Speaker design discussion

[Richard Dunn]

Very interesting, but nothing to do with cubes or internal damping.
Most of that document appears to be discussing speaker positioning within the room.
Designing loudspeakers to account for boundaries is well understood.

Have we moved onto a new topic?

Of course it is to do with Cubes which is why I hi-lited that bit. The 6 was a cube and it was affectionately referred to as "le cube" I couldn't find his tech talk on that and why it is a cube but you seemed to think his designs were to do with Acoustic Research, I was showing it wasn't.

Yes his major work was room boundary effects but as I said a speakers is a box within a box (the room) and this is about the interface between the two boxes and is integral with my other argument about room acoustic treatment and filters. Roy Allison designed speaker to work with the room and not fight it and that is the same principle I work with. You have to tune the speaker position to the room which is far more precise with this type of design, and there isn't an acoustic you cannot tune it to (apart from in alcoves), which probably was the problem at the show, but I was busy and I would have thought others could have done it.

Now not only did he work with external interfaces he worked with internal interfaces and as I say I have taken that further along the same line of thinking that he would have surely used as it is logical progression if he was able to use cabinets like these. If I produced a three way the cabinet would be triangular with a very low bass driver as was the Allison one, maybe one day I will, here is a pic of the Allison 1.

All his designs were semi omni and the other speakers in this pic are like the 6 with top bass/ mid but with other 45deg drivers and I don't know the model.

EDIT other people who used his work were Linn with the Isobarik.

 

[RobHolt]

Again it is interesting reading but veering off topic.
But if you want to extend it.

Perfectly cubed rooms honk and boom dreadfully, and they do so most effectively when driven from the boundary positions.
if you cary this through to the speaker itself, a perfect cube cabinet with a driver mounted at the boundary will have some clear issues, this time further up the frequency range as the speaker box is smaller than the room box.
Potentially worse still if the bass mid driver is running without any filter on it.

In the real world, the room is furnished. Sofas, carpets, curtains, pictures, cabinets, all sorts. These act in complex fashion to absorb, diffuse and reflect at different frequencies.
The inside of the speaker cabinet follows the same rules, and running it bare and empty is akin to listening in a bare unfurnished room. Bloody horrible!

Placing the bass/mid driver on the upper surface of the cabinet so that the front on axis response is reasonably correct also has problems. You are running a driver not intended for full range use wide open and effectively rolling it off by listening to it off axis in the listening position. Sounds like a neat solution until you stop to consider the power response and what happens to all that upper mid/lower treble energy from the bass driver firing up at the ceiling and walls. It still enters the room and contributes to the sound heard at the listening position. Because it isn't directly radiated sound it is of course time smeared by the time it reaches your ears, and the overall in room response will likely be very unnatural, and variable.
That is exactly how the speakers sound IMO

 

[Richard Dunn]

Again it is interesting reading but veering off topic.
But if you want to extend it.

Perfectly cubed rooms honk and boom dreadfully, and they do so most effectively when driven from the boundary positions.
if you cary this through to the speaker itself, a perfect cube cabinet with a driver mounted at the boundary will have some clear issues, this time further up the frequency range as the speaker box is smaller than the room box.
Potentially worse still if the bass mid driver is running without any filter on it.

In the real world, the room is furnished. Sofas, carpets, curtains, pictures, cabinets, all sorts. These act in complex fashion to absorb, diffuse and reflect at different frequencies.
The inside of the speaker cabinet follows the same rules, and running it bare and empty is akin to listening in a bare unfurnished room. Bloody horrible!

Placing the bass/mid driver on the upper surface of the cabinet so that the front on axis response is reasonably correct also has problems. You are running a driver not intended for full range use wide open and effectively rolling it off by listening to it off axis in the listening position. Sounds like a neat solution until you stop to consider the power response and what happens to all that upper mid/lower treble energy from the bass driver firing up at the ceiling and walls. It still enters the room and contributes to the sound heard at the listening position. Because it isn't directly radiated sound it is of course time smeared by the time it reaches your ears, and the overall in room response will likely be very unnatural, and variable.
That is exactly how the speakers sound IMO

Sorry that is gobble-de-gook, an empty speaker cab is nothing like an empty room, it is the other side of the mirror. The driver is in acoustic suspension so it is a pressure wave in the cabinet, the back sound wave dissipates within the cabinet and due to the qualities of acrylic HF is dissipated in the case where as LF is transmitted as a pressure wave into the stand and floor. As I said if I used chipboard cabs I would use some damping but it would still be minimal, you could get away with HDF cabs with no wadding in this configuration but acrylic is better.

You also don't understand the advantages of an upward driver. There are distinct advantages in both planes, down ward it couple bass frequencies directly into the room interface (the stand and floor) as there is no 90deg phase turn for the energy to couple, and there is progressive natural attenuation of high MF and HF as they are progressively directional. The drive unit is also doped to naturally attenuate that and remove nodes mechanically as well, so all the upward HF projection does is increase the height and depth of the sound stage, as long as the speaker is positioned correctly..

Also you don't seem to understand that a cube speaker is not like cube room due to the fact that the sound wave is emitted from a cone that is inside the cabinet, so it does not behave like a cube, if the driver was planar and on the surface of the cab I would understand your argument.

 

[RobHolt]

I understand it perfectly well Richard.

Your understanding of how speaker cabinets work s simply wrong, as the above post illustrates.

 

[Richard Dunn]

I understand it perfectly well Richard.

Your understanding of how speaker cabinets work s simply wrong, as the above post illustrates.

From who's authority, how many speakers cabinets have you designed....

 

[Richard Dunn]

You've always said that Richard, but in my experience (and academic understanding) it is not usually the case. In fact it isn't my experience of filters in general. I think nearly all systems in audio, be they electronic or mechanical are improved by filtering of some sort.

Actually just thinking about it, I guess your amps use feedback of some sort, and isn't that filtering? Have you built a zero-feedback , single-ended amp before, and what did you think?

Only just noticed your question Simon, yes I tried to build a single ended transistor amp with no feedback as a project, but I couldn't make it work and I didn't have the time to pursue it. Tim de P' made one and sold it, really to prove that transistors v valves was not the argument, but circuit topography was.

My circuit is low feedback and my point has always been use the least number of filters as possible, filters are cures for problems, so if you don't have a problem then don't use the cure, but if you have a problem you cannot cure any other way then you have to use them, such as RIAA.

 

[Richard Dunn]

He was probably being nice to you - he's a nice guy.

The simple fact is that he used alternative speakers to run his demo on the day.
He didn't have to, they were sitting in our room as spares for anyone to use.

This is Jims reply to my PM copied here in full with his permission.

************************

I ended up preferring the Dynacos, probably down to the room, gear and music we were playing - and also the Dynacos were easier to drive with the amps we had to hand.

The only times I've preferred the NVA speakers has been when they've been driven by NVA amps - maybe that's a large part of it.

I've never been on Zerogain. I'd much rather not get involved in cross forum arguments. It seems to me that there is far too much of that going on already. I'd prefer the subject dropped please.

Jim

 

[larkrise]

Just noticed the reference to Alison speakers.

I had lots in the 70's/80's and had a pair of Alison Ones right up till 8 years ago.
One of the best speakers ever made - I had a pair which were triamped with Bill Beard power amps - the active crossovers designed by Neil of Albarry Amps.

And here is a pic of my Alison ones - original! Couldn't give them away in the end, however one of the best speakers I have ever heard at any price. i only sold as lack of wall space

Photo uploading wont work for some reason - sorry!

 

[TonyL]

Photo uploading wont work for some reason - sorry!

Just dump it up to www.tinypic.com and copy the BB code link into your post here.

Tony.

 

[sq225917]

Richard I'm afraid you have the wrong end of the Alison stick, and it would appear that you've working under this misapprehension for quite some time. His work regarded the speaker room boundary interface not the internals of cabinet dimensions.

A cuboid is the worst shape for eliminating standing waves, it's a simple matter of symmetry, the more convergent planes of symmetry you have the greater the number of coincident reflections you create focussing all the energy from the back of the driver into a narrow frequency and geometric band within the speaker.

Acrylic is a decent material for self damping, by changing the panel sizes and thickness you can spread the resonant frequencies as you see fit to boost or cut the response from the drivers, as the BBc and Harbeth did/do. Either you hit lucky first time with narrow but deep frequency cut that needed reinforcement from your cuboidal design, or you haven't really thought it through/measured it as a speaker 'designer' would.

If it sounds exactly as you want it to sound, then as a subjective designer there's no reason for you not to be happy with it. What serendipity!

 

[larkrise]

here ya go

 

[RobHolt]

This is Jims reply to my PM copied here in full with his permission.

************************

I ended up preferring the Dynacos, probably down to the room, gear and music we were playing - and also the Dynacos were easier to drive with the amps we had to hand.

The only times I've preferred the NVA speakers has been when they've been driven by NVA amps - maybe that's a large part of it.

I've never been on Zerogain. I'd much rather not get involved in cross forum arguments. It seems to me that there is far too much of that going on already. I'd prefer the subject dropped please.

Jim

Ah, so nothing to do with the speakers, it was the rest of the the kit, the music and the room.

I was given to understand from your previous posts that the rooms doesn't matter, ie your speakers are designed to work with the room. This appears to be another area of design that needs some thought.

On the music - you cannot be serious!
A variety was being played, and if you are designing loudspeakers that only suit certain types of music, they are bad designs, clearly.
BBC and other apparently 'complex' loudspeakers have no such limitations, which is interesting.

On the amplifier, that's compete rot, but if that really is your view I'd suggest putting up a list of amplifiers that will magically work with a couple of off the shelf drivers in a box on a minimalist crossover (hardly a demanding load).

No reflection on Jim at all, he clearly doesn't want to upset anyone and you've pushed him into a corner.

 

[Richard Dunn]

Ah, so nothing to do with the speakers, it was the rest of the the kit, the music and the room.

I was given to understand from your previous posts that the rooms doesn't matter, ie your speakers are designed to work with the room. This appears to be another area of design that needs some thought.

On the music - you cannot be serious!
A variety was being played, and if you are designing loudspeakers that only suit certain types of music, they are bad designs, clearly.
BBC and other apparently 'complex' loudspeakers have no such limitations, which is interesting.

On the amplifier, that's compete rot, but if that really is your view I'd suggest putting up a list of amplifiers that will magically work with a couple of off the shelf drivers in a box on a minimalist crossover (hardly a demanding load).

No reflection on Jim at all, he clearly doesn't want to upset anyone and you've pushed him into a corner.

Are you deliberately not understanding what I say in plain English, and what Jim is saying. Anyone who owns or has tried this design, and the designs of RA know that moving the speaker in and out is like using tone control without the side effect as it is not a cure it is natural, and adjusting toe in is like changing the aperture of a camera lens to focus the image. This has to be done with every new acoustic, but there is not an acoustic it cannot be balanced to unlike other speakers. As I said the only problem I have found is setting them back in alcoves either side of a fireplace, they just go dual mono because they need to project.

 

[RobHolt]

English is fine, but your posts seem to just state the blindingly obvious - that moving a speaker around a room changes the sound.
This is hardly a revolutionary discovery and you needn't quote research or application papers verbatim to convince us.

 

[Richard Dunn]

Richard I'm afraid you have the wrong end of the Alison stick, and it would appear that you've working under this misapprehension for quite some time. His work regarded the speaker room boundary interface not the internals of cabinet dimensions.

A cuboid is the worst shape for eliminating standing waves, it's a simple matter of symmetry, the more convergent planes of symmetry you have the greater the number of coincident reflections you create focussing all the energy from the back of the driver into a narrow frequency and geometric band within the speaker.

Acrylic is a decent material for self damping, by changing the panel sizes and thickness you can spread the resonant frequencies as you see fit to boost or cut the response from the drivers, as the BBc and Harbeth did/do. Either you hit lucky first time with narrow but deep frequency cut that needed reinforcement from your cuboidal design, or you haven't really thought it through/measured it as a speaker 'designer' would.

If it sounds exactly as you want it to sound, then as a subjective designer there's no reason for you not to be happy with it. What serendipity!

Sigh! as I have said over and over again, you obviously don't read or it is convenient not to read. If the driver was planar I would agree with you, but it is not, it is a normal cone and the apex of that cone is about 1/3rd distance inside the cabinet, so in relation to the point that is transmitting the signal the last thing the cabinet behaves like is a cube. Wadding is a filter, it is not absolute, it works decreasingly at low frequency so because of that wadding all that comes from inside the cab is waffle. Inside the Cubes the music is music. As I said before put some acoustic wadding or a duster in the bell of a sax and tell me what you hear.

And RA did do cabinet design as should be obvious if you look at his unconventional and lateral thinking designs. Also look at the Isobaric, that was based on some of his ideas as well, but it was thoroughly over damped.

 

[nando]

how bizzare, is it not tone tuning the same via tone controls then messing around with loudspeakers to suit your roon rather then adjusting the tone of a pre-amp to not only change the so dull recordings,specially reggae where most jamaican recordings need such tonal tuning to the preference of the listener, a speaker well desingned should reproduce all sorts of music well with no prejudice, only music has that in the way it was recorded,

 

[Richard Dunn]

English is fine, but your posts seem to just state the blindingly obvious - that moving a speaker around a room changes the sound.
This is hardly a revolutionary discovery and you needn't quote research or application papers verbatim to convince us.

OK back to the beginning. The basis of part of this discussion was the need for filter because speakers could not be fitted into acoustics, so if what you say is true then there is no problem. Obviously this is not the case, speakers struggle with acoustics and you have to often change speakers to get a match. If you understood what RA was saying, he was showing that the problem could be addressed within the design of the speaker cabinet, and that is the design principle I follow. Now of course all speakers change when you move them, but obviously not enough, these designs change enough to fit all acoustics.

 

[RobHolt]

OK back to the beginning. The basis of part of this discussion was the need for filter because speakers could not be fitted into acoustics, so if what you say is true then there is no problem. Obviously this is not the case, speakers struggle with acoustics and you have to often change speakers to get a match. If you understood what RA was saying, he was showing that the problem could be addressed within the design of the speaker cabinet, and that is the design principle I follow. Now of course all speakers change when you move them, but obviously not enough, these designs change enough to fit all acoustics.

No it isn't.

He's talking about integrating the speaker into the room - something well understood by most loudspeaker designers and a good many users.
You began by stating that speakers sound better with the internal wadding removed and that cubed cabinets were best. The RA papers go nowhere near the subject.

 

[Richard Dunn]

No it isn't.

He's talking about integrating the speaker into the room - something well understood by most loudspeaker designers and a good many users.
You began by stating that speakers sound better with the internal wadding removed and that cubed cabinets were best. The RA papers go nowhere near the subject.

I worked for Acoustic Reseach I read his papers on cabinet design, but I don't have them now, and a major part of it was minimal filtering mechanical or electrical. The marketing emphasis was on room boundary effect as that was to do with the customer interface and problem solving, his cabinet work was not marketed but he knew perfectly well what he was doing. Go look at that picture of the Allison ones, no stand, bass driver right next to the floor, MF and HF drivers at the top of the cab, a triangular cabinet with equal dimensions, none of the drivers pointing at you. It goes so far against British convention and BBC design as you can get, and yet as Larkrise says people who own them have *never* replaced them because they cannot find better, people lovingly re-surround and coddle them. There are very few here in the UK because *yet again* it didn't fit into the flat earth bullshit, and the FE idiots didn't even realise the Isobarik used some of his design principles. Bespoke Audio who also were involved in creating Albarry (Albert and Barry the owners of Bespoke Audio) were also importers of Allison speakers, and much good it did them in the face of the Linn and FE bullshit onslaught, yet another brilliant range of products doomed to failure in the UK market, much to our loss.

He left Acoustic research because of his design work as the company had been taken over by Teledyne who were only interested in the conventional and even cocked that up over the next ten years.

 

[RobHolt]

I worked for Acoustic Reseach I read his papers on cabinet design, but I don't have them now, and a major part of it was minimal filtering mechanical or electrical. The marketing emphasis was on room boundary effect as that was to do with the customer interface and problem solving, his cabinet work was not marketed but he knew perfectly well what he was doing. Go look at that picture of the Allison ones, no stand, bass driver right next to the floor, MF and HF drivers at the top of the cab, a triangular cabinet with equal dimensions, none of the drivers pointing at you. It goes so far against British convention and BBC design as you can get, and yet as Larkrise says people who own them have *never* replaced them because they cannot find better, people lovingly re-surround and coddle them. There are very few here in the UK because *yet again* it didn't fit into the flat earth bullshit, and the FE idiots didn't even realise the Isobarik used some of his design principles. Bespoke Audio who also were involved in creating Albarry (Albert and Barry the owners of Bespoke Audio) were also importers of Allison speakers, and much good it did them in the face of the Linn and FE bullshit onslaught, yet another brilliant range of products doomed to failure in the UK market, much to our loss.

He left Acoustic research because of his design work as the company had been taken over by Teledyne who were only interested in the conventional and even cocked that up over the next ten years.

Yes I know Richard, I have ARs.

The fact remains they are stuffed to the nines with wadding, aren't cube shaped, and work best close to the back wall. We know all this.
Go look on the vintage loudspeaker forum if you seriously believe that these old designs were plain empty boxes, or cubes.