Announcement

Collapse
No announcement yet.

aperiodic vent modelling?

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

  • aperiodic vent modelling?

    Is there any way to model or estimate the effect of using an aperiodic vent? Or do you just have to build and test?

  • #2
    Re: aperiodic vent modelling?

    I built and tested with my ear but you will have better results with proper measuring equipment.

    Here is a good read on aperiodic vents

    http://diyaudioprojects.com/Technica...udspeakers.htm

    PWR knows his stuff about aperiodic vents, hopefully he will chime in.

    Comment


    • #3
      Re: aperiodic vent modelling?

      Originally posted by donprice View Post
      Is there any way to model or estimate the effect of using an aperiodic vent? Or do you just have to build and test?
      An Aperiodic Vent is called such because it is intended to behave like a leak without a periodic resonance, which it may not do perfectly, but is probably close enough.

      When you understand its behaviour it's actually pretty easy to model. What it does is give you a very leaky closed box. It is pretty accurately modeled by entering a value of around 3 for Ql in a box modeling program that is complete enough to include Ql, Qa, Qp, etc. A value of Ql indicates a very low Q due to a high degree of leakage in the box.

      It basically increases the compliance of the air spring of the air inside the box by letting some of the pressurization escape. The result isn't the same as if you had a larger box, like some people say, it's more like if you had a woofer with a lower Qts instead.

      I hope this helps.

      Jeff B.
      Click here for Jeff Bagby's Loudspeaker Design Software

      Comment


      • #4
        Re: aperiodic vent modelling?

        does it have a known transfer function? At 18dB / octave it will be third order. I've never seen one and google search comes up empty for me.

        Comment


        • #5
          Re: aperiodic vent modelling?

          Hi,

          If you search "my" threads on the project QA-25, you can follow my findings on aperiodic vents.

          http://techtalk.parts-express.com/album.php?albumid=445

          Regards//lasse
          Perry Mason talking to his dentist:

          "Do you swear to take the tooth, the whole tooth and nothing but the tooth, so help you God?"

          Comment


          • #6
            Re: aperiodic vent modelling?

            I like speakers because of the math. There is a certain joy to creating a series of equations that have not been explored and turning it into something you can touch. I never realized before you started this thread that the aperiodic is a third order transfer function. I have never worked with equations that assume the third order component is pre-determined. It's an interesting idea.

            If you ever see any equations or math calculations about an aperiodic that would be awesome! TIA.

            Comment


            • #7
              Re: aperiodic vent modelling?

              Originally posted by moron#99 View Post
              I like speakers because of the math. There is a certain joy to creating a series of equations that have not been explored and turning it into something you can touch. I never realized before you started this thread that the aperiodic is a third order transfer function. I have never worked with equations that assume the third order component is pre-determined. It's an interesting idea.

              If you ever see any equations or math calculations about an aperiodic that would be awesome! TIA.
              That is another misconception, the aperiodic enclosure is not third order, it is still second order. Even a simple thought experiment will reveal this, as there is no mechanism for an additional pole to introduce an additional 6dB/octave roll-off. It simply behaves as a damped second order system.

              Even a vented box is really just two second order systems combined; both the cone's output and the vent output roll-off second order. What makes the summed roll-off fourth order is due to the output of the vent moving to a more and more out-of-phase relationship with the cone. This causes the two to begin to cancel and resultant is a fourth order roll-off.

              Again, an aperiodic box is second order system.

              Jeff B.
              Click here for Jeff Bagby's Loudspeaker Design Software

              Comment


              • #8
                Re: aperiodic vent modelling?

                Originally posted by Jeff B. View Post
                That is another misconception, the aperiodic enclosure is not third order, it is still second order. Even a simple thought experiment will reveal this, as there is no mechanism for an additional pole to introduce an additional 6dB/octave roll-off. It simply behaves as a damped second order system.

                Even a vented box is really just two second order systems combined; both the cone's output and the vent output roll-off second order. What makes the summed roll-off fourth order is due to the output of the vent moving to a more and more out-of-phase relationship with the cone. This causes the two to begin to cancel and resultant is a fourth order roll-off.

                Again, an aperiodic box is second order system.

                Jeff B.
                I read somewhere that all aperiodics have an 18dB roll-off. I have been racking my brains trying to figure out how the vent could introduce a single pole. My best guess to date was that Smalls equations combine the driver compliance with the compliance of trapped air in box =>> Cat ~ Cas*Cab. When you combine these two compliances into one then the equivalent circuit becomes second order. However, if the vent placed a finite resistance across the Cab compliance then it would not longer be like two capacitors in series. One of the capacitors would have a resistor across it and bleed off. Therefore, our second order equation would no longer be valid. Our new circuit model would have two caps, one coil, and two resistors.

                but I like Pete's idea better. It's simpler. he says it's really a bass reflex tuned to either zero or infinity or somewhere else with a really wide Q. I'm dying to plug this into small equations and see what pops out the other side.

                Comment


                • #9
                  Re: aperiodic vent modelling?

                  Originally posted by moron#99 View Post
                  I read somewhere that all aperiodics have an 18dB roll-off. I have been racking my brains trying to figure out how the vent could introduce a single pole. My best guess to date was that Smalls equations combine the driver compliance with the compliance of trapped air in box =>> Cat ~ Cas*Cab. When you combine these two compliances into one then the equivalent circuit becomes second order. However, if the vent placed a finite resistance across the Cab compliance then it would not longer be like two capacitors in series. One of the capacitors would have a resistor across it and bleed off. Therefore, our second order equation would no longer be valid. Our new circuit model would have two caps, one coil, and two resistors.

                  but I like Pete's idea better. It's simpler. he says it's really a bass reflex tuned to either zero or infinity or somewhere else with a really wide Q. I'm dying to plug this into small equations and see what pops out the other side.
                  No offense, but if that is what Pete said, then that is incorrect. If you enter zero for Fb into Small's polynominal for a vented box terms cancel out and it reduces to the math for a sealed box. In fact, it will produce the indentical response curve as that of a second order system polynomial.

                  The correct way of looking at an aperiodic system is by looking at the impedance plot. This will reveal that it really is aperiodic, with no periodic resonance frequency, and with no impedance peak associated with one. What you will find is that the impedance peak is flattened significantly - in same way as if Qms were lowered, which is essentially what is happening - the system's mechanical Q is being damped by the leakage. This is how Ql models it in the Benson model, and that's why using a Ql of around 3 gives you a very accurate model of an aperiodic system. It's really pretty simple.

                  Jeff B.
                  Click here for Jeff Bagby's Loudspeaker Design Software

                  Comment


                  • #10
                    Re: aperiodic vent modelling?

                    Originally posted by Jeff B. View Post
                    No offense, but if that is what Pete said, then that is incorrect. If you enter zero for Fb into Small's polynominal for a vented box terms cancel out and it reduces to the math for a sealed box. In fact, it will produce the indentical response curve as that of a second order system polynomial.

                    The correct way of looking at an aperiodic system is by looking at the impedance plot. This will reveal that it really is aperiodic, with no periodic resonance frequency, and with no impedance peak associated with one. What you will find is that the impedance peak is flattened significantly - in same way as if Qms were lowered, which is essentially what is happening - the system's mechanical Q is being damped by the leakage. This is how Ql models it in the Benson model, and that's why using a Ql of around 3 gives you a very accurate model of an aperiodic system. It's really pretty simple.

                    Jeff B.
                    I read in a couple places that aperiodics display 18db slopes. If true, then the A/S model isn't a good approximation. I don't think Pete meant absolute zero. I think the idea is that it was so far removed from the tested area that it did not show up.

                    Comment


                    • #11
                      Re: aperiodic vent modelling?

                      The 18dB thing IS misinformation (and unfortunately exists in the first article on google) -

                      Dickason, however, notes that the effect is equitable to simply stuffing a box.
                      I am trolling you.

                      Comment


                      • #12
                        Re: aperiodic vent modelling?

                        Originally posted by moron#99 View Post
                        I read in a couple places that aperiodics display 18db slopes. If true, then the A/S model isn't a good approximation. I don't think Pete meant absolute zero. I think the idea is that it was so far removed from the tested area that it did not show up.
                        Well, I'll bow out at this point, but I assure you there are a lot of misconceptions in speaker design, and audio in general and this is one of them. I looked into this extensively and it is simply a damped second order system with no additional pole to create another 6dB of roll-off. The thrist order roll-off doesn't even make sense if you understand box modeling math. As with any bass alignment the key is always in the impedance plot. It tells the story. After all, that's how we measure T/S parameters in the first place. The impedance plot supports what I'm saying. An open baffle speaker will yield an 18 dB/oct roll-off below resonance, but this is not open baffle it is just a leaky box.

                        Jeff B.
                        Click here for Jeff Bagby's Loudspeaker Design Software

                        Comment


                        • #13
                          Re: aperiodic vent modelling?

                          Originally posted by Jeff B. View Post
                          No offense, but if that is what Pete said, then that is incorrect. If you enter zero for Fb into Small's polynominal for a vented box terms cancel out and it reduces to the math for a sealed box. In fact, it will produce the indentical response curve as that of a second order system polynomial.

                          The correct way of looking at an aperiodic system is by looking at the impedance plot. This will reveal that it really is aperiodic, with no periodic resonance frequency, and with no impedance peak associated with one. What you will find is that the impedance peak is flattened significantly - in same way as if Qms were lowered, which is essentially what is happening - the system's mechanical Q is being damped by the leakage. This is how Ql models it in the Benson model, and that's why using a Ql of around 3 gives you a very accurate model of an aperiodic system. It's really pretty simple.

                          Jeff B.
                          Consider what an "aperiodic" system does at DC as compared to an ideal closed box. If you push the cone in with your hand, or simply apply DC to the voice coil, the vent acts as a leak completely out of phase and equal to the cone displacement - the combined output is zero. A vented system also acts this way, and a sealed system does not. It is a highly damped vented system there is no way to dispute this. Some argue that the resistive vent absorbs or dissipates the port output but given the depth of the port and damping I doubt that it is very effective at eliminating the port output even at 30 Hz. The aperiodic vent adds resistive damping and a significant acoustical leak.

                          The impedance plot tells us what the cone motion is doing because the driver is a reciprocal network and motion on the secondary is reflected back into the primary as a voltage. All the impedance plot tells us is what the cone it doing and if the cone is loaded by an acoustical network then that alters the motion. However, the total response of the system the the sum of the volume velocities out of the box as I'm sure you are familiar with per Thiele and Small. The system input impedance tells us nothing of the output from vents and leaks and the complex transfer function involved.

                          Comment


                          • #14
                            Re: aperiodic vent modelling?

                            i have to go but there is a simple way using existing pc software. model a bass reflex cabinet for a 10' Seas woofer in a 1.5Cu Ft box with an 8" diameter port 0.75" long.

                            see what the PC draws and compare it.

                            Comment


                            • #15
                              Re: aperiodic vent modelling?

                              Originally posted by Pete Basel View Post
                              Consider what an "aperiodic" system does at DC as compared to an ideal closed box. If you push the cone in with your hand, or simply apply DC to the voice coil, the vent acts as a leak completely out of phase and equal to the cone displacement - the combined output is zero. A vented system also acts this way, and a sealed system does not. It is a highly damped vented system there is no way to dispute this. Some argue that the resistive vent absorbs or dissipates the port output but given the depth of the port and damping I doubt that it is very effective at eliminating the port output even at 30 Hz. The aperiodic vent adds resistive damping and a significant acoustical leak.
                              At DC it will behave as you say, but so will many "sealed" boxes, because few of them are perfectly sealed. Remember, like T/S parameters, system alignments are defined by their behavior at resonance, not at DC. In this case, the Aperiodic enclosure is defined by the way I described it, and the math supports the measurements and vice-versa.

                              Jeff
                              Click here for Jeff Bagby's Loudspeaker Design Software

                              Comment

                              Working...
                              X