Qb and Vb in acoustic suspension systems

[Pjotr]

Lower Qts by a negative series resistance ...

Yes, that is also an option. But this is an electronic option to lower Qe. Did some experiments with it and yes that works. But it has a big drawback. The voice coil will heat up, rising Qe again (because Re rises). Since you are working now with the difference of resistances, this difference can change severe and is quite audible. A proper tuned system becomes boomy after playing some time at high level because of the heating of the voice coil.

Using a LW transform works much better then or otherwise the way to go can be MFB. But that is not an easy way.

Cheers

[coolkhoa]

Originally posted by sreten


Qb is always greater than Qts.

Your other suppositions are entirely incorrect,
because for a start your equation is wrong :

Qb/Qts = root (Vas/Vb+1)

(Qb/Qts)^2 = Vas/Vb +1
Vas/Vb = (Qb/Qts)^2 - 1
Vb= Vas/ [(Qb/Qts)^2- 1]

the first equation tells you :
for Vb = Vas Qb = 1.414Qts
for Vb = 1/3 Vas Qb = 2Qts.

the other rearrangements tell you the same thing in reverse.

There are no restrictions on the value of Qts,
which in practise can vary between 0.1 and 2,
Though for bass units the range is usually 0.20
to impractically high values above 0.7 up to 1.
(unless open baffle or car free air mounting)

sreten.

Both of our equations are one and the same:

Vb = Vas / [Qb/Qts)^2 - 1]

expands to...

Vb = Vas / [(Qb^2/Qts^2) - 1]

For a flat response the exact value of Qb = (square root 2)/2, which is about 0.707, the commonly accepted value.

[sreten]

Originally posted by coolkhoa



Both of our equations are one and the same:

Vb = Vas / [Qb/Qts)^2 - 1]

expands to...

Vb = Vas / [(Qb^2/Qts^2) - 1]

For a flat response the exact value of Qb = (square root 2)/2, which is about 0.707, the commonly accepted value.

apologies again...., they are equivalent. sreten.

[kelticwizard]

Having heard of negative series resistance, but not knowing what it is, I did a quick look on Google.

After all, it has been mentioned as a way to lower Qts and, by extension, Qtc.

I found that while it is possible to do, if also makes an amp vulnerable to oscillation.

Since the whole idea of lowering Qts is to give a more controlled, snappier sound, it would seem to me that the solution is just transferring the problem from the speaker side to the amplifier side.

Am I right on this?

[Svante]

Originally posted by kelticwizard
Having heard of negative series resistance, but not knowing what it is, I did a quick look on Google.

After all, it has been mentioned as a way to lower Qts and, by extension, Qtc.

I found that while it is possible to do, if also makes an amp vulnerable to oscillation.

Since the whole idea of lowering Qts is to give a more controlled, snappier sound, it would seem to me that the solution is just transferring the problem from the speaker side to the amplifier side.

Am I right on this?

You are right in that a negative output impedance of an amplifier may lead to stability problems, but usually it does not go that far. Obviously, this series resistance may *never* be lower than Re, but that is never required. The main problem is that the resistance of the voice coil is temperature dependent, this leads to a Qes that varies with the temperature (= power) of the voice coil. This is a problem in any speaker, but is worstened by a negative series resistance.

Still, you have the box stuffing option, if Qts is 0.7+

[coolkhoa]

Originally posted by coolkhoa
So we never see a driver with Qts equal to or greater than 0.707. In a sealed box, such a driver will never be able to attain a "flat" response!

Allow me to correct myself. Such drivers do exist, but they are not practical for hi-fi, namely low-fi Goldwoods. That's why Parts Express says that their T/S parameters "make them great for automotive environments."

Consider that some of Goldwood's woofers have Qts of 0.81, 0.83, 0.88, and (get this) 1.69!

But the statement still applies: these drivers will never be able to attain an optimally flat response.

[Ron E]

You can make an amp with negative output impedance and make a "maximally flat" system out of a speaker with Qts greater than 0.7. You can also make a linkwitz transform and change the overall Q and F3 to whatever you want....within reason.

"Optimally flat" has a different connotation to an engineer. Optimum is a relative term, and it depends on the metric used. 10 different people, when asked to come up with an optimum solution, will likely make 10 different choices. I have heard a system with Qtc=1.3 in a car, and it sounded perfectly fine, especially with the techno the guy liked to listen to. The metric in this case was to maximize trunk space....

Expand your horizons....

[phase_accurate]

Optimum is a relative term, and it depends on the metric used.

I fully agree, since the expression "optimum" means the best possible compromise (in terms of cost, size, technical feasibility, taste, ....) for a specific application.

Regards

Charles