Subwoofer components & design questions

[eganders]

I am interested in building a subwoofer. My questions are:

Are the design program and source of supply good starting points
(WinISD (http://www.linearteam.dk/) and Parts Express (http://
www.partsexpress.com))?

Is a crossover necessary when connected to a receiver such as the
AVR-687 which (as I understand it) supplies frequencies that are
designed for the subwoofer frequency range? Maybe one is still
necessary in order to keep away from the speaker resonent frequency.
Other reasons?

I am having a hard time finding the speakers in the WinISD database.
Why is that?

I would like to keep the parts cost for the speaker, amp and crossover
to around $300. Is that resonable for a speaker that is to be put
into a 5400 cu ft room (20 X 18 X 15)? This is a great room with a
high ceiling.

I would appreciate starting suggestions for speakers and amplifiers
and/or manufacturers. I have not built a speaker in about 30 years
and suppliers and designs have changed somewhat over that period of
time (<:


I am an electrical engineer by trade, so I feel confident in building
a subwoofer. I just want to see if it is worth the effort.
I have a Denon AVR-687 receiver
I have the WinISD subwoofer design program
I have started to look at the Parts Express website at the components
they sell
I am a good woodworker with the necessary tools to construct the
housing for the subwoofer

[Paul Stamler]

"eganders" > wrote in message
oups.com...
> I am interested in building a subwoofer. My questions are:
>
> Are the design program and source of supply good starting points
> (WinISD (http://www.linearteam.dk/) and Parts Express (http://
> www.partsexpress.com))?
>
> Is a crossover necessary when connected to a receiver such as the
> AVR-687 which (as I understand it) supplies frequencies that are
> designed for the subwoofer frequency range? Maybe one is still
> necessary in order to keep away from the speaker resonent frequency.
> Other reasons?

If the receiver indeed has an output designed expressly for a subwoofer,
then no additional crossover is needed; it has the crossover built\in. The
crossover for a sub is a low-pass filter; the speaker's resonance is at the
other end of its passband, the bottom.

> I am having a hard time finding the speakers in the WinISD database.
> Why is that?

Note the last term of the URL for WinISD: .dk . That means it comes from
Denmark. Parts Express is an American retailer,
in my remembrance of their catalog, they sell mostly American subwoofer
drivers, including some custom-made for them.

Does WinISD have the capability of entering the parameters of new drivers?
If so, then you can get the parameters of a particular woofer from Parts
Express and enter them.

> I would like to keep the parts cost for the speaker, amp and crossover
> to around $300. Is that resonable for a speaker that is to be put
> into a 5400 cu ft room (20 X 18 X 15)? This is a great room with a
> high ceiling.

It depends how much you spend on wood, of course, but yes, you should be
able to build a decent powered subwoofer for that price.

> I would appreciate starting suggestions for speakers and amplifiers
> and/or manufacturers. I have not built a speaker in about 30 years
> and suppliers and designs have changed somewhat over that period of
> time (<:

No explicit suggestions (I've been away from that market for about five
years), but look for drivers with long Xmax.

Peace,
Paul

[Scott Dorsey]

eganders > wrote:
>I am interested in building a subwoofer. My questions are:
>
>Are the design program and source of supply good starting points
>(WinISD (http://www.linearteam.dk/) and Parts Express (http://
>www.partsexpress.com))?

Probably, although you might want to just get the equations out of the
Vance Dicksason Loudspeaker Cookbook and do some plugging and chugging
with a hand calculator.

>Is a crossover necessary when connected to a receiver such as the
>AVR-687 which (as I understand it) supplies frequencies that are
>designed for the subwoofer frequency range? Maybe one is still
>necessary in order to keep away from the speaker resonent frequency.
>Other reasons?

If the receiver has a built-in crossover, that's fine. BUT, you need
to make sure it's a real crossover, with a high-pass on the main output
as well as just a low-pass on the subwoofer output. You also want it
to be adjustable in frequency.

>I am having a hard time finding the speakers in the WinISD database.
>Why is that?

I don't know, I never used WinISD. But you can select a driver you like,
get the T-S parameters off the datasheet, and plug them into any application
you like.

>I would like to keep the parts cost for the speaker, amp and crossover
>to around $300. Is that resonable for a speaker that is to be put
>into a 5400 cu ft room (20 X 18 X 15)? This is a great room with a
>high ceiling.

That depends on what you want it to sound like. How low do you want to go
and how loud do you want to be? Do you care about actual device impulse
response because you're listening to acoustic bass, or do you just want it
thumpy and loud?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Chris Hornbeck]

On 1 Apr 2007 09:54:43 -0700, "eganders" > wrote:

>I am interested in building a subwoofer.

>I would appreciate starting suggestions for speakers and amplifiers
>and/or manufacturers. I have not built a speaker in about 30 years
>and suppliers and designs have changed somewhat over that period of
>time (<:

Far and away the most significant change is in the move
from the old model of trying to build a speaker-in-a-box
with flat enough frequency response to the new model of
a speaker-in-a-box with electronic correction for their
frequency response.

The best woofers I've heard (judgment call; natch) were
opposed pairs in boxes so small that their fundamental
resonance is above their operating range. Simple (single
op-amp) correction and big amplifiers do the rest.

The key here is to remember that magnitude response and
output capability are *completely unrelated*.

Extrapolating further, removing the speakers from the box,
getting out the emergency checkbook, and building a dipole
array can give a greater potential than any box design.
Here output capability and checkbook capability reach
Baghdad levels of difficulty.


>I have a Denon AVR-687 receiver

Does this model have automated EQ? If so, you might be
able to use it.

>I have the WinISD subwoofer design program

All the speaker design stuff I've seen is hopelessly old-
fashioned. Haven't seen it all; haven't seen this one.
But the thinking is really more for woodworking kinda
guys than for engineering kinda guys. May not be for *you*
is all I'm saying.

Good fortune; sounds like an excellent project. Everyone
should build speakers. It's too hard to grow food but ya
gotta do something important,

Chris Hornbeck

[eganders]

On Apr 1, 3:29 pm, (Scott Dorsey) wrote:
> eganders > wrote:
> >I am interested in building a subwoofer. My questions are:
>
> >Are the design program and source of supply good starting points
> >(WinISD (http://www.linearteam.dk/) andParts Express(http://
> >www.partsexpress.com))?
>
> Probably, although you might want to just get the equations out of the
> Vance Dicksason Loudspeaker Cookbook and do some plugging and chugging
> with a hand calculator.
>
> >Is a crossover necessary when connected to a receiver such as the
> >AVR-687 which (as I understand it) supplies frequencies that are
> >designed for the subwoofer frequency range? Maybe one is still
> >necessary in order to keep away from the speaker resonent frequency.
> >Other reasons?
>
> If the receiver has a built-in crossover, that's fine. BUT, you need
> to make sure it's a real crossover, with a high-pass on the main output
> as well as just a low-pass on the subwoofer output. You also want it
> to be adjustable in frequency.
>
> >I am having a hard time finding the speakers in the WinISD database.
> >Why is that?
>
> I don't know, I never used WinISD. But you can select a driver you like,
> get the T-S parameters off the datasheet, and plug them into any application
> you like.
>
> >I would like to keep the parts cost for the speaker, amp and crossover
> >to around $300. Is that resonable for a speaker that is to be put
> >into a 5400 cu ft room (20 X 18 X 15)? This is a great room with a
> >high ceiling.
>
> That depends on what you want it to sound like. How low do you want to go
> and how loud do you want to be? Do you care about actual device impulse
> response because you're listening to acoustic bass, or do you just want it
> thumpy and loud?
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

I am just experimenting and want to pick a speaker that will
ultimately work well in a subwoofer configuration, but I will be
testing it and comparing it to the mathematical results to see how
well you can choose the speaker and box dimensions using a design
program. I would like the speaker to have a good, non-boomy sound,
but I would like to have the response be as low as practical. It
looks like 30 to 45 Hz is about the best I can expect with 12 inch
speakers in a closed box. I can get down to about 25 to 30 Hz with a
vented box with the same speaker. These are just my preliminary
findings. The Dayton RSS315HO-4 (a 12 inch driver with a Qts=.4,
Fs=26, Qes=.44, SPS=88 and a Pe=700 for about $130) looks very
interesting as a vented driver. If someone would comment on that
driver, it would be appreciated. I might be interested in a good
lower cost (about $45) driver to compare what the extra $$ buys. So
far one of the big things I see is that it looks like additional $$
will allow me to get a driver that will sound better in a smaller
housing. I don't have a feeling what group delay, phase and other
results will do to the sound. I guess getting 2 or 3 samples of
speakers with different characteristics would help me understand that
better.

[Scott Dorsey]

eganders > wrote:
>I am just experimenting and want to pick a speaker that will
>ultimately work well in a subwoofer configuration, but I will be
>testing it and comparing it to the mathematical results to see how
>well you can choose the speaker and box dimensions using a design
>program. I would like the speaker to have a good, non-boomy sound,
>but I would like to have the response be as low as practical. It
>looks like 30 to 45 Hz is about the best I can expect with 12 inch
>speakers in a closed box.

You can get way lower than that if you pick a driver with a low enough
Fs. The driver resonance is what is limiting the low frequency corner
in these cases. The problem you run into, of course, is that with a
12" driver you can't move a huge amount of air, so a 12" driver that
is tuned very low won't be very loud. NHT, though, makes a 10" and a
12" driver with Fs down below 20 Hz. Some other folks do, but they
aren't pro audio drivers.

> I can get down to about 25 to 30 Hz with a
>vented box with the same speaker. These are just my preliminary
>findings.

Right. With the vented box, you are generating a big box resonance
superimposed on the driver resonance... so as the frequency drops and
the response from the driver drops off, the response from the box picks
up. (Okay, that's a very crude way of thinking about it and Dick Pierce
wouldn't approve, but it will do for now.)

The Dayton RSS315HO-4 (a 12 inch driver with a Qts=.4,
>Fs=26, Qes=.44, SPS=88 and a Pe=700 for about $130) looks very
>interesting as a vented driver. If someone would comment on that
>driver, it would be appreciated.

Roughly, the Fs tells you how low you can make the thing operate, the
Vas gives you a rough idea of how big a box you will need for a given
alignment, and the Xmax multiplied by the cone area tells you how much
air you can move with it. Those are the three things you really care
about. (Again, this is sort of a cheesy generalization but it will do
for the time being.)

The numbers tell you everything you need to know about the driver.

I might be interested in a good
>lower cost (about $45) driver to compare what the extra $$ buys. So
>far one of the big things I see is that it looks like additional $$
>will allow me to get a driver that will sound better in a smaller
>housing.

What does "better" mean? All of these devices have a compromise between
level and response.

If you are currently looking through the Parts Express catalogue, I urge
you also to get the Madisound catalogue which has some higher quality
home audio drivers. For the most part the pro audio drivers are designed
for moving a lot of air not for going to the lowest possible frequency.

I don't have a feeling what group delay, phase and other
>results will do to the sound. I guess getting 2 or 3 samples of
>speakers with different characteristics would help me understand that
>better.

For the most part, phase response is so room-dependant anyway that I wouldn't
worry about it for the time being. I will say that the impulse response of
a sealed box is apt to be better than on any vented box system without
careful engineering. So I would tend to advocate a sealed box as your
first try, even if the cabinet will need to be larger for the same resonant
point.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[eganders]

On Apr 23, 9:23 am, (Scott Dorsey) wrote:
> eganders > wrote:
> >I am just experimenting and want to pick a speaker that will
> >ultimately work well in a subwoofer configuration, but I will be
> >testing it and comparing it to the mathematical results to see how
> >well you can choose the speaker and box dimensions using a design
> >program. I would like the speaker to have a good, non-boomy sound,
> >but I would like to have the response be as low as practical. It
> >looks like 30 to 45 Hz is about the best I can expect with 12 inch
> >speakers in a closed box.
>
> You can get way lower than that if you pick a driver with a low enough
> Fs. The driver resonance is what is limiting the low frequency corner
> in these cases. The problem you run into, of course, is that with a
> 12" driver you can't move a huge amount of air, so a 12" driver that
> is tuned very low won't be very loud. NHT, though, makes a 10" and a
> 12" driver with Fs down below 20 Hz. Some other folks do, but they
> aren't pro audio drivers.
>
> > I can get down to about 25 to 30 Hz with a
> >vented box with the same speaker. These are just my preliminary
> >findings.
>
> Right. With the vented box, you are generating a big box resonance
> superimposed on the driver resonance... so as the frequency drops and
> the response from the driver drops off, the response from the box picks
> up. (Okay, that's a very crude way of thinking about it and Dick Pierce
> wouldn't approve, but it will do for now.)
>
> The Dayton RSS315HO-4 (a 12 inch driver with a Qts=.4,
>
> >Fs=26, Qes=.44, SPS=88 and a Pe=700 for about $130) looks very
> >interesting as a vented driver. If someone would comment on that
> >driver, it would be appreciated.
>
> Roughly, the Fs tells you how low you can make the thing operate, the
> Vas gives you a rough idea of how big a box you will need for a given
> alignment, and the Xmax multiplied by the cone area tells you how much
> air you can move with it. Those are the three things you really care
> about. (Again, this is sort of a cheesy generalization but it will do
> for the time being.)
>
> The numbers tell you everything you need to know about the driver.
>
> I might be interested in a good
>
> >lower cost (about $45) driver to compare what the extra $$ buys. So
> >far one of the big things I see is that it looks like additional $$
> >will allow me to get a driver that will sound better in a smaller
> >housing.
>
> What does "better" mean? All of these devices have a compromise between
> level and response.
>
> If you are currently looking through the Parts Express catalogue, I urge
> you also to get the Madisound catalogue which has some higher quality
> home audio drivers. For the most part the pro audio drivers are designed
> for moving a lot of air not for going to the lowest possible frequency.
>
> I don't have a feeling what group delay, phase and other
>
> >results will do to the sound. I guess getting 2 or 3 samples of
> >speakers with different characteristics would help me understand that
> >better.
>
> For the most part, phase response is so room-dependant anyway that I wouldn't
> worry about it for the time being. I will say that the impulse response of
> a sealed box is apt to be better than on any vented box system without
> careful engineering. So I would tend to advocate a sealed box as your
> first try, even if the cabinet will need to be larger for the same resonant
> point.
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

I really like your cheezy explanations! When I approach a product,
concept or whatnot, a nice cheezy description beats a complex one ANY
day (as long as it is presented as not being fully descriptive)

Along with that, I think real experience also beats just going through
the figures. I am a competent woodworker with a well endowed shop. I
can turn out boxes and jigs like they were on a production line. S-o-
o-o... I would like to get some speaker samples (at least 2 or 3) that
represent strengths in different areas. I don't want any one of them
to be over $100 or $125, but I want to experiment with them to see
what I am looking for from doing the tests empirically.

I also need a plate amplifier that I can ultimately use in my final
design and has features I can use for my experimentation. The Bash
300 from Parts Express looks like it would be a good choice. Any
other suggestions?

[Arny Krueger]

"eganders" > wrote in message
oups.com
> I am interested in building a subwoofer. My questions are:
>
> Are the design program and source of supply good starting
> points (WinISD (http://www.linearteam.dk/) and Parts
> Express (http:// www.partsexpress.com))?

They are OK, but maybe not always the best possible.

> Is a crossover necessary when connected to a receiver
> such as the AVR-687 which (as I understand it) supplies
> frequencies that are designed for the subwoofer frequency
> range?

No external crossover should be necessary. If you buy a subwoofer with a
crossover built-in, set the crossover frquency as high as you can, and rely
on the bass management in the receiver. The receiver is designed to
optimize the main, surround, and subwoofer for each other.

> Maybe one is still necessary in order to keep
> away from the speaker resonant frequency.

Not at all. The speaker should be designed to work with the crossover in
the receiver.

Other reasons?

An external crossover and/or equalizer may be helpful if you are designing
your own subwoofer.

> I am having a hard time finding the speakers in the
> WinISD database. Why is that?

Things are chaning so fast that it is hard for everybody to keep up. There
might be a way to enter manufacturer-supplied Thiel/Small parameters into
WinISD for any speakers you are interested in.

> I would like to keep the parts cost for the speaker, amp
> and crossover to around $300. Is that resonable for a
> speaker that is to be put into a 5400 cu ft room (20 X 18
> X 15)? This is a great room with a high ceiling.

Depends on how agressive you are about loudness and bass. Depends on the
furnishings in the room.

> I would appreciate starting suggestions for speakers and
> amplifiers and/or manufacturers. I have not built a
> speaker in about 30 years and suppliers and designs have
> changed somewhat over that period of time (<:

Good call. My friends who are more into subwoofers than I am recommend the
following sites:

http://www.electronichouse.com/article/all_about_subwoofers/

http://www.nousaine.com

http://ibsubwoofers.proboards51.com/index.cgi

http://www3.sympatico.ca/bonaz/IB-Sub/IBdrv-01.htm

http://pages.sbcglobal.net/ginmtb/woofer_comparison_chart.htm

Specific drivers that have been recommended to me for performance and value:

The value leader is the Mach5Audio http://www.mach5audio.com/ MJ-18

More performance, more money: Ficaraudio 18". See www.ficaraudio.com.