Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Hornresp
Hornresp
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 31st December 2017, 03:20 PM   #7981
Circlomanen is offline Circlomanen  Sweden
diyAudio Member
 
Join Date: Sep 2004
Location: Göteborg
After some months of experiments I had to make some measurements of my ROAR12 design.

ROAR12-lossyLE-spl.png

ROAR12-lossyLE-xmax.png

ROAR12-INPUT.png

With 28,3 volts input Hornesp predicts 4,4 mm one way cone excursion without lossy LE and 5 mm one way cone excursion with lossy LE enabled.

I used the signal generator in REW and a Tamp E800 to drive my B&C 12PS100 with 28,3 volts over the speaker terminals. I lightly pushed the the backside of a pencil held between my hand and the speaker magnet, against the cone and slowly increased the volume to 28,3 volts. The cone pushed the pencil outwards until it stopped buzzing. I tried very gently several times to get the end of the pencil closer to the moving cone without any buzzing noise.
With this simple crude measurement I could measure the cone excursion to slightly less then 1,0 mm. I round this measurement upwards to 1,0 mm to not exaggerate the difference between the Hornresp sim and my real world results.

I have been playing sine wave sweeps from 20 Hz to 90 Hz and I could not find this hump in the cone excursion anywhere. The ROAR design really holds the cone very firmly.

It seems Hornresp exaggerates the predicted cone excursion by 440% (8,8/2 mm) at 53 Hz and 28,3 volts (100 watt in 8 ohms) in my ROAR design. The difference grows to 500% with the use of Lossy Le enabled in Hornresp.

We have been testing several different drivers in several variations of my TPQWR designs and we have seen this behavior over and over again.

Yesterday we pushed a cheap 500 watt (peak) car audio driver to 2000 watts in a TPQWR and we could not get more then 10 mm p-p cone excursion. Hornresp predicts close to 60 mm p-p cone excursion. We did not measure the spl but this cheap car audio 12 inch driver made everything inside a 1000 m2 store rattle and buzz with ease.

This is the reason I asked if Hornresp could factor in non-linear aspects of large "slugs" of air in large ducts earlier in this thread. This seems to be a prime example of the non-linear behavior at high sound pressure levels.

Cheers,
Johannes
__________________
We live in a horizontal world, why use vertical topologies???
  Reply With Quote
Old 31st December 2017, 03:31 PM   #7982
Brian Steele is offline Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by Circlomanen View Post
After some months of experiments I had to make some measurements of my ROAR12 design.
Did you measure the impedance curve for your design? This should be the first step in trying to compare the results to a Hornresp sim. Typically the impedance peaks and dips should be close to or at the same frequencies in the Hornresp sim. The height of the peaks will be lower (Hornresp does not include the effect of box losses), and the position of the lowest impedance peak might be a bit different, but if the rest lines up, there's a pretty good chance that you've built what was sim'd.

As for excursion, it basically tracks the impedance curve. However, Hornresp assumes a flat BL curve (unlimited Xmax), so its predictions, particularly when excursion is approaching the driver's rated Xmax, should be taken with a grain of salt .

Did you do any distortion measurements for your build?
__________________
www.diysubwoofers.org
  Reply With Quote
Old 31st December 2017, 03:44 PM   #7983
Circlomanen is offline Circlomanen  Sweden
diyAudio Member
 
Join Date: Sep 2004
Location: Göteborg
Quote:
and the position of the lowest impedance peak might be a bit different, but if the rest lines up, there's a pretty good chance that you've built what was sim'd.
I have not measured the impedance peaks, but it does not seem to have a large effect wherever the peaks are. There is no measurable peaks in the cone excursion at 28,3 volts over the terminals over 35 Hz. Below 35 Hz the cone excursion starts to increase slowly and below 25 Hz it increases fast.

Cheers,
Johannes
__________________
We live in a horizontal world, why use vertical topologies???
  Reply With Quote
Old 1st January 2018, 04:11 AM   #7984
David McBean is offline David McBean  Australia
diyAudio Member
 
Join Date: Jun 2007
Default Hornresp Update 4240-180101

Hi Everyone,

BUG FIX

Further to post back loaded horn design, if a new data file was created by selecting the File > New menu while the Delete button on the Input Parameters window was enabled, the Delete button would remain enabled even though the newly-created file only contained a copy of the default record - see Attachment 1.

A fatal run-time error would be generated if the default record was then deleted by clicking the Delete button - see Attachment 2.

To ensure that this potential problem cannot occur, the Delete button is now always disabled following the creation of a new data file.

Kind regards,

David
Attached Images
File Type: png Attach_1.png (54.0 KB, 187 views)
File Type: png Attach_2.png (23.9 KB, 184 views)
__________________
www.hornresp.net
  Reply With Quote
Old 1st January 2018, 07:29 AM   #7985
David McBean is offline David McBean  Australia
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by Circlomanen View Post
It seems Hornresp exaggerates the predicted cone excursion by 440% (8,8/2 mm) at 53 Hz and 28,3 volts (100 watt in 8 ohms) in my ROAR design.
Hi Johannes,

"Small signal" driver parameter values are used in your ROAR design simulation. 28.3 volts is not a small input signal :-).

Back in 2008 Dr Bjørn Kolbrek showed quite conclusively using very accurate measurement techniques that when an appropriate input signal is used (in his case 2.83 volts rms), "the Hornresp predictions are very close to reality".

Hornresp

Kind regards,

David
__________________
www.hornresp.net
  Reply With Quote
Old 1st January 2018, 10:23 AM   #7986
Circlomanen is offline Circlomanen  Sweden
diyAudio Member
 
Join Date: Sep 2004
Location: Göteborg
Thank you David McBean!

Yes, I know Hornresp is very accurate at 1 watt aka small signal. I don't mean to criticize Hornresp in any way.
The reason I post this is to highlight the max spl prediction capability of Hornresp.
Usually it seems to be quite accurate and correct., but here it seems we have found a design that departs for this normal behavior in a very interesting way.
I find this discrepancy between Hornresp and my ROAR design very interesting. I am surprised by the stellar performance of the ROAR design at high power levels.

I do understand that I probably need to use COMSOL or something similar to properly simulate my ROAR design at high power levels. We are actually looking into this, but COMSOL has a quite steep learning gradient and a discouraging price tag.

Regards,
Johannes
__________________
We live in a horizontal world, why use vertical topologies???
  Reply With Quote
Old 1st January 2018, 07:25 PM   #7987
mwmkravchenko is offline mwmkravchenko  Canada
Mark Kravchenko
diyAudio Member
 
mwmkravchenko's Avatar
 
Join Date: Jan 2002
Location: Perth Canada
Hornresp
Quote:
Originally Posted by Circlomanen View Post
Thank you David McBean!

Yes, I know Hornresp is very accurate at 1 watt aka small signal. I don't mean to criticize Hornresp in any way.
The reason I post this is to highlight the max spl prediction capability of Hornresp.
Usually it seems to be quite accurate and correct., but here it seems we have found a design that departs for this normal behavior in a very interesting way.
I find this discrepancy between Hornresp and my ROAR design very interesting. I am surprised by the stellar performance of the ROAR design at high power levels.

I do understand that I probably need to use COMSOL or something similar to properly simulate my ROAR design at high power levels. We are actually looking into this, but COMSOL has a quite steep learning gradient and a discouraging price tag.

Regards,
Johannes

Your observations are visual?

If they are visual at what frequencies are you doing the observations?

Below 50 hertz you will be able to see the movement easily.

And this might help you get an accurate idea of the movement.

wedge-micro.gif
__________________
Mark
www.kravchenko-audio.com
  Reply With Quote
Old 2nd January 2018, 09:11 AM   #7988
Kolbrek is offline Kolbrek  Norway
diyAudio Member
 
Kolbrek's Avatar
 
Join Date: Aug 2007
Quote:
Originally Posted by Brian Steele View Post
As for excursion, it basically tracks the impedance curve. However, Hornresp assumes a flat BL curve (unlimited Xmax), so its predictions, particularly when excursion is approaching the driver's rated Xmax, should be taken with a grain of salt .
Also keep in mind that driver suspensions are nonlinear and gets stiffer as excursion increases. One reason is to prevent the coil from leaving the gap.

Hornresp does not model driver nonlinearity, which can have quite a large effect on the performance at high power levels. Suspension nonlinearity, BL nonlinearity, increase in Re with temperature, modulation of Le etc means that the small signal parameters will not give a good representation of driver behaviour at high levels.

My guess is that the difference between simulated and measured performance is most likely due to the shift in driver parameters, rather than nonlinear acoustics.
__________________
Dynamics to the people!
  Reply With Quote
Old 4th January 2018, 07:42 AM   #7989
Circlomanen is offline Circlomanen  Sweden
diyAudio Member
 
Join Date: Sep 2004
Location: Göteborg
Quote:
Below 50 hertz you will be able to see the movement easily.

And this might help you get an accurate idea of the movement.
Thanks for a brilliantly simple and useful idea!

I will try this as soon as I can. I hope I can make a video of the cone excursion at 28,3 watts using this simple little measuring method.

Quote:
Suspension nonlinearity, BL nonlinearity, increase in Re with temperature, modulation of Le etc means that the small signal parameters will not give a good representation of driver behaviour at high levels.
I don't believe a B&C 12PS100 has 7 dBs of power compression or Bl non-linearity at 100 watts. If so, it would be a very crappy driver for pro audio use. 28,3 volts into a nominal 8 ohm driver and 2 mm peak to peak cone excursion is not enough cone excursion to push a B&C 12PS100 into heavy suspension based mechanical limiting and compression or BL nonlinear compression.

I mention the cheap car audio driver and 2000 watts because I am impressed with the performance. This driver did show signs of thermal compression and BL nonlinearity, along with a bad smell of a red hot voice coil.

These measurements of cone excursion was done with my ROAR12 and my B&C 12PS100, not the cheap car audio driver.

Cheers,
Johannes
__________________
We live in a horizontal world, why use vertical topologies???

Last edited by Circlomanen; 4th January 2018 at 07:45 AM.
  Reply With Quote
Old 4th January 2018, 05:24 PM   #7990
NEO Dan is offline NEO Dan  United States
diyAudio Member
 
NEO Dan's Avatar
 
Join Date: Jun 2008
Location: N.E. Ohio
An LED flashlight with strobe mode can also be useful.
__________________
Regards,
Dan
  Reply With Quote

Reply


HornrespHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Hornresp Class FlipC Subwoofers 11 4th June 2016 06:33 AM
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 09:27 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.79%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki