Hornresp

To do something like that, you need to define what the goal is. Those "design equations", when given a driver's t/s parameters, provide a box size and tuning that's supposed to give the flattest response in the passband - that is the goal of those equations is. IRL though, using a maximally-flat build with a low Qts driver like what you see in pro audio is going to produce sub-optimum results.

So, if you want an optimization routine, you need to define what the goal is. Maximally-flat is probably not the best goal to aim for. So what should goal be?

Hi Brian,

In light of your comments, is there any point in me trying to include something in Hornresp? I was thinking that I could perhaps use the closed box and bass reflex formulas from your "The Subwoofer DIY Page" for the 'optimisation' tools requested by Marcelo. Would the results generated possibly lull users into a false sense of security though?

I won't proceed with the project if you feel it would be a waste of time... :).

Kind regards,

David
 
Interesting...

Those equations don't take into consideration Le, (and obviously don't consider semi-inductance either) and assume (in the case of the equations for the vented alignment) a Ql of 7.

To be honest, I haven't used them in several years, and was considering assigning them (and my ported.xls) spreadsheet to the archives as part of my ongoing site update :). I basically use Hornresp for all design work now.

I suppose that the equations could be used in Hornresp as a "starting point" for a simulation. i.e. if the user selects to do a vented enclosure:

1. Set starting Vrc to be 20*Qts^3.3*Vas
2. Set starting Lrc = Vrc^0.33 (assuming a square box)
3. Set starting Ap to be 1/3rd Sd
4. Set Fb to be (Vas/Vrc)^0.31*Fs
5. Set Lp to (23562.5*2*((Ap)/PI)/(Fb^2*Vrc))-(0.732*2*((Ap)/pi)^0.5)

#5 is likely to be the most troublesome one to implement, as the sim'd Fb might not work out to the the same as the calculated Fb (there have been some discussions in the past that the "standard" vent length equation results in a vent length that's longer than it actually needs to be). Another approach may be to set Lp to achieve the target Fb in Hornresp, but that might require some sort of "regression analysis" procedure to be used, which might be more trouble than it's worth.
 
Hi Brian,

Many thanks for your comments.

I am now of two minds - I will have to give it some serious thought before deciding where to go from here :).

A question for you:

I notice that you use Vb = 20 * Vas * Qts ^ 3.3 rather than the (seemingly) more common Vb = 15 * Vas * Qts ^ 2.87. What is the reason for this?

Kind regards,

David



Hi David,All

My take:

b:)
 

Attachments

  • 1_DaytonAudio_RSS390HO-4.jpg
    1_DaytonAudio_RSS390HO-4.jpg
    931.9 KB · Views: 172
  • 2_DaytonAudio_RSS390HO-4.jpg
    2_DaytonAudio_RSS390HO-4.jpg
    936.5 KB · Views: 153
Hi David,

Got a error repport ! :p
Strange result with wizard when changing FR1 and FR3 from 0 in this simple CH 6th BP parrallel.

I too got strange result somewhere else using multiple passive radiator, in wizard VS in calculated figures. I'll give some example later, but from what i recall , one of the two resulting frequency response seems to have 1 more PR.

:)
 

Attachments

  • strangeCHfilling.txt
    1 KB · Views: 38
  • 3.PNG
    3.PNG
    29.1 KB · Views: 150
  • 2.PNG
    2.PNG
    32.2 KB · Views: 160
  • 1.PNG
    1.PNG
    30.4 KB · Views: 178
Hi papasteack,

Strange result with wizard when changing FR1 and FR3 from 0 in this simple CH 6th BP parrallel.

Many thanks for the feedback. I see what you mean, there is definitely something wrong somewhere :). I need to investigate...

I too got strange result somewhere else using multiple passive radiator, in wizard VS in calculated figures. I'll give some example later, but from what i recall , one of the two resulting frequency response seems to have 1 more PR.

A specific example in due course would be very helpful, thanks.

Kind regards,

David
 
Ok, i can give you the clue for a first problem with response between wizard and calcul using passive radiators.
The problem is when adding mass to multiple passive radiators.
For this example i add 51g on two passive radiator. The result in wizard is the same as adding when 101g when calculating.
First capture is wizard, moving added mass from 51g (grey result) to 101g.
Second capture is calculated result.
From help, added mass is added mass to each cone. Which result is the good one ? 51g to each cone or 101 g ?

...next, another repport about filling, again ^^
 

Attachments

  • Wizard.PNG
    Wizard.PNG
    76.7 KB · Views: 54
  • Calculate.PNG
    Calculate.PNG
    40.9 KB · Views: 52
  • PR.txt
    1 KB · Views: 52
And with same record, i got strange result using filling, but i though at first it came from the use of false closed back with 0.01cm² pipe. But here i found the same result moving this port from 0,01cm² to 128-210 cm²... ~ 164db @ 180 hz...wow :rolleyes:
 

Attachments

  • largeshematic.PNG
    largeshematic.PNG
    57.6 KB · Views: 54
  • large port.PNG
    large port.PNG
    55.3 KB · Views: 50
Hi papasteack,

I have found what is causing the problem with the filling results. It is great that you picked it up, as the bug would have been there for quite some time.

The following statement appears in several parts of the program:

If CompoundHorn Or CH1Flag And N = 4 Then

For the logic to be correct the statement should actually have been written as:

If (CompoundHorn Or CH1Flag) And N = 4 Then

It will be fixed in the next release.

That's how easy it can be to wreck the results :).

I will have a look at the passive radiator added mass problem as soon as I can, but unfortunately Hornresp Santa won't be bringing you any other presents this year :).

Selecting a passive radiator in place of a segment by double clicking on L12-L45 would be just too messy to implement. Things are complicated enough already...

Kind regards,

David
 
Don't know about any AES/whatever paper, but he published 'Confessions of a Loudspeaker Engineer' with all this and much more plus basic XO filter design in the 8/78 issue of AUDIO.

GM

edit: Fortunately it's available: https://www.americanradiohistory.com/Archive-Audio/70s/Audio-1978-08.pdf





I totally missed this Greg!


Thanks for posting this. What a blast from the past. Even Neville Thiele has an article in this issue. I bought those and read them voraciously when Richard Heyser and Don Keele were doing loudspeaker testing for them.
 
Hi bjorno,

Many thanks for the great information - more things for me to take into account! If I was to include something in Hornresp, what would be the best alignment to choose? There seem to be so many options :).

Kind regards,

David

Hi David, once for Bass Reflex there are many options, you could start the analysis for closed box, it's simpler and may give you better indication of hornresp integration codes. If you approve the results then you cold focus next on BR (BP4, BP6 :D - improvements never ends)

Regards,

Marcelo
 
I notice that you use Vb = 20 * Vas * Qts ^ 3.3 rather than the (seemingly) more common Vb = 15 * Vas * Qts ^ 2.87.

In a quick scan didn't see where this came from, so FWIW: AES E-Library >> Personal Calculator Programs for Approximate Vented-Box and Closed-Box Loudspeaker System Design

I assume the 'why' is buried in the text; me, I prefer it since it specs a larger cab, which is far easier to 'shrink' if need be than if left 'wanting'. If nothing else it ~accounts for a certain amount of thermal power distortion [~10% IIRC] ........

GM