Using Thiele Small to estimate *upper bandwidth of woofers & mids

[rick57]

Suggested to me recently:

“Fmax is . . the frequency where the distortions become acceptably inaudible

fmax (Hz) = Re*1000/ (Le * 2 * Pi)
is a good rule, even for published specs”

(Obviously it ignores beaming, and cone material)

I’ve calced a number of drivers with this formula. Some ‘fmax’ upper cutoffs sound about right on, others not so . .

Anyone heard of this formula, or have a better rule of thumb?

Cheers

[pinkmouse]

References? It's new to me, and I must admit, I'm doubtful.

[infinia]

Hi
So upper end usable response of a driver is simply a ratio of Re/Le times a constant (Fmax = 159*Re/Le)? maybe Le in mH?

Maybe someones crude rule of thumb for (sub) woofers only and might be OK for them. But I don't think it's possible to relate T/S to midrange and tweeters upper end responses and qualities there. It would be nice if there was one tho. Seems T/S is only usefull for determining the lower end response possiblities.

[mazurek]

Along those lines, if you can get an adequate picture of the inductance or the impedence plot of the woofer, you can estimate where the first order inductive lowpass is taking place, which may or may not be the upper limit of the response, but you would do well to take the phase shift into account. Linkwitz does something analogous on his Thor, where he equalizes the directional rolloff to eliminate the phase shift when crossing over to mains at higher frequencies.

[rick57]

Al,
> References?
None published. It was posted in the Yahoo Groups Lambda Drivers forum recently, and I am attempting to verify it, or even it’s basic logic - that Inductance relative to Re indicates upper Hz for undistorted sound.


Infinia,
It was suggested in the context of (Lambda) midwoofers ~ how high can they playundistorted, not subwoofers. No suggestion re tweeters. Mids??


mazurek,
Yes, Linkwitz refers to Fmax www.linkwitzlab.com/thor-measmt.htm apparently where the inbox resonance is, ie a different measure.

Regardless of the Linkwitz measure or the Lambda formala, anyone have any thoughts on paramaters or ratois that inducatie bandwith?

Or does anyone else believe that inductance indicates distortion in the upper usable bandwidth?

Cheers

[pinkmouse]

Originally posted by rick57
Regardless of the Linkwitz measure or the Lambda formala, anyone have any thoughts on paramaters or ratois that inducatie bandwith?

The only real way I can see is to measure.

[rick57]

pinkmouse

my suggestion is not that it is highly accurate, but that other things being equal, it could act as a guide to what drivers may be able to do, ie with other things, a pointer to better buying

[sreten]

Hi,

not sure what we are talking about here ?

An upper limit for distortion ? frequency response ? or both ?

I've not seen a driver that distorts more as you go up in frequency,
other than highly resonant drivers (metal cone) that exxagerate
the intrinsic distortion of a frequency due to huge response peaks.

I think the whole idea is far too unclear to be any practical use.

BTW, what does the principle indicate about full range drivers ?

sreten.

[infinia]

Rick, If a rule is to be valid, should it not apply to any driver? Otherwise, I can maybe see the validity of the formula if only applied to a narrow spectrum of driver types, sharing similar motor structures, cone/geometries, and materials when based on empirical data. FWIW Le is usually characterized at small signals and can also be one source of distortion of many, as its' value is a modulating influence during large signals.

[rick57]

I had serious doubts about it, you guys have convinced me. I was meant to indicate upper usable Hz.

As I went to sleep, I was thinking there would need to be different values for materials, sizes, etc


Thanks for your info

Cheers