step response that this was a minimum phase design".
      Cool - but what should I (we) look for?

      I think I hear a difference - wow, it's amazing!" Ethan Winer: audio myths
      "As God is my witness I'll never be without a good pair of speakers!" Scarlett O'Hara

      High value, high quality RS150/TB28-537SH bookshelf - TARGAS NLA!
      SB13/Vifa BC25SC06 MTM DCR Galeons-SB13-MTM
      My Voxel min sub Yet-another-Voxel-build

      Tangband W6-sub

      Comment


      • #4
        This is also one of those graphs that I sort of understand, but don't. I'm looking forward to a long explanation.
        - Ryan

        CJD Ochocinco ND140/BC25SC06 MTM & TM
        CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
        CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
        CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

        Comment


        • #5
          Look at a few reviews from Stereophile. In the measurements you'll find the step response discussed.

          I can't tell from your images, I need more resolution of it, but the step response tells you about driver polarity, time alignment and phase matching. I use the LM-1 a lot because I made it specifically as an educational tool, so take a look at the step response (last image on the page) from the measurement's page.

          The two peaks from left are the tweeter and woofer. It shows:
          • Both are aligned in the same, positive polarity,
          • The tweeter leads the woofer
          • Excellent phase matching since the end of the tweeter's step blends smoothly with the beginning of the woofer's output.

          Of course, for the simulation to be accurate, you must have input the correct relative driver offsets.

          There are probably other things you can tell from this chart I'm not aware of also. To look at many more discussed, I encourage you to look at Stereophile. If you want to see the step response for time/phase co-incident speakers look for Thiel, VanDersteen and Dunlavy speakers.
          My Little Soapbox The industry, commercial and DIY, has gone to trying to use purely theoretical environments to judge their driver and sp...

          Comment


          • #6
            The impulse is just the frequency response expressed in a different way. Obviously, the frequency response, after the Impulse is convolved is more useful. The Impulse itself allows you to see when, in time, reflections enter into the data, which will show up as later ripples in the impulse. This allows you to set the gate time prior to the reflections, but it also limits the low frequency extension of the data.

            The Step Response serves the purpose of showing the acoustic polarity (upward or downward peak) and the time relationship between two drivers. A time coherent two-way, for example will have only one peak to the step because the impulse from both drivers will reach the mic at the same time. Typical two-ways will have two-peaks because the tweeter's output will arrive slightly earlier than the woofer's.
            Click here for Jeff Bagby's Loudspeaker Design Software

            Comment


            • #7
              Originally posted by Brian Steele View Post
              Well, the title of the topic is basically the question - how important is the predicted impulse and step response, when designing a crossover?

              ​I'm working on another x-over design at the moment, and XSim predicts the following for the impulse and step response. What are these graphs telling me about the x-over and how the end result will perform?
              Hi Brian,

              Frequency response, Phase Response, Impulse Response, Step Response, CSD plots, Wavelet Analysis - all of these, if applied correctly, reveal some important aspects of the loudspeaker systems.
              Typically, we examine or use all of these tools, because each one allows you to probe and focus on somewhat different aspect of system performance, even though they are all related.

              For example, http://www.bodziosoftware.com.au/Fre...Evaluation.pdf

              Best Regards,
              Bohdan

              Comment


              • #8
                Originally posted by ErikSquires View Post
                Look at a few reviews from Stereophile. In the measurements you'll find the step response discussed.

                I can't tell from your images, I need more resolution of it, but the step response tells you about driver polarity, time alignment and phase matching. I use the LM-1 a lot because I made it specifically as an educational tool, so take a look at the step response (last image on the page) from the measurement's page.

                The two peaks from left are the tweeter and woofer. It shows:
                • Both are aligned in the same, positive polarity,
                • The tweeter leads the woofer
                • Excellent phase matching since the end of the tweeter's step blends smoothly with the beginning of the woofer's output.

                Of course, for the simulation to be accurate, you must have input the correct relative driver offsets.
                ​Thanks. Curiously enough though, even though I'm getting the same little peak at the beginning of the step response, for this particular build the woofer's acoustic center is actually closer to the mic, by about 0.82 inches (worked it out via WinPCD). This is because the "tweeter" is a CD in an APT150 horn and the woofer is front-mounted on the baffle.
                Brian Steele
                www.diysubwoofers.org

                Comment


                • #9
                  The group delay of the woofer + low pass filter is considerable, thus it still shows up after the tweeter. If you have a group delay chart examine it.

                  Comment


                  • #10
                    So here is the impulse response from a speaker I built. What do the peaks and dips and spacing of it mean? I understand the ripples all along the way are reflections, right? This was taken in a small room at about 2m.
                    Attached Files

                    Comment


                    • #11
                      That's the measured impulse response. You can tell from the blips in the graph how long it takes for reflected sound from the speaker to reach the listening position. In your case it looks like the first reflection occurs about 1.4 mS after the initial sound from the speaker.

                      ​However, your question is slightly different to the one I raised. I raised a question about the "predicted" impulse response from the x-over design, and what does it tell me about how my design will perform.

                      Brian Steele
                      www.diysubwoofers.org

                      Comment


                      • #12
                        Hi Dynamo,

                        Nice speakers! I love OmniMic too.

                        The impulse response is the raw test signal. It gets converted into the FR you see above it. The color of the impulse respons3e indicate how much is being included in the "gated" portion of the display. If you select "Only to" you'll see the FR that corresponds to the red part of the impulse response. The longer the gate, the lower the frequency you can measure, but the more likely you are to run into reflections which muddy your result.

                        Hope this helps!

                        E

                        Comment


                        • #13
                          Originally posted by dynamo View Post
                          Good question, I don't understand either of these things or how to interpret the graphs. So whoever answers please speak slowly and use small words for me :D

                          Hi dynamo,

                          Some useful information can be found on pages 10-14.

                          http://www.troelsgravesen.dk/downloa...hnAtkinson.pdf

                          Best Regards,
                          Bohdan

                          Comment


                          • #14
                            Great, thanks for the responses guys. That is all very helpful. This is one of those things that I will have to reread the info several times for it to sink in better, but at least so far I see how to use it as a tool to set the gating (I think).

                            Comment


                            • #15
                              Originally posted by bohdan View Post


                              Hi dynamo,

                              Some useful information can be found on pages 10-14.

                              http://www.troelsgravesen.dk/downloa...hnAtkinson.pdf

                              Best Regards,
                              Bohdan
                              Great resource. It has been years since I read Sterophile. JA is the amazingly knowledgable and while somewhat criptic at times, a straight shooter. His expertise is worth the subscription cost. Thanks for posting the link.
                              - Ryan

                              CJD Ochocinco ND140/BC25SC06 MTM & TM
                              CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
                              CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
                              CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

                              Comment

                              Working...
                              X