Akabak & stuffing

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
just slight offtopic in my own thread.... I compared the akabak scripts for a tapped horn with 2 drivers, 1 with the 2 assumed at the same location (as it is in Hornresp) and 1 with the 2 drivers at separate locations (as the real thing)

this is with 1 shared location
 

Attachments

  • akabak spl v2 (2 drivers op zelfde plek).png
    akabak spl v2 (2 drivers op zelfde plek).png
    72.7 KB · Views: 528
I know MJK's sheets, but thanks anyway...

I started to look more into Akabak because it is able to simulate stuff even Martin's sheets cannot do, like a tapped horn and systems with mulitple drivers. (+ that Martin has withdrawn his sheets from the public domain, which leaves us with only Hornresp in combination with Akabak)

I also asked this question to Joerg panzer, the author of the program, and this is his reply:
Hm, difficult question. You see, the simulator Akabak assumes a more
abstract level. You can take into account losses by lowering the quality
factors as provided in the Enclosure and Duct components. However, the
correspondence between the effect of material and these quality factors is
not provided. Typically it is regarded as a difficult issue and in many
cases best controlled by your experience. The isothermal effect can be
taken into account by slightly increasing the volume of the component.
Theoretically the upper bound is 1.4.

Rule of thumb: Run the simulation with low losses in order to see
resonances clearly. Then play with the quality factors to learn about the
sensitivity of the response to these parameters.

PS: yes, line shape and driver parameters are more important, but the location and amount of stuffing can be the key to success, especially with a full range application
 
The standard viscosity is 1. When using a larger number you increase the viscosity.

I lined the throath and first section of the horn with polyester batting of around 2 cm thickness. When comparing the measured results with the calculations the 50 setting gave the best results. I also used a small amount of stuffing in the mouth area (20)

With these setting you can see a little bit the effect of the damping but as always you have to check the reality.
 
Sorry to resurrect a forgotten thread, but.....

I am using Akabak (excellent program, steep learning curve) to model some weird-ish stuff and I face the same problem with including stuffing in the simulations.

Is it kosher to bust the model into little segments and play with the visc parameter? I am anticipating making some physical mules to test this, and maybe work backwards to some viable approximation, but I do not wish to go into the wheel reinvention business. That area of activity is fully subscribed already....

Can anybody shed any definitive light on this subject?

I have done quite a bit of comparison between Akabak-with-viscosity and MJK worksheets (and hornresp, and generic T/S models too) modelling the same simple enclosures and I have achieved no sense of convergence at all. :confused:

I know Akabak works well to expose the underlying operational modes of more complex enclosures, but there comes a time when a little more subtlety is required and one of my 'thought experiments' really does rely on reasonably trustworthy modelling of damped volumes.

Any assistance is very welcome! :)

Keith Arnold
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.