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

[maxime.levesque]

I am looking at multi way DIY enclosure designs that have crossovers,

and I was wondering if the crossovers can be eliminated if I use an amp with DSP sych as hypex plate amps.

Thanks

 

[Doodski]

Using a DSP based active crossover will enable bypassing the hardware crossovers. Newer DSP based crossover gear sometimes comes with a parametric EQ too. The Hypex plate amps will suffice and you will need one channel of amplification for each speaker driver.

 

[maxime.levesque]

Thanks, is it easy to figure out the appropriate DSP configuration to upload to the hypex, given the specs of the crossover the DSP will be replacing ?

 

[Doodski]

Which models are you looking at?

 

[maxime.levesque]

I've been looking at a few,

and this one in particular: https://www.audiosciencereview.com/...ds/celuaris-spk5-purifi-review-speaker.20125/

Given that I'm pretty settled on the amp, and much less on the speaker kit, I'm wondering if
I can just consider any kit designed with crossovers, and assume the DSP can replace them with the proper configuration.

 

[Doodski]

I've been looking at a few,

and this one in particular: https://www.audiosciencereview.com/...ds/celuaris-spk5-purifi-review-speaker.20125/

Given that I'm pretty settled on the amp, and much less on the speaker kit, I'm wondering if
I can just consider any kit designed with crossovers, and assume the DSP can replace them with the proper configuration.

As long as the output channels of the plate amp have enough channels for say a 2 way speaker or 3 way then you could use the plate amp and DSP crossover. Which Hypex plate amp with DSP are you considering?

 

[restorer-john]

You need to consider the following:

Elevated residual noise, particularly in the tweeter. The amplifier's entire residual noise spectrum will be audible in both drivers (in a two way).

Direct connection of a tweeter to an amplifier capable of full range/high power without comprehensive switch on/off and protection/muting (other than the amplifier's own protection) is very risky. Tweeters can only absorb ~5 watts or so before burning out.

 

[Doodski]

You need to consider the following:

Elevated residual noise, particularly in the tweeter. The amplifier's entire residual noise spectrum will be audible in both drivers (in a two way).

Direct connection of a tweeter to an amplifier capable of full range/high power without comprehensive switch on/off and protection/muting (other than the amplifier's own protection) is very risky. Tweeters can only absorb ~5 watts or so before burning out.

Perhaps putting a capacitor inline on the tweeters will help? I ran a tri-amp'd system with about 1000 watts and yes I popped one KEF 107.2 tweeter that I bought from KEF with a strange tiny little squeek/peep sorta sound and then it was dead. It's the risk one takes when running direct to the amp with a active crossover.

 

[maxime.levesque]

You need to consider the following:

Elevated residual noise, particularly in the tweeter. The amplifier's entire residual noise spectrum will be audible in both drivers (in a two way).

Direct connection of a tweeter to an amplifier capable of full range/high power without comprehensive switch on/off and protection/muting (other than the amplifier's own protection) is very risky. Tweeters can only absorb ~5 watts or so before burning out.

Isn't lowering volume (and power by extension) part of what the DSP can do ?

This kit seems to have an amp output directly connected to the tweeter: http://www.troelsgravesen.dk/Ekta-2D.htm

 

[gene_stl]

You should have both a capacitor in series (or better yet two , one on each lead of the tweeter ) as well as a fuse in series with the tweeter. One quarter amp fast blow should help protect the tweeter though tweeters are often known to protect fuses.

Factory built speakers today are very overpriced. DIY can save you a lot of money. MultiAmping also is usually a great improvement. Needs to be done carefully and correctly.

If the DSP is part of the amplifier it may have a power limiter feature available. Crown DSP inputs have these some with great sophistication, ie modeling probable voice coil heat and keeping a history of the recent music and varying the amount of limiting. This should all be a no brainer for just about any DSP setup.

 

[Doodski]

@restorer-john was mentioning any spike, random power surge or overpowering the tweeter without a passive crossover inline to provide some isolation and filtering.

Yes the drivers connect directly to the Hypex plate amp.

 

[maxxevv]

I am looking at multi way DIY enclosure designs that have crossovers,

and I was wondering if the crossovers can be eliminated if I use an amp with DSP sych as hypex plate amps.

Thanks

You can have a good read on @dualazmak 's project using a 8 channel DAC and software.

https://www.audiosciencereview.com/...oftware-crossover-and-multichannel-dac.12489/

He used the Okto 8, but if your demands are not so high, or you just need a 2 way control per speaker, the Mini DSP could probably work too.

https://www.audiosciencereview.com/...sp-shd-review-updated-dsp-dac-streamer.18681/

 

[dualazmak]

Hello maxxevv,
Thank you for your referring to my project thread.

Hello maxime.levesque,
My project thread is rather long and intensive. If you would like to have short-cut to my recently completed system, you would please visit my post #311 there and thereafter for the details.

 

[maxime.levesque]

Hello maxxevv,
Thank you for your referring to my project thread.

Hello maxime.levesque,
My project thread is rather long and intensive. If you would like to have short-cut to my recently completed system, you would please visit my post #311 there and thereafter for the details.

Hi dualazmak, I see that you are exploring the extreme of the audio range segmentation (and specialisation).

I am a total ignorant in accoustics (or a complete beginer if I want to be more charitable), and I am wondering how the law of diminishing returns applies to range segmentation of the sound spectrum.

What I mean by range segmentation is the idea of having dedicated drivers (and possibly their own amplifier) for each range of the spectrum, I guess it is called "multiway" in the jargon.

So with zero segmentation, you have a full range driver, then you split the range in two with a woofer and tweeter, then you split it in three, then four, then five, and this is where your project is at...

So my question is: after how many split does one get diminishing returns, or perhaps inverse returns ?

For the "single full driver crowd", any range split is more than they can handle !

It seems to me that within the "audio geeks" crowd, you have the minimalists at one extreem, the "2 wat SET amp + single driver" guys, and at the other end you have maximalists. But then again, on the minimalist vs maximalist axis, everyone will say they are at "their" minimum !

 

[abdo123]

So my question is: after how many split does one get diminishing returns, or perhaps inverse returns ?

once you get intermodulation distortion below 1% at your listening level.

intermodulation distortion happens when one diaphragm tries to produce two tones at the same time.

this is an example of IMD, when trying to produce two tones, 50Hz and 200Hz (not equal amplitudes, 1:10 ratio) by a 2-Way speaker.

Adding a subwoofer into that setup would make it much less messy. Same concept goes to multiple drivers.

 

[maxime.levesque]

Interesting !

Questions:

1) Suppose that one splits the range in 10 (with a 10 way system),

and within one of the segments (ex. 200 and 300), you have two notes (ex: 211 and 290),

I suppose the IMD would only be emited by the driver handling the 200-300 range ?

2) In a musical instrument capable of polyphony (for example a violin) when two notes are played at the same time, isn't their also IMD, given that the sound from the string is amplified by the same (wood resonating) surface ?

Could we not say that in this specific instance, the IMD is part of the music ?

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

 

[abdo123]

Interesting !

Questions:

1) Suppose that one splits the range in 10 (with a 10 way system),

and within one of the segments (ex. 200 and 300), you have two notes (ex: 211 and 290),

I suppose the IMD would only be emited by the driver handling the 200-300 range ?

2) In a musical instrument capable of polyphony (for example a violin) when two notes are played at the same time, isn't their also IMD, given that the sound from the string is amplified by the same (wood resonating) surface ?

Could we not say that in this specific instance, the IMD is part of the music ?

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

1) yeah that makes sense, however crossovers are usually not perfectly splitting the spectrum like you said but rather have declining/increasing slopes, you can also make very dramatic slopes via DSP but the industry didn't adopt that (yet?).

the closer the tones are to each other the less IMD you have (and the more likely the distortion will be masked by the fundamental tones).

2) the IMD of the instrument is part of the music, the IMD of your speakers is not.

3) I think the ears have different receptors for different frequencies, I'm also not sure if IMD works the same for recording, it is usually concerned with reproduction of tones.

 

[Chrispy]

Curious

What I mean by range segmentation is the idea of having dedicated drivers (and possibly their own amplifier) for each range of the spectrum, I guess it is called "multiway" in the jargon.

!

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

 

[Speedskater]

You can't just take the values off a loudspeaker's schematic, then plug them in to a 'DSP' and expect the same results. There is a whole lot going on that doesn't show-up on the schematic.
'Dave Rich' technical editor for thew old 'Audio Critic' magazine wrote:

Crossovers
“In the old days, there was a lot of trial and error in designing a crossover. So, adding more than a few components was futile, since it wasn’t practically possible to optimize them. Simplicity yielded better optimizations”.
“The individual crossover components usually don’t have functions in a way that can be isolated. A crossover is a filter network that implements a transfer function as a result of all of its parts. Sometimes it is possible to generalize and say, Oh, this resistor adds damping. Or, This network compensates for a resonance. But, really, the better a crossover design is, the more the parts work together, symbiotically”.
Dave Rich

 

[abdo123]

You can't just take the values off a loudspeaker's schematic, then plug them in to a 'DSP' and expect the same results. There is a whole lot going on that doesn't show-up on the schematic.
'Dave Rich' technical editor for thew old 'Audio Critic' magazine wrote:

Crossovers
“In the old days, there was a lot of trial and error in designing a crossover. So, adding more than a few components was futile, since it wasn’t practically possible to optimize them. Simplicity yielded better optimizations”.
“The individual crossover components usually don’t have functions in a way that can be isolated. A crossover is a filter network that implements a transfer function as a result of all of its parts. Sometimes it is possible to generalize and say, Oh, this resistor adds damping. Or, This network compensates for a resonance. But, really, the better a crossover design is, the more the parts work together, symbiotically”.
Dave Rich

you're very likely to reach better results though with backyard quasi-anechoic measurements and DSP.

I believe the Directiva ASR speaker project will surpass any reasonably priced speaker made in the 70s or 80s thanks to the NFS.

sure you probably can't get the exact same thing but you can do so much better than that.