Go Back   Home > Forums > Loudspeakers > Multi-Way

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 26th April 2013, 02:00 PM   #491
diyAudio Member
 
Join Date: Apr 2005
Location: Sitting behind the 'puter screen, in Illinois, USA, planet earth
Hi Guys
So far as horn “pattern flip”, there isn’t a great deal written about it as it is a flaw or wart which is nearly unavoidable and not a “selling feature”.
An examination begins with Don Keele’s realization that the horn wall angle and dimension govern where (what frequency) it will lose pattern control. Also, the Horizontal and Vertical angles interact through this handy thumb rule relationship.

http://www.xlrtechs.com/dbkeele.com/...xp%20Horns.pdf

The upshot is that for a given straight walled horn (makes it easy to define the horn wall angle), if you make it half the angle, it will lose pattern control an octave higher.
That second angle is needed to avoid the narrowing he mentions as the frequency falls and you approach the pattern loss point.

What is less clear is that if one makes a simple horn, one is limited to a symmetric case unless one accepts pattern flip.
When the loudspeakers go places where you count on a radiation pattern shape, pattern flip is bad JuJu.

It is absolutely ludicrous in commercial sound that loudspeakers are given a pattern angle, which may only be achieved at one Frequency over it’s entire band (similar to how some rate sensitivity too).

The problem is clear when you do the math. What if you want a 20 degree by 80 degree horn?
No problem, you make a rectangular pyramid shape horn. This gives you the familiar wide short horn shape. Do the math though and one finds if you want to avoid pattern flip, the mouth height has to be 4 times the width while the real thing is more like the opposite of that, wide and short.

The problem is that if you loose pattern control in one plane, the control in the other plane pushes the energy into the uncontrolled direction. When in the flipped domain, the pattern will be narrow horizontally and wide vertically, even though the horn shape looks like it would do the opposite.
I have tried to avoid geometry which causes this problem but there are times when the package size and arrangement requires a compromise. For me, I more or less limit the asymmetry to around 1.6:1 because greater amounts produce much more severe pattern flip.
I said this is not a selling feature but a wart and while ubiquitous, is rarely discussed because of the problem fixing it.

You can see what a mild case looks like if you look at the clf file for an sh-95. This is a simple asymmetric horn which had to be small.

Open the CLF file, grab the radiation balloon with the cursor and swing it around and examine it on axis, pointed at you.

Notice up high, it is rectangular where the horn is large enough to set that pattern.

Now, step the frequency down from 1.6Khz and watch the shape change as it losses pattern control. Note that at 500-600Hz, the pattern is “flipped”.

Now, while this is a mild case, it can be very severe depending on the horn asymmetry. It is why the proper mounting for a T-35 horn is up and down if you want wide dispersion horizontally, it’s pattern flip begins 8-10Khz.

Hey Earl,
Sebastian has taken the first quick pass converting the spherical polar data to a + - 90 degree map style display.
I am asking him to see if he can get the raw data out of the CLF file because this has 1/3 octave averaging / smoothing and I would like to see less. Also, we don’t need the display to the low corner, the directivity down to 100Hz would be plenty. This is a 3dB step each color.
I haven’t looked at that many of this spectrogram style “maps” but to me it looks like it has a decent amount of directivity.
Best,
Tom
Attached Images
File Type: png Tom_contour_sh50.PNG (130.0 KB, 413 views)
__________________
Bring back mst3k and futurama
  Reply With Quote
Old 26th April 2013, 03:24 PM   #492
gedlee is offline gedlee  United States
diyAudio Member
 
Join Date: Dec 2004
Location: Novi, Michigan
Hi Tom

What you talk about is precisely why I resist using exponential waveguides in my designs despite the fact that some have been strongly critical of round and claim that elliptical would be better. My analysis says otherwise for precisely the reasons that you state. How sound radiation really works and how people think that it sould work are often quite different things.

Your map is a start, but 1/3 octave smoothing is way too broad. Everything will look great done that way. I use only a minimal smoothing, about 1/20th octave (but I use critical band smoothing that varies with frequency.) Also there are better ways to deal with the polar aspects than just interpolation, which has inherent errors. If you look at my data you will see that I have 2 degree resolution in angle and 1/20th octave in frequency. This is very high resolution and takes some doing to achieve.
  Reply With Quote
Old 27th April 2013, 05:59 PM   #493
winslow is offline winslow  United States
diyAudio Member
 
Join Date: Aug 2005
And I was just getting ready to ask how do you calculate when pattern flip occurs.
  Reply With Quote
Old 5th May 2013, 05:29 PM   #494
diyAudio Member
 
Join Date: Mar 2011
Location: Rochester, MN
Quote:
Originally Posted by natehansen66 View Post
Hopefully by Sunday I'll have both Paralines modified to have a reflector running nearly to the phase plug, following the driver exit angle and beginning the 1/4" internal height in the comp driver itself. It's going to be difficult to get the driver lined up over the reflector properly..... We'll see. If I have to I'll pull the back chamber/diaphragm assembly and mount the magnets first so I can see down the throat. I've got a cad drawing I'll post up later today showing what I'm up to.
So I'm not lazy and did actually do this. The driver mounting plate smoothly transitions to the CD throat angle, with the Paraline reflector continuing up into the CD like in my drawing. I wasn't very scientific about this process and made several changes all at once. I added a 45° chamfer on the "eye" layer to better match the outside eye 45° chamfers like Paul W noticed and keep the inside height from pinching down to less than .2". There were a few slight mismatches in the layers at the bend that I smoothed over, using a file and Danley's tip about sealing the Paraline. I sealed it with paint and wood glue.

I must admit that while I was hoping for improvement, I wasn't sure how it would result. Here's an on axis measurement pre- and post modification. Measured on different days but with the mic at the same height and distance from the exit reflector of the Paraline. Nevermind the levels on the graph, and I did offset one so they would line up since I don't keep track of the levels I'm testing at. Purple is before the mod and green is after. Without any eq they're both pretty ugly and I can't say that from these measurements there was any improvement. Polar measurements did seem a bit more consistent but I can't see why that would be. Most likely just measurement conditions as I'm doing these in my living room and it isn't exactly large .

The other day I decided to go a step further and continue the CD throat reflector right up to the phase plug itself. This was pretty difficult work but I think it came out ok. Lots of trial and error! I measured with the horn on the stand and the mic in the same position before and after this mod and the result was negligible at best.
Attached Images
File Type: jpg throat mods.jpg (75.4 KB, 314 views)
  Reply With Quote
Old 5th May 2013, 09:50 PM   #495
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
Quote:
Originally Posted by Tom Danley

Do the math though and one finds if you want to avoid pattern flip, the mouth height has to be 4 times the width while the real thing is more like the opposite of that, wide and short.
Not a lot of people know that, well they do now

Years ago Motorola used to recommend that their rectangular Piezo horns be mounted verically for correct dispersion. At the time it seemed odd, as no other horn manufacturer, i was aware of anyway, suggested this ? At least somebody @ Motorola knew their onions, & that was over 30 years ago

I guess because Piezos got a bad name, due to most people directly wiring them to the Amp, instead of using a power resistor & Xover, as i did, they didn't give the mounting advice much creedence ? Or never read it as they ignored Piezos ?

I'm not saying Piezos are All that, but they do sound a Lot nicer Xover properly The Motorola ones were the best of their type. When the patent ran out & copies appeared they wern't as good !
  Reply With Quote
Old 6th May 2013, 12:28 AM   #496
diyAudio Member
 
Join Date: Jan 2008
Quote:
Originally Posted by Zero D View Post
Not a lot of people know that, well they do now
Years ago Motorola used to recommend that their rectangular Piezo horns be mounted verically for correct dispersion. At the time it seemed odd, as no other horn manufacturer, i was aware of anyway, suggested this ? At least somebody @ Motorola knew their onions, & that was over 30 years ago
The rectangular Motorola Piezo horn was very similar to the Electro-Voice T-35 tweeter horn, which had recommended vertical mounting decades before.

Pattern flip is not a new concept, but without measurement is "hard to get".

Below are an equalized 90 degree horizontal "non divergent" Paraline, the horizontal dispersion shows very uniform (excellent) pattern control to below 500 Hz.
The vertical control goes from about 16 dB down at 16K 30 degrees off to only about 1dB down at 500 Hz, from a narrow beam to a nearly omnidirectional radiation pattern.

Art
Attached Images
File Type: png Paraline Horizontal dispersion.png (134.8 KB, 288 views)
File Type: png Paraline vertical dispersion.png (114.7 KB, 288 views)
  Reply With Quote
Old 6th May 2013, 11:28 AM   #497
Jmmlc is offline Jmmlc  France
R.I.P.
 
Join Date: Oct 2005
Quote:
Originally Posted by Patrick Bateman View Post
So - here's the idea I had:

Click the image to open in full size.

What about a horn that's LeCleach in the vertical dimension, and conical in the horizontal dimension?

Plus it would look cool.

Plus no one has ever done this before.

Thoughts?


If this is a stupid idea I could simply build a box that looks like a VTC EL210 but 4' tall. A mouth that tall would knock the directivity down another octave in the vertical plane.

Hello Patrick,

I used to study a horn having a better control directivity in the horizontal plane than the axisymetriacl Le Cléac'h horn while maintaining the most acoustical resistive load to the driver on most of its useful frequency range.

Here is the result:

Too bad no one built it since today...

Best regards from Paris, France

Jean-Michel Le Cléac'h
Attached Images
File Type: gif JMLC_HCD_elliptique.gif (140.2 KB, 88 views)
  Reply With Quote
Old 6th May 2013, 12:10 PM   #498
Zero D is offline Zero D  United Kingdom
diyAudio Member
 
Join Date: Aug 2009
Quote:
Originally Posted by weltersys

The rectangular Motorola Piezo horn was very similar to the Electro-Voice T-35 tweeter horn, which had recommended vertical mounting decades before.
Ahh, the T-35, i remember it, but didn't realise about the mounting details. See my screnie for the V/H beamwidth, which "seem" very close. I found a www with a number of other similar EV horns http://hf-antenna.com/Flotsam/EV/How...pkr_system.pdf

Quote:
Pattern flip is not a new concept, but without measurement is "hard to get".
Apparently, not new, but not well known. But it might improve from now on

@ Jmmlc

Stunning designs & concept

Quote:
Too bad no one built it since today...
I recently discovered the iwata range, which are based on yours Horns

Apart from the other benefits, what "appears" incredible to me, is the "apparently" seemingly shorter lengths, compared to other horns, but inspite of that, go down lower !
Attached Images
File Type: png US-Beam.png (66.9 KB, 93 views)
File Type: png T-35.png (36.6 KB, 64 views)
  Reply With Quote
Old 7th May 2013, 12:38 AM   #499
diyAudio Member
 
Join Date: Jan 2008
Quote:
Originally Posted by Zero D View Post
Ahh, the T-35, i remember it, but didn't realise about the mounting details. See my screnie for the V/H beamwidth, which "seem" very close.

@ Jmmlc

Stunning designs & concept!

I recently discovered the iwata range, which are based on yours Horns

Apart from the other benefits, what "appears" incredible to me, is the "apparently" seemingly shorter lengths, compared to other horns, but inspite of that, go down lower !
The T-35 horizontal dispersion goes from 110 degrees at 3kHz to 85 at 10kHz, a 25 degree range, the horizontal from 160 to 60, a 100 degree range. That "seems" a rather large difference to me.

As far as the Iwata horns, if one is not concerned with constant directivity, a relatively low cutoff is not hard to achieve.
The Paraline which I posted polars for in post 496 has usable response down to 167 Hz (EV DH1AMT driver), and is only about 11.5 inches (29 centimeters) in length. It holds a uniform in horizontal response of 90 degrees from 500 -16 kHz, the much longer Iwata 300 goes from about 100 degrees to 15 degrees over the same range.

Art
Attached Images
File Type: png Paraline raw.png (59.0 KB, 61 views)

Last edited by weltersys; 7th May 2013 at 12:53 AM.
  Reply With Quote
Old 7th May 2013, 12:43 AM   #500
diyAudio Member
 
Join Date: May 2009
Location: Wellington
Quote:
Originally Posted by Jmmlc View Post
... Here is the result:

Too bad no one built it since today...
Wow. That would make a bold design statement if painted to resemble a pair of lips.
  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
My Wave Isn't Square. Troncones Tubes / Valves 153 20th October 2011 07:54 AM
EICO 666 meter needle moves too fast/pegs too hard...help? jasonguitar Tubes / Valves 3 15th September 2010 01:58 AM
What gives with this square wave? martyh Tubes / Valves 23 21st January 2010 09:22 PM
Why are almost all sub boxes square elambert Subwoofers 12 28th May 2005 08:49 PM
Square wave Prune Everything Else 11 28th June 2004 06:17 AM


New To Site? Need Help?

All times are GMT. The time now is 01:46 PM.


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