"Unitized" Image Control Waveguide

inside the box:

from point 8:
 

Attachments

  • 20230220_174051.jpg
    20230220_174051.jpg
    216 KB · Views: 226
  • 20230222_171852.jpg
    20230222_171852.jpg
    216.7 KB · Views: 239
  • confronto aperta chiusa 50cm.jpg
    confronto aperta chiusa 50cm.jpg
    75.1 KB · Views: 215
  • confronto aperta chiusa 100cm.jpg
    confronto aperta chiusa 100cm.jpg
    75.6 KB · Views: 192
  • Like
Reactions: 2 users
I have a few questions:
What is the limiting factor for the output on this horn? Is it the tweeter or the midranges?
I've been toying with the idea of a compact unity horn for a while, but my main issue has been my SPL target. I am trying to hit 105db at 10 ft.
According to the tweeter's native sensitivity, (about 88db), and power handling (30W), at 10 ft we should start to run out of steam at 95db. Now of course, a horn will help increase that, but does it actually give another 10-15db of sensitivity here?

Same thing with the midranges, they seem to have a bit more capacity than the tweeter, but how much is the bandpass / horn helping here with regards to sensitivity / power handling?

And lastly:
What kind of support / infill / wall / layer height settings did you use to print the original .stl file? How did you deal with the midrange ports specifically?

Thanks and have a nice day!
 
Hi, I finally decided on the sealed box and to start with I built Patrick's crossover in which I only replaced the 2.7ohm resistor with a 2.2ohm resistor.
crossover.png

At this point I also inserted the minidsp for the low range in which I inserted a low pass for the woofers at 450Hz 12db/oct and a high pass for the Gento at 450Hz 6db/oct.
I used Dacron wads in the throats, I noticed that the best linearity is achieved by not filling the throats is by having it stay up to the surface of the waveguide as seen in the graph.

filtro 22 ohm al variare del dacron.png


The woofers were 6db less efficient than the mid-highs so the effective crossover frequencies became 550Hz and 1750Hz.
I also applied a little equalization:
Correzione DSP.png

This was the result:
complessiva da vicino con correzione DSP.png

This is the response up close(about 20cm)
I then made dispersion measurements outside in 15° steps with the speakers positioned at 135cm height and the microphone about 150cm away, centered on the center of the horn.
risposta 0 60 1,5m aperto.png

What do you think about?
 
  • Like
Reactions: 2 users
The low efficiency of the woofers bothers me, but their choice was made because they are suitable to be brought closer to the center of emission, and for the same reason their emission is channeled into a throat. So I decided to do some tests to see if I would benefit from direct emission.
Here is the result: in a sealed box (12l):

-->DIRECT

20230527_123719.jpg
Re = 5,9 ohm Fc =78,7 Hz Qtc = 0,98
Comfronto woofer diretto.jpg


---->SLOT

20230527_130308.jpg
Re = 5,9 ohm Fc = 68,6 Hz Qtc = 0,74

Confronto woofer slot.jpg


----->LAT 250

Looking for suitable woofers, I found LAT250 woofers that have an interesting shape and also given the fairly high RE I could use a pair in parallel:

20230527_133122.jpg

Re = 7,8 ohm Fc = 98,2 Hz Qtc = 1,05

Confronto woofer LAT.jpg


-------------->Here is the comparison chart:

confronto risposte.jpg


The slot configuration does not seem to be less efficient than the direct configuration. Certainly with the LAT 250s I could get a speaker with a narrower front and maybe even a few db more efficient.
Better 4 woofers in series/parallel or two LATs in parallel placed as the woofers above and below the waveguide?
What do you guys think?

Best regards.
Guglielmo
 
Hello,
This evening I tried adjusting the phase between woofer and midrange following Patrick's directions:
"The way that I get the phase correct (and it might work for you too) is like this:

I get the response of the mids and the tweeters to be flat in their passband, like the last post
I flip the polarity on the tweeter
I measure the combined response, noting the xover point in particular
I tweak the delay with the goal of making the dip as deep as possible
Once the depth is as deep as possible, I flip the polarity. (The reason I measure inverted is because when the dip is the deepest, flipping the polarity will generally get you the best phase response.)
"
inversione fase woofer senza ritardo picco massimo.jpg

I got the best result without adding any delay.
Ciao.
Guglielmo
 
  • Like
Reactions: 3 users
Greets!

Correct, presumably originally based on this Altec doc's simple horn alignment procedure for 2nd order XOs, which in turn has its beginnings with W.E. two way cinema horn alignments:
 

Attachments

  • Polarity and Phase_TL_AN-9_phase.pdf
    314.5 KB · Views: 81
I had attempted to make the Bwaslo synergy, but gave up on the complexity of the horn and sold the parts on DIYAudio. Now, I have an Ender 3 V2 printer and have found this project suitable for my next project. First a huge THANK YOU to all involved in this thread for the development and improvements. I first heart the synergy concept with the Unity by Lambda Acoustics. Been looking to replicate that sound ever since I heard it before.

My speaker will use Pelanj's modified waveguide. The midwoofers will be 4xFaital Pro 6FE100's in a sealed cabinet and narrow the beamwidth with the DNA concept.

The plan is to 3D print the waveguide with thin walls and use a gyroid infill so that the waveguide and be filled with epoxy. Shown in the photos was my first attempt with a 1.0mm nozzle and PLA as a proof of concept. The gyroid infill was awesome to allow the epoxy infill, unfortunately, the print leaked the epoxy through some voids.

The next print is in progress using an 0.8mm nozzle and PETG (shown attached in black). I setup the alternating walls function in the Cura slicer and ran calibration prints before this one. The print quality is much better with 1.0mm vs 0.8mm. I went for 8% gyroid infill and a tree support, it seems to be doing really well. I'll fill the print with water (since PETG is not affected by water) to find the needed volume of epoxy and do a full epoxy gel coat to seal it before printing.

I am going to use the DNA concept for narrowing the beamwidth. I'll look into making a 3D print back cup for the midranges.
 

Attachments

  • Screenshot from 2023-10-02 09-26-11.png
    Screenshot from 2023-10-02 09-26-11.png
    288.8 KB · Views: 156
  • Screenshot from 2023-10-02 09-26-02.png
    Screenshot from 2023-10-02 09-26-02.png
    228.9 KB · Views: 154
  • Screenshot from 2023-10-02 09-25-32.png
    Screenshot from 2023-10-02 09-25-32.png
    413.5 KB · Views: 154
  • Screenshot from 2023-10-02 09-25-23.png
    Screenshot from 2023-10-02 09-25-23.png
    350.1 KB · Views: 153
  • Like
Reactions: 1 user
Thanks! I am going to setup the side woofer ports for an MTM configuration - I believe the will narrow the horizontal directivity when spaced for a specific range? For the vertical speakers, I need to space the woofers closer to the waveguide (one on top, one on bottom). Are there any tricks with DSP that can be used simulate the two woofers farther away without physically spacing them farther?
EDIT - found your post on FollGott from https://www.diyaudio.com/community/threads/combating-pattern-flip.325278/page-3#post-5517055
 
Last edited:
Very cool.

I found four more of these in my garage so I'll be listing those on eBay soon
I read here that you use many of your horns in your car. Do you have any pictures or measurements of horns/waveguides in a car?
And what do you think is the best so far? I'm currently running small elliptical horns in my sail panels as well but would like to upgrade them as they are too directional and have problems in the top octave. Also, where do you find the room for bigger rectangular horns in a car?
Here's pics of my current setup:
 

Attachments

  • 20230831_145018 (1).jpg
    20230831_145018 (1).jpg
    494.3 KB · Views: 107
  • 20230831_044032 (1).jpg
    20230831_044032 (1).jpg
    417.3 KB · Views: 103
I am surprised that more attention has not been given to Apple Homepod drivers for car stereo, horn loaded applications. You have a BEM optimized horn loaded wide range array that outputs 7 modules in an omnidirectional pattern, with each module covering a 40 degree spread and covers 500-20000hz. The developer of that concept made one of the first amateur unity style speakers on this forum.
I had the idea of an 84 degree speaker pod array (2 modules per side). The hard part of designing the waveguide has already been done, just basically cut an 84 degree section of the homepod with the two drivers, turn it upside down, add a 84 degree pizza pie slide on top to simulate the surface the bottom of the homepod and your done. (si first patent concept drawing, just turn that upside down and recess down into the dash) with two drivers per corner, you can control the cross firing and forward firing modules separately and basically equalize for constant amplitude between the speakers across the soundstage. (This takes after the DBX Soundstage concept). See photos.
 

Attachments

  • Screenshot from 2023-10-05 11-57-03.png
    Screenshot from 2023-10-05 11-57-03.png
    95.4 KB · Views: 108
  • Screenshot from 2023-10-05 11-55-49.png
    Screenshot from 2023-10-05 11-55-49.png
    5.1 KB · Views: 111
  • Screenshot from 2023-10-05 11-55-32.png
    Screenshot from 2023-10-05 11-55-32.png
    312.3 KB · Views: 115
The homepod array horn looks like a nice design, but It's not cheap or practical to buy two homepods just to put them in a car.
I wonder what sort of efficiency and power handling that array has though...

I would much rather buy and mod a cheap horn, or make or 3d print one and use after-market drivers to be able to change or replace drivers.
Not like apple to sell replacement parts to consumers.

I was looking at the new Celestion 1412 driver. Looks to have a very smooth top end and it's much smaller than my current PRV D280ti.
Although I'd love a driver that can play down to 500Hz like my midrange, they almost always sacrifice top end so I would still need a super tweeter and the higher crossover would be harder to integrate.
Some sort of synergy horn would be ideal but I don't know where I could fit them in the car. My current mids would work very well in a synergy horn though. Eminence alpha 6cbmra.
 
Hmm, using a 500 Hz Fs driver: sqrt(500*20000) = 3162 Hz mean/Fhm = ~reactance annulled rear chamber Fb and Qts' = 2*500/3162 = 0.312, so doable in theory. Anyway, the math to juggle some actual small 'FR' driver specs for a truncated (large throat to keep from 'choking' the HF) (under?) dash/windshield FLH. That or do an RCA horn in horn.

(Qts'): (Qts) + any added series resistance (Rs)
 
Last edited:
I finished filling up my horns with epoxy, they are rock solid! I was scratching my head figuring how the SB19 was to fit when it dawned upon me that Pelanj's waveguide follows guglielmope's design with the NE19VTS-04 tweeter.... (Doh!) Well, the Gento's seem to fit. However, it looks like I will either order the NE19VTS-04, or keep the SB19 by filing down the waveguide and since its solid epoxy and just tap directly into it...

(Edit - looks like I am going the SB19 route since the NE19's are out of stock..)
 
I have conducted some measurements on the original and modified UICW to compare them. Specifically, I measured from the beginning of the waveguide to the end, which is from the flat throat to the end of the horn.

The following are the differences I found:

- Guigliemope/Pelanj NE19: Length = 80.47mm* / 23.6mm
(The distance to the driver mount area was 82.47mm, but there was a lip where the true termination was an additional 1.67mm in. I subtracted that to get 80.47mm.)
- Patrick's SB19: Length = 80.47mm / Diameter = 21.9mm

Based on my findings, I plan to belt sand the throat down by 1.67mm. Then, I will fill in the throat area with bondo and sand the throat for a smooth transition to 21.9mm. Since it is solid epoxy, longer SB19 sheet metal screws will be used directly into the waveguide. Do you think this is a feasible way to salvage the horns that I have recently made?
 
Hey Patrick,

I observed some white putty in one of your earlier photos around the SB19. I wonder if the putty serves to fill the void around the dome flange since the SB19 has a small void from the fillet?

I sanded down the horn for the SB19 and used JB weld to fill in the hole with a 3d printed cone mold. The outcome was exceptionally solid! I barely added any JB weld, and it may suffice without filling in the opposite horn.
 

Attachments

  • IMG_1316.JPEG
    IMG_1316.JPEG
    736.8 KB · Views: 88
  • IMG_1315.JPEG
    IMG_1315.JPEG
    926.6 KB · Views: 90