Inspired by this kit: Point P25, I recently began researching 10" pro-sound coax drivers. What I found is that most of the available options still seem to exhibit severe discontinuities between CD, horn throat/center of the pole piece and cone – especially when compared to the highly optimized throat geometries and mouth terminations of e.g. the latest KEF and TAD/Pioneer HiFi coax drivers.
Some of the newer designs (e.g. Radian Audio | 5210 Coaxial Speaker or 0080 10CX650 - Coaxials 10CX650 : Eighteen Sound - professional loudspeakers) claim a smooth transition from CD to throat to cone, but they either show only heavily smoothed FR graphs that don't prove anything (Radian) or the graphs still seem pretty rough (18Sound) – and none of them has any detailed description or diagrams of the driver construction.
Does anyone have a recommendation for a driver that performs (at least reasonably) well in said respects? Or can someone point me to some more detailed information about the design and performance of the more well-known drivers of PHL, Radian, 18Sound et al.? Or maybe there's someone that even has measurements of them that are a bit more meaningful than the factory-provided ones?
Some of the newer designs (e.g. Radian Audio | 5210 Coaxial Speaker or 0080 10CX650 - Coaxials 10CX650 : Eighteen Sound - professional loudspeakers) claim a smooth transition from CD to throat to cone, but they either show only heavily smoothed FR graphs that don't prove anything (Radian) or the graphs still seem pretty rough (18Sound) – and none of them has any detailed description or diagrams of the driver construction.
Does anyone have a recommendation for a driver that performs (at least reasonably) well in said respects? Or can someone point me to some more detailed information about the design and performance of the more well-known drivers of PHL, Radian, 18Sound et al.? Or maybe there's someone that even has measurements of them that are a bit more meaningful than the factory-provided ones?
"Smoothest throat transition/frequency response in large pro coaxes? "
For a coaxial speaker, if the cone is used as the horn, then the woofer pole piece must be machined to the correct throat shape.
If a separate horn is used in front of the cone, then the bottom throat of that horn can provide the correct shape directly down to the compression driver.
The Radian 5210, 5212, and 5215 coaxial datasheets document:
1) The compression driver mates directly to an identical diameter horn throat machined into the center of the LF pole piece. The horn flare is completed by the cone of the woofer, resulting in excellent polar response at mid range frequencies and beyond.
2) 90º Conical Dispersion
None of the other vendor commercial cone-horn datasheets mention a conical horn throat machined into the center of the LF pole piece.
SO, until specs of another coaxial speaker with a machined throat pole piece arrives, the Radian 5210 seems the one to beat.
For a coaxial speaker, if the cone is used as the horn, then the woofer pole piece must be machined to the correct throat shape.
If a separate horn is used in front of the cone, then the bottom throat of that horn can provide the correct shape directly down to the compression driver.
The Radian 5210, 5212, and 5215 coaxial datasheets document:
1) The compression driver mates directly to an identical diameter horn throat machined into the center of the LF pole piece. The horn flare is completed by the cone of the woofer, resulting in excellent polar response at mid range frequencies and beyond.
2) 90º Conical Dispersion
None of the other vendor commercial cone-horn datasheets mention a conical horn throat machined into the center of the LF pole piece.
SO, until specs of another coaxial speaker with a machined throat pole piece arrives, the Radian 5210 seems the one to beat.
If Danley is using B&C coax, they are not very smooth.
The graph below shows the 8" coax DSL SH100 frequency response with "15% smoothing", and my measured response without smoothing.
The 12" coax SM80 uses 18% smoothing, but looks even more ragged than the SH100.
Attachments
There is a video of the manufacturing plant here in NC out on the web, and you can clearly see the neo B&C 12s...and he has mentioned several times that when he was looking for a coax, the B&C's had the best horn to cone transition he had found at that time.
He does use the BMS 5 for another horn though.
Search his name and coax...it's out there.
He does use the BMS 5 for another horn though.
Search his name and coax...it's out there.
Who is Danley Sound Labs - YouTube
Go to 4:08 mark and tell me that's a BMS coax...lol
And the SH-Mini has a 100 degree coverage pattern...same as the B&C, not BMS that is 90.
Go to 4:08 mark and tell me that's a BMS coax...lol
And the SH-Mini has a 100 degree coverage pattern...same as the B&C, not BMS that is 90.
Last edited:
He wrote about his reasons for choosing the B&C Coax for the SH100 in a number places, including this: http://www.diyaudio.com/forums/multi-way/219579-help-me-pick-coaxial-speaker.html#post3162446 – there he says: "At the time I did the comparative measurements of the available driver choices, the B&C 8CX21 had the smallest acoustic / horn discontinuity between the cone and compression driver portion."
Weltersys's graph just confirms my initial impression – if that's what a coax with "the smallest acoustic / horn discontinuity between the cone and compression driver portion" gets us, then the state of things isn't really good in pro-sound coax world.
On the other hand, I think I read somewhere that Tom Danley made his comparisons back in 2008, so maybe things have changed since then?
My point exactly – and the less smoothed graphs of newer designs from 18Sound et al. still seem pretty ragged.
Although not pro coax's the unobtanium coaxial drivers used in the Genelec monitors have smooth transition from dome to cone as the cone is the waveguide. KEF has something similar too. 454Casull has a thread where he is making an OS waveguide driver cone with CD - http://www.diyaudio.com/forums/multi-way/261511-18-90-deg-os-wg-coaxial-design.html. Haven't seen anything commercially available though.
Last edited:
I doubt it, physics have not changed since the first coax. The Altec and Tannoy 15" coax drivers from the 1950's have similar response to the modern coax, modern coax just handle more power.
Coax are kind of painted in to a corner, with a choice of two compromises:
1) Make a curved cone to extend the HF horn, which has problems because the horn "wiggles" in use. With modern driver cones having 6-8 mm or so Xmax, there is a definite discontinuity that changes forward and backwards with each wave so the problem is worse than with older, limited Xmax designs. The curved cone required for a horn extension has more breakup problems, since it is not as stiff per weight as a conical cone.
2) Make a stiff cone, and design a decent horn to "nest" in front, which causes midrange diffraction and reflection, screwing up dispersion and temporal response, and the small horn compromises crossover point selection.
To some degree the problems in #2 can be addressed with DSP. Of interest, when I did the test on the DSL SH100, we compared it to the client's Mackie HD1521. The HD1521 uses "Gunness Focusing", it has bone flat frequency and phase response, though it uses a rather ordinary 15" and separate HF horn/driver. Although the SH100 sounded quite good, the HD1521 sounded better, and the smoother response was apparent, especially with gain before feedback with live microphones.
Dave Gunness left EAW/Mackie/Loud and now has a company called Fulcrum Acoustics, with at least two different powered coax offerings using his DSP magic, getting rave reviews. Though I have not heard them, just based on Dave's past work (and the spec sheets) I'd bet they are probably the best sounding high powered coax going, but the driver used (and his horn design) is only half the story.
Art
Not available as separate drive units, but these Tannoy large format coaxes are as smooth as KEF/TAD/Pioneer 'home' units with the added dynamics and pattern control due to their size.
Tannoy 7900 1289 Combo Driver For VXP12HP, 7900 1289 | Full Compass
High powered tannoy 12" $400 off buying a complete cabinet.
High powered tannoy 12" $400 off buying a complete cabinet.
Yes, these would be pretty cool in a DIY design (resp. are, as e.g. in this and this design). A good replacement might be the SEAS magnesium coax, recently tested here.
But the point for me was to see if there are similarly well-developed designs in high-efficiency land…
We can address that problem, too – just add a dedicated woofer and HP the midrange driver of the coaxial at a frequency where cone travel isn't a problem (the horn doesn't "wiggle") anymore. Earl Geddes, Andrew Jones et al. have been discussing this over at http://www.diyaudio.com/forums/multi-way/261555-small-coax-vs-waveguides.html.
Have the Tannoys really profited that much from the "Tulip waveguides"? I seem to remember photos of Tannoy drivers where the throat and the cone expand at very different rates/angles…
An externally hosted image should be here but it was not working when we last tested it.
This is how it should be done. In contrast other pro drivers try to integrate an off-the-shelf compression driver, not so successfully.
The compression driver and tulip design seems to vary a lot with different models. I have two drivers from the polyprop studio monitor series, SYS600 (the driver shown in daniels post) and SYS1200, and the horn section is different in the larger driver in that it has much higher compression ratio. I also own 15" paper cone drivers (model 3836 used in CPA15 PA monitors) which have a very similar horn section but a steeper geometry at the base of the cone. All drivers are from roughly the same design period for the tulip waveguides with rather curved contours for the vanes.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.