First trials of active horn speaker

manisandher

Well-Known Member
Feb 7, 2011
243
217
950
www.the2ndtier.com
Is that a B&C compression tweeter?

Mani.
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Is that a B&C compression tweeter?

Mani.
Hi Mani,
No, it is a Beyma CP755Ti. Huge beast (7Kg). Initial listening suggests that this driver with this horn is smooth baby smooth.
 

manisandher

Well-Known Member
Feb 7, 2011
243
217
950
www.the2ndtier.com
Ah OK. Thanks.

Look forward to hearing more about these speakers...

Mani.
 

Robh3606

Well-Known Member
Aug 24, 2010
1,471
461
1,155
Destiny
"Crossed at 650hz with 2nd order Butterworth. Horn gain is reduced by 14db compared to 96db Fostex driver! "

Hello morricab

Nice! So what type of horn flare?? You doing any horn compensation on the compression driver??

Rob :)
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
"Crossed at 650hz with 2nd order Butterworth. Horn gain is reduced by 14db compared to 96db Fostex driver! "

Hello morricab

Nice! So what type of horn flare?? You doing any horn compensation on the compression driver??

Rob :)
The horn flare is an elliptical profile designed for constant coverage and minimal diffraction. Not sure what you mean horn compensation on the compression driver.

Did some more experiments yesterday and found 600hz 2nd LR works even better. This is the low end of the range for both driver and horn but at my volume levels there shouldn’t be any stress. I will hopefully make some measurements soon to see how well everything is really blending. This CP755ti is really smooth sounding and I don’t hear coloration from the horn, which is important.

I also have a pair of Beyma CP350TI drivers with elliptical horns that I might try as an alternative but only crossed above 1.5khz.

Later I can see what sounds better but so far with this bigger horn/driver a lower xover sounded better.
 
  • Like
Reactions: Robh3606

Robh3606

Well-Known Member
Aug 24, 2010
1,471
461
1,155
Destiny
Hello morricab

Depending on horn type CD or Exponential, as examples, the Exponential would be relatively flat on axis because it uses the DI ,directivity index ,which increases with frequency to compensate for the mass roll off of the compression driver. You are flat on axis and rolled off off axis. The off axis roll off gets more pronounced the higher you go.

A modern CD horn has a much more constant DI with frequency so the on axis response follows the mass roll off. You typically get a peak in output around 1-3K and a roll off above. This will be consistent off axis as well because of the more constant DI. On a CD horn you would use either passive or active EQ to flatten both the on and off axis response. That EQ is the horn compensation I was curious about.

The amount of roll off in a CD horn gets less the narrower the dispersion pattern is. So a 60X40 would require less EQ, horn compensation, compared to say a 60X100 as an example. The only way to know for sure is look at the on axis response curve for your particular compression driver and horn combo as different responses are certainly possible on the same horn with a different compression driver.

I as a rule use CD type horns so all of my systems have horn compensation built into their passive crossovers. Even my active set-ups use passive compensation as all of my active crossovers are simple analog. I enjoy designing the passive networks even though it can be much more time consuming than using DSP active filters.

With you using a digital active it should be quite simple for you to dial them in if you feel the need. Modern digital DSP is a mighty handy tool. One of these days!

Rob :)
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Dear OP,

Please can you tell us more about how the mini DSP is working out for you as I deliberated it for my project but decided to keep purist (but still my get one to experiment with). Also what was the design criteria for selecting the Fostex? Did you consider trying to back load that to get sensitivity up a lot to avoid curtailing the compression driver so much?
 

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
Curtailing was a bad choice of word - I meant attenuating to be precise.
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Hello morricab

Depending on horn type CD or Exponential, as examples, the Exponential would be relatively flat on axis because it uses the DI ,directivity index ,which increases with frequency to compensate for the mass roll off of the compression driver. You are flat on axis and rolled off off axis. The off axis roll off gets more pronounced the higher you go.

A modern CD horn has a much more constant DI with frequency so the on axis response follows the mass roll off. You typically get a peak in output around 1-3K and a roll off above. This will be consistent off axis as well because of the more constant DI. On a CD horn you would use either passive or active EQ to flatten both the on and off axis response. That EQ is the horn compensation I was curious about.

The amount of roll off in a CD horn gets less the narrower the dispersion pattern is. So a 60X40 would require less EQ, horn compensation, compared to say a 60X100 as an example. The only way to know for sure is look at the on axis response curve for your particular compression driver and horn combo as different responses are certainly possible on the same horn with a different compression driver.

I as a rule use CD type horns so all of my systems have horn compensation built into their passive crossovers. Even my active set-ups use passive compensation as all of my active crossovers are simple analog. I enjoy designing the passive networks even though it can be much more time consuming than using DSP active filters.

With you using a digital active it should be quite simple for you to dial them in if you feel the need. Modern digital DSP is a mighty handy tool. One of these days!

Rob :)
Ok, I understand you well now. Thanks for clarifying. So far, I have not made measurements as I didn't really have much free time. I have been tuning by ear but will put the measurement tools to use soon and see how close I got "by ear" to a reasonable response. I did do a bit of boost above 5Khz because that is where the driver itself starts to droop with a steady shallow slope to 20Khz. I put in about 3-5 db of boost and left the rest flat (the driver actually measures very flat from about 600Hz up to 4-5Khz); however, I don't know yet what the horn is contributing because the measurements from Beyma were in a Beyma horn. Once I measure a bit on and off-axis I will be able to give you a better idea what this driver is doing in this horn.

I will likely eventually build a passive xover that will replace the miniDSP but it is just too easy for developing not to use one! I also have a Behringer DCX2496 that I could have used but it would never be a permanent solution whereas the miniDSP could conceivably be a final stop. I might also, for better FR control, use my Behringer DEQ2496 digital in/digital out prior to going into the miniDSP and using it only as a raw crossover and driver level matcher...let's see. Going the miniDSP route all digital means I needed two DACs (three if I ever go three way!) and two amplifiers (still shopping for the "final" amp for the horn).

The little Metrum DAC I am using on the mid/bass driver is really quite good considering how low money it cost me! The Monarchy is simply a great DAC that will go toe to toe with 5K+ DACs any day of the week...love it.
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Dear OP,

Please can you tell us more about how the mini DSP is working out for you as I deliberated it for my project but decided to keep purist (but still my get one to experiment with). Also what was the design criteria for selecting the Fostex? Did you consider trying to back load that to get sensitivity up a lot to avoid curtailing the compression driver so much?

So far it works very well. It does precisely what I wanted from it, which is xover, attenuation level matching, time delay and some eq. I didn't want one with analog out (I already have that with a Behringer DCX2496) because I wanted to use high quality DACs. The other amp I will get will need to be integrated because then I will do the level matching in the analog domain with amp gain rather than with digital attenuation (not sure how much that is degrading the sound with this box at this time). I might eventually go passive with it but this is a great design tool and the SQ from only one speaker so far (need that other amp first) is really quite good and measurements should help me dial it in further both in frequency and time.

The bottom part is a commercial speaker (Decware HDT) so that design was settled on a long time ago. It actually gets a lot more bass out of this Fostex (a proper 38hz) than any of the backloaded designs. The cabinet design is using multiple tuning with three internal ports, a transmission line of sorts and passiver radiators on the sides. The damn thing is very fast and the cone barely moves...like in a horn! Horn loading would boost maybe a few db but sacrifice the bass punch. 96db is pretty good and low level listening its forte. As mentioned above, I want to eventually have analog gain level matching rather than digital attenuation.

That being said, I will replace them with Supravox single driver speakers (99db and no whizzer cone) in a TQWT loading. Evenutally I will transplant that driver into a new DTQWT with 2 x 10 or 2 x 12 inch drivers on the backside (like Horning is doing). The Supravox driver will also feed the TQWT then and contribute to the bass along with the two dedicated bass drivers. This would be a variation on what Troels Gravessen has done in DIY (but later as a kit from Jansen). I will use a preferred midbass and true horn upper mid/high rather than his midbass choice and a normal softdome tweeter in a waveguide (not my first choice for HF drivers).
 
  • Like
Reactions: Audiophile Bill

Audiophile Bill

Well-Known Member
Mar 23, 2015
4,293
4,093
675
So far it works very well. It does precisely what I wanted from it, which is xover, attenuation level matching, time delay and some eq. I didn't want one with analog out (I already have that with a Behringer DCX2496) because I wanted to use high quality DACs. The other amp I will get will need to be integrated because then I will do the level matching in the analog domain with amp gain rather than with digital attenuation (not sure how much that is degrading the sound with this box at this time). I might eventually go passive with it but this is a great design tool and the SQ from only one speaker so far (need that other amp first) is really quite good and measurements should help me dial it in further both in frequency and time.

The bottom part is a commercial speaker (Decware HDT) so that design was settled on a long time ago. It actually gets a lot more bass out of this Fostex (a proper 38hz) than any of the backloaded designs. The cabinet design is using multiple tuning with three internal ports, a transmission line of sorts and passiver radiators on the sides. The damn thing is very fast and the cone barely moves...like in a horn! Horn loading would boost maybe a few db but sacrifice the bass punch. 96db is pretty good and low level listening its forte. As mentioned above, I want to eventually have analog gain level matching rather than digital attenuation.

That being said, I will replace them with Supravox single driver speakers (99db and no whizzer cone) in a TQWT loading. Evenutally I will transplant that driver into a new DTQWT with 2 x 10 or 2 x 12 inch drivers on the backside (like Horning is doing). The Supravox driver will also feed the TQWT then and contribute to the bass along with the two dedicated bass drivers. This would be a variation on what Troels Gravessen has done in DIY (but later as a kit from Jansen). I will use a preferred midbass and true horn upper mid/high rather than his midbass choice and a normal softdome tweeter in a waveguide (not my first choice for HF drivers).

Hi,

Thanks for the detailed response - it makes sense.

I thought the Aries horn is a back loaded design?
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Hi,

Thanks for the detailed response - it makes sense.

I thought the Aries horn is a back loaded design?

It is but it is a pretty long horn (over 3 meters). Also, the construction is quite difficult as there are no sharp corners...it is all machined as a smooth folded contour. I don't have the tools to make something so nice like that...many speaker builders do not have it either. Any horn I would build would be compromised too much and so a TQWT is a good and relatively simple compromise...
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Slight update: I hooked up the system with some silver/palladium ribbon loudspeaker cables (still only one speaker) and WOW this driver/horn combo is good! So smooth yet dynamic and lovely tone. The little MasterSound does well in this role. Bass control was a bit looser than with the Vacuumstate copper ribbons but higher resolution and better blend with the horns.

Finally got to play it loud and it rocks but so smooth and resolved at the same time.

Currently the setup:
Tweeter: Beyma CP755Ti/18 sound XT1464
Bass/mid: Decware HDTS
Minidsp all digital
Midbass Dac:Metrum “Octave Dac”
Tweeter Dac: Monarchy M24
Amp: Mastersound Dueundici
Crossover: 2nd order LR 600Hz
 
  • Like
Reactions: Audiophile Bill

DaveC

Industry Expert
Nov 16, 2014
3,899
2,141
495
Hmm, if you're going to xo @ 600 Hz you really don't need a single driver. I'd consider a larger woofer that will match your horn's directivity at xo... maybe an Acoustic Elegance TD12 or similar.

Right now it's likely the dispersion of the single driver and horn do not match so you have a step in the polar response which is nice to avoid if possible.
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Hmm, if you're going to xo @ 600 Hz you really don't need a single driver. I'd consider a larger woofer that will match your horn's directivity at xo... maybe an Acoustic Elegance TD12 or similar.

Right now it's likely the dispersion of the single driver and horn do not match so you have a step in the polar response which is nice to avoid if possible.
It’s what I had handy to play with :).

Not sure why the polars would be so out of whack as the single driver would not be very directional and this frequency...the horn will be less constant down there though. I tried 800hz and found 600 better (the horn/driver is really open and transparent). I will play more a bit higher up where the horn has true constant directivity (from 800hz).
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
Latest update: It lives! Picked up a nice little 5 watt (Class A Triode) EL84 PP amp to go with the Mastersound Dueundici and I now have both channels up and running.
Setup is as follows: Midbass Decware HDTs fed from Metrum Octave DAC and Dueundici amp. Horn is Beyma CP755TI + 18 Sound XT 1464 fed from Monarchy Audio M24 DAC into a WLM Minueta in triode mode.
Crossover: minidsp nanodigi : 12db/Octave LR design at 700HZ. Mild eq on bass hump at 90hz and mild eq on horn to boost frequency above 7khz, which rolls off a bit too fast (mainly for a bit more “air”).
AAC98830-90A1-4474-9863-2C882049EE17.jpeg
 

Folsom

VIP/Donor
Oct 25, 2015
6,024
1,490
520
Eastern WA
So does it sound right at all having two different DAC's and amplifiers?
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,986
978
Switzerland
So does it sound right at all having two different DAC's and amplifiers?

Yup , seems to work very well. I will swap amps around at some point to see if the SEP works better on the horns than the PP triode but the DACs will stay as is because the Monarchy has a better high frequency and spatial presentation.
 

Robh3606

Well-Known Member
Aug 24, 2010
1,471
461
1,155
Destiny
Since you are losing 14db have you tried the opposite approach of rolling off your lower end 600 and up? This way you bleed off the difference with minimal HF boost. That is the worst place to add power to the driver as it is already putting out higher thd there from the controlled resonanses and break up in the last octave or so. This will simulate what your passive crossover will do. You can use the voltage drive from the DSP as a reference to design and build your crossover.

Rob:)
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing