AVS Forum banner
1 - 20 of 37 Posts

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #1 ·
Say one has three sealed subs using Shiva drivers, with enclosure Q = 0.50, 0.70, and 1.00.




The difference in amplitude response can be seen as well as heard.


Additionally, reports are that the overdamped enclosure (q = 0.50) will sound subjectively "cleaner" but not have as much "punch", while the underdamped enclosure (q=1.00) will have subjectively more "punch", but may sound "a little flabby" because of ringing. Hence, the advice for most folks to start with a Q = 0.70 enclosure for subs that will be used for a combination of movies and music.


QUESTION 1: Is this all correct? Explain.


In the second part, Linkwitz Transform EQ is applied to these three enclosures in order to provide for identical frequency response from each enclosure.




The difference in amplitude response appears to have been eliminated.


QUESTION 2: Despite having equal response in amplitude, these are still three *very* different enclosures. Will the subjective sound differences between the overdamped and underdamped enclosure remain? Explain.



Thanks in advance.

 

·
Registered
Joined
·
3,527 Posts
While you could technically build a 48 cu ft enclosure for a Shiva-x, I wouldn't. Most drivers with a powerful motor can approach a smaller QTC in much smaller enclosures.


While you could eq / boost them to have the same overall response, despite the actual enclosure volume, the amount of power required to drive the proposed driver in a large box to xmax will be less than the small box.


When you cut back on the amount of power flowing through the coil, you lessen inductance change vs power [ Le(i) as per Klippel. ]


Take a look at the impedance traces to verify the change relative to enclosure size.


If one box required 500 watts to get to xmax, and the other requires 1kw, which one will be more linear? ( assuming proper motor design, shorting rings )
 

·
Registered
Joined
·
283 Posts

Quote:
Originally Posted by michael hurd /forum/post/17014281


While you could technically build a 48 cu ft enclosure for a Shiva-x, I wouldn't. Most drivers with a powerful motor can approach a smaller QTC in much smaller enclosures.


While you could eq / boost them to have the same overall response, despite the actual enclosure volume, the amount of power required to drive the proposed driver in a large box to xmax will be less than the small box.


When you cut back on the amount of power flowing through the coil, you lessen inductance change vs power [ Le(i) as per Klippel. ]


Take a look at the impedance traces to verify the change relative to enclosure size.


If one box required 500 watts to get to xmax, and the other requires 1kw, which one will be more linear? ( assuming proper motor design, shorting rings )


Which question were you answering?

Quote:
Will the subjective sound differences between the overdamped and underdamped enclosure remain?
To me He's asking if the typical "ringing" you hear from underdamped enclosures/subs still be there, despite having the response smoothed to look the same as a critically damped one.


I don't know the answer, because I haven't tried it. Despite the fact that yes, it would require more power due to the L/T, I'm not sure it would 'sound' different. Is the ringing due to the peaky response, or does it have to due with the pressure in the smaller box or as Michael seemed to be alluding to, but maybe not, was that the extra current through the coil has something to due with the typical "too small box" sound. Maybe bosso will chime in, as he deals with small sealed enclosures and L/T "all the time".
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #4 ·

Quote:
Originally Posted by mgboy /forum/post/17014685

To me He's asking if the typical "ringing" you hear from underdamped enclosures/subs still be there, despite having the response smoothed to look the same as a critically damped one.

you have keyed in on the essence of the op. that is, do the subjective differences associated with various tunings (q's) remain after equalization?


(also, a general comment: please disregard any comments particular to the shiva. i chose it merely because i needed to choose _a_ driver to illustrate the question graphically.)
 

·
Registered
Joined
·
6,980 Posts

Quote:
Originally Posted by LTD02 /forum/post/17014905


you have keyed in on the essence of the op. that is, do the subjective differences associated with various tunings (q's) remain after equalization?

If the examples are kept within say 1/2 of Xmax and given no amplifier clipping in both listening cases (be cautious in this assumption for real world listening), they will sound more similar than different, where differences could be slightly noticeable at low levels, and more identifiable as levels were pushed. Michael was correct to point to impedance curves, as many forget that when comparing 2 sealed box alignments you are always going to be comparing the resulting changes to BOTH the response and the impedance curve. The sealed box with a Q=0.5 will also have a different modulation/change in response with higher excursion (dropping BL & stiffer Cms) than the box with a Q=1.0 where both the Q and Fb are different. This difference will most certainly be audible. There are other factors as well, where some will be dependent on the strengths and weaknesses of the driver being used.
 

·
Registered
Joined
·
2,260 Posts
I agree with Mark entirely that for practical use, you may not hear a lot of difference unless you are way the heck out. Driver means a lot more.


One more case of "quit typing and go listen to the music!"
 

·
Registered
Joined
·
2,260 Posts
Thinking way back, as we don't see this problem that much any more, a very high Q box when pushed hard the driver will actually follow twice the input frequency. "doubling" we used to call it. This was very common in the pre T/S days when so many manufactures had no clue what they were doing even with the best of intentions.
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #9 ·

Quote:
Originally Posted by Mark Seaton /forum/post/17015007


If the examples are kept within say 1/2 of Xmax and given no amplifier clipping in both listening cases (be cautious in this assumption for real world listening)

yes. let's try to hold as many other things equal as possible in order to isolate the effect.

Quote:
they will sound more similar than different, where differences could be slightly noticeable at low levels, and more identifiable as levels were pushed.

so, there is either a non-linearity in the system or in our hearing? is this a primary or secondary consideration? please explain.

Quote:
Michael was correct to point to impedance curves, as many forget that when comparing 2 sealed box alignments you are always going to be comparing the resulting changes to BOTH the response and the impedance curve.

in smaller enclosures the impedance peak moves up as q increases (of course), they are located at 20hz, 30hz, and 40hz, for the 0.5, 0.707, and 1.0 tunings.


this doesn't seem like a primary consideration.

Quote:
The sealed box with a Q=0.5 will also have a different modulation/change in response with higher excursion (dropping BL & stiffer Cms) than the box with a Q=1.0 where both the Q and Fb are different. This difference will most certainly be audible.

is this a primary consideration, or is it more of a truth, but a secondary consideration?
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #10 ·

Quote:
Originally Posted by tvrgeek /forum/post/17015119


I agree with Mark entirely that for practical use, you may not hear a lot of difference unless you are way the heck out. Driver means a lot more.


One more case of "quit typing and go listen to the music!"

interesting, you are saying that after applying eq, the 0.50 and 1.00 will sound "more or less" the same?


any empirical data (yours, others, anecdotes, or otherwise)?
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #11 ·

Quote:
Originally Posted by tvrgeek /forum/post/17015150


Thinking way back, as we don't see this problem that much any more, a very high Q box when pushed hard the driver will actually follow twice the input frequency. "doubling" we used to call it. This was very common in the pre T/S days when so many manufactures had no clue what they were doing even with the best of intentions.

the fundamental disappeared? or, the second harmonic just shot super high?
 

·
Registered
Joined
·
2,260 Posts
Usually the fundamental did suppress if not fall out completely. We are talking really pushing one. If you backed off, then it would just have been 100% distortion. Like I said, I have not heard anything this bad in years. It was very common in the 70's. After Marty G. (Ohm) read Dick Small's book and introduced the C2, everyone else took notice and got their act together.
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #13 ·

Quote:
Originally Posted by tvrgeek /forum/post/17015286


Usually the fundamental did suppress if not fall out completely. We are talking really pushing one. If you backed off, then it would just have been 100% distortion. Like I said, I have not heard anything this bad in years. It was very common in the 70's. After Marty G. (Ohm) read Dick Small's book and introduced the C2, everyone else took notice and got their act together.

fascinating piece of history...too bad most of this stuff becomes lost with time.
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #15 ·

Quote:
Originally Posted by NEO Dan /forum/post/17016292


The BIG box would have internal standing wave issues that would likely color the sound a bit.

good point. however, if we could, let's try to keep such issues out of the mix. while true, they distract from the essence of the op.


i not championing actually building any of these designs. they are presented merely to illustrate the point.
 

·
Registered
Joined
·
4,575 Posts
A high Q sub offers free output for equal input by transferring the stored energy at the higher resonant frequency into the motion of the driver when the frequency of motion is equal to the resonant frequency.


When you L/T the high Q you get free headroom vs the critically damped box, in this case, from 40Hz to X-over.


The .5 box requires 1dB of boost where the 1.0 box requires 2dB of cut. That's equivalent to having 2 of the 1.0 boxes vs 1 of the .5 boxes for the same input.




For the sake of staying on topic, I'll ignore what's happening below the knee and look only at 40Hz through X-over. The question is; do you hear a difference between the .5 box L/T'd to .7 and the 1.0 box L/T'd to .7.


No.


Specifically, as Seaton points out, at moderate levels relative to the Pe of the system, there is no audible difference in sonic signature from the 2 systems because the differences in the non-linear phenomena due to headroom and the non-linearity of the stored energy transfer are too small.


At higher levels, the smaller box has more headroom and will exhibit a proportional drop in THD to the L/T curve.


Will that be audible?


Maybe.


Depends on the rest of the system, calibration, in-room response and how that in-room response is arrived at.


In perfect conditions under critical listening... Yes.


Bottom line; IMHO, the answer to the question lies in the balance of the transfer of store energy. As I understand it, the balance at the resonant frequency is perfect, but, as you move above and below that frequency the balance is not perfect, as is the case in any resonant system.


I believe that the differences in balance are not great enough to be audible when the advantages of free headroom are weighed in, in the case of a sealed sub. I also believe that stuffing the box helps to equalize the imbalances, however small the percentage may be.


Exaggeration for illustrations sake is not a bad thing, but in the real world, no one would consider the differences in a 1 cube box vs a 50 cube box as a reasonable comparison (except maybe Ilkka). The real world comparison will be more like .8 vs .7 vs .6, where the differences are smaller vs the weight of advantages elsewhere.


Building the 3 boxes and graphing and listening to the results is a simple enough way to get your answer, in which case you'll probably get an equal split in subjective preference, except under the most extreme presentation of the system.


Bosso
 

·
Registered
Joined
·
123 Posts
To put it another way, If the 3 transfer functions are eq'd the same, they will have identical impulse responses and will sound the same Any audible differences with variations in power applied would suggest variations in the transfer functions at those levels.


C
 

·
Registered
Joined
·
5,551 Posts
What about increased ASD at high drive levels in the smallest box? The high Q box having a bit of extra gain at the resonance has been mentioned as a free lunch, but what about the extra EQ and thus power that will be required compared to the bigger boxes below the knee? Is this not the same thing, or are we not counting that because of the excursion demands down low.



Illka's LMS test showed worse distortion for the 75L (2.65ft) smaller box everywhere below about 50hz. I guess it could be argued that the box is too small but it would be a Q of about 0.64. Perhaps these were the effects of ASD or moderate amplifier clipping.

http://www.hometheatershack.com/foru...cial-test.html


I still agree that you will be very hard pressed to tell the difference by listening to boxes of 0.6, 0.7, and 0.8 qtc. Perhaps if they are pushed to the limit the differences will become more apparent through the different distortion profiles and the different levels or power and compression.
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #19 ·

Quote:
Originally Posted by bossobass /forum/post/17019783


For the sake of staying on topic, I'll ignore what's happening below the knee and look only at 40Hz through X-over. The question is; do you hear a difference between the .5 box L/T'd to .7 and the 1.0 box L/T'd to .7.


No.

this is a surprising conculsion.


thanks for the extensive reply bosso. very much appreciated.
 

·
Registered
Joined
·
21,345 Posts
Discussion Starter · #20 ·

Quote:
Originally Posted by cc00541 /forum/post/17020114


Any audible differences with variations in power applied would suggest variations in the transfer functions at those levels.

doesn't the resonant behavior of the high q system show up in the time domain though? in other words, frequency response alone doesn't tell the whole story?
 
1 - 20 of 37 Posts
Top