Hi, I can't find any reliable info on how to calculate qtc for vented boxes and the diferent types of alignment derived from it. Any help will be very apreciated.
Qtc does not exist for vented boxes.
All you can really do is specify Vb, Fb and plot response shape.
If you want to optimize transient response, tuning lower is better , and avoiding peaks is always a good idea. 😉
All you can really do is specify Vb, Fb and plot response shape.
If you want to optimize transient response, tuning lower is better , and avoiding peaks is always a good idea. 😉
if you don't mind a begginer like me asking, why is tuning lower better? also, how can you predict the transient response of a wofoer in any box? any links would be helpful as i'm sure the second question would be a very long answer.
Chris,theChris said:
Here's a link to a very good primer on loudspeaker and x-over design. Take a look, it's pretty good for the beginner.
http://www.loudspeakers101.com/
Happy New Year,
Rodd Yamashita
The reason I brought up transient response is because that is one of the things that is affected by Qtc in sealed boxes and may have been what the original poster was after. The other thing that is obvious about changes in Qtc is the response shape.
I gave a vague answer, hoping to draw the original poster out and see why they were wanting to calculate Qtc for a vented box.
Tuning way too low is not a good thing. Tuning lower is like approaching a sealed box, and if you were to tune a woofer whose normal tuning was 40Hz to 20 Hz, you would find that the response would look very much like a sealed box above 20Hz. This kind of defeats the purpose in having a vent in the box in the first place, doesn't it? In fact, if you could tune a vented box to 0Hz, it would respond the same as a closed box.
In short, tuning lower gets you better transient response, but the transient response doesn't get that much better and you should really be more worried about response shape, mainly avoiding peaks....
Calculating transient response is not trivial, if you could understand a short explanation, you would already know how to do it. Unibox, by Kristian Ougaard calculates step response, but interpreting these graphs is not really something for people who don't know how to calculate it...😉 Basically you want to minimize the ripples that come after the step (ringing), but if they are at a low enough frequency they wil not be audible....
I gave a vague answer, hoping to draw the original poster out and see why they were wanting to calculate Qtc for a vented box.
Tuning way too low is not a good thing. Tuning lower is like approaching a sealed box, and if you were to tune a woofer whose normal tuning was 40Hz to 20 Hz, you would find that the response would look very much like a sealed box above 20Hz. This kind of defeats the purpose in having a vent in the box in the first place, doesn't it? In fact, if you could tune a vented box to 0Hz, it would respond the same as a closed box.
In short, tuning lower gets you better transient response, but the transient response doesn't get that much better and you should really be more worried about response shape, mainly avoiding peaks....
Calculating transient response is not trivial, if you could understand a short explanation, you would already know how to do it. Unibox, by Kristian Ougaard calculates step response, but interpreting these graphs is not really something for people who don't know how to calculate it...😉 Basically you want to minimize the ripples that come after the step (ringing), but if they are at a low enough frequency they wil not be audible....
That's why I'm posting this, In a book by Charles-Henry Delaleu I found this formula to calculate the Q of a vented box:
S=(Vb/Vas)/qts^2. In that book the Q for vented boxes is called S.
Also in programs like WinISD and especially Speaker Box there's Q values involved, as well as three types of alignments:
SBB4/BB4, QB3/SQB3, SC4/C4.
Well, transient response is in fact important, but my intention is to get a better understanding of the factors involved in the design of a vented box. I've found to many discrepancies between diferent programs and/or books, what I want is a clearer picture, for instance what those three different alignments mean?, what are its advantages and how are calculated?.
As a general guideline, transient response is better for vented boxes withe speakers with low Qts and boxes that are less than Vas. As the Qts of the speaker gets higher and the box volume gets larger in relation to Vas, the transient response gets worse.
I believe S is related to the Q of the vented box resonance itself - or a function of the response shape of the vent output. This is not really important in designing vented boxes, and in the form you show, it is the undamped case and does not take into account losses in the box due to damping material and losses in the port itself.
This factor is perhaps useful for bandpass boxes in getting the response shape you want.
Thiele redefined vented box design by defining them in terms of electrical filter analogies (the alignments) and devising ways of measuring simple parameters for design purposes.
Alignments are a dead end. I was once curious about them and found out how to calculate them in a series of papers by Richard Small in the JAES. They are mathematical abstractions that were useful before the age of computers because they ensure that you get a certain response shape. You can now plot with a computer and see the response shape dynamically.
In order to apply alignment theory to speakers you have to make a number of simplifying assumptions and the full model cannot be defined in terms of these simple filter analogies.
No matter what any book tells you, the differences between the alignments you specify are very small and probably not worth the distinction. The alignments do serve as useful starting points or reality checks in design
FWIW,
QB3 - Quasi-Butterworth 3rd order (Qts<0.4) has the lowest F3 for a given box size of the standard alignments, but the advantage is small.
SQB3 is an extension for Qts>0.4 that peaks before rolling off and is of limited usefulness unless you want bom.
C4 is Chebychev 4 response - equiripple
Qts>~0.4
SC4 is an extension of the C4 alignment for Qts<0.4 - it offers no real advantage over QB3 type alignments
The QB3 and C4 alignments intersect at the B4 or butterworth alignment.
I do not know how the BB4/SBB4 alignments are calculated, but they have the same dividing line at Qts~0.4
Any other questions?
This factor is perhaps useful for bandpass boxes in getting the response shape you want.
Thiele redefined vented box design by defining them in terms of electrical filter analogies (the alignments) and devising ways of measuring simple parameters for design purposes.
Alignments are a dead end. I was once curious about them and found out how to calculate them in a series of papers by Richard Small in the JAES. They are mathematical abstractions that were useful before the age of computers because they ensure that you get a certain response shape. You can now plot with a computer and see the response shape dynamically.
In order to apply alignment theory to speakers you have to make a number of simplifying assumptions and the full model cannot be defined in terms of these simple filter analogies.
No matter what any book tells you, the differences between the alignments you specify are very small and probably not worth the distinction. The alignments do serve as useful starting points or reality checks in design
FWIW,
QB3 - Quasi-Butterworth 3rd order (Qts<0.4) has the lowest F3 for a given box size of the standard alignments, but the advantage is small.
SQB3 is an extension for Qts>0.4 that peaks before rolling off and is of limited usefulness unless you want bom.
C4 is Chebychev 4 response - equiripple
Qts>~0.4
SC4 is an extension of the C4 alignment for Qts<0.4 - it offers no real advantage over QB3 type alignments
The QB3 and C4 alignments intersect at the B4 or butterworth alignment.
I do not know how the BB4/SBB4 alignments are calculated, but they have the same dividing line at Qts~0.4
Any other questions?
Ron, thank you very much. I'm now more confident with computer generated results and less worried with that Q issue.
- Status
- Not open for further replies.