Double 10 and Horn

Peter Morris

Senior
May 8, 2011
1,062
163
63
Australia
I have been asked by a few people if I could put up some details of my dual 10” and horn speaker I mentioned in the DIY Mid Hi thread. I built the original design 18 years ago, it’s now very similar to an active version of JTR’s - 3TX.

So here goes…

It uses the same horn as the DIY double 12” - the RCF HF950, a BMS4594 and the BMS passive crossover. My original boxes used two propriety Australian made 16 ohm Lorantz 10” speakers with an EV HP94 horn.

In 2015 I would suggest something like 18sound’s 10NMB420 or B&C’s 10NDL64, but there are other drivers that would be suitable.

The trick with this design is to cross the two 10” drivers low enough so they act as more or less one source. To do that you need to find a horn the correct size with a low cut off frequency and combine it with a compression driver that is happy to operate that low. This is where the 4594and HF950 are perfect.

I used 650Hz – 24dB LR active crossover plus a few parametric EQs on the HF and LF and some delay to align the drivers at the crossover.

It works perfectly on top of a double 18 sub crossed at about 110Hz. While it has some serious output it’s probably about 10dB less than the DIY double 12 stick horn (D12SH), but is much lighter and easier to build.

Box dimensions

Height – 870 mm
Depth – 375 mm
Width - front - 325mm
Width - back – 200mm

The horn is in a separate enclosure leaving about 40L for the two 10” drivers. If you select a different driver that needs more volume you can include the horn volume i.e. about 60L

Porting – 2 tuned to suit drivers - 65Hz for 18sound and B&C

HF driver – BMS 4594 + crossover

LF drivers - 18sound’s 10NMB420 or B&C’s 10NDL64

If anyone else is interested I will post some more details.

http://www.eighteensound.it/PRODUCTS/Products/CatID/8/ProdID=28#.Vg0NQGkiPaQ
http://www.bcspeakers.com/products/lf-driver/10-0/10ndl64
http://www.bmsspeakers.com/index.php?id=bms_4594nd
http://www.rcf.it/products/precision-transducers/horns/hf950
 

Attachments

  • image_12410.jpg
    image_12410.jpg
    412 KB · Views: 531
  • image_12411.jpg
    image_12411.jpg
    386.2 KB · Views: 560
Last edited by a moderator:
Re: Double 10 and Horn

Hi Peter, Would a less powerful hf driver work in this case since it outrun the low? Thanks
Before the BMS I was using an Eighteen Sound NSD1480N. It was a very good driver but not happy going below about 800 – 900 Hz. At that frequency the two 10” drivers are still not playing that well together, it works, but it’s not best solution.

A couple of examples of dual 10” cabinets that are very good are EAW’s KF394 and JTR Noesis 3TX. Both of these use BMS dual diaphragm compression drivers and cross them very low.

The trick is to get down to about 650Hz or less so the two 10” drivers more or less act as one driver and do not phase cancel each other when you are off axis. To do that you probably need a 4” diaphragm driver.

I think all of these drivers will work:

· Eighteen sound NSD4015N – 1.4 or ND4015Ti2 - 1.4
· RADIAN 951PB - 1.4
· RCF ND950 - 1.4
· I think B&C are going to release a 1.4” version of their DE1080 TN ... I would check this one out if the do.

They will of course need quite a bit of CD horn correction EQ etc.

I know RCF crosses the ND950 over at 650Hz their TT5a. It uses the same horn as the Dual 10 (HF950). I have also AB my Dual 12 and Dual 10 against the smaller RCF TT 2a. The TT is an excellent speaker, up there with the best that money can buy, but the 4594 sounds better. FWIW the TT2a uses a ND850 compression driver on the HF950 horn.
 
Last edited by a moderator:
Re: Double 10 and Horn

Great to see this thread. I was curious about hearing your dual 10's when I saw them. Are they passively crossed or fully active?

I ask because I was looking at building something very similar. I have four 10's that "may," work but they do like larger enclosures. I have the Beyma SM110/N. SO while they may need a larger volume I was looking at still building a 2 way with the right horn combo with them if I can. These speakers will be playing double duties for my living room and outdoors. Thats also why I was curious to know if your dual 10 was passively crossed.

Will have to check out which CD is the best value for me. While it might not be the best option such as the BMS at least it will get the project completed and playing. :D I should try and visit your place again and check out all the CD's and other speakers you have. :)
 
Re: Double 10 and Horn

The trick is to get down to about 650Hz or less so the two 10” drivers more or less act as one driver and do not phase cancel each other when you are off axis. To do that you probably need a 4” diaphragm driver.

Why not add a passive LPF the bottom 10" below say 250hz, making a "2.5 way" box, and reducing cancellations (and they are on -axis too) between the 2 x 10s? The additional coherency could allow a higher active cross to the HF device.
 
Last edited by a moderator:
Re: Double 10 and Horn

I am not a fan of the 2.5 way crossovers.
10s will couple below 250hz and be essentially 3db louder from 250hz down. Meaning you will need to eq that excess efficiency down and trowing it away. So either use bigger HF that can cross lower or make the box in MTM configuration with the horn in the middle.
 
Re: Double 10 and Horn

The earlier versions of my double 10 used standard compression drivers crossed at 900 to 1200 Hz. It didn’t work as well as the BMS version crossed at 650Hz. Once you get up around 1000Hz the 10’s don’t play that well together, it works but its not as good. Some of the new 4" compression drivers from B&C, 18sound, RCF etc. may be an option crossed at 700 - 800Hz.

You could make the box as a MTM with the horn in the middle as Marjan suggested. I think this would work acoustically better. I decided to put the horn on top simply because it made it easy to get the HF above the audience’s head height.

Like Marjan, I’m not a fan of 2.5 way crossovers. The 250 Hz crossover on one driver creates phase cancelation and directivity issues that can be difficult to deal with.
 
Last edited by a moderator:
Re: Double 10 and Horn

We will agree that all engineering involves compromise. There are compromises we can live with, and those we cannot.

Why the desire for a 2 x 10, when interaction between the 2 cones introduces fundamentally unresolveable issues? There are new generation 12"/13"/14" that offer reasonable piston area and power handling without the propagation interference issues of dual drivers..

The 2.5 way phase/directivity issues can be filtered to occur across a narrower bit of bandwidth than the "natural" interactions of unbridled 2 x 10s.

As far as the +3db issue goes, the .5 xover freq choice should in actuality be much further down-spectrum, where the additional swept cone aids tailing response and excursion limits. IOW: Where it can really be of help.
 
Last edited:
Re: Double 10 and Horn

We will agree that all engineering involves compromise. There are compromises we can live with, and those we cannot.

Why the desire for a 2 x 10, when interaction between the 2 cones introduces fundamentally unresolveable issues? There are new generation 12"/13"/14" that offer reasonable piston area and power handling without the propagation interference issues of dual drivers..

The 2.5 way phase/directivity issues can be filtered to occur across a narrower bit of bandwidth than the "natural" interactions of unbridled 2 x 10s.

As far as the +3db issue goes, the .5 xover freq choice should in actuality be much further down-spectrum, where the additional swept cone aids tailing response and excursion limits. IOW: Where it can really be of help.

If the 10s are crossed low enough then they will sum very well, which they do.

The crossover frequency is 650Hz. Its wavelength is 528mm, more or less twice the spacing of the 10’s.

Now because the low frequency source is not circular and is twice as high as it is wide its directivity is a better match for the 90 x 50 HF horn
.
With two drivers you can achieve an efficiency greater than 99dB w/m with 700 watts of power handling. The box can also be quite small with a very useful form factor.

The 2.5 way design has a lot of issues in my opinion. The phase shift caused by the low pass filter means that the LF 10 will not combine correctly with extended range 10 unless you implement some fancy all-pass correction. In addition one 10 is doing much more work than the other; that is you have not used all the power available.

There is also no reason that you could not use one of the new 14’s or 15’s that are available. They will give you more low frequency but the trade-off will be efficiency in the low and upper mids.
The double 10 was designed as a light weight (25 kg) mid/hi box not a full range cabinet … it just depends on what you want.
 
Last edited by a moderator:
Re: Double 10 and Horn

The crossover frequency is 650Hz. Its wavelength is 528mm, more or less twice the spacing of the 10’s.

I would argue (and god knows I argue too often), that this spacing is unhelpful in the upper range of the dual 10 operation, just on its face. Measurement may prove otherwise, but why build and hope?
 
Re: Double 10 and Horn

I would argue (and god knows I argue too often), that this spacing is unhelpful in the upper range of the dual 10 operation, just on its face. Measurement may prove otherwise, but why build and hope?

This is what happens when you model 2 point sources spaced a 10 inches @ 630 Hz ... just what I wanted :razz:
(HP filter - 100Hz)
 

Attachments

  • 630 dbl10.jpg
    630 dbl10.jpg
    79.6 KB · Views: 121
Re: Double 10 and Horn

This is what happens when you model 2 point sources spaced a 10 inches @ 630 Hz ... just what I wanted :razz:
(HP filter - 100Hz)

Predictive software makes things appear so easy.

Now add the 3rd point source (horn).

Now adjust to assume that listener/mic isn't at absolute X/Y center.

Now introduce filtering to all 3 devices to mimic real-world response any passive + active inputs + plus component behavior.

Only then do we get closer to a CAD rendered idea of pattern.

Then build trial box, measure, adjust, measure again...
 
Re: Double 10 and Horn

Predictive software makes things appear so easy.

Now add the 3rd point source (horn).

Now adjust to assume that listener/mic isn't at absolute X/Y center.

Now introduce filtering to all 3 devices to mimic real-world response any passive + active inputs + plus component behavior.

Only then do we get closer to a CAD rendered idea of pattern.

Then build trial box, measure, adjust, measure again...

Hi Jim,

I’m not sure I understand your point completely ...

The original design is about 15 years old now, with at least 60 boxes being made and used on a huge number of shows very successfully. The original version was switchable passive / active.

This is the latest version which has greatly improved performance over the original. Not being a symmetrical design it does compromise the directivity a little, but it makes getting the HF above the audiences head a little easier.

Its light, measures very well, sounds great and goes surprisingly loud when combined with a good double 18.

Basically it my version of JTR Noesis 3TX, except its active, uses slightly better components and was initially designed in about 2000.
 
Re: Double 10 and Horn

Hi Peter,

I'm very interested in this design, although for slightly different reasons, particularly form factor.

We do a lot of wedding ceremonies where I live in NZ. I've been on the lookout for a more suitable (and higher SQ) option for some time. The requirements necessitate ground placed speakers only (cabs on sticks are considered an absolute no-no visually). Up till now we've been making do with what we have - a nothing special pair of 12" & horn actives, generally placed slightly wider than the seating, about halfway between the bridal party and the audience, with toe-in for coverage. It's less than ideal - the horn is much lower than desirable, and I doubt the speaker holds pattern very well, contributing to feedback issues with the lapel mics. Content is primary about the vocals (lapels on groom + celebrant, + a mic on stand) and typically includes recorded processional / registry signing / recessional music, or occasionally live music (say solo guitar / vocal) instead.

I'm interested in adapting your design to provide 'just' enough half-space bass response to be usable / sound passable for the recorded / live music in this situation (no-one is expecting epic bass here - just not blatantly missing / tinny sounding), while still being able to double duty as cabs on sticks for speeches / background music, and as kick-arse P.A. tops.

The quasi-fullrange SPL requirements for ceremonies would be pretty moderate (the front row is generally pretty close to the speakers). The TMM format would certainly help get the horn in a better location (I could even afford to go a little taller). I'm aware of the issues with driver unloading below port tuning freq, and consequently tuning may have to be kept a bit lower, almost a kind of EBS alignment perhaps? I'd like to maintain as much sensitivity above 90Hz as possible, and squeeze out just enough extension in 1/2 space.

I'd likely run them bi-amped, with typical 'affordable' processing (dcx, PA2, bunker etc). I'd dearly love to get 4594's for them, but may have to settle for something like the radian 951, with the option to upgrade to 4594 / tri-amping later.

I'd really appreciate your thoughts & suggestions on this, as you seem to be well ahead of the game!

Kind regards,
Boyd
 
Re: Double 10 and Horn

Hi Peter,

I'm very interested in this design, although for slightly different reasons, particularly form factor.

We do a lot of wedding ceremonies where I live in NZ. I've been on the lookout for a more suitable (and higher SQ) option for some time. The requirements necessitate ground placed speakers only (cabs on sticks are considered an absolute no-no visually). Up till now we've been making do with what we have - a nothing special pair of 12" & horn actives, generally placed slightly wider than the seating, about halfway between the bridal party and the audience, with toe-in for coverage. It's less than ideal - the horn is much lower than desirable, and I doubt the speaker holds pattern very well, contributing to feedback issues with the lapel mics. Content is primary about the vocals (lapels on groom + celebrant, + a mic on stand) and typically includes recorded processional / registry signing / recessional music, or occasionally live music (say solo guitar / vocal) instead.

I'm interested in adapting your design to provide 'just' enough half-space bass response to be usable / sound passable for the recorded / live music in this situation (no-one is expecting epic bass here - just not blatantly missing / tinny sounding), while still being able to double duty as cabs on sticks for speeches / background music, and as kick-arse P.A. tops.

The quasi-fullrange SPL requirements for ceremonies would be pretty moderate (the front row is generally pretty close to the speakers). The TMM format would certainly help get the horn in a better location (I could even afford to go a little taller). I'm aware of the issues with driver unloading below port tuning freq, and consequently tuning may have to be kept a bit lower, almost a kind of EBS alignment perhaps? I'd like to maintain as much sensitivity above 90Hz as possible, and squeeze out just enough extension in 1/2 space.

I'd likely run them bi-amped, with typical 'affordable' processing (dcx, PA2, bunker etc). I'd dearly love to get 4594's for them, but may have to settle for something like the radian 951, with the option to upgrade to 4594 / tri-amping later.

I'd really appreciate your thoughts & suggestions on this, as you seem to be well ahead of the game!

Kind regards,
Boyd
Hi Boyd,

The double 10 was originally designed as a Mid / Hi box that could go low enough to be used as a vocal PA. The original version crossed to the Horn at 1.2 KHz and was switch-able passive / active.

The next version still used a “normal” compression driver but crossed at 900Hz. This was a big improvement. Once you get pass about 800Hz you will start to have issues with how the 2 x 10” drivers interact with each other. The trick was to find a suitable horn flare, which I did, the RCF 950.

The current version uses the BMS 4594 and crosses to the 10’s at 630Hz. This is even better.

If you want to build something like this I would suggest selecting some 10” drivers that will go lower and give you more bass than what I have used. Select a suitable sized box and tune it as needed. Use the RCF 950 horn with a suitable 1.4” compression driver …. Then the tricky bit … creating some DSP settings :)
 
Re: Double 10 and Horn

Hi Boyd,

The double 10 was originally designed as a Mid / Hi box that could go low enough to be used as a vocal PA. The original version crossed to the Horn at 1.2 KHz and was switch-able passive / active.

The next version still used a “normal” compression driver but crossed at 900Hz. This was a big improvement. Once you get pass about 800Hz you will start to have issues with how the 2 x 10” drivers interact with each other. The trick was to find a suitable horn flare, which I did, the RCF 950.

The current version uses the BMS 4594 and crosses to the 10’s at 630Hz. This is even better.

If you want to build something like this I would suggest selecting some 10” drivers that will go lower and give you more bass than what I have used. Select a suitable sized box and tune it as needed. Use the RCF 950 horn with a suitable 1.4” compression driver …. Then the tricky bit … creating some DSP settings :)

Hi Peter, Thanks for the prompt reply :)

Yes, I understand all of the above (I've also read the entire thread on your dual 12 Box - very nice!).

From what I understand the Radian 951 is good for 500Hz with a -24dB/Oct HPF slope, or 900Hz with a -12dB/Oct HPF slope in a suitable horn, so I'd guesstimate it should work with say, a 3rd order slope for somewhere around 650 - 700Hz. If I can stretch to the BMS coax, I will use it.

The RCF 950 is the obvious choice, as you've already tried and tested it, and 90 deg is probably most suitable for my single box per side purposes. I've no intention to deviate from this design any more than I have to, in order to squeeze out just a little more extension. With a suitable port tuning, and decent xmax drivers in 1/2 space, it seems achievable on the face of it.

Where I fall down here is the lack of modelling software, and proficiency in its use (suggestions welcome). The B&C 10NW64 or Faital Pro 10FH520 seem to me like possible contenders? I'm figuring that 60Hz is probably sufficient, so 1/2 an octave. I really don't mind if the sub 100Hz response is at a reduced level, and relies on 1/2 space loading and a bit of bass boost EQ to get there. For the cabs on sticks / speeches / BG music scenario the additional extension probably isn't necessary, so I doubt the loss of boundary reinforcement would be an issue.

I'm not quite sure what kind of alignment or speaker parameters such as Qts would best help me achieve my goal. I do have a measurement mic, and can figure out some DSP settings OK (though I'm sure nothing like what you can do with a Lake DSP!).
 
Re: Double 10 and Horn

Have you considered an additional 2x 10" module that could be attached to the top of the dbl 10"+horn for higher output applications? Seems to me like it could result in an arrangement similar to your dual 12" horn (assuming the dual 10" sections would play together happily at that separation & crossover freq), and could probably be tested quite easily by stacking another cab on top, and driving only the 10"s in the top cab, and boosting the HF/VHF (c2c would be a bit more for the top cab, but if it works, customised cabs could make it work even better). Just a thought :razz: