Announcement

Collapse
No announcement yet.

Adding a high pass filter to 2 way crossover?

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

  • Adding a high pass filter to 2 way crossover?

    I want to add a 90hz high pass filter to a 2 way crossover. I want to keep lower frequencies from a "smaller" woofer that cannot handle the lower frequencies. Where in the 2 way crossover is it appropriate to wire in the high pass components?

  • #2
    2 part problem:
    a - A 1st order HP (single cap) probably won't do much - due to the fact that you've probably got an impedance "spike" down at the bottom end from the box tuning, or just the Qtc effect of a closed box.
    2 - If you go 2nd order, be forewarned - the co$t will be high due to low XO freq. (low freq. = large component values). While you CAN (and should) consider (cheap) npe caps for the series component, there's no way around the large shunt coil problem. At 90Hz, you're looking at a shunt coil in the 15mH to 30mH range (which PE doesn't even carry), and that's not even considering problem "1". These values are for the "nominal" impedance !

    This is why most folks go active when trying to roll off the bottom end of their mains.

    If you forge ahead, the tweeter is already protected from those freqs, so (in a 2-way "parallel" topology) you could add the filter just to the mid-bass, OR you could hang the entire (small) speaker off of it.

    Comment


    • #3
      Which is why this is usually done with an active filter and sealed mains. One can put a first order cap in the line level feed to the amp. Much smaller as the amp input is probably 100K. What you don't know is the value of the input DC blocking cap in the amp. It has to be part of the equation. ( capacitors in series calculation) They are common in car stereos " bass blockers" .

      Comment


      • #4
        Originally posted by Chris Roemer View Post
        2 part problem:
        a - A 1st order HP (single cap) probably won't do much - due to the fact that you've probably got an impedance "spike" down at the bottom end from the box tuning, or just the Qtc effect of a closed box.
        2 - If you go 2nd order, be forewarned - the co$t will be high due to low XO freq. (low freq. = large component values). While you CAN (and should) consider (cheap) npe caps for the series component, there's no way around the large shunt coil problem. At 90Hz, you're looking at a shunt coil in the 15mH to 30mH range (which PE doesn't even carry), and that's not even considering problem "1". These values are for the "nominal" impedance !

        This is why most folks go active when trying to roll off the bottom end of their mains.

        If you forge ahead, the tweeter is already protected from those freqs, so (in a 2-way "parallel" topology) you could add the filter just to the mid-bass, OR you could hang the entire (small) speaker off of it.
        Also, adding a high pass to the midwoofer which should already have a low pass, makes the crossover for the midwoofer a bandpass which shifts the lowpass point.

        Comment


        • #5
          Originally posted by AEIOU View Post

          Also, adding a high pass to the midwoofer which should already have a low pass, makes the crossover for the midwoofer a bandpass which shifts the lowpass point.
          I have never heard of it being called a bandpass, usually that is reserved to describe a mid range speaker, that does not do the highs of the tweeter nor the bass of the woofer.

          I am not sure what you mean by the bolded part?

          Comment


          • #6
            Originally posted by kevintomb View Post

            I have never heard of it being called a bandpass, usually that is reserved to describe a mid range speaker, that does not do the highs of the tweeter nor the bass of the woofer.

            I am not sure what you mean by the bolded part?
            Lets say you have a midrange with a simple highpass, they do this in very cheap loudspeakers, just a single capacitor in series with the midrange. Then you decide to add an inductor, also in series, you now have a bandpass filter. A bandpass filter is a bandpass filter regardless if it's a woofer, midrange or tweeter. OK, suppose you are lowpassing a midwoofer at 2.5KHz and then you decide to also add some series capacitance to block the lows, at first it was just a lowpass, but now it is technically a bandpass filter. That original 2.5KHz point will shift some, how much may not be significant, but it still occurs. I discussed this a long time ago with Professor Bullock in correspondence with Speaker Builder Magazine. He knew all the math, showed me the formula, results. Try and remember way back then modeling software wasn't readily available and people like Prof Bullock did the calculations the long way.
            Last edited by AEIOU; 09-05-2020, 01:16 AM.

            Comment


            • #7
              Originally posted by by View Post
              Where in the 2 way crossover is it appropriate to wire in the high pass components?
              Usually nowhere. If you're using this with a sub you'd have either a passive or active crossover handling the frequency routing. Just about all modern AVRs have active crossovers built in.

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

              Comment


              • #8
                https://www.parts-express.com/rolls-...SAAEgKNyfD_BwE

                Comment


                • #9
                  Originally posted by AEIOU View Post

                  Lets say you have a midrange with a simple highpass, they do this in very cheap loudspeakers, just a single capacitor in series with the midrange. Then you decide to add an inductor, also in series, you now have a bandpass filter. A bandpass filter is a bandpass filter regardless if it's a woofer, midrange or tweeter. OK, suppose you are lowpassing a midwoofer at 2.5KHz and then you decide to also add some series capacitance to block the lows, at first it was just a lowpass, but now it is technically a bandpass filter. That original 2.5KHz point will shift some, how much may not be significant, but it still occurs. I discussed this a long time ago with Professor Bullock in correspondence with Speaker Builder Magazine. He knew all the math, showed me the formula, results. Try and remember way back then modeling software wasn't readily available and people like Prof Bullock did the calculations the long way.
                  I know what a bandpass filter is, I was asking only cause you referred to it as just a "Bandpass", I did not realize you meant filter.......

                  Comment


                  • #10
                    Originally posted by kevintomb View Post

                    I know what a bandpass filter is, I was asking only cause you referred to it as just a "Bandpass", I did not realize you meant filter.......
                    What did you think I meant, let the Band pass? LOL

                    Click image for larger version  Name:	DSCF1581.jpg Views:	0 Size:	766.2 KB ID:	1450609

                    Comment


                    • #11
                      I'm not going to ask what a highpass is...
                      Francis

                      Comment


                      • AEIOU
                        AEIOU commented
                        Editing a comment
                        Good one! LOL

                      • Steve Lee
                        Steve Lee commented
                        Editing a comment
                        Bogart . . .
                    Working...
                    X