Announcement

Collapse
No announcement yet.

How do you know how low you can cross over a tweeter without damaging it?

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

  • How do you know how low you can cross over a tweeter without damaging it?

    A lot of people mention that you can damage a tweeter by not crossing it high enough. This begs the question: How do you know how low you can crossover a tweeter without damaging it and why does a low frequency signal damage a tweeter? Shouldn't the natural roll off of the tweeter `protect` it?

  • #2
    General rule of thumb.... 2X the tweeters resonant frequency. And no, the natural roll off of the tweeter offers no protection. That's an acoustic roll off, just because they don't make sound doesn't mean the electrical energy isn't there heating the coil to destruction or popping the tiny diaphragm past it's Xmax.

    In detail, there are a few things to consider.
    That resonant frequency impedance peak will make your high pass filter ineffective. As Z goes up, the capacitor(s) / inductor(s) don't attenuate the signal as much. To solve you either cross above that significantly, or add a parallel resistor across the tweeter terminals to bring that peak down.

    More importantly, the distortion will rise as frequency goes down and the tiny suspension gets worked out. That generally happens as you approach Fs as well.
    Electronics engineer, woofer enthusiast, and musician.
    Wogg Music
    Published projects: PPA100 Bass Guitar Amp, ISO El-Cheapo Sub, Indy 8 2.1 powered sub, MicroSat

    Comment


    • #3
      The tweeter naturally resonates at itsFs, somewhat uncontrollably (like the Tacoma Bridge wind collapse). As you near Fs, distortions rise (that usually sounds bad).

      Comment


      • #4
        Originally posted by wogg View Post
        General rule of thumb.... 2X the tweeters resonant frequency. And no, the natural roll off of the tweeter offers no protection. That's an acoustic roll off, just because they don't make sound doesn't mean the electrical energy isn't there heating the coil to destruction or popping the tiny diaphragm past it's Xmax.

        In detail, there are a few things to consider.
        That resonant frequency impedance peak will make your high pass filter ineffective. As Z goes up, the capacitor(s) / inductor(s) don't attenuate the signal as much. To solve you either cross above that significantly, or add a parallel resistor across the tweeter terminals to bring that peak down.
        Makes sense, so a parallel resistor (with relatively high value like 50 Ohms) can be used to attenuate the resonant impedance peak a bit, without affecting the rest of the impedance response significantly.

        In an L-pad configuration, then the series resistor can be used to attenuate the whole impedance curve to level match a woofer, whereas a parallel resistor across the tweeter provides attenuation to the resonance peak.

        Comment


        • #5
          Originally posted by Chris Roemer View Post
          The tweeter naturally resonates at itsFs,. As you near Fs, distortions rise
          No more so than a midrange or woofer. Distortion goes up as frequency goes down primarily as a result of over-excursion.
          a parallel resistor (with relatively high value like 50 Ohms) can be used to attenuate the resonant impedance peak a bit, without affecting the rest of the impedance response significantly.
          There's no benefit from so doing, for the reason above.
          In an L-pad configuration, then the series resistor can be used to attenuate the whole impedance curve to level match a woofer, whereas a parallel resistor across the tweeter provides attenuation to the resonance peak.
          That's also not the case. The primary protection device for a tweeter is the high pass filter section of the crossover. What damages a tweeter is over excursion and over powering. They both go up as frequency goes down, so the better job you do of keeping low frequencies out the more protection you have. One way to maximize protection is to use a higher order high pass filter. The higher the filter order the lower you can run the tweeter.
          www.billfitzmaurice.com
          www.billfitzmaurice.info/forum

          Comment


          • #6
            Originally posted by billfitzmaurice View Post
            No more so than a midrange or woofer. Distortion goes up as frequency goes down primarily as a result of over-excursion.
            There's no benefit from so doing, for the reason above.
            That's also not the case. The primary protection device for a tweeter is the high pass filter section of the crossover. What damages a tweeter is over excursion and over powering. They both go up as frequency goes down, so the better job you do of keeping low frequencies out the more protection you have. One way to maximize protection is to use a higher order high pass filter. The higher the filter order the lower you can run the tweeter.
            Hmm, from my understanding then. High-pass --> primary protection, parallel resistor--> provide some attenuation of the peak to make the high-pass filter curve roll off with a given slope, if the slope without the resistor is undesirable and contains a peak close to the resonant Fr

            Comment


            • #7
              You wouldn't use a parallel resistor, that drops the load seen by the high pass filter, messing up the filter response. There's no reason to be concerned about the Fs peak. There would be if the amp was a current source and the tweeter a current driven device, but they aren't.
              www.billfitzmaurice.com
              www.billfitzmaurice.info/forum

              Comment


              • #8
                Sadly, destructive testing is a skilled discipline for a reason.
                Don't listen to me - I have not sold any $150,000 speakers.

                Comment


                • #9
                  Originally posted by billfitzmaurice View Post
                  You wouldn't use a parallel resistor, that drops the load seen by the high pass filter, messing up the filter response. There's no reason to be concerned about the Fs peak. There would be if the amp was a current source and the tweeter a current driven device, but they aren't.
                  Thanks. I'm now trying to understand the purpose of a parallel resistor across a tweeter. My understanding is that this would modify the impedance curve by lowering the speaker impedance, especially at FR (however this is irrelevant if the high pass is at an appropriate crossover point 2x the FR) . A series resistor clearly will lower the SPL uniformly to match the woofers SPL. So what is the purpose of adding a parallel resistor across the tweeter (I think I have seen this in some designs)?

                  Comment


                  • #10
                    The attached spreadsheet (rename as .xls) may be useful - it allows you to model the excursion and thermal limited power handling (the lesser of each) for 3 drivers at the same time. You could overlay your acoustic XO response for your drivers and see which one will "XMAX out" first. Alternatively, you can see the parameters required to model excursion of a driver and use these in any other xmax program to test excursion limits
                    Attached Files

                    Comment


                    • #11
                      Originally posted by mmu7 View Post

                      Thanks. I'm now trying to understand the purpose of a parallel resistor across a tweeter. My understanding is that this would modify the impedance curve by lowering the speaker impedance, especially at FR (however this is irrelevant if the high pass is at an appropriate crossover point 2x the FR) . A series resistor clearly will lower the SPL uniformly to match the woofers SPL. So what is the purpose of adding a parallel resistor across the tweeter (I think I have seen this in some designs)?
                      It does make the slope of the filter more consistent through Fs even when the filter starts well above. When coupled with a series resistor (L-Pad) you get a more consistent attenuation as well. Really though, many times it's not needed. When playing around with a sim, I'll move resistors before the network, after the network, and across the terminals around to watch how it changes the response, filter slopes, and phase alignment trial and error style.
                      Electronics engineer, woofer enthusiast, and musician.
                      Wogg Music
                      Published projects: PPA100 Bass Guitar Amp, ISO El-Cheapo Sub, Indy 8 2.1 powered sub, MicroSat

                      Comment


                      • #12
                        Originally posted by mmu7 View Post
                        I'm now trying to understand the purpose of a parallel resistor across a tweeter. My understanding is that this would modify the impedance curve by lowering the speaker impedance, especially at FR ?
                        Not by itself. A resistor and cap in series paralled with the tweeter makes an impedance flattener (zobel) which is useful if you're using an off the shelf filter, but you shouldn't do that if you can help it. A resistor, cap and coil in series, that assembly parallel with the tweeter, makes a notch filter, to tame a response peak.

                        www.billfitzmaurice.com
                        www.billfitzmaurice.info/forum

                        Comment


                        • #13
                          Originally posted by billfitzmaurice View Post
                          You wouldn't use a parallel resistor, that drops the load seen by the high pass filter, messing up the filter response. There's no reason to be concerned about the Fs peak. There would be if the amp was a current source and the tweeter a current driven device, but they aren't.
                          mmu7 pretty much has the right idea, but the resistor would need to be a lower value. Maybe 10 ohms. That can make it easier to get the tweeter response to match a target roll-off. The crossover will be designed with this revised load. It will not mess it up. If both responses match the same target roll-off, there is no added protection from the resistor. However, if the response at fs is louder without the resistor, than with it, then there is some additional protection. Might not be a lot though.

                          Basically just look at how loud the tweeter will be playing. Louder means more heat, and more excursion. What you can get away with may depend on how hard you want to push the speaker.

                          Comment


                          • #14
                            An LPad is a much better way of reducing the input power to the tweeter, if that's required. If the tweeter sensitivity is close to that of the woofer sensitivity it wouldn't be.
                            www.billfitzmaurice.com
                            www.billfitzmaurice.info/forum

                            Comment


                            • #15
                              The resistor in parallel with the tweeter becomes a voltage divider. Voltage is constant in a parallel circuit so as the resistance rises in the tweeter current takes the path of least resistance. The impedance peak is still there but the amp or crossover don't see it.
                              "Every man owes a part of his time and money to the business or industry in which he is engaged.
                              No man has a moral right to withhold his support from an organization that is striving to improve conditions within his sphere."
                              -Theodore Roosevelt

                              Comment

                              Working...
                              X