I want to clarify what I said here...
I don't know the why of this.... just that it is.
When I refer to these crossover points, it's really the single crossover point between the bass bin and the HF horn. There is actually some overlap as the HF horn is cutoff electronically lower than the bass bin. I have no idea what the acoustic crossover point is. I had to do a double-take when I was loading the PEQ's into the crossover. I had never heard of a higher driver crossing underneath the lower driver and having an overlap like that.
Then again, I've admitted I don't know too much about the technical stuff. I am simply following the recipe.
I would assume that the overlap is due to both treble and woofer losing output at their frequency extremes and this way they reinforce each other thus avoiding having to use a 3rd way.
It also means that their acoustic slope will be steeper than the 24dB electrical you are using.
I have not heard your speakers and they may well sound excellent.
(Mind you I would expect a few minor miracles from the TAD given that it costs $2.5k!)
However I personally would not design a speaker like that because I'd prefer not to stress any component. I like them all loafing around where they are most comfortable.
That means low-powered amps are out (not a fan of tubes anyways) and I would cross each driver well within its linear range which practically demands a 3way layout in order to meet or exceed the minimum 42Hz - 17kHz coverage I demand from my full-sized speakers.
All those are my preferences and to some extent they are dictated by the fact that I do not have an R&D department at my disposal and my own experiences/prejudices.
After all I am very fond of Tannoy DCs and JBL 44xx neither of which is a million miles from your speakers.
All that said some classic Tannoy DC xovers also had an overlap built in.
It's easiest to see with the low pass which is a simple series coil and parallel cap ie if you look at it it appears to be a straight forward 12dB slope.
But the coil and caps -3dB points do not coincide so the slope starts at 6dB and increases to 12dB an octave up or so. In old brochures Tannoy only ever gave the apparent acoustic xover point.
(Tannoys hi pass is a surprisingly clever piece of engineering and nowhere near as easy to understand so please don't ask! ;-) ).
Active 3-way using digital XO.
Bob Crites Cornwall type C speaker kit: 15” woofer, Selenium D405 compression driver mated to JBL 2380A clone for mids and BMS 4540 ring radiator mated to QSC PL-000446GP waveguide.
XO points 500 Hz and 5kHz using Acourate digital XO Neville-Thiele 2nd order filters.
Amps are DIY Nelson Pass Class A-40 on woofers, and Nelson Pass Class A ACA kits on mid and high freq compression drivers.
Computer audio based playback system using Acourate for digital XO, driver time alignment, and room correction. FIR correction filters hosted in JRiver’s 64 bit Convolution engine.
Article: http://www.computeraudiophile.com/content/556-advanced-acourate-digital-xo-time-alignment-driver-linearization-walkthrough/
Note: not using the HF horns in the pics, but updated with the BMS 4540/QSC’s sitting on top of the cabs which were used for the following measurements:
Frequency response at the listening position 1/6 oct smoothing:
Step response at listening position:
Bob Crites Cornwall type C speaker kit: 15” woofer, Selenium D405 compression driver mated to JBL 2380A clone for mids and BMS 4540 ring radiator mated to QSC PL-000446GP waveguide.
XO points 500 Hz and 5kHz using Acourate digital XO Neville-Thiele 2nd order filters.
Amps are DIY Nelson Pass Class A-40 on woofers, and Nelson Pass Class A ACA kits on mid and high freq compression drivers.
Computer audio based playback system using Acourate for digital XO, driver time alignment, and room correction. FIR correction filters hosted in JRiver’s 64 bit Convolution engine.
Article: http://www.computeraudiophile.com/content/556-advanced-acourate-digital-xo-time-alignment-driver-linearization-walkthrough/
Note: not using the HF horns in the pics, but updated with the BMS 4540/QSC’s sitting on top of the cabs which were used for the following measurements:
Frequency response at the listening position 1/6 oct smoothing:
Step response at listening position:
I'm curious as to how many board members use active crossovers?
At least a 2 way?
How about 3 way systems ?
With a full blown 4 way, being of course, the cat's meow
I'm curious on this because, it seems to me, that around 1980 there were several active crossovers on the market, but these days, other than Marchand, not so much. Or am I missing something ?
Hi Scott,
Yes, definitely active and definitely digital.
In this day and age, going digital only for the purpose of replacing passive crossover seems quite a poor choice.
Yes, you will get improvement due to active amplification, but that’s still a far cry from what a digital system can and should deliver today.
Just for example: The latest Pink Floyd album "The Endless River" has been released in 5.1 PCM, 24bit/96kHz format ( Stephen Hawking has collaborated ). It's an outstanding audio experience. The old 2.0 stereo just does not cut it anymore, and the audio playback system should aim to be capable of surround sound.
Then, there is Binaural Bass Management (BBM) system and CABS system configuration for active room mode suppression.
Also, you can have HBT equalization, which provides flat-line SPL and linear phase response of each driver in the system (or minimum-phase with flat-line SPL if you choose so).
Moreover, going digital, you should expect intelligent room EQ capabilities.
Finally, you should expect your whole system (including amps) to be fully digital, running from single clock source.
All this is readily available today. http://www.bodziosoftware.com.au/
Best Regards,
Bohdan
Bill,
Cool off.
Just as many other readers contributed their system descriptions, I am contributing mine.
Best Regards,
Bohdan
Billshurv, Bohdan,
It is quite clear than Bohdan is not trying to sell anything. Just take a look at the link.
[OK, my tongue is firmly in cheek with what follows, but there is a message in here.]
I am an engineer with over 25 years doing all manner of hardcore signal processing stuff. OK, my personality is more on the creative side than spending endless hours analysing the last tenth of a dB of a thing.
But that web page only told me one thing: You have the "crossover of death" with lots of speakers and some really complex stuff going on in there somehow. Message: If an engineer skims down the page, it is pretty sure that a human being would "change channels" much earlier.
So, now I have done my level best to offend you, what does this beast look like inside?
It is quite clear than Bohdan is not trying to sell anything. Just take a look at the link.
[OK, my tongue is firmly in cheek with what follows, but there is a message in here.]
I am an engineer with over 25 years doing all manner of hardcore signal processing stuff. OK, my personality is more on the creative side than spending endless hours analysing the last tenth of a dB of a thing.
But that web page only told me one thing: You have the "crossover of death" with lots of speakers and some really complex stuff going on in there somehow. Message: If an engineer skims down the page, it is pretty sure that a human being would "change channels" much earlier.
So, now I have done my level best to offend you, what does this beast look like inside?
Hi googlyone,
No problem, here are highlights of what’s under the hood.
1.DSP engine - it’s a 16 channel FIR filter, implemented as 8 or 16 partition, frequency-domain convolution. Data blocks 1024/2048/4096 bins. Sampling 48kHz or 96kHz at 24bits.
2. Software perspective – multi-threaded software, so needs 8-core CPU.
3. Hardware perspective – uses LynxAES16e digital AES/EBU card + miniDSP AES/EBU amplifiers in fully digital configuration. You can also use Delta1010LT card or motherboard codecs, such as ALC1150 for best SNR.
4. Black Box perspective – 1.46Hz bass resolution at 96kHz sampling, 65ms latency in “minimum-phase” mode and 165ms latency in “linear-phase” mode. So, it’s good for audio/video applications. Obviously, there are trade-offs, and you should now where and why.
4. DSP functions
Active system – allows amplifiers to exert maximum control over loudspeaker driver and makes crossover characteristics independent of driver loading.
DSP crossover and voicing – maintains crossover and voicing characteristics with mathematical precision.
Amplitude and phase HBT equalization – results in flat amplitude and phase responses and extends bass response.
Linear phase – maintains 0deg acoustical phase response of each driver, which results in transient-perfect system for correct spatial imaging (realism, depth, resolution, ambience) for multi-channel sound reproduction. Also removes “flabby bass”.
Minimum-phase/Linear-phase room EQ – removes most obnoxious room modes.
Available with BBM bass management and CABS active suppresor of room modes. Explanatory links are on my website.
Remote control – via 2.4GHz wireless keyboard and mouse.
Built-in WAV Player - with this option, there is no need for any sound cards at all. UE will create and play "playlists" with the standard CD-style WAVE files (16-bit/2-channels/44.1kHz) using motherboard codecs. UE7 has WAV Player with 5.1 HT, 24bit/96kHz capability.
Use Windows Media Player, or install JRivers Player – to create audio server and state-of-the-art loudspeaker management system (user-friendliness, power, flexibility, cost efficiency and sound quality) on one HTPC.
Let me know if you need further data.
Best Regards,
Bohdan
lol.
I use an amp either side of a 250Hz lower crossover with line level crossovers (passive) built in to the amps. Passive speaker level gives me fine tuning for these and a passive upper crossover.
I find conventional active crossovers don't give the finer control that my measurements call for.
DSP would probably take the fun out of it so I ask, is it necessary? I think I can do better with passive.
I use an amp either side of a 250Hz lower crossover with line level crossovers (passive) built in to the amps. Passive speaker level gives me fine tuning for these and a passive upper crossover.
I find conventional active crossovers don't give the finer control that my measurements call for.
DSP would probably take the fun out of it so I ask, is it necessary? I think I can do better with passive.
Yes. 2-way (4.0) desktop system. Includes: PC; $30 sound card; S/DIF (optical) cable; software-based xover and EQ; 5.1 receiver. Plays everything (ie online media and games too). My main system will evolve into a hybrid of the above (minus the sound card hopefully).
Was using a MiniDSP 2x4, external dac and receiver for desktop duties previously.
Was using a MiniDSP 2x4, external dac and receiver for desktop duties previously.
Last edited:
I am, and have been since the 80's when I had Linn Saras, and later DMS's. Nearly every other speaker I've had in the interim was active at some stage, especially if it was my design.
I have JBL LSR305s in the bedroom, but these will move to the study when I build the big 4 way (active) design for the bedroom.
My 4 surrounds are 3 way active.
My mains are 3 way active, but I also have 4 21" subs. Xover settings change depending on 2ch or HT use.With a full blown 4 way, being of course, the cat's meow
My friend Terry's system is a 5 way active utilising a pair of DEQX and is one of the best I've ever heard, irrespective of cost.
You're not looking hard enough son. I use modified Behringer DCX2496 and a MiniDSP 4x10, both of which have been around for years. There's the Najda, DEQX, a potentially neat design being developed here and a myriad of pro units that can work well.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.