Very well, thank you for insight.
For now, let's skip TGs' spreadsheet and focus on VituixCad and Hornresp with a test.
I created a new driver in VC enclosure database and filled in everything I could from ScanSpeak datasheet. Same data is in HR.
Now VC gives me qtc 0,57 at 20,6L and HR stays the same as before so 26L.
If I input 26L volume in VC I get qtc 0,53; if I input 20,6L volume in HR I get qtc 0,62
What obvious am I missing here?
For now, let's skip TGs' spreadsheet and focus on VituixCad and Hornresp with a test.
I created a new driver in VC enclosure database and filled in everything I could from ScanSpeak datasheet. Same data is in HR.
Now VC gives me qtc 0,57 at 20,6L and HR stays the same as before so 26L.
If I input 26L volume in VC I get qtc 0,53; if I input 20,6L volume in HR I get qtc 0,62
What obvious am I missing here?
Quick one for @David McBean (or any inclined)
An acoustic horn is represented by what electrical circuit/component(s) ?
An acoustic horn is represented by what electrical circuit/component(s) ?
Hornresp calculates the closed-box system total quality factor Qtc as follows:
Qtc = Sqrt(Mac / Cat) / Ratc
Where:
Mac = acoustic mass of diaphragm and air load for driver on a box (kg/m^4)
Cat = acoustic compliance of diaphragm suspension and enclosure (m^5/N)
Ratc = total acoustic resistance for closed-box system (N.s/m^5)
The method used by VituixCAD to calculate Qtc is obviously different to the one used by Hornresp.
If simplifying approximations are made Qtc can be calculated as follows:
Qtc = Qts * Sqrt(Vas / Vb +1)
Where Qts and Vas are driver Thiele-Small parameters and Vb is the box volume.
Many box programs just use the above simple formula to determine Qtc.
The Hornresp method should in theory give more accurate results.
If the horn is considered to be a non-linear acoustic transmission line then the electrical analogue would be a non-linear electrical transmission line made up of capacitive, inductive and resistive distributed elements. If the horn is considered to be an acoustic transformer then the equivalent would be an electrical voltage transformer.
So it's a math thing. Now I know, haha!
Thank you very much for setting this straight. I've had great results with Hornresp simulating (and building) transmissionline-type enclosures in the past so I will take a leap of faith and trust it with sealed, too.
Have a great weekend!
It is very easy to design a closed-box system using Hornresp. Simply specify a suitable driver, select Tools > Design Wizard > Closed Box from the input parameters window, and if necessary adjust the Vrc slider value to "fine tune" the response to your satisfaction.
Interesting answer. It must a better of perspective, sorta like how FR and Phase are the same thing but different looks
Would I be correct using thee same analogy and call a Bass Reflex an acoustic transformer as well?
I like to look at it, in regards to impedance. Quarter wave and Helm Holtz are both resonating systems that have similar acoustical impedance characteristics, which represent the high energy transfer or coupling of the diaphragm to an airmass.
Last edited:
Acoustically a bass-reflex loudspeaker acts as a Helmholtz resonator. The electrical equivalent would be a RLC resonant circuit as shown below, where:
R = resistance
L = inductance
C = capacitance
Box:
L1 = mass loading on rear side of diaphragm
R1 = box resistance
C1 = box compliance
Port tube:
R2 = resistance of air in port tube
L2 = mass of air in port tube
Port radiation:
R3 = radiation resistance
L3 = radiation mass
Given that the pressure and volume velocity on the rear side of the diaphragm (input) are different to the pressure and volume velocity at the port outlet (output), I guess it could be argued that the values have been "acoustically transformed" from input to output...
The analogy of the horn as a transformer can be easy to misunderstand, as an electrical transformer is a single, relatively wide-band component. But a horn transform impedance over distance, and the transformation is dependent on the wavelength compared to the distance traveled, and the rate of expansion per wavelength. So the transformation ratio is frequency dependent, and you also get other effects you won't see in an electrical transformer, like time delay and reflections from the mouth. So a transmission line model is a better description, even if the impedance is transformed from one end to the other.
If we ignore the fact that waves do not behave in a one-dimensional way in horns at high frequencies, it is possible to set up equations for a transmission line where the characteristic impedance varies along the length (what David called a non-linear transmission line). It can be set up as a set of equations relating pressure and volume velocity at one end to pressure and volume velocity at the other end. (Acoustic impedance equals pressure divided by volume velocity.) But it is also possible to model this as a series of LC lowpass networks (with an R if there are losses), where the L/C ratio varies along the horn. But this model has an upper frequency limit depending on how finely divided it is (LC sections per wavelength). The set of equations work with a continuous model and has no such upper limit (but at high frequencies you run into the limitation that wave propagation in horns is not one-dimensional anymore).
A bass reflex enclosure has only a single LC section: the port (L, air mass) and the enclosure (C, air compliance). So it only "transforms" over a very small frequency range: the port tuning frequency.
If we ignore the fact that waves do not behave in a one-dimensional way in horns at high frequencies, it is possible to set up equations for a transmission line where the characteristic impedance varies along the length (what David called a non-linear transmission line). It can be set up as a set of equations relating pressure and volume velocity at one end to pressure and volume velocity at the other end. (Acoustic impedance equals pressure divided by volume velocity.) But it is also possible to model this as a series of LC lowpass networks (with an R if there are losses), where the L/C ratio varies along the horn. But this model has an upper frequency limit depending on how finely divided it is (LC sections per wavelength). The set of equations work with a continuous model and has no such upper limit (but at high frequencies you run into the limitation that wave propagation in horns is not one-dimensional anymore).
A bass reflex enclosure has only a single LC section: the port (L, air mass) and the enclosure (C, air compliance). So it only "transforms" over a very small frequency range: the port tuning frequency.
@David McBean Thank you so much. Can you display the Quarter Wave circuit, in the same format, for comparison?
The electrical equivalent of an acoustic quarter wave resonator would be an open circuit transmission line stub as shown below. If lossless there would be no resistive elements.
If you require further information, Google is your friend.
(Hornresp does not use electrical transmission line models).
If you require further information, Google is your friend.
(Hornresp does not use electrical transmission line models).
For a bass reflex system, the chart shows the acoustical radiation impedance loading the front side of the diaphragm. The frequency at which resistance Ra equals reactance Xa depends on the radius of the diaphragm and the solid angle into which the driver radiates. The diaphragm would need to have an unrealistic radius of 1.7 metres and radiate into 0.5 x Pi space for the frequency to be reduced to 22 Hz, as shown in the attachment. System topology has no effect on radiation impedance - the only parameter values that matter are those of Ang and Sd.
For a horn system, the chart shows throat acoustical impedance, quite a different thing.
🤦♂️Ty I should of known. But I cannot figure this one, I bet its easy for you
In the above graph in a Vented line, I only changed the sd. One is 600 and one is 1200... why do we see this higher efficiency of the 600cm2 example ~15hz-34hz?
Below shows 1200sd on top of 600sd
Here with damping material, the relationship changes....
Below, without damping material, the huge peak at tuning.
In the above graph in a Vented line, I only changed the sd. One is 600 and one is 1200... why do we see this higher efficiency of the 600cm2 example ~15hz-34hz?
Below shows 1200sd on top of 600sd
Here with damping material, the relationship changes....
Below, without damping material, the huge peak at tuning.
Last edited:
Changing the value of Sd changes the acoustical impedance loading on the diaphragm. The effect that this has on power conversion efficiency depends upon the frequency and the values of all the other driver and enclosure parameters defining the system.
Increasing the value of Sd could be expected in general to improve the conversion efficiency, but in the vicinity of the system resonance frequency the situation can be reversed, as you have observed.
The calculations required to determine power conversion efficiency are extensive and complex, and everything interacts, so it is not possible to manually identify a specific reason why the results are what they are, for any particular case.
Hi,
I am trying to do a simulation on a Uluwatu type speaker with multi driver TL with an offset port at the middle. Is this something that Hornresp can do? I am only a basic user and new to this. If someone can point me to some posting on how this can be done, it would be very helpful....
Many thanks in advance
Oon
I am trying to do a simulation on a Uluwatu type speaker with multi driver TL with an offset port at the middle. Is this something that Hornresp can do? I am only a basic user and new to this. If someone can point me to some posting on how this can be done, it would be very helpful....
Many thanks in advance
Oon