[ChrisCoaster]

I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

[geoff]

ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

MDF.

The world didn't end here first ! I/ 3/4 af the way past 22/05 .

geoff

[Scott Dorsey]

ChrisCoaster wrote:
I want to concentrate on building speakers!

You have a long, slow road ahead of you. Buy the Vance Dickason loudspeaker
cookbook... and realize that is just a cookbook and really none of the theory
is in there. Read it cover to cover. Build a lot of prototypes and expect
most of them not to be very good.

Speaker building is one of the last places in the audio design field that
remains as much art as science.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

Seriously. One of my obsessions in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor -- minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

Resonance is important, but there seem to be other important factors.

Back in 1980, I was doing repair work at Chestnut Hill Audio in
Philadelphia. The Dahlquist rep stepped in with "something interesting" to
show us. Jon Dahlquist was still at work on the DQ-6, a speaker that never
saw the light of day in its intended form. (Jon was very much the Orson
Welles of speaker design -- he could never finish anything.)

The rep told us about a Weird Occurrence. One of two "identical" prototypes
sounded quite different, even though they used the same components.
Switching the components (drivers + crossover) did not switch the
better-sounding speaker -- it remained with the cabinet.

It turned out the "better" speaker was painted with 3M Nextel. It was used
solely for cosmetic reasons. The "nappy" paint was apparently suppressing
radiation from the cabinet surface.

You should have heard the demo speakers. They generated a truly 3D
soundfield. You heard instruments not only "outside" the speakers, but in
front of and behind them. I've never heard anything like that since.

If I were designing my own cone-type speakers, I would do the following...

Use the thickest, densest particle board, and brace the hell out of the
cabinet.
Design to cabinet and position the drivers to minimize diffraction.
Paint all cabinet surfaces -- including the rear -- with Nextel.
Cover the face of the cabinet -- except for the drivers, of course! --
with Fibreglas.

The goal is to get zero radiation from the cabinet surfaces and edges.
Another approach would be to free-mount the drivers that don't need
baffling.

Looking at the other postings, I don't see any bad suggestions.

[Sean Conolly]

"ChrisCoaster" wrote in message
...
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

I've been thinking about a speaker build myself, but I'm thinking this isn't
really the forum to discuss ideas. I just want to try my best and see how
good it ends up - not reference monitors, but better than typical home
stereo speakers.

Maybe...

Sean

[Peter Larsen[_3_]]

Sean Conolly wrote:


I've been thinking about a speaker build myself, but I'm thinking
this isn't really the forum to discuss ideas.

We used to talk about that over in rec.audio.tech, and it still is a very
good place for it.

Sean

Kind regards

Peter Larsen

[alex]

Il 22/05/2011 4.20, ChrisCoaster ha scritto:
Now That 5/21 Passed without incident . . .
I want to concentrate on building speakers!

wrong, you started to build speakers, the world will never be te same. :-)

alex

[Les Cargill[_4_]]

ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.


It doesn't work that way. The cabinet is part of the deal. You can
build large, open back baffles but that just makes the "cabinet"
variable size.

http://www.linearteam.dk/default.aspx?pageid=winisd

(doesn't cover open back designs)

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?


MDF is allegedly the least self-resonant material that'll fit on a
table saw. At least in guitars, metal is used because
it's resonant.

-ChrisCoaster

--
Les Cargill

[ChrisCoaster]

On May 22, 3:59*pm, Les Cargill wrote:
ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. *One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

It doesn't work that way. The cabinet is part of the deal. You can
build large, open back baffles but that just makes the "cabinet"
variable size.

http://www.linearteam.dk/default.aspx?pageid=winisd

(doesn't cover open back designs)

How critical is that last factor - minimizing resonance? *And should I
build them out of 1" thick steel plate or oak/other wood?

MDF is allegedly the least self-resonant material that'll fit on a
table saw. At least in guitars, metal is used because
it's resonant.

-ChrisCoaster

--
Les Cargill
______________________

I was just being sarcastic about the steel plate. LOL! Of course I
would use wood, it's just that I've heard the sound of too many
bookshelf speakers and gotten fatigued by the presence of low-mid
"tub"(150 - 300Hz). I want to avoid that.

Other ?? I have:

Other than the cabinet, how DO mfgs. achieve a relatively flat
response - no more than +-3dB from 50~20,000Hz/+-5dB 20~20k?

Is there some calibration done to the drivers themselves? To the
crossovers? The materials of the drivers themselves, Or, is it the
way they wind the copper coils?

And what combination of above? I've been told repeatedly that I "will
not like" the sound of flat speakers, but my rebuttal time and time
again is, my ears and the room are at fault for that!

-ChrisCoaster

[g]

On May 22, 8:13*pm, ChrisCoaster wrote:
On May 22, 3:59*pm, Les Cargill wrote:



ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. *One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

It doesn't work that way. The cabinet is part of the deal. You can
build large, open back baffles but that just makes the "cabinet"
variable size.

http://www.linearteam.dk/default.aspx?pageid=winisd

(doesn't cover open back designs)

How critical is that last factor - minimizing resonance? *And should I
build them out of 1" thick steel plate or oak/other wood?

MDF is allegedly the least self-resonant material that'll fit on a
table saw. At least in guitars, metal is used because
it's resonant.

-ChrisCoaster

--
Les Cargill

______________________

I was just being sarcastic about the steel plate. LOL! *Of course I
would use wood, it's just that I've heard the sound of too many
bookshelf speakers and gotten fatigued by the presence of low-mid
"tub"(150 - 300Hz). *I want to avoid that.

Other ?? I have:

Other than the cabinet, how DO mfgs. achieve a relatively flat
response - no more than +-3dB from 50~20,000Hz/+-5dB 20~20k?

Is there some calibration done to the drivers themselves? *To the
crossovers? *The materials of the drivers themselves, Or, is it the
way they wind the copper coils?

And what combination of above? *I've been told repeatedly that I "will
not like" the sound of flat speakers, but my rebuttal time and time
again is, my ears and the room are at fault for that!

-ChrisCoaster

The specific woofer and box tuning is going to affect that high bass
peak the most. The speakers designed to be on stands away from the
wall typically have more bloom in that region, and if put them against
the wall !!!

Some of the most successful designs are two ways.

For road work I like plywood with proper reinforcements and padding.
It's lighter. Plexaglass is not bad for cabinets. If done right.

Corning made the best fiberglass a few years ago, itch less. I still
have reserve stock, for when you want to get the most out of a very
small closed box. Makes the box bigger. Other materials don't do that
as well.

I more often than not, attach frontal absorbent material around
tweeters. Tweeter rings. It makes a big difference in sound anomalies
caused by refraction. I go as needed, listening to pink noise. That's
for a typical dome tweeter.

Afte r dozens of speaker boxes, I have yet to use MDF because of
factors including hauling it home it's good stuff.

I built one speaker column out of oak. I didn't really make it, had it
made from raw wood stock cut down about 1 inch thick, braced and
padded. I gave it to my brother for a wedding present. Still sounds
great after17 years!!

Sometimes I leave the crossover out of the box. I started doing this
so I could change values and not have to play take the box apart
again. Sometimes many attempts. Sometimes it takes me weeks to adjust
values after weeks of listing.

Greg

[Scott Dorsey]

ChrisCoaster wrote:
I was just being sarcastic about the steel plate. LOL! Of course I
would use wood, it's just that I've heard the sound of too many
bookshelf speakers and gotten fatigued by the presence of low-mid
"tub"(150 - 300Hz). I want to avoid that.

That's usually not a cabinet resonance, though. That's most often built
into the alignment.

You need to stop listening to crappy speakers.

Other than the cabinet, how DO mfgs. achieve a relatively flat
response - no more than +-3dB from 50~20,000Hz/+-5dB 20~20k?

They don't. I have never seen any speaker whose narrowband response
is that flat, and I have seen some multi-hundred-thousand dollar monitor
systems.

Is there some calibration done to the drivers themselves? To the
crossovers? The materials of the drivers themselves, Or, is it the
way they wind the copper coils?

And what combination of above? I've been told repeatedly that I "will
not like" the sound of flat speakers, but my rebuttal time and time
again is, my ears and the room are at fault for that!

Well, since they don't exist it's moot.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[PStamler]

On May 22, 9:40*pm, (Scott Dorsey) wrote:

And what combination of above? *I've been told repeatedly that I "will
not like" the sound of flat speakers, but my rebuttal time and time
again is, my ears and the room are at fault for that!

Well, since they don't exist it's moot.

But some speakers get remarkably close within a limited range of
frequencies; the BBC LS3/5a monitor, for example, can measure pretty
damn flat from 125Hz up to about 10kHz in a decent room. Of course,
there's almost nothing below 100Hz...and these little half-cubic-foot
speaker systems have some remarkably sophisticated engineering in
them, including selected custom drivers, a crossover which can match
sensitivities in unusual ways, construction from Baltic Birch multi-
layer plywood (sometimes sold in the USA as Finnply) with panels
damped by pads made from asphalt-impregnated felt, diffraction control
around the tweeter...on and on. The thing is, for a really good system
*everything* matters.

Scott's right -- go get Vance Dickason's book and read it several
times. Then look at Martin Colloms's "High Performance Loudspeakers";
it's a book with issues, but he also documents some very good designs
from British manufacturers. Then look at Siegfried Linkwitz's designs.
Then...

....it's never-ending. Fun, though.

Peace,
Paul

[geoff]

PStamler wrote:
On May 22, 9:40 pm, (Scott Dorsey) wrote:

And what combination of above? I've been told repeatedly that I
"will not like" the sound of flat speakers, but my rebuttal time
and time again is, my ears and the room are at fault for that!

Well, since they don't exist it's moot.

But some speakers get remarkably close within a limited range of
frequencies; the BBC LS3/5a monitor, for example, can measure pretty
damn flat from 125Hz up to about 10kHz in a decent room. Of course,
there's almost nothing below 100Hz...and these little half-cubic-foot
speaker systems have some remarkably sophisticated engineering in
them, including selected custom drivers, a crossover which can match

My LS3-5As are MDF and lined with 1/8" lead on the inside ! And I made the
crossover with the real auto-transformer rather than the resisitive version
;-)


geoff

[William Sommerwerck]

"Scott Dorsey" wrote in message
...

Other than the cabinet, how DO mfgs. achieve a relatively flat
response - no more than +-3dB from 50~20,000Hz/+-5dB 20~20k?

They don't. I have never seen any speaker whose narrowband response
is that flat, and I have seen some multi-hundred-thousand dollar monitor
systems.

Scott, you are so rarely wrong about anything that I had to mull this one
over overnight before commenting.

Ignoring the issue of /how/ one measure's a speaker's response, and what
that measurement "means"... *

Drivers have improved drastically in the past 20 years. A speaker that can't
get from, say, 50Hz to 15kHz, within a 6dB envelope, is a pretty bad design.
I'd expect a really good speaker's response to fall within a 4dB envelope,
or even 3dB.

For example, B&W claims +/- 3dB from 56Hz to 22kHz -- on an unspecified
measurement axis -- for one of its CM-series mini-monitors. The response for
the 800 Diamond is spec'd at +/- 3dB from 32Hz to 28kHz.

http://www.bowers-wilkins.com/Downlo...info_sheet.pdf

I'm reminded of the Thiel CS-5 which, almost 20 years ago, had a response
that looked as if it had been drawn with a ruler. Say what you like about
Thiel -- their products are the epitome of "blah"-ness -- but no one ever
accused Thiel of lying.

* Stereophile's recent review of a Linn speaker is an excellent example of
how measurements don't always correlate with what you hear. Of course, I
wouldn't own a Linn product unless I was paid a lot of money. And I doubt
I'd actually listen to it...

[g]

On May 21, 10:20*pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. *One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? *And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.


Some have mentioned books. I think I got most of what I know from my 4
foot high set of Speaker Builder mags. I started reading s friends set
in the early 80's and I was hooked. Many of the top people like Wrote
articles over the years. You should be able to get a hold of
collections. Not mine.

[Don Pearce[_3_]]

On Mon, 23 May 2011 11:07:46 -0700 (PDT), G wrote:

On May 21, 10:20*pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. *One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? *And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.

What does that do, and how does it work? I can't picture this any
other way than completely wrong.

d

[PStamler]

On May 23, 1:13*pm, (Don Pearce) wrote:
On Mon, 23 May 2011 11:07:46 -0700 (PDT), G wrote:
On May 21, 10:20 pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.

What does that do, and how does it work? I can't picture this any
other way than completely wrong.

The idea is to use elastomeric gaskets between the driver and the
cabinet, and attach the driver via screws and T-nuts which are
decoupled from the cabinet by similar elastomeric suspensions. The
intent is to prevent vibration from being conducted from the driver
frame to the material of the cabinet. Since most cabinet vibration
doesn't come from that source, but from the pressure changes in the
cabinet (aka "sound") the decoupling devices create only a small
improvement if any.

Peace,
Paul

[Don Pearce[_3_]]

On Mon, 23 May 2011 12:09:41 -0700 (PDT), PStamler
wrote:

On May 23, 1:13*pm, (Don Pearce) wrote:
On Mon, 23 May 2011 11:07:46 -0700 (PDT), G wrote:
On May 21, 10:20 pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.

What does that do, and how does it work? I can't picture this any
other way than completely wrong.

The idea is to use elastomeric gaskets between the driver and the
cabinet, and attach the driver via screws and T-nuts which are
decoupled from the cabinet by similar elastomeric suspensions. The
intent is to prevent vibration from being conducted from the driver
frame to the material of the cabinet. Since most cabinet vibration
doesn't come from that source, but from the pressure changes in the
cabinet (aka "sound") the decoupling devices create only a small
improvement if any.

Peace,
Paul

Ah, a gasket is somewhat different to grommets. But that still makes
no sense; the last thing you want is the driver floating in the
cabinet. Rigid mounting is most definitely what is needed.

d

[PStamler]

On May 23, 2:34*pm, (Don Pearce) wrote:
On Mon, 23 May 2011 12:09:41 -0700 (PDT), PStamler



wrote:
On May 23, 1:13 pm, (Don Pearce) wrote:
On Mon, 23 May 2011 11:07:46 -0700 (PDT), G wrote:
On May 21, 10:20 pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.

What does that do, and how does it work? I can't picture this any
other way than completely wrong.

The idea is to use elastomeric gaskets between the driver and the
cabinet, and attach the driver via screws and T-nuts which are
decoupled from the cabinet by similar elastomeric suspensions. The
intent is to prevent vibration from being conducted from the driver
frame to the material of the cabinet. Since most cabinet vibration
doesn't come from that source, but from the pressure changes in the
cabinet (aka "sound") the decoupling devices create only a small
improvement if any.

Peace,
Paul

Ah, a gasket is somewhat different to grommets. But that still makes
no sense; the last thing you want is the driver floating in the
cabinet. Rigid mounting is most definitely what is needed.

Well, I agree, but there was a time in the 1980s when there was a fad
for using elastic gaskets and rubber grommets to float the driver.
Again, the supposed justification was to keep vibrations from the
driver from being conducted to the cabinet.

Personally I favor a nice lossy cabinet made from MDF or thin-ply
plywood such as Finnply, and absorption on the panels to damp down
their vibrations, plus strategically-placed bracing. Given good
attention to those factors, conduction from the drivers to the cabinet
will have little effect.

Peace,
Paul

[Bill Graham]

Don Pearce wrote:
On Mon, 23 May 2011 11:07:46 -0700 (PDT), G wrote:

On May 21, 10:20 pm, ChrisCoaster wrote:
I want to concentrate on building speakers!

Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should
I build them out of 1" thick steel plate or oak/other wood?

-ChrisCoaster

One trick that has been done many time is decouple the driver with
some rubber like mounting grommets.

What does that do, and how does it work? I can't picture this any
other way than completely wrong.

d

Why not hang the speaker from 4 or 6 short bungee cords? That should
effectively decouple it from the surrounding cabnetry.

[Ben Bradley[_2_]]

On Sat, 21 May 2011 19:20:07 -0700 (PDT), ChrisCoaster
wrote:

I want to concentrate on building speakers!

I've only built a few pair over the last several decades, with, in
retrospect, hardly acceptable results. I've quite often wanted to do
more, but the more I've read and researched the topic, the more
knowledge it takes to do it well (like most anything else). I've seen
speaker designing and building come up occasionally on RAP as well as
of course rec.audio.tech, so check the archives in that "Google
Groups" interface.


Seriously. One of my obsession in this regard is building a speaker
where the only part you hear sound from are the transducers/drivers/
cones/whatever, NOT the material they're mounted in/bolted into.

How critical is that last factor - minimizing resonance? And should I
build them out of 1" thick steel plate or oak/other wood?

Not mentioned in the thread so far is each cabinet wall/floor/top
made of two layers of wood/MDF/whatever with sand filled into the
space in between. What others say about damping internal reflections
still applies. That's one reason cabinet dimensions are never a cube
(the internal reflections have the same resonant frequency in all
three dimensions!), and are generally some nonharmonic ratios (such as
1 : cube root of 2 : square of the cube root of 2).

Having "researched" so much audio stuff (yet admittedly only
applied very little), I've found interesting parallels such as studio
room design (now being touched on in the "cost to build studio"
thread) follows pretty much the same rules as a speaker cabinet,
though the "room nodes" are of course at lower frequencies due to the
longer distances involved.

I've read the already-mentioned Dickason and Collums books, and
have a collection of the the Speaker Builder magazines, as well as the
earler (and concurrent) Audio Amateur magazines which included speaker
building before the publisher decided to spin off that topic into its
separate magazine. They apparently got integrated back in the more
recent incarnation of AudioXpress and whatever name change it went
through, but I don't have many issues of that and I don't know how
much it covers speaker building, if at all.

There's lots of stuff not mentioned in the thread so far. I think
just about everything is covered in the above material, but there's
also lots more freely available on the Web.

Yet another detail is the baffle step, which is described (among
many other places) he
http://sound.westhost.com/bafflestep.htm
That's a pretty decent site, it's surprising how much audio
electronics stuff he's written up, not to mention the page on CF
lamps.

For good online resources, I stumbled on http://diyaudio.com a year
or two ago, and it's got lots of people with lots of knowledge and
experience, including some recognizable names, authors and designers
in the audio field. There's surely some 2-way versus 3-way arguments
there, and there's a separate forum for designing speakers with
full-range drivers.

For crossovers, line-level and power audio elctronics has gotten so
cheap it's almost a no-brainer to do active crossovers, instead of the
traditional passive crossover in the cabinet. The filters are more
easily done at line level, nd each driver can be run directly with its
own amplifier.

Also with good forums and presumably good off-the-shelf
cabinet/crossover designs for the drivers they sell is Parts Express:
http://techtalk.parts-express.com/
I just glanced at Madisound (the other big driver mail-order supplier
in the USA), I see a forum there but it's much smaller.

There's several DIY loudspeaker mailing list(s) and forums that
have been aroud for many years - just google.

Almost forgot - namebrand specific, but several posters are former
employees and designers, with a lot of good technical info such as the
reasons for various design decisions:
http://www.audioheritage.org/vbulletin/


-ChrisCoaster