Hornresp - Page 180 - diyAudio
Go Back   Home > Forums > Loudspeakers > Subwoofers

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
Reply
 
Thread Tools Search this Thread
Old 3rd April 2011, 11:56 AM   #1791
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by David McBean View Post
Hi Brian,

You obviously used a release of Hornresp older than Version 28.00 when doing your comparisons :-).

Kind regards,

David
Actually I'm using version 28.00, but I was using a compound horn config and S5-S6 set to 140 cm^2 to simulate the "vent" (see images below). Shouldn't this produce the same results as using Apt and Lpt? The schematic is basically identical.
Attached Images
File Type: jpg 20110403-vented-01.jpg (64.1 KB, 119 views)
File Type: jpg 20110403-vented-02.jpg (82.7 KB, 117 views)
  Reply With Quote
Old 3rd April 2011, 03:26 PM   #1792
tb46 is offline tb46  United States
diyAudio Member
 
Join Date: Jan 2006
Location: Texas
Default Helmholtz freq

Hi David,

When trying to duplicate Brian's results from Post #1788 the nice little "Helmholtz freq is ...Hz" (your Post #1790) disappeared when going from Lpt 96.90 to Lpt 97.00. This is with the newest version of Hornresp (2830-110403). In an older version (2830-110124) this was the point where this display changed to "Ltc is 0.00cm".

Hadn't noticed that before.

Regards,
Attached Images
File Type: jpg 2011_Apr03_Helmholtz_1.jpg (55.6 KB, 110 views)
__________________
Oliver
  Reply With Quote
Old 3rd April 2011, 07:43 PM   #1793
diyAudio Member
 
Join Date: Jan 2008
Quote:
Originally Posted by David McBean View Post
Hi Oliver,

Your comment above prompted me to do some further validation tests.

The SPL response of two William Cowan 60 hertz tapped horns connected in parallel and radiating into 2 x Pi half-space was compared against the results for a single similar system radiating into 1 x Pi quarter-space, and also against the results for a single similar system with two drivers connected in parallel and the cross-sectional areas doubled, radiating into 2 x Pi space. Eg was set to 2.83 volts and Rg set to 0 ohms in all three cases.

The SPL predictions are identical for the three different configurations. This would suggest to me that there can’t be too much wrong with Hornresp tapped horn multiple speakers model :-).

Kind regards,

David
I agree, not much wrong with the TH multiple speakers model, other than it does not agree with actual measurements of real cabinets in which the LF corner does not change in multiples.

Art
  Reply With Quote
Old 4th April 2011, 07:22 AM   #1794
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by Brian Steele View Post
I was using a compound horn config and S5-S6 set to 140 cm^2 to simulate the "vent" (see images below). Shouldn't this produce the same results as using Ap and Lpt?
Hi Brian,

Prior to Version 28.00, the results were the same. A tube specified by Ap and Lpt was assumed in all cases to act as a cylindrical horn, rather than possibly as a lumped air mass. Unfortunately this meant that predictions for conventional bass-reflex systems in particular, were not entirely accurate.

To address this shortcoming Hornresp now checks the Ap and Lpt dimensions and subject to certain conditions being met, assumes that the air in the tube is acting as a lumped mass. In such cases a Helmholtz frequency value will be given. Where the conditions are not met Hornresp assumes that the tube is acting as a cylindrical horn, as before.

When S5 and S6 are used, Hornresp assumes that the tube is always acting as a horn regardless of the S5, S6 and L56 dimensions.

As you have observed the results are different for the two cases.

With a bit of work I could probably make it so that the results for the two scenarios are consistent again, with the tube acting as a lumped mass rather than a cylindrical horn where appropriate - independent of how it is specified - but I am not sure that it is really worth the effort.

I guess I will have to think about it :-).

Kind regards,

David
__________________
www.hornresp.net
  Reply With Quote
Old 4th April 2011, 07:44 AM   #1795
diyAudio Member
 
Join Date: Jun 2007
Hi Oliver,

Quote:
Originally Posted by tb46 View Post
When trying to duplicate Brian's results from Post #1788 the nice little "Helmholtz freq is ...Hz" (your Post #1790) disappeared when going from Lpt 96.90 to Lpt 97.00.
This is because the tube length relative to cross-sectional area is such that the tube now acts more as a cylindrical horn than as a lumped air mass - see my reply to Brian above.

Quote:
Originally Posted by tb46 View Post
In an older version (2830-110124) this was the point where this display changed to "Ltc is 0.00cm".
I decided to remove the "Ltc is 0.00cm" message from this view because it is not really relevant and could be confusing to some users, since the throat chamber sliders are not present.

Quote:
Originally Posted by tb46 View Post
Hadn't noticed that before.
I'm impressed - you are very observant indeed :-).

Kind regards,

David
__________________
www.hornresp.net
  Reply With Quote
Old 4th April 2011, 08:27 AM   #1796
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by weltersys View Post
I agree, not much wrong with the TH multiple speakers model, other than it does not agree with actual measurements of real cabinets in which the LF corner does not change in multiples.
Hi Art,

I really don't understand why the results for multiple tapped horns should be any different to those for other speaker types.

Do you have an explanation for the difference - what reason(s) could there be for the "LF corner" not to move for multiple tapped horns, when apparently it does for all other speaker types?

Please excuse me for being a bit wary - at one stage a number of users believed that the Hornresp tapped horn diaphragm displacement predictions were also incorrect. Bjørn Kolbrek did some very careful and accurate test measurements to prove that this was simply not the case.

Kind regards,

David
__________________
www.hornresp.net
  Reply With Quote
Old 4th April 2011, 11:33 AM   #1797
diyAudio Member
 
Join Date: May 2007
Quote:
Hi Lars,

Some observations...

For identical inputs, the calculated Hornresp and subbashorn.xls spreadsheet results are effectively the same. The formula used to calculate system Vas in your spreadsheet is an approximation only - that is why there are slight differences in our results.

The spreadsheet assumes that the Mmd value in Hornresp will be the same as the driver Mms value. This is not the case.

Your spreadsheet requires the user to initially nominate values for m, Fl and Fh and then manually optimise to achieve an acceptable result. Hornresp does this automatically.

Kind regards,

David
Hi David,

Moved this from the other thread as it for some reason is closed.

I admit the Spreadsheet is wrongly written about Mmd in the export to Hornresp. Mmd should best be calculated in Hornresp. So I will remove the speaker data and only export the horn and volume data. But as in Marshall Leach AES paper, Mms is what I use in the spreadsheet. Maybe some light should be spread about this. Checked the paper a few days ago as my speakerinterest was newborn.

I admit, when you use Fl and Fh from the spreadsheet the results are identical and should be. But the Design function forces you, what values to be used and that is where I found it less useful. My spreadsheet might be less userfriendly but really optimizes values directly with the T and Fl you choose.

About Vas to my knowledge is not an approximation as it relies on the same Small/Thiele formulas. But I have also a version where you input Vas and can´t see any difference. I have also checked the calculated Vas against many speakers and they seem to be right.

Last edited by revintage; 4th April 2011 at 11:36 AM.
  Reply With Quote
Old 4th April 2011, 12:21 PM   #1798
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by David McBean View Post
Hi Brian,

Prior to Version 28.00, the results were the same. A tube specified by Ap and Lpt was assumed in all cases to act as a cylindrical horn, rather than possibly as a lumped air mass. Unfortunately this meant that predictions for conventional bass-reflex systems in particular, were not entirely accurate.

To address this shortcoming Hornresp now checks the Ap and Lpt dimensions and subject to certain conditions being met, assumes that the air in the tube is acting as a lumped mass. In such cases a Helmholtz frequency value will be given. Where the conditions are not met Hornresp assumes that the tube is acting as a cylindrical horn, as before.

When S5 and S6 are used, Hornresp assumes that the tube is always acting as a horn regardless of the S5, S6 and L56 dimensions.

As you have observed the results are different for the two cases.

With a bit of work I could probably make it so that the results for the two scenarios are consistent again, with the tube acting as a lumped mass rather than a cylindrical horn where appropriate - independent of how it is specified - but I am not sure that it is really worth the effort.

I guess I will have to think about it :-).

Kind regards,

David
Ah, I figured it was something along those lines.

Perhaps having HornResp give a warning that the "lumped mass" model might give better results once specific conditions are met might suffice. Or even an option to switch to the lumped-mass calculations (similar to the option to switch between Con, Hyp and Par calculations), might be enough.

Another approach might be a best-fit curve for the "end correction" that gives results more in line with the lumped-mass model when the conditions calls for it.

Bear in mind though that even the calculation of Fb for the lumped-mass model might be a little off, as emperical testing on another site suggests that the equations give values for Lv that may be as much as 20% too long. I aim to do my own emperical testing of predictions vs actual results once I get the opportunity, as that difference is a bit too much for my liking (unfortunately I'm way too busy these days to even look at the wood languishing in my garage awaiting my attention ).

Finally, this might be pure coincidence, bit I've noticed that there is much closer alignment between the horn model and the lumped mass model when the space is switched to 0.5*PI. I don't know if that's useful or not.
  Reply With Quote
Old 4th April 2011, 02:35 PM   #1799
diyAudio Member
 
Join Date: Jan 2008
Quote:
Originally Posted by David McBean View Post
Hi Art,

I really don't understand why the results for multiple tapped horns should be any different to those for other speaker types.

Do you have an explanation for the difference - what reason(s) could there be for the "LF corner" not to move for multiple tapped horns, when apparently it does for all other speaker types?

Please excuse me for being a bit wary - at one stage a number of users believed that the Hornresp tapped horn diaphragm displacement predictions were also incorrect. Bjørn Kolbrek did some very careful and accurate test measurements to prove that this was simply not the case.

Kind regards,

David
David,

I am far more an empiricist than a theorist, I only know what I have measured or have seen others measure. Thus far, I have seen no actual measurements indicating that the low corner changes on multiple TH, other than what would be accounted for by the increased frontal area multiple cabinets afford.

I have also confirmed that the Hornresp TH displacement predictions are correct, other than below Fb (Fc?) where the speaker suspension limits excursion to lesser values than predicted at high power, but that would apply to any design simulation.

Your response to Brian Steele in post 1794 may explain the difference between a TH and a regular horn far better than any explanation I could come up with.

It appears that TH as generally built may be more like two tubes acting as a lumped mass rather than a cylindrical horn.

TH-Tapped Horn, DDD-Dual Driven Duct, DLM-Dual Lumped Mass, name that cabinet....

Calling a tiny TH a horn seems to make about as much sense as calling a flute a trumpet, but once named, the name sticks whether it makes sense or not.

“Classic” horns drop at around 12 dB per octave below Fc, “classic” Theile-Small BR alignments drop at around 24 dB per octave below Fb.

Most TH below Fb (Fc?) seem to drop at an even steeper rate than TS BR alignments.
TH seem to share more in common with BR ( ducts with lumped air mass) than normal horns.

Like “TH”, BR cabinets low corner (F3) does not drop in multiples.


Art Welter
  Reply With Quote
Old 4th April 2011, 02:47 PM   #1800
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by weltersys View Post
Your response to Brian Steele in post 1794 may explain the difference between a TH and a regular horn far better than any explanation I could come up with.
If that was the case, HornResp's impedance response predictions will not match real life measurements for THs. I can confirm that this is not the case - the predicted impedance response is a very close match to what I've actually measured.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Hornresp Class FlipC Subwoofers 8 3rd November 2008 06:23 PM
Some questions about hornresp brsanko Full Range 4 18th October 2008 09:36 PM
Hornresp help / JX150 316a Multi-Way 0 11th February 2004 03:56 PM


New To Site? Need Help?

All times are GMT. The time now is 01:53 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2