
Home  Forums  Rules  Articles  The diyAudio Store  Gallery  Blogs  Register  Donations  FAQ  Calendar  Search  Today's Posts  Mark Forums Read  Search 

Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving 

Thread Tools  Search this Thread 
30th August 2012, 10:23 PM  #141 
diyAudio Member
Join Date: Jan 2006

while you can use a th higher than two octaves from the low corner and the response is alot cleaner than hornresp suggests, its still less than ideal sound quality wise. see the impulse spectogram (ctrl+h in spl response window) to see what i mean. compare it to a closed box for giggles. just from experience, i didnt mind using my 30hz th's up to 200hz, they measured fine too, but listenening to them alone did sound slightly weird "hollow" with higher tones.
also, you set the segments to PAR, which should be quite difficult to realize. i prefer conical for the straight walls. 
31st August 2012, 07:18 AM  #142  
diyAudio Member
Join Date: Jun 2007

Quote:
Just to clarify, if a rectangular crosssection horn segment has two parallel straight walls and two nonparallel straight walls, then the actual area expansion will be parabolic not conical. It is therefore more accurate to specify Par rather than Con in this case. Kind regards, David
__________________
www.hornresp.net 

31st August 2012, 11:16 AM  #143 
diyAudio Member
Join Date: Jan 2006

thanks
then i have to say, it doesnt seem to matter, since i did it wrong with several tapped horn subwoofers and didnt realize it, despite measuring alot. i guess room modes >> build accuracy, even in nearfield. edit: let me use this opportunity to say thank you for your amazing program and your support! both has helped me so much with this hobby. Last edited by MaVo; 31st August 2012 at 11:22 AM. 
1st September 2012, 06:16 AM  #145  
diyAudio Member
Join Date: Jun 2007

Hi MaVo,
Quote:
Quote:
Kind regards, David
__________________
www.hornresp.net 

2nd September 2012, 08:02 AM  #146  
diyAudio Member

Quote:
Could you please clarify why the area expansion is parabolic? If S is the cross section area and a (constant) and b (linearly expanded) are the rectangular dimensions, then: S=a*b. If b=k*x (where k is a constant and x the horn axis) then S=a*k*x, but a*k is also a constant so it appears to me that S expands conically? What I am missing? Thank you, 

2nd September 2012, 09:06 AM  #147 
diyAudio Member
Join Date: Jun 2007

Hi ConExp,
Sorry, my mistake  I was not thinking too clearly at the time :). It is the equivalent axisymmetric radius that has a parabolic expansion rate, not the area. The plane crosssectional area of an axisymmetric parabolic horn will expand conically with axial length  the same as for a rectangular horn segment having two parallel straight walls and two nonparallel straight walls. This is why I suggested using Par rather than Con. My apologies for any confusion caused. Kind regards, David
__________________
www.hornresp.net 
3rd September 2012, 04:50 AM  #148  
diyAudio Member
Join Date: Jun 2007

Quote:
Attachment 1 shows how the plane crosssectional area of a conical horn varies nonlinearly with axial length. S = S1 * (x ^ 2) Where S is the area at distance x from the cone origin or vertex, and S1 is the area at distance x = 1 from the origin. Attachment 2 shows how the plane crosssectional area of a parabolic horn varies linearly with axial length. S = S1 * x Where S is the area at distance x from the parabola origin, and S1 is the area at distance x = 1 from the origin. Kind regards, David
__________________
www.hornresp.net 

3rd September 2012, 09:05 AM  #149 
diyAudio Member
Join Date: Jul 2004
Location: Scottish Borders

David,
my head must have got twisted back to front when I pulled on my pullover this morning. Could you explain that for me.
__________________
regards Andrew T. 
3rd September 2012, 05:15 PM  #150  
diyAudio Member
Join Date: Nov 2007
Location: München, Bavaria, Germany

Quote:
As the area of your conical horn increases by a linear factor of the length in x dimesion, and in y dimension, the area increases by a multiple of this, therefor a quadratic factor of the length. In order for the area to increase linearly, you have to decrease the rate of expansion the further out you move, in an (inverse) parabolic fashion. As hornresp simulates the horns as having circular cross sections, you need this parabolic expansion (that is: width expands with length^0.5 and height expands with length^0.5, leading to a total expansion rate constant with length^1). Another way to achieve this expansion rate is to have one dimesion (eg height) expand linearly with the length, but keep the other dimesion constant (width of the horn stays the same). Total area expansion is then still linear. Therefore to have the same rate of area expansion for these typical horns with 2 parallel sides, you need to simulate them as Par.
__________________
Regards, James 

Thread Tools  Search this Thread 


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Tapped Horn For Car  mwmkravchenko  Subwoofers  365  17th November 2011 10:22 AM 
help! LARGE tapped horn  bubblersound  Subwoofers  3  18th June 2009 05:17 PM 
Tapped Horn behavior  screamersusa  Subwoofers  32  21st October 2008 03:31 AM 
New To Site?  Need Help? 