Parts Express offers a "Woofer Selection Guide" that states a recommended box internal volume Vb for a closed-box alignment of their woofers for sale. Among 8 inchers, some are recommended to be acoustically suspended (alpha or Vas/ Vb equal to or greater than 3) while others are not.
They say that the alignments of their selection guide are derived from the (software program) BassBox. Besides being too cheap right now to buy BassBox, I think that there must be a simple way to look at the specs of a woofer and be able to decide if power handling will be seriously compromised if a given woofer isn't acoustically suspended.
For two woofers of the same size, certainly relative VAS should say something about the matter. The woofer with the smaller VAS has less compliance of its spider and surround and thus would be the more likely candidate to be able to live without acoustic suspension. This generality however really doesn't get me anywhere in making a decision.
What is needed, probably, is an equation taking VAS, Xmax and sensitivity into account. Probably those who created BassBox know this equation.
Does anybody know of this equation (if it exists) or perhaps can think of a few simple calculations that might let me look at a spec sheet for a woofer and be able to tell which side of the alpha = 3 divide the woofer stands on?
They say that the alignments of their selection guide are derived from the (software program) BassBox. Besides being too cheap right now to buy BassBox, I think that there must be a simple way to look at the specs of a woofer and be able to decide if power handling will be seriously compromised if a given woofer isn't acoustically suspended.
For two woofers of the same size, certainly relative VAS should say something about the matter. The woofer with the smaller VAS has less compliance of its spider and surround and thus would be the more likely candidate to be able to live without acoustic suspension. This generality however really doesn't get me anywhere in making a decision.
What is needed, probably, is an equation taking VAS, Xmax and sensitivity into account. Probably those who created BassBox know this equation.
Does anybody know of this equation (if it exists) or perhaps can think of a few simple calculations that might let me look at a spec sheet for a woofer and be able to tell which side of the alpha = 3 divide the woofer stands on?
A good paramater to look at is Qts. If it is below roughly .4, your resonant frequency in box goes way up. You get very little bass response in a sealed enclosure.
www.carstereo.com/help/Articles.cfm?id=30
Here is a link to a box calculator, you can punch in some numbers from different drivers and you should get the idea.
I believe there is a simple formula for determining if a driver is better in a vented or sealed box, but I don't remember it offhand.
JJ
www.carstereo.com/help/Articles.cfm?id=30
Here is a link to a box calculator, you can punch in some numbers from different drivers and you should get the idea.
I believe there is a simple formula for determining if a driver is better in a vented or sealed box, but I don't remember it offhand.
JJ
MCPete said:
EBP (Efficiency Bandwidth Product)
EBP = Fs/Qes
However.. Qts, Mms and/or sensitivity, and maybe a quick glance at Fs usually tells me enough.
Also, I think after modeling so many hundreds of woofers over time, you start to know what to expect. Well, I guess it doesn't take THAT long. But after a while, you'll see what I mean if you continue to model woofers constantly. I often find myself guessing in my head real quick at the F3 and box size for .707 Qtc while entering parameters. I'm usually pretty close, and often spot on. I definitely couldn't do this just a couple years ago.
still puzzled
Thanks for all of the replies so far. I guess so far nobody reading my post knows exactly how to go about solving my problem.
Modelling software programs certainly are convenient. I wouldn't be across the board opposed to using one. But I think there is a danger of letting the computer do all of the work, ending up with a good speaker system, and being at a loss to explain why the design is good. So as much as possible, I like to get to know the math behind the solutions.
EBP or efficiency bandwidth product I believe is used in deciding the greater suitability of a woofer for a closed-box or vented alignment.
Qts by itself doesn't indicate what the box resonance frequency or Fc will be. If free air resonance frequency or Fs is sufficiently low, then even if alpha must be 3 or greater, then Fc won't be excessively high. Also, Fc = sqrt(alpha +1)Fs, so rising Fc is at a slower rate than alpha.
Thanks for all of the replies so far. I guess so far nobody reading my post knows exactly how to go about solving my problem.
Modelling software programs certainly are convenient. I wouldn't be across the board opposed to using one. But I think there is a danger of letting the computer do all of the work, ending up with a good speaker system, and being at a loss to explain why the design is good. So as much as possible, I like to get to know the math behind the solutions.
EBP or efficiency bandwidth product I believe is used in deciding the greater suitability of a woofer for a closed-box or vented alignment.
Qts by itself doesn't indicate what the box resonance frequency or Fc will be. If free air resonance frequency or Fs is sufficiently low, then even if alpha must be 3 or greater, then Fc won't be excessively high. Also, Fc = sqrt(alpha +1)Fs, so rising Fc is at a slower rate than alpha.
Hi,
Vas, Fs and Qts are all inextricably linked, and given these parameters
the box size / bass extension pretty much depends on the wanted
efficiency, the higher this is, the bigger box or less bass extension.
(cone mass for driver size is a good efficiency pointer)
Often ignored (completely) is that Vb<Vas linearises the suspension
at low frequencies, Vas non-linearity is a major cause of distortion
below ~ 100 Hz. You cannot guarantee high quality bass with
just box modelling from small signal parameters.
/sreten.
Vas, Fs and Qts are all inextricably linked, and given these parameters
the box size / bass extension pretty much depends on the wanted
efficiency, the higher this is, the bigger box or less bass extension.
(cone mass for driver size is a good efficiency pointer)
Often ignored (completely) is that Vb<Vas linearises the suspension
at low frequencies, Vas non-linearity is a major cause of distortion
below ~ 100 Hz. You cannot guarantee high quality bass with
just box modelling from small signal parameters.
/sreten.
MCPete said:
Fc/Fs = Qtc/Qts = sqrt (Vas/Vb+1)
What this tells you is that For Vas/Vb = 3, Fc/Fs = Qtc/Qts = 2
So if you want to design a speaker with Qtc= 0.7 and acoustic suspension, you need a woofer with Qts less than or equal to 0.35
Qtc is the response shape factor.
When designing a sealed woofer, you specify wither Qtc or Vb and the rest falls into place. While a formula could be derived for calculating excursion vs power, and is quite simple actually, there is no need for it.
At DC:
i = V / Re - here V is the voltage input (power = V^2/Re)
F = Bl * i (current)
F = k*X - you can calculate k from Cms. ( k = 1/Cms - watch your units!) see the wikipedia Thiele small article....
Re: still puzzled
It's nice to know and understand the math behind it but it's not really necessary - the modeling software will play with the numbers. All you need to do is learn how to use it properly.
You don't need to understand how your operating system works on your computer to use it; you don't need to understand how your car converts the gasoline it guzzles into forward motion to drive to the beer store.
For most drivers with a higher Qts (suitable for closed box), the Vb will be less than VAS. This is normal...
Am I seeing this right?
Seems to me that acoustic suspension can be tried with any driver with a large VAS. Large VAS equals a loose suspension and would greatly benefit from the air spring support of the smaller box.
F3 will suffer a bit, but this would be a small price to pay for a smaller box and less bass distortion.
MCPete said:
It's nice to know and understand the math behind it but it's not really necessary - the modeling software will play with the numbers. All you need to do is learn how to use it properly.
You don't need to understand how your operating system works on your computer to use it; you don't need to understand how your car converts the gasoline it guzzles into forward motion to drive to the beer store.
sreten said:
For most drivers with a higher Qts (suitable for closed box), the Vb will be less than VAS. This is normal...
Am I seeing this right?
Seems to me that acoustic suspension can be tried with any driver with a large VAS. Large VAS equals a loose suspension and would greatly benefit from the air spring support of the smaller box.
F3 will suffer a bit, but this would be a small price to pay for a smaller box and less bass distortion.
I can't communicate?!
I like the discussion so far, but it's off the mark from what I am essentially driving at. As I strayed myself in my second post I'm not going to complain.
Basically what I'm saying is that it is very likely equation(s) are behind BassBox specifying one driver to be acoustically suspended and another one not, and I would like to see them. Also because I am cheap, I want to be able to know whether or not a woofer must be acoustically suspended by looking at the spec sheet for the woofer and not having to rely on a computer program. I think that probably most of us with more interest than usual in speaker systems want to understand what makes them "tick" (this is probably a really outdated expression nowadays). The computer programs can be useful to check the correctness of our understanding, but I think that they can be counterproductive if relied on too much.
You know very possibly also there are some criteria for specifying alpha one way or the other for a particular woofer that aren't etched in stone. If you don't know what those criteria are, then you can't be sure that you can agree with the final assessment.
I like the discussion so far, but it's off the mark from what I am essentially driving at. As I strayed myself in my second post I'm not going to complain.
Basically what I'm saying is that it is very likely equation(s) are behind BassBox specifying one driver to be acoustically suspended and another one not, and I would like to see them. Also because I am cheap, I want to be able to know whether or not a woofer must be acoustically suspended by looking at the spec sheet for the woofer and not having to rely on a computer program. I think that probably most of us with more interest than usual in speaker systems want to understand what makes them "tick" (this is probably a really outdated expression nowadays). The computer programs can be useful to check the correctness of our understanding, but I think that they can be counterproductive if relied on too much.
You know very possibly also there are some criteria for specifying alpha one way or the other for a particular woofer that aren't etched in stone. If you don't know what those criteria are, then you can't be sure that you can agree with the final assessment.
Well, I just happen to have a driver that is well suited to acoustic suspension. It's a Max Pentivent 12" woofer - PV-1230.
Looking at the specs, I see low Qms and high VAS. Qts indicates it's best alignment is sealed.
Input the parameters into Unibox, and I get the best result for this driver in a sealed box. 95 litres for Vb, alpha at 2.9, Qtc .706
Looking at the specs, I see low Qms and high VAS. Qts indicates it's best alignment is sealed.
Input the parameters into Unibox, and I get the best result for this driver in a sealed box. 95 litres for Vb, alpha at 2.9, Qtc .706
Attachments
Re: I can't communicate?!
No such formula exists. Bassbox is not some be all and end all of loudspeaker design. I explained how design of sealed boxes is managed in very simple terms. Either Qtc or Vb is specified. There is no magic spec that says: "this driver should have a Qtc of 1 because it needs to be acoustically suspended".
It is rather ironic that you decry the use of software, yet the whole aim of this thread is to reverse engineer a piece of software. 
MCPete said:
No such formula exists. Bassbox is not some be all and end all of loudspeaker design. I explained how design of sealed boxes is managed in very simple terms. Either Qtc or Vb is specified. There is no magic spec that says: "this driver should have a Qtc of 1 because it needs to be acoustically suspended".
It is rather ironic that you decry the use of software, yet the whole aim of this thread is to reverse engineer a piece of software.
Attached is a bit of data analysis for you from parts express' speaker selection wizard. It appears on a quick look that the software, which you so respect and just have to understand, with few exceptions selects an "optimal" Qtc of 0.74.
Note that speakers with alpha greater than 3 have a Qtc/Qts of less than 0.37, (half of 0.74), just as predicted by my simple math above.
Fascinating...
I bet if you asked David Harris nicely, he would tell you the secret.
Note that speakers with alpha greater than 3 have a Qtc/Qts of less than 0.37, (half of 0.74), just as predicted by my simple math above.
Fascinating...
I bet if you asked David Harris nicely, he would tell you the secret.
Replying to Ron E,
Here is my idea. If nothing else, this will hopefully make clear what my concern is.
First I would modify your equation for excursion, X, so that it applies to a closed-box system,
X = Bl*i *Cat
where Cat is total acoustic compliance of driver and enclosure.
Applying that equation to a particular driver with a specified max. power handling, determine Cat such that X doesn't exceed Xmax.
Knowing Cat according to the above and also knowing Cas or the acoustic compliance of the driver suspension makes it possible to solve for Cab or the acoustic compliance of the air in the box. That is,
Cat = (Cab*Cas)/(Cab + Cas)
Multiplying Cab by density of air * the speed of sound squared would define Cab as a volume of air (not positive about this). So now we have a recommended Vb from which a recommended alpha can be derived.
By the way, thanks for your posts past and present. They have been very helpful to me.
Pete
Here is my idea. If nothing else, this will hopefully make clear what my concern is.
First I would modify your equation for excursion, X, so that it applies to a closed-box system,
X = Bl*i *Cat
where Cat is total acoustic compliance of driver and enclosure.
Applying that equation to a particular driver with a specified max. power handling, determine Cat such that X doesn't exceed Xmax.
Knowing Cat according to the above and also knowing Cas or the acoustic compliance of the driver suspension makes it possible to solve for Cab or the acoustic compliance of the air in the box. That is,
Cat = (Cab*Cas)/(Cab + Cas)
Multiplying Cab by density of air * the speed of sound squared would define Cab as a volume of air (not positive about this). So now we have a recommended Vb from which a recommended alpha can be derived.
By the way, thanks for your posts past and present. They have been very helpful to me.
Pete
Well, now that we have danced around the subject of bassbox and some notion of an acoustic suspension "litmus test", if all you want are excursion vs power equations, you can develop them from the information at www.diysubwoofers.org Brian Steele at diysubwoofers has a spreadsheet that codes up the equations for you - I believe it is locked with a null password. Or you can buy Benson's Theory and Design of Loudspeaker Enclosures and wade through all the math yourself.
One of the true benefits of acoustic suspension is that the spring of the box linearizes (somewhat) the spring of the driver. This has nothing to do with power handling as such. The difficulty is in knowing the force vs excursion curve of the driver.
Attached is a spreadsheet investigation I did a _long_ time ago. With it you may decide for yourself if the linearizing capacity of the air spring is significant at alpha = 3 or not.
One of the true benefits of acoustic suspension is that the spring of the box linearizes (somewhat) the spring of the driver. This has nothing to do with power handling as such. The difficulty is in knowing the force vs excursion curve of the driver.
Attached is a spreadsheet investigation I did a _long_ time ago. With it you may decide for yourself if the linearizing capacity of the air spring is significant at alpha = 3 or not.
Re: Re: still puzzled
Hi,
Large Vas does not imply non-linear Vas, and probably the problem
is worse for small Vas, if it is non-linear it is more difficult to correct.
Many drivers are now designed for vented loading. For sealed you
can reduce Fs and increase Vas with little effect on response, but
for vented low Fs / high Vas means poor subsonic power handling,
you want suspension stiffness to be as high as you can get away
with. This difference IMO sorts the sealed from the vented. You
can vent some drivers suitable for sealed but Fvent is very low.
Just because a driver can be vented does not mean it should be.
IMO the majority of HT vented small boxes would be better off
being sealed and built to approximately half internal volume.
(And then properly filtered of course .......)
As for looking at the specification sheet without analysing it ... ?
It does not contain any magic numbers meaningful in isolation.
/sreten.
MJL21193 said:
Hi,
Large Vas does not imply non-linear Vas, and probably the problem
is worse for small Vas, if it is non-linear it is more difficult to correct.
Many drivers are now designed for vented loading. For sealed you
can reduce Fs and increase Vas with little effect on response, but
for vented low Fs / high Vas means poor subsonic power handling,
you want suspension stiffness to be as high as you can get away
with. This difference IMO sorts the sealed from the vented. You
can vent some drivers suitable for sealed but Fvent is very low.
Just because a driver can be vented does not mean it should be.
IMO the majority of HT vented small boxes would be better off
being sealed and built to approximately half internal volume.
(And then properly filtered of course .......)
As for looking at the specification sheet without analysing it ... ?
It does not contain any magic numbers meaningful in isolation.
/sreten.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.