• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

DIY Mini-Unity Horn. Am I crazy?

Stupid (is it?) question: Why not just buy a couple of Uni-Q drivers from the KEF R-series and just attach enough bass drivers for your output needs?
Because that is not the same. It's still not a perfect point source. Also the shape of the mouth is all wrong for a horn.

But yes, you can use a coaxial. Danley does that in several models as well, like here:
index.php

You could probably 3D print such an adapter to mate with the reso of the horn, but I haven't seen any drawing of how exactly that adapter works.
 
Personally i would either go with a Uni-Q or a full range driver with a mechanical crossover like Lil Audio F15 (In case you're interested in sending the driver to Erin for Klippel testing, i can pitch in some money for that).

a DIY synergy horn will not only look awful but is not honestly necessary and will end up being a huge investment.

F-15-45.jpg


%E8%AF%B4%E6%98%8E%E4%B9%A6%E7%AE%B1%E4%BD%9311.jpg
 
Personally i would either go with a Uni-Q or a full range driver with a mechanical crossover like Lil Audio F15 (In case you're interested in sending the driver to Erin for Klippel testing, i can pitch in some money for that).
Uni-Q does not have the efficiency that you reach with a Synergy, nor does it hold the pattern that low.
100 dB scale on those graphs :facepalm:. Let's see off-axis, it's probably not gonna be very pretty. Would love to see this Kippeled ;).
 
OP does not have neither of these things as requirements in their design btw.
But he does want a Unity horn ;) That sort of implies efficient and pattern control.
 
I should talk. I have a large Synergy that I built 6 years ago and recently purchased a pair of KEF R3s as a quick way to approach its performance in a small footprint. But that isn't because I thought the KEFs would be as good. I was just trying to simplify and fit a smaller space.

There is absolutely no doubt in my mind as to which one plays louder or cleaner or over a wider range. With the ATH tool for refining the horn profile and 3D printing, a new Synergy design could/should come out smoother and with better polars and more dynamic than a Uni-Q. The question still open in my mind is to what extent the midport holes compromise those polars. There is a lot of fud around about their effect but anyone who has heard a well done Synergy knows its not a significant problem - if it were, they wouldn't sound so good.

I'll give KEF kudos for having the best coax drivers available but I contend that a better sounding/performing Synergy is within DIY reach.
 
Stupid (is it?) question: Why not just buy a couple of Uni-Q drivers from the KEF R-series and just attach enough bass drivers for your output needs?

This is like way cheaper and way easier and meets the goals.
To answer this question, I actually have a pair of KEF Q100s, and have heard everything from the 150s to the R-11s and the blade 2. None of them were as dynamic or "lifelike" as even the JBL studio 580s or HDI-series, which use compression driver tweeters.

With this design, I want to target 105 db playback at >10ft listening distance, at least from 200hz up. (external bass drivers can be added for larger spaces).
The idea being that theater reference level is 105db @ 0dbfs. Basically every coaxial speaker out there (excluding pro audio drivers) can't hit that volume at >10 ft. Pro drivers have nasty frequency response, so they are out too.

Hence my arrival at the idea of a synergy horn. It'll be point source, high output, and controlled directivity.

Edit: I'm not saying KEF has bad drivers, when operated at lower levels, the kef coaxials are really good. But when you crank it, they compress too much. And by too much I mean they compress at all. A compression driver tweeter and horn loaded midrange will not compress until crazy output levels are reached.
 
As an alternative you might consider just a large horn and a large driver, like JBL M2. You can make it smaller for sure, like 10” woofer with a good 1.4” exit CD. It would still give directivity to just under 1 kHz, and is a much simpler design. The 105dB should be easily reached.
 
True, I had considered that as a potential candidate. It would definitely be easier to build and design, but then I would run into the issue with vertical lobing around the crossover frequency right?
I had considered using a large horn and multiple smaller drivers for the midbass, but I still think I would have issues with lobing.
Also... I kinda want to do something different... I mean, there are tons of monitors like that out there, but there aren't really tons of (very) compact synergy horns out there.

Speaking of the synergy horn design, I'm thinking about using 4x 5" drivers paired with a 1 or 1.4" CD. I still need to model it in CAD to see if the drivers'll fit, but it seems that 4 smaller drivers will give me better bass output while still hitting 105db up to around 1khz.
 
To answer this question, I actually have a pair of KEF Q100s, and have heard everything from the 150s to the R-11s and the blade 2. None of them were as dynamic or "lifelike" as even the JBL studio 580s or HDI-series, which use compression driver tweeters.

With this design, I want to target 105 db playback at >10ft listening distance, at least from 200hz up. (external bass drivers can be added for larger spaces).
The idea being that theater reference level is 105db @ 0dbfs. Basically every coaxial speaker out there (excluding pro audio drivers) can't hit that volume at >10 ft. Pro drivers have nasty frequency response, so they are out too.

Hence my arrival at the idea of a synergy horn. It'll be point source, high output, and controlled directivity.

Edit: I'm not saying KEF has bad drivers, when operated at lower levels, the kef coaxials are really good. But when you crank it, they compress too much. And by too much I mean they compress at all. A compression driver tweeter and horn loaded midrange will not compress until crazy output levels are reached.
I personally don’t agree with your opinion that the KEF R9/R11 midrange cannot handle your output needs.

Even the Genelecs 8361A has a similar sized midrange and is rated for 116 dB.
 
I personally don’t agree with your opinion that the KEF R9/R11 midrange cannot handle your output needs.

Even the Genelecs 8361A has a similar sized midrange and is rated for 116 dB.
A 4” driver needs about 1.5mm of excursion to reach 110dB at 1m, so it should be doable.
 
best 4" driver for synergy mid I know of is B&C 4NDF34 with Xmax = 3.8
 
True, I had considered that as a potential candidate. It would definitely be easier to build and design, but then I would run into the issue with vertical lobing around the crossover frequency right?
I had considered using a large horn and multiple smaller drivers for the midbass, but I still think I would have issues with lobing.
Also... I kinda want to do something different... I mean, there are tons of monitors like that out there, but there aren't really tons of (very) compact synergy horns out there.

Speaking of the synergy horn design, I'm thinking about using 4x 5" drivers paired with a 1 or 1.4" CD. I still need to model it in CAD to see if the drivers'll fit, but it seems that 4 smaller drivers will give me better bass output while still hitting 105db up to around 1khz.

What about the B&C coax compression driver? Put it in a horn and you should be good.
 
What about the B&C coax compression driver? Put it in a horn and you should be good.
The goal was a mini-horn, not a 15- to 18” wide monster. If you don’t load this thing all the way down, performance will suffer greatly.
 
Edit: RE to the 2 posts above: Although the B&C coaxes are pretty neat, they are also crazy expensive. My hopes are for the speaker (single, excluding amps / x-over) to come in around the cost of 1 of those drivers lol...

After doing some research on possible drivers to use in this project, I think either a B&C DE250 or JBL 2409h for the CD, with 4x 5" drivers to handle the mid-bass region (Probably Eminence Alpha-5). These drivers are cheap-ish, and seem to be good quality, and should work in this application.
I'm thinking an x-over around 1.2-1.4 khz, with the horn being a 90-degree pattern with a 12" square mouth.
The 4 woofers should be able to keep up with the CD, although if space is too small, I may end up with 2x 6.5" woofers instead.

I'll probably try modeling something in CAD later today to see if this combo will even fit into the space constraints I have, then I'll have to figure out how to use hornresp / the synergy calc spreadsheet someone posted earlier.
 
Last edited:
DE250 is well regarded and known to be OK with low crossover points, JBL 2409h though seems like an odd choice?

Wide dispersion unity horns force a lower HF/MID crossover point due to the increased spacing of the mids from the throat. I think also mids larger than 4" will struggle with the C2C spacing for 1.2-1.4kHz. For domestic use even a single B&C 4NDF34 probobly has enough output, you would however need another driver for bass.
 
I just picked the 2409 as an option if I could ever find data on it. It is the driver used in the 705 and 708 monitors, and apparently has really good HF extension due to the ring radiator design.
The only issue I have is that IDK how low it goes. In the 708p, it is crossed at 1.7khz, so it may not work here when I need an xover as low as possible.

Using dual 4" mids I could probably get away with around 1.4khz x-over frequency.
The injection ports would have to be within 2" of the CD, not impossible with 4" drivers.

Then I would need to integrate some kind of bass driver into the enclosure...

Edit: Another crazy option might be to have a slightly longer path length for the injection ports, and to just delay the CD with DSP in order for the alignment to work... If the models worked, then I could theoretically have the injection ports located right next to the CD, and could run a higher crossover. (Of course, it would have to be 3d-printed)
 
Last edited:
Edit: RE to the 2 posts above: Although the B&C coaxes are pretty neat, they are also crazy expensive. My hopes are for the speaker (single, excluding amps / x-over) to come in around the cost of 1 of those drivers lol...

After doing some research on possible drivers to use in this project, I think either a B&C DE250 or JBL 2409h for the CD, with 4x 5" drivers to handle the mid-bass region (Probably Eminence Alpha-5). These drivers are cheap-ish, and seem to be good quality, and should work in this application.
I'm thinking an x-over around 1.2-1.4 khz, with the horn being a 90-degree pattern with a 12" square mouth.
The 4 woofers should be able to keep up with the CD, although if space is too small, I may end up with 2x 6.5" woofers instead.

I'll probably try modeling something in CAD later today to see if this combo will even fit into the space constraints I have, then I'll have to figure out how to use hornresp / the synergy calc spreadsheet someone posted earlier.
There's a 4" B&C coax under $140. The cone would need horn loading to bring the efficiency up, but, heck, you're doing that.
 
There's a 4" B&C coax under $140. The cone would need horn loading to bring the efficiency up, but, heck, you're doing that.
The slightly bigger BMS coaxes like the ones used by Danley can work quite well.
https://www.diyaudio.com/community/threads/bookshelf-synergy-with-a-6-5-coax-in-a-reflex-box.353906/

I think either a B&C DE250 or JBL 2409h for the CD, with 4x 5" drivers to handle the mid-bass region (Probably Eminence Alpha-5). These drivers are cheap-ish, and seem to be good quality, and should work in this application.
I'm thinking an x-over around 1.2-1.4 khz, with the horn being a 90-degree pattern with a 12" square mouth.
The 4 woofers should be able to keep up with the CD, although if space is too small, I may end up with 2x 6.5" woofers instead.
It is worth to run potential candidates through Hornresp as drivers that seem similar on manufacturers datasheets model very differently in a band pass arrangement. It is hard to make a woofer go both low and high as the bandpass tends to want to restrict both. The shape of the cone and the amount of air it traps plays a part together with the length of the port. A short port and a small volume of air is needed to get higher in frequency.
From memory the Faital 5FE120 driver had good parameters for working over a wide range.

Another thread to go through
https://www.diyaudio.com/community/...ay-unity-waveguide-home-audio-speaker.312414/
 
Back
Top Bottom