Subwoofer FAQ (just the begining stages)

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

glen

Lifer
Apr 28, 2000
15,995
1
81
Originally posted by: thomsbrain
of course CD's are only encoded and decoded from 20 Hz up, so anything below that is a mute point (pardon the pun), unless you're playing old records which were recorded with all analog equipment (which basically means nothing since 1980 or so). but having speakers that extent below 20 is nice because it means you probably have less drop-off on the frequencies it will actually be fed.


This is actually funny because I have heard this from salesmen at different stereo shops. I suspect it is a "talking" point they are taught from the marketing reps because the wording is so simillar. It sounds resonable. How many people can challenge such a statement? - Only people who know the Red Book standard for CD's, or folks who have the test equiptment to do the analysis.
 

MikeMike

Lifer
Feb 6, 2000
45,885
66
91
here is my arguement for why you cant say 8's are tighter than 15's

they make different motor structures for a reason!!!! just like in a car, a v8 was made to power a bigger car, while an i4 was for a small compact car.

thus, they made a stronger, quick, motor for larger subs.

MIKE
 

Shockwave

Banned
Sep 16, 2000
9,059
0
0
Originally posted by: glen
Originally posted by: thomsbrain
of course CD's are only encoded and decoded from 20 Hz up, so anything below that is a mute point (pardon the pun), unless you're playing old records which were recorded with all analog equipment (which basically means nothing since 1980 or so). but having speakers that extent below 20 is nice because it means you probably have less drop-off on the frequencies it will actually be fed.

no.

This guy simply analysed the data and listed its frequency and how loud it is recorded.
Top 30 Bass Cd's
frequncy bits from max recordable level track number time

So, for the first in the list, it is ranked #1, the frequency is 5hz, it is recorded at 13.7 db from the max recordable level on a cd it is track number 1 and occurs between 3:30 and 8:10 name of CD etc...


1 5hz -13.7 1 3:30-8:10 Battle Music of Beethoven and Liszt Beethoven Telarc CD-80079
2 5hz -16.1 4 5:14-5:15 Big Notes Flim & The BB's DMP CD-454
3 7hz -14.5 1 12:30-15:10 1812 Overture Tchaikovsky Telarc CD-80041
4 7hz -22.1 35 ALL Sonic Booms 2 Mobile Fidelity Bainbridge BCD 6285
5 8hz -20.1 10 4:15-4:16 Toxic Bass Toxic Bass I.B.P. IBP-0008-2
6 9hz -14.8 8 0:20-0:40 Sonic Booms 3 Mobile Fidelity Bainbridge
7 10hz -11.8 20 3:10-3:20 Great Fantasy/Adventure Album Erich Kunzel Telarc CD-80342
8 10hz -15.5 21 ALL Great Fantasy/Adventure Album Erich Kunzel Telarc CD-80342
9 10hz -19.5 1 0:20-0:30 Food For Woofers Volume #3 B&W KOTF I.R.S. FFW 003 CD
10 10hz -20.5 5 2:20-2:40 Fantastic Journey Erich Kunzel Telarc CD-80231
11 12hz -14.9 12 3:10-3:20 Illegal Bass Bass Outlaws Newtown NTN 2210
12 13hz -17.3 1 ALL Time Warp Erich Kunzel Telarc CD-80106
13 16hz -11.7 16 3:00-3:40 Romantic Organ Music Volume 2 Peter Hurford London 421 296-2
14 16hz -11.9 18 0:12-END It Came From Outer Bass 2 Techmaster P.E.B. Newtown NTN 2211
15 16hz -12.5 2 3:20-END Quad Maximus Bass Mekanik Pandisc PDD-8848
16 16hz -14.5 3 ALL Bass Box Seventh Order Hip Rock DIDX 018599
17 16hz -16.7 10 0:38-5:06 Tyranny For You Front 242 Epic/Sony EK 46998
18 16hz -21.8 1 0:48-0:50 Joshua Judges Ruth Lyle Lovett Curb / MCA MCAD-10475
19 17hz -8.6 1 0:10-0:20 Chiller Erich Kunzel Telarc CD-80189
20 18hz -7.1 1 3:50-3:55 Illegal Bass Bass Outlaws Newtown NTN 2210
21 18hz -14.5 3 1:30-1:43 Bass Ecstacy Bass Erotica Neurodisk NRD-31005
22 19hz -6.7 7 0:00-0:10 Drivin Bass Bass Connection Neurodisk CD-31004
23 19hz -9.6 6 1:10-END More Bass Boom Bottom Power Supply Hip Rock DIDX 022094
24 19hz -9.9 2 ALL Cyberbass Virtual Reality Bass Syndicate D M Records
25 20hz -4.5 5 1:10-1:20 Bass Society Bass Society Majammy E-SA-1010
26 20hz -7.9 1 ALL It Came From Outer Bass 2 Techmaster P.E.B. Newtown NTN 2211
27 20hz -8.1 12 0:00-0:10 Bass Is Loaded Bass Hit Neurodisk CD-31002
28 20hz -10.5 9 ALL Bass Overload Bass Alliance D M Records DMR-41269
29 22hz -4.3 11 ALL Drivin Bass Bass Conection Neurodisk CD-31004
30 22hz -7.4 12 0:40-END Bass Overload Bass Alliance D M Records DMR-41269


Frequency: lowest significant frequency with potential to be percieved.
DBfs: decibels below maximum digital record level.
Time: point(s) during track to find lowest frequency (ALL=various points in track).
CD's were played on a SONY CDP-610es (confirmed flat to below 2hz) and Measured on a Tektronix 2630 FFT analyser.

:heart: Glen
:heart: the data
:heart: bass
 

Triumph

Lifer
Oct 9, 1999
15,031
14
81
Originally posted by: Viper GTS
Originally posted by: Triumph
Ci = Ca (eq 3) or i :: a (eq 4)

Maybe my 4 years of engineering school were wasted, but that's just bad math. It should be C1i = C2a, and thus the C's do not cancel out. C2 = m, and since it doesn't divide out, m is still in there. So let's go back to their previous equation:

BLi = ma (eq 2)

Assume you have two speakers with equivalent magnetic field and voice coil lenght. BL = constant. Lets assume BL = 1 to make it easy. Now if they are two different speaker sizes, 12" and 15", the masses will be different. So we have:

12" driver: i = m1*a1
15" driver: i = m2*a2
m2 > m1

Input the same current i, then

m1*a1 = m2*a2 or
a1 = (m2/m1)*a2

If m2 > m1, then a1 > a2, hence the 12" driver has a higher acceleration than a1.

The only inhibitions I have about this is because you posted a link, which is from a seemingly reputable source, and it's possible that I'm wrong because I'm not specifically a speaker expert. But I just don't see how you can "eliminate" the mass, B, and L, simply because they are time invariant. C != C, so they don't cancel out! I'd like to see another engineer's analysis of this conundrum.

Interesting problem, nonetheless. I would love to see what is wrong with my argument, but I'll be away from AT for a few days. That'll give you guys plenty of time to write up a response.

One major flaw I see in your reasoning that I see:

You assumed current would be constant, which it obviously would not be. The question is not which can respond faster at a given current. Two different drivers playing the same frequency at the same pressure will use two different currents.

Viper GTS

Well that only negates the 2nd part of my argument, if it is correct. The first part still remains: BLi = ma cannot be reduced down to simply i = a, as the linked article contends. The C's are different and can't be divided out like that. Therefore, acceleration of the driver is still inversely proportional to the mass.
 

glen

Lifer
Apr 28, 2000
15,995
1
81
The "acceleration" of the driver is a moot point and can be derived from the frequency the driver is playing.
If it accelerates "slower" then the frequency is lower, if it gets faster, the frequency gets faster.

This talk of motors and physics and magnets is as preposterous as arguing which car is faster, the small one with this engine, or the other one with that engine, COMPLETLY missing the point that they just drag raced so you have the 0-60 times imperically, and the theoretical 0-60 pointless to argue about.
 

Howard

Lifer
Oct 14, 1999
47,982
11
81
Originally posted by: Triumph
Originally posted by: Viper GTS
Originally posted by: Triumph
Ci = Ca (eq 3) or i :: a (eq 4)

Maybe my 4 years of engineering school were wasted, but that's just bad math. It should be C1i = C2a, and thus the C's do not cancel out. C2 = m, and since it doesn't divide out, m is still in there. So let's go back to their previous equation:

BLi = ma (eq 2)

Assume you have two speakers with equivalent magnetic field and voice coil lenght. BL = constant. Lets assume BL = 1 to make it easy. Now if they are two different speaker sizes, 12" and 15", the masses will be different. So we have:

12" driver: i = m1*a1
15" driver: i = m2*a2
m2 > m1

Input the same current i, then

m1*a1 = m2*a2 or
a1 = (m2/m1)*a2

If m2 > m1, then a1 > a2, hence the 12" driver has a higher acceleration than a1.

The only inhibitions I have about this is because you posted a link, which is from a seemingly reputable source, and it's possible that I'm wrong because I'm not specifically a speaker expert. But I just don't see how you can "eliminate" the mass, B, and L, simply because they are time invariant. C != C, so they don't cancel out! I'd like to see another engineer's analysis of this conundrum.

Interesting problem, nonetheless. I would love to see what is wrong with my argument, but I'll be away from AT for a few days. That'll give you guys plenty of time to write up a response.

One major flaw I see in your reasoning that I see:

You assumed current would be constant, which it obviously would not be. The question is not which can respond faster at a given current. Two different drivers playing the same frequency at the same pressure will use two different currents.

Viper GTS

Well that only negates the 2nd part of my argument, if it is correct. The first part still remains: BLi = ma cannot be reduced down to simply i = a, as the linked article contends. The C's are different and can't be divided out like that. Therefore, acceleration of the driver is still inversely proportional to the mass.
It did not say equal to, merely proportional to.
 

Howard

Lifer
Oct 14, 1999
47,982
11
81
Yep. Frankly, I'm still not convinced mass is meaningless.

If I can see a proof that 2 cones of different masses accelerate at the same rate with the same moving force, then I'll be convinced.
 

NightCrawler

Diamond Member
Oct 15, 2003
3,179
0
0
Originally posted by: glen
Speakers are advertized as 20 to 20khz
I would much rather have speakers do 5hz to 12,000hz


Not getting anything above 12,000 would sound pretty dreadful. I put a filter in at 12,000 and the music became lifeless. Also 5 hertz is completely not neccessary since music musical instrutments don't play that low. I believe the lowest is 16 hertz and most rock and rap music won't go much below 30 hertz.


 

Apex

Diamond Member
Oct 11, 1999
6,511
1
71
www.gotapex.com
Originally posted by: NightCrawler
Originally posted by: glen
Speakers are advertized as 20 to 20khz
I would much rather have speakers do 5hz to 12,000hz


Not getting anything above 12,000 would sound pretty dreadful. I put a filter in at 12,000 and the music became lifeless. Also 5 hertz is completely not neccessary since music musical instrutments don't play that low. I believe the lowest is 16 hertz and most rock and rap music won't go much below 30 hertz.

Not everyone avoids home theater completely. :) Speakers are used for more than just musical instruments.
 

NightCrawler

Diamond Member
Oct 15, 2003
3,179
0
0
Originally posted by: Apex
Originally posted by: NightCrawler
Originally posted by: glen
Speakers are advertized as 20 to 20khz
I would much rather have speakers do 5hz to 12,000hz


Not getting anything above 12,000 would sound pretty dreadful. I put a filter in at 12,000 and the music became lifeless. Also 5 hertz is completely not neccessary since music musical instrutments don't play that low. I believe the lowest is 16 hertz and most rock and rap music won't go much below 30 hertz.

Not everyone avoids home theater completely. :) Speakers are used for more than just musical instruments.

Yeah but to get loud 5 hertz bass that you could actually feel would require a really outrageous subwoofer setup. Most DVD's get filter below 30 hertz as to not blowup cheaper subwoofers. Oddly enough THX setups get bass below 35 hertz cut so it doesn't become booming cause of room gain.

 

Woodchuck2000

Golden Member
Jan 20, 2002
1,632
1
0
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?
 

Shockwave

Banned
Sep 16, 2000
9,059
0
0
Originally posted by: Woodchuck2000
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?

Cone breakup, beaming etc etc.
 

Woodchuck2000

Golden Member
Jan 20, 2002
1,632
1
0
Originally posted by: Shockwave
Originally posted by: Woodchuck2000
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?

Cone breakup, beaming etc etc.
Um, Thanks :confused:
 

NightCrawler

Diamond Member
Oct 15, 2003
3,179
0
0
Originally posted by: Woodchuck2000
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?


1. Sound Quality: a single 15 would strain trying to reproduce the entire spectrum.
2. Cost: Actually easier and cheaper to build a 2-way or 3-way system then one perfect 15".


 

Woodchuck2000

Golden Member
Jan 20, 2002
1,632
1
0
Yeah, but <why> would a 15-inch driver strain at low frequencies and <why> does it need to be that big in the first place?
 

Viperoni

Lifer
Jan 4, 2000
11,084
1
71
A 15" wouldn't strain at low frequencies unless it's being driven really hard, of which many factors are involved, none-the-least being the box and power input.
A bigger driver tends to move more air in a single sweep, creating more bass :)
 

Howard

Lifer
Oct 14, 1999
47,982
11
81
Originally posted by: Woodchuck2000
Originally posted by: Shockwave
Originally posted by: Woodchuck2000
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?

Cone breakup, beaming etc etc.
Um, Thanks :confused:
Well, you need lots of displacement to create loud bass. stroke * surface area of the diaphragm (cone) = displacement.

Now, if you have a large surface area and a relatively small stroke, you can still create bass, but your moving mass will be too high to create the very high frequencies (i.e. moving the cone back and forth 20 000 times a second). Then you can use a small surface area and REALLY long stroke to do the bass, but it would be tremendously difficult to engineer a motor that can make the cone that far. This also necessitates a very loose suspension, and I'm not too sure about this, but SQ would generally suffer. There are wideband drivers out there, but they don't do the lower frequencies all too loudly.

This is all assuming the cone doesn't start to flex at higher frequencies... and if it does, you get a hell of a lot of distortion.
 

Shockwave

Banned
Sep 16, 2000
9,059
0
0
Originally posted by: Howard
Originally posted by: Woodchuck2000
Originally posted by: Shockwave
Originally posted by: Woodchuck2000
Can someone explain mathematically why you can't just use a single 15" driver to reproduce the entire frequency range? Why do you need to switch to smaller drivers at higher frequencies? Could a 20mm tweeter reproduce frequencies down to 100Hz?

Cone breakup, beaming etc etc.
Um, Thanks :confused:
Well, you need lots of displacement to create loud bass. stroke * surface area of the diaphragm (cone) = displacement.

Now, if you have a large surface area and a relatively small stroke, you can still create bass, but your moving mass will be too high to create the very high frequencies (i.e. moving the cone back and forth 20 000 times a second). Then you can use a small surface area and REALLY long stroke to do the bass, but it would be tremendously difficult to engineer a motor that can make the cone that far. This also necessitates a very loose suspension, and I'm not too sure about this, but SQ would generally suffer. There are wideband drivers out there, but they don't do the lower frequencies all too loudly.

This is all assuming the cone doesn't start to flex at higher frequencies... and if it does, you get a hell of a lot of distortion.


Some large diameter driver can run WELL beyond a 1000 hertz. So, its close to doablke.... 1 guy I had read about was working (planning) on trying a 10 or 12" driver and a tweeter. Thats it. Dunno if it really ever went anywhere though.
 

Woodchuck2000

Golden Member
Jan 20, 2002
1,632
1
0
So if you take a particular driver, put a sine wave in at constant power, vary the frequency and record the output SPL, you'd get a nice bell curve? There'd be a peak effeciency somewhere in the middle tailing off either side?
 

MikeDub83

Member
Apr 6, 2003
96
0
0
Originally posted by: nan0bug
Dont they have a car audio forum somewhere that you guys can all go jerk each other off in?

This is the funniest post I have ever read on Anandtech.
 

JonnyBlaze

Diamond Member
May 24, 2001
3,114
1
0
i owned a cerwin-vega stroker 15" speaker. it cost around $750. i had it in a box i built that was around 5 cubic feet with a huge rectangular port. i thru in this bass mekanic cd that had a track from 1hz to 100 or something like that. it was amazing too see everything in my car shake before you could even hear anything.

I agree tho. 4 15's in a sealed box is what i would prefer tho. that would do the trick.


JBLaze
 

sharkeeper

Lifer
Jan 13, 2001
10,886
2
0
i owned a cerwin-vega stroker 15" speaker. it cost around $750. i had it in a box i built that was around 5 cubic feet with a huge rectangular port. i thru in this bass mekanic cd that had a track from 1hz to 100 or something like that. it was amazing too see everything in my car shake before you could even hear anything.

.01Hz to 20Hz 5 SEC Sweep

24bit/96KHz DAC required.

Cheers!