Why do you need a speaker crossover - discuss

[Dr Bunsen Honeydew]

This is another way of saying that the crossover is a sticking plaster to try and hide the fact that inadequate drivers have been used.

Many driver manufacturers have now learned (or should that be re-learned) the mechanical techniques necessary to avoid the need for complex crossovers.

Exactly, spot on.

[Dr Bunsen Honeydew]

I have heard good and awful speakers using both simple/none and complex crossovers. It all depends on the drivers and the way they are used.

I think the point is being lost here somewhere down the line. Do you accept that any component in the signal path will damage the music? If so by altering the selected driver mechanically to remove that component can only be a good thing - agreed? So you have heard good speakers with a crossover, and so have I, but to follow the argument the speaker would have been even better without those components in the signal path and if the job had been done mechanically.

[Marco]

I think that point is pretty much indisputable

But then, there are also other variables in the equation, in terms of the final design of any loudspeaker, and I'm not a fan of absolutes.....

Marco.

[Dr Bunsen Honeydew]

I think that point is pretty much indisputable

But then, there are also other variables in the equation, in terms of the final design of any loudspeaker, and I'm not a fan of absolutes.....

Marco.

Tell me what the variables are - there is only mechanical and electrical. A speaker in its simplest form is a mechanical beast, apart from the voice coil. All additional electronics does is get in the way of the music and the job it is trying to do, get the mechanics right and the sticking plaster is not required. Cures always have side effects, it is much better not to have the illness in the first place.

Mechanical is not just selecting driver design and user doping. It is damping (a mechanical filter) and cabinet design and material and driver positioning as well.

[anthonyTD]

Tell me what the variables are - there is only mechanical and electrical. A speaker in its simplest form is a mechanical beast, apart from the voice coil. All additional electronics does is get in the way of the music and the job it is trying to do, get the mechanics right and the sticking plaster is not required. Cures always have side effects, it is much better not to have the illness in the first place.

Mechanical is not just selecting driver design and user doping. It is damping (a mechanical filter) and cabinet design and material and driver positioning as well.

Agreed!
A...

[MartinT]

This is another way of saying that the crossover is a sticking plaster to try and hide the fact that inadequate drivers have been used.

I don't disagree, Chris. I'm not a mechanical engineer and don't know what compromises might have to be made to create a driver that doesn't require a filter to remove frequencies it can't handle, as opposed to one that is created 'wide open' but needs a filter to limit its bandwidth.

[Paul Hynes]

I use full range loudspeakers without crossovers and would not sacrifice the coherent presentation I am getting with these. During the 1970’s I used Eagle FR8 full-range drive units loaded with sewer pipes, tuned to the drive units, driven with Linsley Hood Class A amplifiers (the original simplest version in turn driven by Hood’s phono stage) and really enjoyed the musical presentation these items achieved even though the measured response could have been criticised. Everyone who heard this system was in agreement that musically it was exceptional compared to what was generally available at the time for similar levels of expenditure.

Historically I have had a lot of involvement with loudspeaker design, designers and manufacturing. I have tried most of the design options available for cabinet loading and crossovers, both passive and active through the 1980’s and early 1990’s. Whilst some of these designs were impressive for one reason or another, few approached the coherence provided by my 1970s system so I went back to full range, and I have been there ever since.

Much has changed since the 1970s but loudspeaker design is still governed by the laws of Physics. Disregarding the recording chain and replay electronics, which have their own issues that need care and attention in design, recreating a convincing illusion of a live musical event requires more than most commercial loudspeaker designs can achieve.

For instance, a classical concert performance or a rock concert will require movement of air that no single small drive unit would ever be able to realistically achieve. Think about it, you cannot reproduce the full energy of a tympani drum or a kick drum realistically with an 8 inch drive unit in a small cabinet. If you mess up the phase response through the audio band the fundamental notes and their harmonics will not be reproduced with the correct time relationship causing much musical confusion. Squash the dynamic scale and the performance fails to excite as much as a live event. I agree that a flat frequency response is not as important as these three key attributes. Anything that happens to the sound in the airspace of a live acoustic musical event will be perceived as a natural occurrence, and your brain will accept this as normal if it occurs in a sound system provided the effect is not exaggerated. Allowing a loudspeaker to generate effects that are not normally perceived at a live acoustic event can be disruptive to the illusion you wish to create. This goes for other equipment in the chain of reproduction too. Measurements are useful initially to get some idea of the overall electrical/mechanical performance but should never be solely relied upon. I have heard so many designs that measure well but sound wrong and some designs that do not measure to well by accepted standards that sound great. Your ears are your best asset when designing or choosing equipment.

Paul

[Dr Bunsen Honeydew]

I use full range loudspeakers without crossovers and would not sacrifice the coherent presentation I am getting with these. During the 1970’s I used Eagle FR8 full-range drive units loaded with sewer pipes, tuned to the drive units, driven with Linsley Hood Class A amplifiers (the original simplest version in turn driven by Hood’s phono stage) and really enjoyed the musical presentation these items achieved even though the measured response could have been criticised. Everyone who heard this system was in agreement that musically it was exceptional compared to what was generally available at the time for similar levels of expenditure.

Historically I have had a lot of involvement with loudspeaker design, designers and manufacturing. I have tried most of the design options available for cabinet loading and crossovers, both passive and active through the 1980’s and early 1990’s. Whilst some of these designs were impressive for one reason or another, few approached the coherence provided by my 1970s system so I went back to full range, and I have been there ever since.

Much has changed since the 1970s but loudspeaker design is still governed by the laws of Physics. Disregarding the recording chain and replay electronics, which have their own issues that need care and attention in design, recreating a convincing illusion of a live musical event requires more than most commercial loudspeaker designs can achieve.

For instance, a classical concert performance or a rock concert will require movement of air that no single small drive unit would ever be able to realistically achieve. Think about it, you cannot reproduce the full energy of a tympani drum or a kick drum realistically with an 8 inch drive unit in a small cabinet. If you mess up the phase response through the audio band the fundamental notes and their harmonics will not be reproduced with the correct time relationship causing much musical confusion. Squash the dynamic scale and the performance fails to excite as much as a live event. I agree that a flat frequency response is not as important as these three key attributes. Anything that happens to the sound in the airspace of a live acoustic musical event will be perceived as a natural occurrence, and your brain will accept this as normal if it occurs in a sound system provided the effect is not exaggerated. Allowing a loudspeaker to generate effects that are not normally perceived at a live acoustic event can be disruptive to the illusion you wish to great. This goes for other equipment in the chain of reproduction too. Measurements are useful initially to get some idea of the overall electrical/mechanical performance but should never be solely relied upon. I have heard so many designs that measure well but sound wrong and some designs that do not measure to well by accepted standards that sound great. Your ears are your best asset when designing or choosing equipment.

Paul

Couldn't have put it better.

[anthonyTD]

I use full range loudspeakers without crossovers and would not sacrifice the coherent presentation I am getting with these. During the 1970’s I used Eagle FR8 full-range drive units loaded with sewer pipes, tuned to the drive units, driven with Linsley Hood Class A amplifiers (the original simplest version in turn driven by Hood’s phono stage) and really enjoyed the musical presentation these items achieved even though the measured response could have been criticised. Everyone who heard this system was in agreement that musically it was exceptional compared to what was generally available at the time for similar levels of expenditure.

Historically I have had a lot of involvement with loudspeaker design, designers and manufacturing. I have tried most of the design options available for cabinet loading and crossovers, both passive and active through the 1980’s and early 1990’s. Whilst some of these designs were impressive for one reason or another, few approached the coherence provided by my 1970s system so I went back to full range, and I have been there ever since.

Much has changed since the 1970s but loudspeaker design is still governed by the laws of Physics. Disregarding the recording chain and replay electronics, which have their own issues that need care and attention in design, recreating a convincing illusion of a live musical event requires more than most commercial loudspeaker designs can achieve.

For instance, a classical concert performance or a rock concert will require movement of air that no single small drive unit would ever be able to realistically achieve. Think about it, you cannot reproduce the full energy of a tympani drum or a kick drum realistically with an 8 inch drive unit in a small cabinet. If you mess up the phase response through the audio band the fundamental notes and their harmonics will not be reproduced with the correct time relationship causing much musical confusion. Squash the dynamic scale and the performance fails to excite as much as a live event. I agree that a flat frequency response is not as important as these three key attributes. Anything that happens to the sound in the airspace of a live acoustic musical event will be perceived as a natural occurrence, and your brain will accept this as normal if it occurs in a sound system provided the effect is not exaggerated. Allowing a loudspeaker to generate effects that are not normally perceived at a live acoustic event can be disruptive to the illusion you wish to great. This goes for other equipment in the chain of reproduction too. Measurements are useful initially to get some idea of the overall electrical/mechanical performance but should never be solely relied upon. I have heard so many designs that measure well but sound wrong and some designs that do not measure to well by accepted standards that sound great. Your ears are your best asset when designing or choosing equipment.

Paul

Intersting acount paul,
it would seem that there are quite a few of us here that have experienced what good full range drivers are capable of, as you say, its not all about measurments, its about getting a speaker and system to portray music in a way that excites and keeps one's interest.
nice to see you posting again.
Anthony,TD...

[AlexM]

It was stated earlier in the thread that a flat frequency response, or a reasonable approximation is not important as the brain 'corrects' issues here. I don't think that I can agree with that POV as, at least for me, one of the goals of HiFi reproduction is to reproduce a recording accurately and therefore get as close as possible to the recorded representation of the original performance and the tonality of the instruments.

Deviation from the recorded tonal balance as a result of dips or peaks in the frequency response is introducing colouration, and is altering the reproduction of the harmonics of the instrument. The problem I see with speakers that are inherently coloured is that they may may work better for some programme material than others, or be inconsistent in how the tonality of instruments is reproduced based on the pitch of the note played and that isn't what I personally am looking for, albeit that you can get used to it to an extent.

As has been said already, there are many ways to skin the cat and many of the problems of using passive cross-overs is a consequence of the design decisions that result in a particular choice of driver passbands and where this determines that crossover points need to be. I feel that the right choices of mechanical drive unit parameters in conjunction with cabinet and electrical design of the cross overs can ensure that the drivers work as effectively as possible, distortion is minimised, and dispersion across the frequency range is as consistent as possible with the aim of having as neutral a sound as possible.

I don't think that this is at odds with using high quality drivers with a minimum of 'nasties' that need crossover help to mask, and/or minimising the number of passive components in the crossover network.

Regards,
Alex