As the title says, I'm on the search for information about coaxial drivers. I'd like to get a better understanding of the characteristics of the different types. And is there a design that is better than another?
My current speakers use a coax driver with a separate HF wave guide for the coax dome tweeter. Then I see there are also coax speakers use the woofer cone as the wave guide for a dome tweeter. And there are coax drivers that use a compression driver.
I am playing with the idea of adding a coax driver to the project I'm working on (home stereo, sound quality prioritized over spl). If I choose a coax driver, I would be looking for a 6" to 8". The coax driver would be for mids/ highs, and it will be crossed to a 12" woofer at around 300hz.
Is there a particular advantage to a coax driver using a compression driver for HF? I don't have any experience with horn speakers in hifi speakers, so I don't know what to expect. I've heard people discussing that the HF wave guide in front of the mid driver causes distortion, depending on the size of the wave guide. And people talk about how cone excursion is a problem when using the woofer as the HF wave guide.
I'd like to hear what you guys have to say on the subject.
My current speakers use a coax driver with a separate HF wave guide for the coax dome tweeter. Then I see there are also coax speakers use the woofer cone as the wave guide for a dome tweeter. And there are coax drivers that use a compression driver.
I am playing with the idea of adding a coax driver to the project I'm working on (home stereo, sound quality prioritized over spl). If I choose a coax driver, I would be looking for a 6" to 8". The coax driver would be for mids/ highs, and it will be crossed to a 12" woofer at around 300hz.
Is there a particular advantage to a coax driver using a compression driver for HF? I don't have any experience with horn speakers in hifi speakers, so I don't know what to expect. I've heard people discussing that the HF wave guide in front of the mid driver causes distortion, depending on the size of the wave guide. And people talk about how cone excursion is a problem when using the woofer as the HF wave guide.
I'd like to hear what you guys have to say on the subject.
Hi,
I compression drivers in coax: the ones i liked the most until now Tannoy system 15dmtII and System 1200 both used it.
I own System 800 and in direct comparison the 1200 had more 'weight'/ 'mater' if that means something to you. The one i've heard where more true in term of dynamic and impact ( but they are 12" and 15" and higher efficiency than typical domestic loudspeakers which can play a role too).
I can see a technical advantage in the fact compression drivers have far more ouput potential ( spl) and the high range is usually where there is large dynamic needs ( crest factor is higher on this part of spectrum so high output potential= more headroom).
It's a mater of preference in the end. I can live with both.
The 'protubering' horn will generate issues. I was sceptical about principle ( i've heard Urei monitors previously, liked them without being blown away as i was with System15) until i heard in past years Presonus Scepter8 which use Fir to correct the issue ( Fulcrum audio licenced the technology to Presonus).
There is video describing the principle they implemented. If i understood corectly it could be implemented ( not easily through).
On my System 800 ( direct radiator used as waveguide) i heard intermodulation issues when pushing them. Once highpassed at 250/275hz it is far less noticable.
In my view coax ask for three or more ways...
Your plan sound good to me.
6? If you can accept 5,5"... there is Kef ( but dome...). Once i've looked at measurements of r300/800 i needed a pair. So my next one are going to be dome based again ( but they offer other interesting features for me).
Adason: it depends. Latest generation 6,5" are great (the one used in Revolution xt iirc). Even the 6.5" of system 600 is great IF implemented using the active circuitry of the A version ( Active). It is the same with mine: there is physical offset between both driver in Z axys and it needs to be adressed or there is a 'hole' in the fr around 1,5khz ( in mine, in system600 it must be higher i think). The passive version suffer from it, active with allpass filters implemented cure it.
I compression drivers in coax: the ones i liked the most until now Tannoy system 15dmtII and System 1200 both used it.
I own System 800 and in direct comparison the 1200 had more 'weight'/ 'mater' if that means something to you. The one i've heard where more true in term of dynamic and impact ( but they are 12" and 15" and higher efficiency than typical domestic loudspeakers which can play a role too).
I can see a technical advantage in the fact compression drivers have far more ouput potential ( spl) and the high range is usually where there is large dynamic needs ( crest factor is higher on this part of spectrum so high output potential= more headroom).
It's a mater of preference in the end. I can live with both.
The 'protubering' horn will generate issues. I was sceptical about principle ( i've heard Urei monitors previously, liked them without being blown away as i was with System15) until i heard in past years Presonus Scepter8 which use Fir to correct the issue ( Fulcrum audio licenced the technology to Presonus).
There is video describing the principle they implemented. If i understood corectly it could be implemented ( not easily through).
On my System 800 ( direct radiator used as waveguide) i heard intermodulation issues when pushing them. Once highpassed at 250/275hz it is far less noticable.
In my view coax ask for three or more ways...
Your plan sound good to me.
6? If you can accept 5,5"... there is Kef ( but dome...). Once i've looked at measurements of r300/800 i needed a pair. So my next one are going to be dome based again ( but they offer other interesting features for me).
Adason: it depends. Latest generation 6,5" are great (the one used in Revolution xt iirc). Even the 6.5" of system 600 is great IF implemented using the active circuitry of the A version ( Active). It is the same with mine: there is physical offset between both driver in Z axys and it needs to be adressed or there is a 'hole' in the fr around 1,5khz ( in mine, in system600 it must be higher i think). The passive version suffer from it, active with allpass filters implemented cure it.
Last edited:
The foolish plan I have in mind is either a 3.5way or 4 way design, depending on how you view the world. I started with the idea of replicating a Genelec W371A, which is a low frequency cabinet designed to augment their "The Ones" monitors.
Now I'm asking about coax drivers, because as I stare at the test enclosure I hastily slapped together, I'm realizing it would take almost zero extra effort to add a coaxial driver to the cabinet.
Why coax? I guess this project is about doing something a little different, and using a coax driver is an interesting idea to me.
I said 6-8" driver, meaning smaller than a 12" or 15". A little smaller than 6" is fine.
Discussions I have read so far indicate that the woofer cone as wave guide is primarily a problem when there is a lot of excursion. So it seems to me that using a 3-way design will eliminate this problem? Or is there another problem with this design?
A poorly done coaxial can have very strong cancellations and voids in the frequency response. KEF of all producers are periodically guilty of this, but their drivers are top notch. My current shooting star is the Q100 driver. Small enough to produce midrange at a wide angle, and their tangerine waveguide really does keep the high-end consistent. I want to match it with some serious 8" drivers or an array of 6.5"
Best IME is $$ Altec/GPA $$, though not available in <12" frames, but all things considered, whizzer cone coax's for the upper mids with the dustcap as tweeter is what many of us have used over the decades and there's a number of good ones in the 6-8" range, so best to research/query the Full Range Forum.
Yes intermodulation is a potential issue but high pass to keep only mid on coax woofer and it is greatly reduced.
So if Kef i would use same one i ordered: SP1753 ( the one used in R300 and up) as it is crossed at 400hz in commercial range, but could go down to 300hz with low issues. And while at it order the 'ring' they call 'shadow flare', it act as an aditional waveguide and should help with diffraction.
Sealed it require low box volume. Maybe 2/3liters. Maybe even less.
The real issue cloning a 'one' is the 8361's box is designed to lower diffraction to minimum and is a waveguide at same time.
It'll ask for efforts to build something inspired at home.
But maybe it is doable.
I'll follow your journey. Have you documented your w371a project? I'm interested in.
D1sco: it seems we all want to do the same more or less... 🙂
Thanks. I'll take a look. I don't know what to think about whizzer cones, since I have never heard one myself, atleast not that I am aware of.
I was browsing "pro" coaxial drivers, and there is a decent selection to choose from on parts express. Most of them are from B&C, and most have a compression driver. A compression driver seems like a good way to implement a coaxial driver, but I do not have any experience with those either..
I'm focusing more on the W371a claims to do, and less on the 8361 monitors. The Genelec aluminum cabinets are nice, but I suspect that placing them on a big rectangular box creates some reflections.
I was planning on using my existing monitors at first. But as I think about it, I keep coming back to the idea that a box sitting on another box is going to create reflections that wouldn't occur if it was a single full range cabinet. Genelec has business reasons for sticking to the good small-ish monitor that can be augmented with the W371 cabinets. But for me, it may make more sense to combine it all into a single cabinet. I'm not sure what advantage there would be in stacking multiple cabinets?
I did start a thread in subwoofers. "Not your average subwoofer". So far I just confused everyone by struggling through simulations while making mistakes...but I will have a more substantial update soon. At the moment I have a mock-up with drivers installed, but the DSP unit was DOA. That is being returned. So I'm just waiting at the moment.
Any stand being a loudspeaker or not, is going to generate reflection and diffraction. The thing is to understand what they are, where they are located in spectrum and if they can be managed or not.
It doesn't stop with diffraction: the way the low end interact with the room change when on stand located or with a solid plane going down to floor and or ceiling ( faceplate).
Iow time to simulate... then prototype it.
And to look at what is done by commercial manufacturer about this: Kef white paper are worth a read, all of them...
It might ask different strategy depending on your own application that the one they used though. But most working strategy against edge diffraction they use on one model or another.
You talked about studio monitor use, what are your requirements for spl @ listening point?
Yes try with what you have at hand, it is the point with prototype: check if the global idea ( acoustic design) 'works' in real life.
I can see why i would like a a system based upon a 'mini monitor' that could be expanded ( this is what i'm going to do) but your needs are probably differents than mine. Whatever suit you is fine, you are the end user and 'custom' is the point of diy in my view.
Will take a look thanks.
You've got multi in out interface? No need to wait for a dsp... use a dsp freeware meanwhile...
Last edited:
It's easy to come to the same conclusions when you follow the science
There really needs to be quality third party measurements of these compression drivers. A horn is theoretically a good place to disperse complex signals, but it could also be disastrous. Think Unity Horn vs a college project. In concerts there's so much cancellation anyways I don't know if I could tell they were helping or hurting in their assumed role.
My SPL requirements are fairly mild. My usual listening is maybe 70db. Occasionally I will crank it, but I don't know how loud that actually is. Maybe up to 100db for short periods.
In my simulations I'm at 105db on the 10" subwoofer at peak amplifier power, And more like 110db for the 12" woofers. I figure that's plenty!
I playing music on the prototype speaker right now. Left is the 12", and right is the 10" subwoofer. No EQ or crossovers at all. I'm pleasantly surprised. The 12" w/ PR extends higher than I expected.
I did a couple sweeps in REW and played some tones...I'm thinking I can shrink the cabinet without losing any useful sound info. I wanted an F3 below 20hz just because, but with the help of a tone generator I'm realizing 20hz really isn't audible to me. I'll reserve judgement until I can add some EQ to get a flatter response, but for now it seems that if I designed for an F3 of 23hz, I'd shrink the cabinet by 20L and not lose any audible content.
But for now, this is the prototype cabinet. It would be so easy to add a coax driver to the top and make it truly full range! (This is a 12" driver w/ matching 12" PR. The subwoofer is rear firing.)
I did find this bench test of a B&C 5" coaxial:
https://audioxpress.com/article/test-bench-b-c-speakers-5cxn44-pro-sound-coaxial-driver
I'm not very familiar with this kind of data, so it'll take me some effort to work through it. But at first glance it seems decent.
Last edited:
Now we have pictures: why bother with a small diameter coax?
I mean, your box is already wide enough and would accept a 12" with minimal baffle so not bad from an edge diffraction point of view.
A 12" could be crossed at 1,2khz, some could even accept 1khz. They will already be in 'control' of directivity at this freq too.
Given you have a sub you could run the box sealed and x around 280hz/300hz. This will ask for a small box volume and a linkwitz transform but as efficiency could already be high to boot with... nothing problematic imo.
If adventurous you could even have coax woofer cardiod passive too... it could make a very interesting loudspeaker imo.
I mean, your box is already wide enough and would accept a 12" with minimal baffle so not bad from an edge diffraction point of view.
A 12" could be crossed at 1,2khz, some could even accept 1khz. They will already be in 'control' of directivity at this freq too.
Given you have a sub you could run the box sealed and x around 280hz/300hz. This will ask for a small box volume and a linkwitz transform but as efficiency could already be high to boot with... nothing problematic imo.
If adventurous you could even have coax woofer cardiod passive too... it could make a very interesting loudspeaker imo.
Hmmm....
Just for the record, the little black subwoofer next to the cabinet is going away. The prototype cabinet has it's own 10" subwoofer in the back:
Now, are you suggesting I add a 12" coaxial driver in addition to the 12" woofer and 12" PR? That's a stack of 3 12" speaker cones. It feels slightly ridiculous, but I kind of like it.
I quick sketch on the computer indicates the center of the coax woofer center would be ~35" high with close spacing. That's almost exactly the same tweeter height I have had my speaker set at. And the Total cabinet height would be ~43" with that same spacing.
The original idea is that the subwoofer covers from 20hz to ~200hz (if possible..), and the 12" woofer overlaps the subwoofer from~60hz to 200+ hz.
And you're proposing crossing from the 12" woofer to a 12" coax at ~300hz, instead of getting a "proper" midrange driver? (Sorry if I am being tedious, but I'm just wrapping my mind around the idea. This definitely isn't what I imagined, but it is interesting.
Do you have a coax woofer in mind?
Just for the record, the little black subwoofer next to the cabinet is going away. The prototype cabinet has it's own 10" subwoofer in the back:
Now, are you suggesting I add a 12" coaxial driver in addition to the 12" woofer and 12" PR? That's a stack of 3 12" speaker cones. It feels slightly ridiculous, but I kind of like it.
I quick sketch on the computer indicates the center of the coax woofer center would be ~35" high with close spacing. That's almost exactly the same tweeter height I have had my speaker set at. And the Total cabinet height would be ~43" with that same spacing.
The original idea is that the subwoofer covers from 20hz to ~200hz (if possible..), and the 12" woofer overlaps the subwoofer from~60hz to 200+ hz.
And you're proposing crossing from the 12" woofer to a 12" coax at ~300hz, instead of getting a "proper" midrange driver? (Sorry if I am being tedious, but I'm just wrapping my mind around the idea. This definitely isn't what I imagined, but it is interesting.
Do you have a coax woofer in mind?
Here's a quick sketch if the above concept:
It works on paper. But is it a good idea for high quality audio? I'm not sure. But it's fun to think about.
It works on paper. But is it a good idea for high quality audio? I'm not sure. But it's fun to think about.
Eminence 12" coax was the most affordable on Parts Express, and I found this article that is interesting through an independent search:
https://eminence.com/blogs/blog/great-uses-for-coaxial-products
Apparently one of their engineers designed a home theater system setup using their coax drivers, and he developed a crossover specifically for a home theater use. And now it's an off the shelf component. That's interesting. Learning how to design crossovers for this project is a bridge too far, but using a pre-made design or possibly copying it with better components, that's definitely a possibility.
https://eminence.com/blogs/blog/great-uses-for-coaxial-products
Apparently one of their engineers designed a home theater system setup using their coax drivers, and he developed a crossover specifically for a home theater use. And now it's an off the shelf component. That's interesting. Learning how to design crossovers for this project is a bridge too far, but using a pre-made design or possibly copying it with better components, that's definitely a possibility.
Coaxials with compression driver from Sica and Faital are great. Stay away from Radian. I have a smaller B&C 4CXN36-8/16, which is usable but has low efficiency for a pro driver. Anyways, I would not look for KEF or Tannoy, but for a pro driver. When played at HiFi levels, they are bound to distort considerably less than HiFi speakers at the same SPL.
However, with all these drivers, they don't come with straight FR's. So you need to solve it in the crossover, but frankly: coaxials require more compensations than are easily doable with analog circuitry. So going active is the only practical solution imo.
However, with all these drivers, they don't come with straight FR's. So you need to solve it in the crossover, but frankly: coaxials require more compensations than are easily doable with analog circuitry. So going active is the only practical solution imo.
Yes this is what i think about.
Xover at 200hz should be ok, it'll ask a bit more to the woofer part of coax ( it'll move a bit more than cut higher for a given spl) but nothing to worry about. It'll give an advantage regarding integration though: for both drivers to be heard as one and only you should satisfy 1/4 wavelength ( at xover freq) for center to center distance, at 200hz it is 43cm (~17").
While at it let's talk about BSC: with 15" width it place center freq of baffle step compensation shelve ( 4 octave wide) at something like 290hz. It means the woofer part of coax will have at most 3/4db of gain 'robbed' by bsc. From circa 70hz the sub will have the heavy lifting in charge by itself, from 1200hz the compression driver/horn is outside bsc compensation range AND there is already directivity control from horn: less bafle edge diffraction. Overall if you start with a 'PA, 12" you can expect 96db efficiency. So overall you could have something around 93db once bsc applied. Still correct efficiency.
For the 'proper' midrange driver, could you give me an example of what you call a proper midrange driver? ( i'm teasing but i'm serious)?
Yes i have reference in mind but let's first adress your concerns...
No it will be horrible. The worst you'll ever listen to.
That is why, historically (in pro circles) peoples kept on trying with this kind of idea:
https://www.hifido.co.jp/sold/12-67924-49210-00.html?LNG=E
https://images.app.goo.gl/ajkKDGgyHzZfda8DA
This is a Buckingham: 1x10" coax 2x12" woofer.
Some people felt the form factor didn't fit theyr taste so they rearanged the layout:
https://images.app.goo.gl/d78KM5h8PrCotnBG9
Hmm... doesn't it look familiar?
In the same theme but for the 'big boys' Tannoy produced a scaled up version:
https://images.app.goo.gl/mpkptbBYnxiqiyyB8
1x15 coax + 2x15". It is big and loud. Few were builts. Probably because of costs.
I know a pair of this is used at 200km of my location:
http://studioblackbox.fr/Studio_Black_Box_Francais/Black_Box_-_Cabine.html
Other user included Vangelis and iirc 'Blade Runner' soundtrack was composed/mixed on them.
So i repeat the concept of a coax xossed to woofer will be horrible for sure.
About the size of driver.
12" you shouldn't use. It is a bad compromise for sound quality between the tiny coax and the 15 or 18".
So bad it is a surprise that albums like 'Consumed' from Plastikman was ever produced ( composed/mixed on System12dmt2).
An other example of bad 12" sound is Tony Maserati. His work is awful. Really. This is all because of his System1200!
Since then he fears Beyonce will hit him in stomach because of his choice of monitor! Or JayZ's homies ( including Kanye West!) stalking him. This is a serious treat given it include 'Ye' ( Keny West)... 😉
So, principle doesn't have been used by manufacturers, nor have been succesful in term of rendering. Move away! 🙂
Of course i recommend listening to either of the album or artists mentioned to see if this kind of loudspeakers can bring good audioquality. 😉
So this bring to the choice of driver.
From the 'recent' offer i've heard BMS, some B&C, probably some Faitals too. I really liked BMS.
In older drivers i heard old ( 20 years old) Radian 12"/15" that i liked.
Eminence coax are enjoyable as wedge on stage. Can make a great party loudspeaker too. They are 'cheap'(?) but not accurate in my view.
Beyma has 12cx30 which is very nice in my view.
I agree with Vacuphile, the real issue is you need active filtering ( analog or dsp) with this 'pa' coax. But it's already the plan no? 😉
Last edited:
It seems the Beyma is eol and so prices have doubled on this one. Bummer.
If you are interested in this kind of arrangement (which is known to 'work' if implemented carefully) you are in known territory. A bit of search about spec of the monitors mentioned should give you trends about xover freq used.
Given you have opened the door to cardiod active for the sub, i would investigate passive cardio for the woofer part of the coax.
I've met that from M. Gunness of Fulcrum Acoustics : https://www.google.com/url?sa=t&sou...wQFnoECBUQAQ&usg=AOvVaw3ZLDIogiHrpr2hRDXUtC_J
Worth a read.
I think you just gave me enough inspiration for the next 2 years, i just finished plans for my next build (soon to be started, i'm collecting parts) and was in fear lacking inspiration for something next! Thank you.
If you are interested in this kind of arrangement (which is known to 'work' if implemented carefully) you are in known territory. A bit of search about spec of the monitors mentioned should give you trends about xover freq used.
Given you have opened the door to cardiod active for the sub, i would investigate passive cardio for the woofer part of the coax.
I've met that from M. Gunness of Fulcrum Acoustics : https://www.google.com/url?sa=t&sou...wQFnoECBUQAQ&usg=AOvVaw3ZLDIogiHrpr2hRDXUtC_J
Worth a read.
I think you just gave me enough inspiration for the next 2 years, i just finished plans for my next build (soon to be started, i'm collecting parts) and was in fear lacking inspiration for something next! Thank you.