Another set of interesting results.
The TL looks a bit oversized in image 1 (BS022), a bit oversized in image 2 (MK2006), undersized (again) in image 3 (MK2021A), and Ok in images 4 and 5 (MK2021B and MK2021C). In fact, what's happening in 1 is along the lines of what I suspect would happen, because the equations for Vb that I've been providing assume a Ql of 7, and the Hornresp sim does not include the impact of losses, so output around Fb would be a bit more in the sim than what would be predicted by the equations.
I'll bet that using Vb=20*Qts^3.3*Vas for my method might produce better results in that first image.
I plotted all those responses in one graph using REW:
...and again, with the offsets slightly modified to smooth the passband response...
The "sweet spot" seems to be somewhere between BW=2 and BW=3 Above that the TL appears slightly undersized and/or tuned too low and below that slightly oversized and/or tuned too high. Hmm, maybe the ideal box volume or Fb does not remain constant as the taper is increased? Interesting. And I'll bet that has something to do with the changing value for the end correction.
if you reducing an area when velocity is greater vs reducing area when pressure is…🤷wasted use of space?
kind of the same as stuffing polyfil , change in shape needs to be done on the ars of greatest ‘bottle neck affect’ so to speak (its the pmax/vmax for the #3 harmonic )
a constant taper is not the best idea nor is random locations of folding it seems. a great example of this is either expansion or reduction where the two initial holds are held to a certain cross-section and the third is the only one that changes.
This becomes even more obvious when using the other side of the driver in pH one mode and its ability to allow distinct offsets in both sides be studied
kind of the same as stuffing polyfil , change in shape needs to be done on the ars of greatest ‘bottle neck affect’ so to speak (its the pmax/vmax for the #3 harmonic )
a constant taper is not the best idea nor is random locations of folding it seems. a great example of this is either expansion or reduction where the two initial holds are held to a certain cross-section and the third is the only one that changes.
This becomes even more obvious when using the other side of the driver in pH one mode and its ability to allow distinct offsets in both sides be studied
The fs, Vas and Qts values are calculated from the electro-mechanical driver parameter values entered on the main input parameters window. The Thiele-Small parameter values shown are rounded to two decimal places. The actual values used in the BS 2022 calculations are:
fs = 28.0001303902894
Qts = 0.394752840577462
Vas = 71.02529767152
Using the above values in Vb = 20 * Qts ^ 3 * Vas:
Vb = 87.3813700326159 (or 87.381 litres when rounded to 3 decimal places)
For BW = 1 the transmission line volume shown on the schematic diagram is 87.382 litres.
For BW = 5 the transmission line volume shown on the schematic diagram is 87.380 litres.
Both are pretty close to the target of 87.381 litres calculated above. The slight differences in the third decimal place are due to the rounding of the S1, S3 and L13 input values used to calculate the schematic diagram volume.
The actual electro-mechanical driver parameter values being used are:
Sd = 255.00
Bl = 7.64
Cms = 7.66E-04
Rms = 1.11
Mmd = 39.83
L= 0.29
Re = 3.30
How are you determining the "actual" Fb?
By inserting the schematic diagram volume as Vb into equation Fb = (Vas / Vb) ^ 0.31 * Fs, or by looking at the diaphragm displacement?
All three alignments in MK 2021 use common code, the only thing that differs is the Vance Dickason table data. MK 2021 A uses the data from Table 2.3 (SBB4 and BB4 QL = 15). The data has been double-checked to ensure that it was transcribed correctly. No errors were found.
You guessed correctly. Do you still want me to change back to the original Vb=20*Qts^3.3*Vas equation and generate a new set of results?
Rear chamber acoustical lining is specified using Fr and Tal.
Alternatively, rear chamber resonances can be "hidden" by selecting the 'resonances masked' option.
I would like to suggest a new feature:
In the "Loudspeaker Wizard" window, I think it would be very convenient to have the option to see simultaneously the graphs of the first drop-down, for example to see at the same time the "Shematic" and "Power" graphs.
Nowadays the screens of our computers are large and with good resolution, maybe the "Loudspeaker Wizard" window could be converted into something more open, and in this way quickly observe many graphs.
Since I have dared to suggest, I think it would be good to have the alternative to vary directly with the keyboard the values of the sliders.
Thanks.
In the "Loudspeaker Wizard" window, I think it would be very convenient to have the option to see simultaneously the graphs of the first drop-down, for example to see at the same time the "Shematic" and "Power" graphs.
Nowadays the screens of our computers are large and with good resolution, maybe the "Loudspeaker Wizard" window could be converted into something more open, and in this way quickly observe many graphs.
Since I have dared to suggest, I think it would be good to have the alternative to vary directly with the keyboard the values of the sliders.
Thanks.
I'm wondering if anyone can help, I've been able to get a response and electrical impedance but have noticed for all my designs the acoustic impedance is right in the operating range, I am unsure how I shift this to either reduce or move to outside the operating range. Is it something I can shift by changing the horn geometry or is this related to the driver?
Furthermore, if anyone is able to explain the effect of acoustic impedance on the SPL as if I understand correctly acoustic impedance should reduce the cone displacement and thus create a dip in the SPL, i have noticed this doesn't reflect on the SPL graph on horn resp so wonder if its the biggest issue or not??
Any help is greatly appreciated
TIA
Furthermore, if anyone is able to explain the effect of acoustic impedance on the SPL as if I understand correctly acoustic impedance should reduce the cone displacement and thus create a dip in the SPL, i have noticed this doesn't reflect on the SPL graph on horn resp so wonder if its the biggest issue or not??
Any help is greatly appreciated
TIA
Where do we go from here?
Is Qts^3.3 going to be used rather than Qts^3?
Are there any other changes that need to be made before the new BS 2022 method can be released?
Do you perhaps want more time to think about things further, before an update is released?
Sorry, it's not going to happen 🙂.
That can already be done. Mouse-click on the slider to give it the focus, and then type the value and press the enter key.
I think that would be useful, yes. If MK's 2021 method uses the LDC tables, the results should be pretty close to the ones predicted by the MK method.
Google 'loudspeaker acoustic impedance'.
You can safely disregard the acoustical impedance chart in Hornresp if that makes things easier.
Ok great that was my concern whether the acoustic impedance would affect the horns function. thanks for the help
Let's see what changes using Qts^3.3 brings.
I'm curious about the response changes as taper is increased, but not concerned enough about them to put a pause on things because the changes in question are fairly small in the TL's usable passband.
I suspect the MK2021 method would show similar results too as it also calls for net size and Fb to remain unchanged as the taper rate and length of the TL is changed.
I will have a closer look at it when I get the chance, and if necessary I will suggest a modified method that makes slight adjustments to Vb as the taper is increased, as I suspect the differences might be due to changes in "effective volume", i.e. actual net volume + the volume of any "end correction" that needs to be applied. It's the volume of that "end correction" that changes as the TL's taper increases, so I suspect it has something to do with the response changes. I might be wrong though, and it won't be the first time 🙂.
As a quick summary, the resistive part of the acoustic impedance (the black line) is where the acoustic power is dissipated, i.e. the useful radiated energy. The reactive part (red line) mainly acts as added mass at low frequencies. The shape of the impedance curve depends on the shape of the horn.
But how this affects the response also depends a lot on the driver parameters. For a simple front loaded horn, you may imagine it as a voltage divider, where the loudspeaker driver makes up the series arm, and the horn load makes up the shunt arm. Depending on how the driver impedances (electrical and mechanical) and the acoustic impedance vary, the "voltage division ratio" will vary. (Although in this case, it wouldn't be voltage but pressure.)
Just saying "acoustic impedance should do this or that" isn't really meaningful, as it depends on the context. Like the statement "inductance should reduce HF output". It is true if the inductance is in series, but not if it's in parallel with the driver, and it also depends on the value of the inductance.
What the effect of the acoustic impedance is depends on how it interacts with the rest of the system. Also, a high acoustic resistance may reduce cone displacement, but it also enables the driver to deliver more power with less excursion, so the effect you see on the response curve may be small. The change in efficiency will be more obvious.
As David says, there's no need to worry about the acoustic impedance in most designs. Unless you're familiar with how it interacts with the rest of the system, the most useful application is probably to check the resonance frequencies.