PMA - 60; New arrayable double 12 and horn.

Here is a rough sketch of the plans for the PMA60 (have to reload sketch up) - There are a couple of things to note;

1) To minimize its size and keep the 2 x 12" speakers as close as possible to each other, the horn flare covers part of the 12" driver and shares one bolt / screw hole. (see earlier pictures).
2) The material I used was 16mm ply
3) The rear port also serves as a handle or grab point, so there is a small lip at the top of the port to help.
4) When arraying two boxes the DSP settings can be modified with a couple of filters - a 5 to 6 dB shelving filter to adjust the coupling of the 12's and a PEQ to adjust the coupling of the horn at its lower operating range,
 

Attachments

  • dbl12 PMA60.jpg
    dbl12 PMA60.jpg
    468.3 KB · Views: 263
This is outstanding. Truly. Peter some time ago you mentioned auditioning the Ciare horn, how do you like it compared to the 18sound?

Im dying to get a CAD file for these. This is exactly what the doctor ordered.

Did you play with the 675 XO any to see what a lower selection would yield? It just seems a shame to leave that couple hundred hz on the table. How would you think a DE990 would play in lieu of the coax at a 700hz Xover?

This is a lost art, flying and splaying high quality point source boxes.

Love it Peter.


Sent from my iPhone
 
I think the Ciare PR614 sounds slightly better than the 18sound, and the pattern does not narrow so much in the VHF region, but it does not go quite as low. From memory the horn crossover is about 670Hz (and about 3300Hz on the VHF), but the 12" go up to 900Hz. This produces a bump around 800Hz which is EQ-ed on the input. I found this gave smoother vertical off axis responses.

I have to say the sound absolutely beautiful and go hard for such a small light box.

I'm sure the DE990 would work ... but I don't know how well.

Here is the frequency response (ignore the response below 200Hz) and step response which is almost text book for what it is.

It was measured on a much better system than mine using a B&K mic etc.

https://audioxpress.com/article/testing-loudspeakers-which-measurements-matter-part-2
 

Attachments

  • PMA 60 frequency response and step response.jpg
    PMA 60 frequency response and step response.jpg
    232.4 KB · Views: 120
  • step resp.jpg
    step resp.jpg
    93.3 KB · Views: 117
Thanks Peter. I’m dropping my current design and picking this up.

I’m going to follow the recipe to the T. The box design sketchup you made, thats for the 40degree splay?

Can you share the Xovers youre using all the way up (675,900,3300)? Im most likely going to use a Venue360, doesnt have the filtering your ASC48 has but it will do the job.

Now I just need to sell these 14NDL88s to someone, and switch gears. I think a 14” driver is probably pushing the limits of 675hz crossover point for lobing.

The HPL is one heck of an efficient driver. I’m comparing it to the 12NDL76 and the NDL brings equally (1db more 300-900hz) more sensitivity and a 50watt increase of power with little to no increase in price. The 88 version adds 1/4 more to the cost (and weight) and while it handles twice as much power, the sensitivity comprimises by almost 3db so, its somewhat of a moot point.

I think the NDL76 will be a fine fit as is the HPL. Cant beat the db/w vs weight numbers on a flyable box.

B&C released the passive network for the Coax also btw.

Thanks and again, your work is just phenomenal.


Sent from my iPhone
 
Last edited:
Its designed for a maximum 50 degree tight pack splay (25 + 25) but it can also be tight packed with 30 and 40 degrees by swapping the boxes around (see my first post) This is handy if you want to fly the boxes.

If I didn't have the 12" speakers in stock etc. I would probably use some B&C 12NDL88 of Faital Pro 12FH520.

I'm happy to share the settings for my ASC48, but they will not translate to a Venne360. They use FIR filters programmed with FIR Designer. Having said that, developing suitable settings is not that difficult in this case, much easier than the PM60/90.
 
Last edited:
Question:

I notice upon close examination of the cabbnet design and photos that the Ciare horn is not centered in the width of the front baffle with the 12” drivers, it seems maybe 2-2.5cm off to ine side from the centered 12s. I assume its for the Comp driver against the 25degree side? What consequences does this offset pose to the patterns between the LF and Coax and cohesiveness between them?

My other question is:

When you splay three boxes at the 25 degrees per side, what happens to the overall pattern that is closer and less than the 60degree horizontal pattern? If the signals are overlapping and not seperated by a full 60degrees (30 degrees per side) will there be any comb filtering? Is the overall coverage in the horizontal then 150 degrees?

Thanks Peter.


Sent from my iPhone
 
Hi Chris, The horn is not in the centre imply because it will not fit any other way, in fact the side wall of the box had to be ground down to about 6-8mm thickness just to get the compression driver in.

When you array any two horns you generally find there are some issues. If you array this box at at total of 60 degrees there is a little bit of a drop with the HF in the middle; 50 degrees seems to work quite well without too many issues, its not perfect, but much better than most, and the two 12" drivers seem to combine very well.

In this case the pattern is somewhere around 110 degrees for two boxes (60 + 60 - 10 degrees of overlap), you can of course open them up a little to give a total coverage of 120 degrees but its not as even.

The idea was a general purpose small light 60 degree box that could be array if needed and still work well.
 
Hi Chris, The horn is not in the centre imply because it will not fit any other way, in fact the side wall of the box had to be ground down to about 6-8mm thickness just to get the compression driver in.

When you array any two horns you generally find there are some issues. If you array this box at at total of 60 degrees there is a little bit of a drop with the HF in the middle; 50 degrees seems to work quite well without too many issues, its not perfect, but much better than most, and the two 12" drivers seem to combine very well.

In this case the pattern is somewhere around 110 degrees for two boxes (60 + 60 - 10 degrees of overlap), you can of course open them up a little to give a total coverage of 120 degrees but its not as even.

The idea was a general purpose small light 60 degree box that could be array if needed and still work well.

Thanks Peter. Appreciate the response. Im ready to start cutting panels so I just want to ensure Ive got this nailed down before I commit.

How did the 40 degree * 3 splay play out in testing? I imagine the increased overlap of 10 more degrees was still acceptable? The overall coverage then would be 160 degrees? If you ever have a chance to do a field full range sweep of one and then the two and three together to see the combined responses that would he fantastic!

So it would then be acceptable to center the 12” drivers over and under the centerline of the Ciare? I noticed in the other box 3d design it was this way.

So if we came straight back 90 degrees square off the baffle 2.5-5cm (min length needed)and then angled 25 and 15 to keep the drivers centered on the baffle, will this be detrimental to a defree noticable due to upsetting the close proximity of those horns in your splay coherence with those horns nearly side to side?

I’m totally fine with building a pair each side of these for 100degrees as exactly per your design, I dont think Id ever have a use for the triple per side for 160degrees, although what s great is I can always add side fills or build another pair for an outdoor wide venue.

Im going to test the first box built with the 464 and also a 4594 (These I have) and also going to pick up a DE1090 and the new Eminence 314X. Id like to see the response and performance with all three. The two standard comp drivers are happy at a 700hz XO, and Ive heard fantastic reviews of that new 314X. Of course the two coaxials will always sing better than a mono driver, but I just want to evaluate the difference.

Going with 12NDL88s.


Sent from my iPhone
 
3 x boxes at 40 degrees worked quite well but because I only built 4 boxes I did not do any settings for 3 boxes, just 2 per side, and yes you will get good output to about 150 - 160 degrees as a guess.

The reason I put the 12" drivers in the centre of the box and not directly under the horn was so that they remained the same distance from the 12's in the adjacent box no matter which way they were stacked. i.e. the centre to centre distance remains at 360mm ... what you need for the two to combine OK is less than 1/2 wave length spacing centre to centre (1/4 wavelength is optimum.) This will get you up to about 475Hz, so its OK but not perfect ... but way better than most; and more to the point it works in practice :) ... this is why it uses a low crossover frequency crossover point around 700Hz, preferably less but the HF horn is a bit small to go much below 700Hz.

I have done a SIM on the 12NDL88 ... they look great in this box, just add an 80hz 12dB Butterworth HP crossover and the acoustic response will be very close to a 24dB LR.
 
Last edited:
Perfect Peter great info thank you!!!

I think the DCX and the NDL88 is going to be w winning combo, I just really want to get that crossover point lower to get out of that 88 much lower than 700hz and put some more vocal in the horn. I may toy with going back to the XT1464 in a wider full width prototype just to see how it pans out getting maybe 3/4 of an octave lower in the horn if I can get the depth in the cabinet.

I’ll post progress here as it goes.

Thanks again!


Sent from my iPhone
 
Hi Chris,

This started out as building a 60 double 10", but I figured I could get 2 x 12" in the same size box, which should help with their ability to array (in practice its more than just the centre to centre distance of the drivers that matters). Anyway the problem is getting the side angles steep enough, the correct box volume and then getting the horn & driver in the box without making the box too wide and spacing the drivers too far apart, compromising their ability to array .... and then ... ... getting the vertical spacing of the drivers (and crossover frequency) in the MTM configuration such that the vertical directivity is still OK.

Its all a juggling act ... the smaller 4594HE should help in this respect but I did not have 4 spare 4594HE's at the time.
 
Peter Im going to build this exactly per spec. I can see the magic and uniqueness of this comprimise and balancing act. If I can get the 4594 to fit would centering the horn on he baffle have little to no negative effects (I dont think so but wanted to check)

Unfortunately I have the older 4594 (non HE). Its the same dimensions though as the HE.

Whee do you have your grills made Peter, or do you do them yourself? They always look first class.


Sent from my iPhone
 
There are lots of drivers that will work in this box - these look very good, and I'm sure there are many more options.

Caire NDH12.4
FaitalPRO 12HF520
B&C 12NDL88

Peter before I pull the trigger on the NDLs, it seems to me the Ciare 12” is probably the best choice, and what I read about it says it has great upper octave response as well. I wanted to ask, Im partial to B&C products, and really like that 88, I already own a pair of 14s, but wanted to ask what you think.

Why though did they attach the leads on the cone outside the cap like a 12 dollar full range clock radio transducer?? Man thats hokey looking.

The thing is, either the NDL88 or the Ciare, I’d want to be out of either by 800hz max, esp the Ciare as some breakup looks like it occurs. The 12NDL76, has better support above 800hz and supprts the cross up to 1.2 or so with less issues.

Sent from my iPhone
 
Last edited:
A couple of answers best I can -

“the older 4594 (non HE). Its the same dimensions though as the HE” yes except the phase plug on the HE extends out of the throat of the driver by about 20 mm.

Which is the best driver … short answer is I don’t know. From what I have read they all perform very well. As a general rule any of the new drivers from these manufactures are excellent. All of these have been using FEA and Kippel measurements to improve their designs, and the results have been excellent.

https://audioxpress.com/article/Test-Bench-B-C-Speakers-12NDL88-12-Woofer

FWIW This is a test looking for the best 10” driver; I suspect its representative of these manufactures 12” drivers of the same series. https://www.diyaudio.com/forums/multi-way/324260-war-10-monster-midbasses.html

Al is usually correct with his comments … https://usspeaker.com/ciare 12ndh4-1.htm
 

Attachments

  • 4594HE.jpg
    4594HE.jpg
    57.2 KB · Views: 17
Went with the Ciare 12s. Users who have i tegrated them have had great feedback, even from OEMs who have said its the best deiver out there for clarity and detail. It has LOTS of headroom as well.

Peter those box numbers are based off of 16mm birch? I cant find 16 here, I can find 12 and 18. I’ll keep calling around but I’ve made quite a few calls and nobody has it or in some cases heard of it. Maybe thats a European cut?


Sent from my iPhone
 
Hi Chris, I think you will love the Ciare's ... 16mm is easily available in Australian (not birch) and I suspect Europe - try 5/8" in the US = 15.9mm

Ended up getting 18mm. Its either that or 12 around here, esp with Covid supplies are limited. Id rather a little more weight than less rigidity. Im going to try and recalculate the measurements and try and keep the chamber the same internal volume.

Drivers and horns arrived today.

Whats the overall net volume and tune Peter? I calculated the optimum per driver for the Ciare, just want to compare. The port dimensions are obvious, whats the square equivalent port length?

Sorry for all the questions Peter, this is a big deal and investment for me so want to ensure I “measure twice and cut once” as they say.


Sent from my iPhone
 
Last edited:
It was about 43 - 44 L allowing for the, horn and diver volumes - I did not calculate an equivalent square port area, but he box needs to be tuned to 65 - 75 Hz depending on the driver.

Doing the ports the way I did braced the box walls, in addition the port exit is a bit small but doing it this way minimized port noise. The port tuning frequency is also below the crossover point to the subs; and the port resonant frequency above the crossover frequency to the horn
.