Can crossovers be replaced by DSP adjusted signals in any speaker design ?

[maxime.levesque]

Curious

How would you do this bypassing of a passive crossover without separate amps for each driver, tho?

The hypex plate amp I was talking about is multi chanel, each DSP channel has it's own amplifier.

 

[Chrispy]

The hypex plate amp I was talking about is multi chanel, each DSP channel has it's own amplifier.

Okay, was confused when you said "possibly their own amplifier". Such a multi channel amp would just have their own amplifier, perhaps a shared power supply, tho. I'd prefer racked amps in any case rather than a plate amp.

 

[dualazmak]

Hello maxime.levesque and friends,

Looks many people already responded so nicely; I am very happy reading all of them.

Let me add some other points and my thoughts hopefully this evening; the time here in Japan is 6:25 AM, and I am rather busy today on my main mission related to COVID-19 vaccination. (I am not MD but Ph.D. pharmaceutical/epidemiology scientist...)

Until then, you would please read these posts on "my policy and approach" in my multichannel multi-driver multi-amplifier project; post #30, post #332 and post #341.

Will get back to you hopefully this evening!

 

[dualazmak]

Hello maxime.levesque and friends,

I believe now you understand well about the policy and approaches in my project.

At the start of my project, I firmly decided to keep the SP drivers and enclosures remain unchanged, and I moved step-by-step towards the final full active multichannel system.

In addition to the nice responses and suggestions already given above by our ASR friends...

Based on my rather long and intensive exploration, I would like to suggest you as follows:

Have your reference sound system to start with
I highly recommend you to have your reference (base) passive sound system before going into multichannel exploration.

In my case the passive system, single amplifier (Accuphase E-460) + renovated outer LC-network box (inductors, capacitors, attenuators) + Yamaha NS-1000 + horn super-tweeter + L&R sub-woofers, has been my reference sound system throughout the project, and I can/could easily roll back to this reference sound system any time during the project for comparative listening sessions.

Step-by-step progress with only one parameter modification in each step
If you would change multiple parameters at once, then you would easily get into confusion.

Very carefully eliminate LC-network and attenuators and mimic/simulate these by digital crossover and EQ
I assume you well understand that the inductors, capacitors, resistors and attenuators in passive LC-network consume/waste considerable portion of "energy" provided by the amplifier before the SP level signal would finally go into the SP driver. The elimination of LC-network and attenuator, therefore, is significantly effective in making the amplifier to drive SP driver much more easily, effectively and efficiently.

As @Speedskater nicely pointed in his post #19, however, the process is not straightforward, not simple. You definitely need your passive reference system, I believe. Even the simple attenuators are part of the LC-network having certain variable impedance (resistance) in the SP line as I shared in detail at here and here.

Avoid, as far as possible, magnetizable (magnetic susceptible) metals in the SP lines
This is really important and critical in high-end HiFi audio system to keep the cleanliness of the sound, as I shared here and here.

Major issues are always interdependent with each other
As I wrote in the very first post on my project thread, the following major issues in multichannel multi-driver multi-amplifier system are always interdependent with each other;

- Digital I/O for software crossover within PC
- Delay, latency and synchronization issues
- Phase issues
- Master volume and gain issues

Very carefully evaluate and select amplifiers, in your system in your room
I believe you would use quite HiFi grade SP drivers, and these drivers very much efficiently respond to the "nature" of the dedicated amplifier directly connected without LC-network.

Even though I know you already have, or soon to have, Hypex plate multi-channel amplifier(s), I strongly suggest you also to try/test some of (affordable) Class-AB and/or Class-A HiFi grade "power amplifiers" and "integrated amplifiers", in your system.

As I shared here, I intensively evaluated and tested YAMAHA MX-A5200 (fairly nice 11-channel amplifier) in my project, but the total sound quality with MX-A5200 was considerably inferior to my reference passive system with Accuphase E-460. That was, therefore, the start of my long and intensive explorations on search for amplifiers in my multichannel system.

In my post #311, I wrote;
Furthermore, throughout my amplifier exploration, I well experienced and learnt that we should never exclude high quality Hi-Fi "integrated amplifiers" to be possibly implemented in this type of multichannel multi-amplifier project.

The durability, warranty, repair possibility are also important factors in amplifier(s) selection.


In any way, I believe you can get much valuable insights for your possible multichannel DSP project, after you would kindly and carefully read all the way through my project thread in which many of our ASR friends are cooperatively participating.

 

[Speedskater]

One neat thing about some DIY 'DSP' crossovers is, you can program different crossover setups in as scenes. Then have someone else select a random scene. Then do your listening tests (in effect a blind test).

 

[dualazmak]

One neat thing about some DIY 'DSP' crossovers is, you can program different crossover setups in as scenes. Then have someone else select a random scene. Then do your listening tests (in effect a blind test).

Yes, I fully agree with you, and this is another reason and rationale for my using software digital crossover-EQ "EKIO". EKIO can save any of my XO-EQ configuraions and re-load it; EKIO even has ABx comparative listening features, which I seldom use, though.

I recently intensively evaluated three slightly different EKIO configurations with -12 dB/OCT, -24 dB/OCT and -48 dB/Oct slopes for High-Cut (Low-Pass) LR filters at 25 kHz to cut-off the possible HF noises in near ultrasound - ultrasound Fq zone. I could do it in quasi-blinded ABCx comparative listening sesseions. If you would be interested, please visit my posts here to here.

 

[pozz]

3) On the receiving end (the ear membrane), is their a similar phenomenon, when it is hit by two notes of dissimilar amplitude ?

It is much more complex for the ear. It depends on the relative amplitude and difference between frequencies. The processing by the brain will also combine events in left and right ears.

Mead Killion of Etymotic has stated that the inherent IMD of the ear is around 7% in a presentation on otoacoustic emissions. I don't have a good answer as to why we can hear events well below that figure, but it's likely to do with adaptation and processing of frequency bands according to the power spectrum of the sounds we hear. Distortion is correlated to music, so in general it is easier to ignore or miss unless it's due to a persistent mechanical issue, like a resonance, which will decay at its own pace instead of that of the music signal.

 

[Razorhelm]

https://www.audiosciencereview.com/...-crossovers-via-rephase-and-camilladsp.19910/

This is the system i built over lockdown the dsp is in the pc and i use a multichannel dac to send it to the amps but the concept is the same and it works great.

I highly recommend giving it a go.

Good luck with your build