Hornresp

[revintage]

Think the guy also had a TH calculator in the same spreadsheet. Will search!

EDIT: Sorry I was wrong, it only had f1 and f2 below the parameter input.

http://www.diyaudio.com/forums/subwoofers/176089-easy-horn-excel-spreadsheet-5.html#post2375408

 

[David McBean]

http://www.diyaudio.com/forums/subwoofers/176089-easy-horn-excel-spreadsheet-5.html#post2375408

Many thanks Lars.

On first glance, it does seem that this spreadsheet can in fact solve for f2, given T and f1. I will study the contents with great interest.

Kind regards,

David

 

[David McBean]

On first glance, it does seem that this spreadsheet can in fact solve for f2, given T and f1. I will study the contents with great interest.

Hi Lars,

I must be working too hard...

It turns out that I was the one who originally suggested to soho54 how to automatically derive f2 from f1 and T .

http://www.diyaudio.com/forums/subwoofers/176089-easy-horn-excel-spreadsheet.html#post2345159

I thought that the spreadsheet looked familiar .

Kind regards,

David

 

[revintage]

Probably it works like my basic but soho54 has taken it a big step forward with the "Solve HC" function.

 

[revintage]

As I remember it my sheet was OK against HR.

Also found the VASversion you asked for earlier:

http://www.revintage.se/subbashornVAS.xls

 

[David McBean]

Also found the VASversion you asked for earlier:

Many thanks Lars.

The next release of Hornresp will include the f1 + T input option. The required value of f2 will be automatically calculated using an iterative process. T input values will be restricted to the range 0.5 to 0.8 - designs become increasingly impractical outside these limits.

Kind regards,

David

 

[revintage]

 

[Kravchenko_Audio]

From the recently swept dark recesses of David's mind!

Constantly impressive work.

Nice.

 

[AndrewT]

When does D.McB get time to live?
For the last couple of years I have only been downloading about 10% of the updated software.
Thanks again.

 

[mashaffer]

Great tool this. Thanks David. Sometimes I end up getting to bed way too late due to spending long hours what if'ing and learning in the process. A few observations.

- It seems like all of the cone drivers I have modeled really like a lot of compression in order to get smooth response and any kind of high end response (at least with respect to mid-range horns). SD to throat ratios on the order of five or ten to one seem common. Nothing at all like all those pictures I see of horns just touching the outer edges of the cones.

- Tractrix horns seem to be really, really, picky.

- The compression chamber and rear chambers can make a huge difference in the shape and smoothness of the pass band.

- The loudspeaker wizard is fabulous. It is a great way to see how some of the myriad of different design parameters affect the results. It would be great if it could be used with exponential and tractrix horns too. BTW, is it possible to get a little bit longer x axis on the SPL graphs in the wizard so that the effects on the upper mid-range could be seen?

Again David thank you so much for you indispensable tool and your over the top devotion to it's growth.

 

[revintage]

It seems like all of the cone drivers I have modeled really like a lot of compression in order to get smooth response and any kind of high end response (at least with respect to mid-range horns). SD to throat ratios on the order of five or ten to one seem common. Nothing at all like all those pictures I see of horns just touching the outer edges of the cones.

Go for less "Hi-Tech" drivers if we are talking basshorns. There are GPA, Altec, JBL models that don´t need more than ca 2:1. There are also low budget drivers available. Don´t expect any basshorn to reach more than 4-500Hz. Shouldn´t be a problem as midbass horns normally go a lot lower.

If we are talking midbass/midrange horns take a look in the

Best 8" to 12" driver for 130hz tractrix horn

to get a lot of suggestions for low compression drivers.

 

[David McBean]

Hornresp Update

Hi Everyone,

The System Design With Driver tool now has default inputs f1 and f2 and optional inputs f1 and T. The f1 and T input option can be toggled on and off by double-clicking the f2 "Upper rolloff corner frequency" label. My thanks to Lars for suggesting this enhancement.

The latest Hornresp release is Product Number 2870-110913.

Kind regards,

David

 

[David McBean]

Hi Mike,

The loudspeaker wizard is fabulous. It is a great way to see how some of the myriad of different design parameters affect the results. It would be great if it could be used with exponential and tractrix horns too. BTW, is it possible to get a little bit longer x axis on the SPL graphs in the wizard so that the effects on the upper mid-range could be seen?

Thanks for your kind words .

As indicated in the Hornresp Help file, the Loudspeaker Wizard tool is applicable to all multiple segment horns. To model a single segment exponential horn in the Wizard simply insert a very short cylindrical segment at the throat. This then creates a multiple segment horn for the purposes of the Wizard, but the results remain effectively identical to those for a single segment horn.

For example:

Single segment exponential horn:

S1 = 100
S2 = 10000
L12 = 100

Wizard input parameters:

S1 =100
S2 =100
S3 = 10000
L12 =0.01
L23 =100

Of necessity, the Wizard is restricted to plane wavefront models. Calculation times associated with isophase models are just too great for the tool to update correctly, without the sliders "freezing". Tractrix horns are analysed using isophase models, which is why the Wizard cannot work with this horn type.

The Wizard is really intended for use with bass horns. The frequency range has been restricted to 10 - 2000 hertz to ensure that calculation update times are acceptable, and that the abovementioned "freezing" problem does not occur.

When I get a chance, I will have a look at how practical it would be to allow the frequency range to be toggled between 10 - 2000 hertz and 100 - 20000 hertz, and what other problems might arise as a result of doing this. I am not promising anything though .

Kind regards,

David

 

[revintage]

Hello David,
This works great. I get max 2-5% differencies against my spreadsheet trying a few models. Do you have a driver that has 100%(?) right T/S-parameters to try with?

 

[mashaffer]

Thanks for the pointer Lars. I found the thread (in multi way forum) and it looks to be worth some time to digest it. Much appreciated. I was looking at the 300 to 5k range primarily. I have noticed that bass horns seem to need far less compression in general.

BTW in a related item if one were crossing to a bass horn at or under 500Hz would there be any sacrifice in stereo image etc by having the sound source (for the bass) coming out the side or rear of the enclosure rather than the front?

David, thanks for the tips on the wizard tool! I wondered about the calculation speed issue as, at least on my antique computer system, the calculation time for hyp and tra are significantly longer. As to the range of output I wonder if it could be done as a range of say 1.2 decades beginning at a user defined low end. However if your toggle is easier that would be just ducky too. As it is I still found it useful as I could get in the ball park by observing the curve approaching the 2k limit and then tweak from the main screen.

Not being able to do a tractrix in the wizard is not really a hardship as there are tractrix calculators out there that seem to do a pretty good job of providing a reasonable starting point. Also it seems that the exponential has some advantages over the tractrix in BW especially on the bottom end.

 

[revintage]

Not being able to do a tractrix in the wizard is not really a hardship as there are tractrix calculators out there that seem to do a pretty good job of providing a reasonable starting point.

Hornresp calculates Tx, S, L etc. better than any other existing program! It also gives you templates for almost any shape you like, except radialhorns. So forget all the others and dig deeper into HR instead! Remember these horntypes should normally not be used as basshorns.

 

[David McBean]

I get max 2-5% differencies against my spreadsheet trying a few models. Do you have a driver that has 100%(?) right T/S-parameters to try with?

Hi Lars,

I suspect that it may be very difficult (if not impossible) to come up with a set of electro-mechanical parameter values specified to two decimal places that would generate equivalent T-S parameter values exactly matching a given set.

However - if we use the default values in your VAS spreadsheet as a test example, the results become identical to those calculated by Hornresp when the following changes are made:

In Hornresp:

* Internally set Fs, Qes, Qms and Vas to the exact values as given in the spreadsheet.
* Internally set Qmc to the value of 10 as given in the spreadsheet - rather than calculating the value as Sqrt(1 + Alpha) * Qms.
* Change to using a rounded factor of 0.686 rather than the more exact value of 4 / (3 + 2 * Sqrt(2)).

In the spreadsheet:

* Change Fh from 208 hertz to the more exact value calculated by Hornresp.

Kind regards,

David

 

[David McBean]

Hornresp calculates Tx, S, L etc. better than any other existing program.

The Horn Segment Wizard tool can be used to quickly generate a suitable design.

Kind regards,

David

 

[David McBean]

I will have a look at how practical it would be to allow the frequency range to be toggled between 10 - 2000 hertz and 100 - 20000 hertz.

The functionality outlined above will be included in the next release.

Kind regards,

David

 

[mashaffer]

Cool, looking forward to it.