(Kind of) unity horn/akabak gain-delay-phase help needed - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Multi-Way Conventional loudspeakers with crossovers

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 10th February 2016, 09:29 PM   #1
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
Default (Kind of) unity horn/akabak gain-delay-phase help needed

Hi,

I've tryed to sim nearly every to mix 2x4" and a 12" laying around, with hornresp until it offers multi-entry horn. So, the final project wich motivate me a lot :
_Full system around 110-130 liter.
_One 12" woofer in closed back and front bandpass horn loaded, with a long 33cm throat port going to the horn mouth, very near the two 4" mouth.
_The 4" loaded in very small box.
_Shared "horn".
_analog LR2 filters + all pass delay (+2 natural roll off )
_2 way active

With hornresp, i manage to tweak both to get a beautiful joining response with a 350hz LR2 high pass and a 390LR2 low pass, adding to natural roll offs of both kind of loading, but i'd like to adjust phase with delay to take account of the long bandpass path.
Sims lead me to big horn mouth allowing both virtual mouth really near.

Without understanding how to deal with different gain, delay and crossover with akabak now, i came to the sim shown. Up, only woofer connected. Middle, only the high way, and bellow, of course the brut summing. Akabak file joined. Don't understand why need half power with akabak for high...don't know if it deals well with the two speaker exported directly from hornresp.

I volontary roll off speaker response tweaking horn at 10khz, not to excite too much cone resonances, and with the idea of adding a fostex ft17h later.

Closed back bandpass front horn loading allow lower group delay for lows, maybe lower distortion, and . And 130L stay small for a bass cab.

24Vpp 108db => 4,8mm excursion on the 12" of 7mm xmax. 108db given for 12Vpp with Akabak => 1mm for the 4" of the 2,2mm xmax ( 108db with 24Vpp with LR2 in hornresp=> 0,7mm).
108db is only for headroom.

I've measured only the 12", for the 4", i use datasheet. The 4" are kind of full range with phase plug, not measured yet.

I use akabak on a old linux computer with wine, and hornresp on a recent windows.

Thanks for helping !
Attached Images
File Type: png Capture.png (38.8 KB, 63 views)
File Type: png Sans titre.png (326.3 KB, 62 views)
Attached Files
File Type: zip firstmix.aks.zip (1.7 KB, 1 views)

Last edited by papasteack; 10th February 2016 at 09:43 PM.
  Reply With Quote
Old Yesterday, 12:21 PM   #2
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
In Akabak, model each driver and its box and horn if applicable in its own 'System' on same script. In each system, apply the filter of your choice - LR2, BW4, etc high pass or low pass or both - use the manual to see examples. Time alignment can be done with simple delay statement in each system - good for DSP delays. All pass filters are also available but delays on that is tricker to work with. The first term of each filter is the gain. Set to 1 or 0.5 or 2 etc. it's linear gain so 0.5 is -3dB relative to overall global system.

The when plotting result - choose each system as one line red or green and plot total as black. Only 3 lines allowed so max is 2 way with overall or 3 way but no overall.

You can save traces as txt file and plot with another program like REW.

Here is an example of the output I can get for modeling a 2 way point source multiple entry horn as shown in the bookshelf horn thread:

Click the image to open in full size.

2 systems are used and it's the same waveguide in each system but with different drivers active and each with its own filter for the band pass on the woofer and a high pass on the mid/tweeter.

Last edited by xrk971; Yesterday at 12:26 PM.
  Reply With Quote
Old Yesterday, 08:07 PM   #3
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
THANKS !

I just had time to reworked the full script as suggested just after getting my first 3y old son to bed, and until now ready for the 9 month son's milk ^^
Compared to hornresp crossover sim, i just reduced high way gain to 0,8, added a delay coresponding to bandpass front chamber + throat lenght, without any stuffing added. Lot of place for tweaking more
Used same scale as your example to make comparison easier, simed at 24vpp.
Max 112db@woofer xmax 45-20khz +/- 3db @24vrms
Attached Images
File Type: png Capture-2.png (72.2 KB, 37 views)

Last edited by papasteack; Yesterday at 08:12 PM.
  Reply With Quote
Old Yesterday, 08:12 PM   #4
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
That looks better but you have a lot of HF leakage from your woofer. Are you running a low pass filter or just relying on the acoustical bandpass action? Try adding a 12dB/octave Butterworth low pass at circa 250Hz.
  Reply With Quote
Old Yesterday, 08:35 PM   #5
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
Well view ! Ooops, low pass was disactived !
Tweaked a bit and came to LP LR2@450hz HP LR2@280Hz.
Need to deal with natural low pass, so can't go lower than 450hz, nor 280hz. There's a small bandwith windows were it can work flat.
Will have to well "flare" bandpass port to reduce those 2 spikes, and add front and rear volume stuffing.

EDIT : added group delay plot ^^
Edit bis : added phase (even if i don't really know how to interpret...)
Attached Images
File Type: png Capture-3.png (71.0 KB, 33 views)
File Type: png Capture-4.png (74.6 KB, 14 views)
File Type: png Capture-6.png (54.3 KB, 11 views)

Last edited by papasteack; Yesterday at 08:52 PM.
  Reply With Quote
Old Yesterday, 09:03 PM   #6
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Can you show a diagram or sketch of the design? I don't understand why the woofer LP filter is 450Hz when it looks like the natural bandpass is 200Hz? And you say you have a bandpass port spike at 450Hz? I guess you have a rather long bandpass port? What you might want to do is adjust front chamber volume on woofer to be smaller and port length shorter but area smaller. Kind of like how a flat board with a hole is placed in front of a woofer. You can even put a volume filler plug behind baffle and between cone to bring volume lower and push bandwidth extension higher up. That will let you have a smoother Xo workout resonance peaks.
  Reply With Quote
Old Yesterday, 09:54 PM   #7
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
Thanks a lot X ! Partial diagram is shown in first post in hornresp capture screen. About bandpass, it's around 23liter back chamber,
, front chamber is yet only 3cm long for around 17liter, so needs "plug" yet...Port is around 120cm2/30cm, and still is small for tuning/power. Cannot do a lot better without trying to sim a really flared port.
In hornresp sim, it take account of back chamber stuffing wich yet helps dealing with it.

I have to deal with the high woofer inductance, and so the natural bandpass is far from perfect. If i lower the electric low pass, i lost too much bandwith, and cannot compensate with the fullrange (fs=125hz qts=0,4). It's really fast and intuitive tweaking around crossover freq with hornresp, even without at first don't understanding fully what happen ^^ As shown in first akabak sim in first post without any filter, the "response summing windows" is really small.
But as said, i will have to loop between akabak/hornresp for futhermore tweak this all.
Just tested something : custom low pass factor 1, flatten even more response.

Last edited by papasteack; Yesterday at 10:17 PM.
  Reply With Quote
Old Today, 12:34 AM   #8
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Having a hard time understanding the diagram. I see what looks like a tiny chamber followed by a rapidly expanding horn and a huge offset driver connected to the throat where tiny chamber expands to horn?

A tip for plotting Akabak diagrams. In file menu there is a copy to bitmap option that saves a PNG file of active window. Then you don't need to do screen captures which gets the whole desktop - which you probably don't need to show.
  Reply With Quote
Old Today, 08:27 AM   #9
diyAudio Member
 
Join Date: Jan 2012
Location: Limousin
Sorry, i've been a bit lazy doing fast capture ^^
Here a schematic sketch
Attached Images
File Type: png rect3584-8-2.png (47.8 KB, 10 views)
  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
Unity horn script for akabak MaVo Multi-Way 24 22nd December 2012 05:24 PM
OPA549/541 at unity gain - schematic needed. nina Chip Amps 3 24th September 2007 03:46 PM
NI unity gain opamp equivalent but with discretes needed jondoe Solid State 5 10th December 2004 07:45 PM
Unity Gain Unbalanced to Balanced Phase Splitter runeight Tubes / Valves 10 28th August 2004 02:57 PM
Need explanations on WinISD Gain/phase/group delay graph Tired Multi-Way 1 23rd June 2002 06:26 AM


New To Site? Need Help?

All times are GMT. The time now is 12:35 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki