Hornresp

Hey Dirkwright,
All data needed for simulations in Hornresp is in the original datasheet. Did some simulations with different hornmodels and they looked acceptable/useful. I also tried some compression and it looked even better.

How the driver/horn behaves at higher frequencies is already known so concenterate on the behaviour at lower frequencies. If you want the horn to be useful from 500Hz you should go for something like 300-3500Hz Fc.

Do you really think the sensitivity is good enough?

The data sheet I have does not have Cms, or Rms. It also has none of the rear chamber data.
 
D54AF in a 250Hz Tractrix
 

Attachments

  • 250D54.png
    250D54.png
    35.1 KB · Views: 236
I have the plain D54 driver. I bought this horn about 20 years ago from Bruce. He said it was intended for this driver. I believe the D54AF is slightly different than the one I have. I can't get any kind of reasonable SPL response with the data I have. I wonder if I should look for a different driver for this horn.
 
The D54 sims close to the same as the AF.

But if you have the horn and the driver why bother simulating it? Measure it IRL and listen instead;).

Well, I want to understand it. One problem I have is that the Vas isn't listed for this driver, so I can't use the spreadsheet I have to derive the tractrix horn shape. Also, if I put in the throat dimensions, which I assume are close to 2" square, the response in Hornresp is awful. If I reduce the throat size substantially, the response improves. I'm using:
Ang = free space
Expansion is Tractrix
S1 = 28
S2 = 368
F12 = 500
Sd = 28
BL = 8.1
Cms = .0004
Rms = .01
Mmd = 2.78
Re = 4.6
Le = .07
Vre = .01
Lrc = .1
Fr = 1
Tal = .1
Vtc, Atc = 0

The response doesn't look anything like what you showed. Thanks for your help!
 
I can convert it to pints though.
How many pint's before you forget it anyway?:drink:

Nice points you guys are making.

Running simulations on more difficult drivers and the tips to do so are very useful.

And I can vouch for the similarity between the two types of Dynaudio midranges. As far as I can remember the greatest difference was the shape of the face plate. One was flared with a partial horn loading and one used a flat face plate.

Mark
 
Vas=0,1dm3 is what I have found. But using the std formula I get 0,626dm3 (or is 0,0626dm3)?

Thanks so much!

I used 0.1 and got a throat area of 3.6 cm2, mouth area of 368 cm2. I used those in Hornresp and got the same kind of horrible response. If I reduced the throat size to something like 1 cm2, the response improves a lot. I'm pretty sure that Bruce's horn has a 2" square throat. I need to measure this beast in real life.
 
Since I am not of "continental" upbringing, then could you explain what a decimetre has to do with metric.
As far as I know it is neither cgs units (SI) nor MKs units (metric)

If decimetre is allowed as a well known and used but "continental" exception, then does the volume unit become cubic decimetre or dm^3 or something else?
 
How many pint's before you forget it anyway?:drink:

Nice points you guys are making.

Running simulations on more difficult drivers and the tips to do so are very useful.

And I can vouch for the similarity between the two types of Dynaudio midranges. As far as I can remember the greatest difference was the shape of the face plate. One was flared with a partial horn loading and one used a flat face plate.

Mark

I have the D54, which has the partial horn loaded faceplate. The D54 AF has flat faceplate. I have the data sheets for both and the T/S parameters are slightly different.
 
Hi, idont know where to post this
it is for david, ive found a bug, get this error simulating a rear vented horn, at the wizard, selecting the Ang to 0 PI in the slider...
my HR ver 28.40 prod # 2840-110505

Hi polakenfold,

Many thanks for the feedback - you have posted it in exactly the right place :). Amazingly by sheer coincidence this very same bug was reported to me via e-mail on 2 June 2011 by another user. The fault was fixed in Product Number 2840-110603 released on 3 June. All you need to do to overcome the problem is to download a new copy of Hornresp :).

The latest version is Product Number 2840-110609.

Kind regards,

David
 
The latest version is Product Number 2840-110609.

Hi Everyone,

I can’t resist tweaking things - the latest Hornresp release is 2840-110619.

When the Eg slider in the Loudspeaker Wizard is set to zero the Rg slider is now disabled and a "Constant Velocity" reminder message is displayed, as shown in the attached screenprint.

Kind regards,

David
 

Attachments

  • Wizard.png
    Wizard.png
    22.8 KB · Views: 243
David,
I have a question about the port used for rear loading (or the port of a reflex enclosure).

From searching this thread I have determined that end correction is included in the port length displayed. Is this correction included when exporting to AkAbak script? AkAbak does not include end correction by default. It is not required for the "outside" end of the port because the "radiator" takes care of it, but it appears that I need to add end correction for the "inside" end of the port. Do you agree, or have I totally misunderstood the physics?
 
Hi Don,

I have determined that end correction is included in the port length displayed.

The Lpt value shown is the actual physical length - it does not include the "inside" end correction, which is automatically added-on by Hornresp as required, internal to the program.

Is this correction included when exporting to AkAbak script?

No (because as stated above, the Lpt value is the actual physical length, not the length including the "inside" end correction).

AkAbak does not include end correction by default.

Correct.

It is not required for the "outside" end of the port because the "radiator" takes care of it, but it appears that I need to add end correction for the "inside" end of the port.

Correct, when using AkAbak.

Kind regards,

David
 
Thanks, David.
In summary, if I build an enclosure with the port length as specified by Hornresp, it should measure similarly to the simulated enclosure. But in AkAbak, I have to manually add one end correction to the length provided by Hornresp in order to get a correct simulation.

This was one of those "Doh!" moments. (I'm writing a guide which details how to simulate a Weems-style DCR in AkAbak, using a reflex box export from Hornresp as a template. End correction is important to accuracy of the simulation because there are multiple ports.)