Announcement

Collapse
No announcement yet.

Need help understanding passive crossovers power distribution ratios

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

  • Need help understanding passive crossovers power distribution ratios

    This may seem very elementary to most, but I have never personally used a passive crossover. Everything I build uses active crossovers and multiple amplifiers. Where I am hung up is in the following example of using a 3 way build: 15" woofer, 6" midrange, and 2" compression driver horn.

    -The woofer has a sensitivity of 98dB @1w/1m 500w RMS
    -The midrange has a sensitivity of 99dB @1w/1m 250w RMS
    -Compression driver is 111db @1w/1m 100w RMS

    It is very obvious that some attenuation would be necessary on the compression driver, but here is where my question(s) lie.

    1. For example if I were to send 100w RMS to my speaker, does 33.33% go to each of the three (3) drivers? My guess is no, but I don't know why.
    2. If I match the sensitivity of each driver perfectly and the power distribution isn't equal, how do I guarantee I am getting the correct balance of woofer, mid, and tweeter?
    3. If all things were equal in power distribution, what happens when I go past the driver with the lowest power handling (100w)? Is that the limiting factor?

    It seems to me that speakers would sound right at one (1) volume set point, but as the volume is turned up or turned down, things would get out of order. In other words aside from the 'sweet spot' the woofer would get too loud and the mids would be too quiet or vise versa (This is just an example).

    I hope this makes sense and someone can help me make sense of it. And advice is greatly appreciated.
    "I don't know everything and do not claim to. I continue to learn and that is what makes me human."

  • #2
    1. The power splits unequally because most power in music is below 500Hz. That depends on the music of course, but generally the lion's share of power is down there.
    2. You adjust things for fairly flat on-axis response. You may find some variation gives you the sound you want.
    3. All things are seldom equal
    Francis

    Comment


    • #3
      Originally posted by STIchris722 View Post
      if I were to send 100w RMS to my speaker, does 33.33% go to each of the three (3) drivers? My guess is no, but I don't know why.
      The reason why is that power density isn't a constant, it drops by 1/2 (3dB) with each octave increase in frequency. This is why when we test speakers with a broadband signal it's pink noise, which drops in power density as frequency rises by 3dB/octave.

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

      Comment


      • #4
        I'll also note, the power distribution at the listener is the same for passive or active, assuming both are competently achieved.
        Francis

        Comment


        • #5
          Originally posted by billfitzmaurice View Post
          The reason why is that power density isn't a constant, it drops by 1/2 (3dB) with each octave increase in frequency. This is why when we test speakers with a broadband signal it's pink noise, which drops in power density as frequency rises by 3dB/octave.
          Okay that makes sense. The part I'm still hung up on is how speakers can sound good as the volume knob is increased or decreased. It seems to me that speakers would have a sweet spot in volume....So how does a person select drivers for a passive setup vs an active? Active I just adjust the crossover frequencies until it sounds right to me and then I adjust the levels for each driver accordingly. That was the only reason I brought up the active and passive thing. I am trying to build a passive 3 way tower and am confused on driver matching and power handling.
          "I don't know everything and do not claim to. I continue to learn and that is what makes me human."

          Comment


          • #6
            Originally posted by STIchris722 View Post

            Okay that makes sense. The part I'm still hung up on is how speakers can sound good as the volume knob is increased or decreased. It seems to me that speakers would have a sweet spot in volume....So how does a person select drivers for a passive setup vs an active? Active I just adjust the crossover frequencies until it sounds right to me and then I adjust the levels for each driver accordingly. That was the only reason I brought up the active and passive thing. I am trying to build a passive 3 way tower and am confused on driver matching and power handling.
            Passive works pretty much the same way. You match the levels, and when you turn up the volume, everything stays matched, and balanced.

            Power limitations are usually with the woofer. If your woofer can play at 95dB, but distorts badly as you approach 100dB, then that's most likely as much SPL as you are going to get. A 50w 5.25" cone mid can probably keep up with a 10" woofer, although there would be exceptions. If the mid is padded 6dB, then it's getting much less power at any spl than the woofer. Is 6dB half power, or one 4th? I forget sometimes.

            Comment


            • #7
              Originally posted by STIchris722 View Post

              Okay that makes sense. The part I'm still hung up on is how speakers can sound good as the volume knob is increased or decreased. It seems to me that speakers would have a sweet spot in volume....So how does a person select drivers for a passive setup vs an active? Active I just adjust the crossover frequencies until it sounds right to me and then I adjust the levels for each driver accordingly. That was the only reason I brought up the active and passive thing. I am trying to build a passive 3 way tower and am confused on driver matching and power handling.
              I'm not sure it's the speakers which sound good at various volumes, or the recording itself: for example, some music sounds great at any reasonable volume, some only when you turn it up. Deep Purple's 'Made in Japan' sounds good whether it's played softly or at a level at which you annoy the neighbours. The original mix of 'Rock n Roll Animal' by Lou Reed doesn't sound good to me unless it's played loud.

              With classical music - say, Beethoven's Choral Symphony - it's a bit harder as the music has more dynamic range.

              I tend to play an album at the level at which it sounds good, and that varies quite a bit. So to me it's a function of the recording, not the speakers.

              Geoff

              Comment


              • #8
                (NOT an EE, so... )

                When you turn the volume up on an amp, all it does is raise its output voltage. Speakers (typically rated @ 2.83v) have a certain (SPL) output at that given voltage.
                In a proper (passive, but active would also apply) system (esp. easy to comprehend w/a passive "parallel" XO - 'cause the amp is connected to each (of 2 or 3) "filters" - and then to the driver beyond) all legs will see the same voltage increase as the amp gets louder. All the drivers stay "in balance".

                The filters are responsible for adjusting the voltage in diff. freq. ranges (depending on which driver is meant to play which freqs.) and usually use resistance for attenuating the louder drivers.

                Depending on each particular driver "load" (and that of your entire speaker "system"), as the voltage goes up, the (necessary) current demands also rise. Hopefully your amp can supply whatever current your speaker load requires at whatever voltage you crank the knob up to (and hopefully your drivers' power handling relationships will all manage the power okay as well).

                I believe that at least one of the XO-design softwares on here actually estimate power usage across the XO/driver circuits.

                Comment


                • #9
                  Originally posted by STIchris722 View Post
                  The part I'm still hung up on is how speakers can sound good as the volume knob is increased or decreased.
                  Do you have that problem when an active crossover is used? I suspect not so it stands to reason that when a passive crossover is used the same applies.

                  Originally posted by STIchris722 View Post
                  It seems to me that speakers would have a sweet spot in volume
                  They don't, but what we hear changes with SPL due to the limitation and non linearities with human hearing. At very low SPL we don't hear extreme lows and highs so well, this is why there is a Loudness button or control on many home audio components, and at very high SPLs out ears try to protect themselves with a sort of mechanical compression, but in between those extremes hearing is mostly linear.

                  Originally posted by STIchris722 View Post
                  So how does a person select drivers for a passive setup vs an active? Active I just adjust the crossover frequencies until it sounds right to me and then I adjust the levels for each driver accordingly.
                  You could do exactly that and then convert the crossovers and level controls to passive versions. Same basics apply it's just the execution that is different.



                  Paul O

                  Comment


                  • #10
                    Originally posted by STIchris722 View Post
                    So how does a person select drivers for a passive setup vs an active? Active I just adjust the crossover frequencies until it sounds right to me and then I adjust the levels for each driver accordingly.
                    Since you seem to have the necessary gear do the same thing. Once you know what frequency and slope sounds best create a passive that duplicates it, along with LPads for level matching, if required. The only caveat is that you don't want to use an LPad on the woofer, so the sensitivity of the mid and HF drivers must be equal to or higher than the woofer.

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

                    Comment


                    • #11
                      Thank you everyone for the explanations. I guess where I was getting caught up was with ratios: i.e. sensitivity vs. RMS power of a driver.

                      For example:
                      -Woofer is 98dB @1W/1m with 500W RMS
                      -Midrange is 99dB @1W/1m with 150W RMS

                      So with the information above an assuming power is not split 33.33% per driver, how does the volume of the speakers stay linear? I know volume is not linear, but you understand what I am getting at I hope. I am capable of using x-over software and making something on screen that looks pretty, but I don't understand how it works and that is what is eating at me. So forgive my endless questions.

                      I don't have the issues as stated above with an active crossover, but as to why not is where I don't understand. I know that active and passive ultimately do the same thing in the long run with small pros and cons to each.
                      "I don't know everything and do not claim to. I continue to learn and that is what makes me human."

                      Comment


                      • #12
                        Sensitivity tells you how loud a driver gets with 1w input.
                        Power handling will tell you the maximum SPL the driver is capable of generating.
                        For every 3db increase/decrease in SPL you need double/half as much power.
                        Once you level match the drivers that relationship stays the same at all input power levels.

                        In this case that mid driver could be a limiting factor as it won't get as loud as the woofer at full power. Of course after the usual EQ is applied you don't need as much output from the mid to make the system sound natural but it would still get pushed to the max and that is never a good thing from a reliability standpoint, so horn loading the mid or one with more power handling would be necessary IMO. I think the 250w mid used in your first post would be plenty for example.
                        Paul O

                        Comment


                        • #13
                          -Woofer is 98dB @1W/1m with 500W RMS
                          -Midrange is 99dB @1W/1m with 150W RMS

                          How much power a V.C. (or suspension) can take is not a factor AFA level matching is considered.
                          In a passive XO, all that matters is the sensitivity AT THE SAME drive level (in VOLTAGE, not watts).
                          You can't control how many watts a driver gets (remember, 2.83v into 8ohms is 1w, but 2.83v into 4ohms is 2w).
                          In a ("parallel") passive XO, each leg will see the same voltage from an amp (unless it's a multi-amped design). For passive design, it's "standard" to use sensitivities rated at 2.83v at 1m. SOMEtimes (in the past) even PE will forget this and attempt to publish drivers' SPL levels at 1w/m, which tends to lead to many errors and hoses up lots of stuff (as far as passive XO design is concerned).

                          Comment


                          • #14
                            The tweeter needs padded down about 13 db at least so it's literally only ever going to see maybe 2 watts in real music power unless the amp misbehaves.

                            The mid will be fine too, because about half of the energy in music is below 300hz so you're good on sensitivity match and power handling for the mid. So you can be fine to the theoretical power handling of your woofer. I'd rate that system 500W RMS.

                            Comment


                            • #15
                              Okay I think I get it now. This was more or less me making things too complicated it seems. After staring at this stuff for so long though it's easy for me to over-analyze everything.

                              Here is a question: I have put together my x-over as best I possibly can, but I am coming up with some crazy numbers for components in this impedance EQ section. Keep in mind these drivers are totally different from the examples mentioned above. Why does it take this circuit for my trend line to look worth anything? 5300uF Caps??? Those things are the size of half a toilet paper roll and cost 20 to 50 dollars each. Am I doing something wrong here?


                              Click image for larger version  Name:	q9oiefva.JPG Views:	0 Size:	378.1 KB ID:	1430375 Click image for larger version  Name:	wegb.JPG Views:	0 Size:	295.0 KB ID:	1430376
                              "I don't know everything and do not claim to. I continue to learn and that is what makes me human."

                              Comment

                              Working...
                              X