edit: new discussion about issues related to drivers other than just only midranges;
https://www.diyaudio.com/community/...l-x-for-mid-ranges.376164/page-7#post-7703832
Hopefully this is the right subforum for this.
I have a question about the following,
So for loudspeakers there are basically three dominant factors creating distortion/non linear behavior.
There are some other factors, like cone break-up, compression of air etc etc, but for now I am not going to consider those.
So the BL(x) and Cms are both dominant for the lower frequencies, around the Fs up to roughly 2 x Fs.
From that point on the inductance Le(x) is starting to be the more dominant factor.
So let's assume we have a mid-woofer with a Fs around 50Hz.
We are gonna cross this driver (actively) around 100-120Hz (2nd order highpass)
At this point, as far as I understand, it's not really important anymore how (non)linear the Cms and BL(x) is, since the Le(x) is dominant here?
As long as we keep the cone excursion to a minimum, which is most cases is fine since most speakers will run into there max power at this point with a cone excursion of just 1-3mm or so.
https://www.diyaudio.com/community/...l-x-for-mid-ranges.376164/page-7#post-7703832
Hopefully this is the right subforum for this.
I have a question about the following,
So for loudspeakers there are basically three dominant factors creating distortion/non linear behavior.
- The non-linearity of the compliance (1 / stiffness), Cms(x)
- The non-linearity of the BL vs displacement curve, BL(x)
- The non-linearity of the inductance. Le(x)
There are some other factors, like cone break-up, compression of air etc etc, but for now I am not going to consider those.
So the BL(x) and Cms are both dominant for the lower frequencies, around the Fs up to roughly 2 x Fs.
From that point on the inductance Le(x) is starting to be the more dominant factor.
So let's assume we have a mid-woofer with a Fs around 50Hz.
We are gonna cross this driver (actively) around 100-120Hz (2nd order highpass)
At this point, as far as I understand, it's not really important anymore how (non)linear the Cms and BL(x) is, since the Le(x) is dominant here?
As long as we keep the cone excursion to a minimum, which is most cases is fine since most speakers will run into there max power at this point with a cone excursion of just 1-3mm or so.
Last edited:
What precisely is your question? I would agree that Cms and BL nonlinearities (as a function of cone displacement x) aren’t important. But why do you refer to Le(x) as that won’t vary that much either with low displacement values, would it?
Of course this is not a question of 'if', but 'when'. While it would be interesting to detail this, in this position I'd think it would be better to measure.
What precisely is your question? I would agree that Cms and BL nonlinearities (as a function of cone displacement x) aren’t important. But why do you refer to Le(x) as that won’t vary that much either with low displacement values, would it?
Ehm, yes I guess you're totally right when it it Le(x), or Le as function of x

Whoops. Total brainfart
Well, I guess I meant to say that for a midrange just a good low Le, aka plenty of demodulation will be the most important factor.
We can just omit the other two, even more so when Fs is (much) lower than the crossover point.
Last edited:
All midrange speakers sound the same, according some recent thread.
What recent thread are you referring to?
Ah yes, the thread which demonstrates the old concept that as long as two drivers are minimum phase, and pistonic where they have similar dispersion.. and this is fairly typical.. then they will sound the same when equalised to the same response.
This also presumes that small amounts of harmonic distortion are negligible, which has also been demonstrated, however large amounts may change things..
This also presumes that small amounts of harmonic distortion are negligible, which has also been demonstrated, however large amounts may change things..
The current applied to the voice coil causes distortion in of itself. That's where well designed motors with plenty of copper or fixed dummy coils come into play.
Check out the Purifi explanation for more info.
The cleanest high spl midrange (300hz to 2,000hz) I've heard came from the 18Sound 10NDA610. Other drivers don't quite get there in clarity even though they're of similar construction and spl capabilities.
Check out the Purifi explanation for more info.
The cleanest high spl midrange (300hz to 2,000hz) I've heard came from the 18Sound 10NDA610. Other drivers don't quite get there in clarity even though they're of similar construction and spl capabilities.
What recent thread are you referring to?
Probably this one
World's Best Midranges - Shocking Results & Conclusions.
As an attempt to recap: is your question essentially that in case a midrange is used well above Fo and not subject to large, but small excursions (say <1 mm one way), all three non-linearities you mentioned are largely irrelevant?
Because breakup etc is not discussed here, I would be temped to say: yes.
Because breakup etc is not discussed here, I would be temped to say: yes.
Yes that is a proper recap.
As I mentioned in the first post, not taking cone breakup etc into account.
(Since that obviously will add distortion and other unwanted effects)
Btw, the thought behind this question is that it basically concludes that best non compromise approach, is to get something like a 6, 8 or 10 inch as midwoofer (roughly 100Hz and up).
Combined with a tweeter with waveguide (or horn, whatever floats your boat).
This can be a pretty affordable woofer, as long as it has a decent amount of demodulation.
Bigger (8 or 10 inch) is better.
Below 100Hz, just use a multi sub system.
So nice referring to Purify, but I just don't see any benefits at all for these kind of drivers.
Any 8 or 10 inch can do these kind of performance numbers with ease.
For subs you can even just go a little bigger if you want.
As I mentioned in the first post, not taking cone breakup etc into account.
(Since that obviously will add distortion and other unwanted effects)
Btw, the thought behind this question is that it basically concludes that best non compromise approach, is to get something like a 6, 8 or 10 inch as midwoofer (roughly 100Hz and up).
Combined with a tweeter with waveguide (or horn, whatever floats your boat).
This can be a pretty affordable woofer, as long as it has a decent amount of demodulation.
Bigger (8 or 10 inch) is better.
Below 100Hz, just use a multi sub system.
So nice referring to Purify, but I just don't see any benefits at all for these kind of drivers.
Any 8 or 10 inch can do these kind of performance numbers with ease.
For subs you can even just go a little bigger if you want.
There is some demand for small two-ways that can go low. Drivers like the Purify are there to fulfill such demands. If you go three-way there are indeed some parameters that can be relaxed compared to those needed for two-ways.
Regards
Charles
Regards
Charles
Yes, I can see there is a little niche there.
I just wouldn't call that the best solution ever, which they seem to imply.
I just wouldn't call that the best solution ever, which they seem to imply.
yeah, just only a little odd to claim the things they do and people seem to take it like free candy.
The science is very old behind Purifi's reasoning in their motor design. The issues with surrounds is also well documented. They simply put the effort into further refining prior art.
A larger more sensitive driver based on similar design principles would be that much more impressive. Or a lower mass 6" midrange with higher sensitivity would be interesting as well.
They are too expensive and low sensitivity for my uses but I understand where they're coming from.
A larger more sensitive driver based on similar design principles would be that much more impressive. Or a lower mass 6" midrange with higher sensitivity would be interesting as well.
They are too expensive and low sensitivity for my uses but I understand where they're coming from.
Last edited:
I would find it extremely impressive to see those improvements or a part of them in more conventional speakers on a bigger scale.
Things like progressive spiders are not much more expensive to make.
Those are just stamped out of a mold, until the mold is just worn out.
Demodulation rings can be found in drivers or like 15-30 bucks as well.
If you have experience with OEM custom orders, almost everything can be done for just a very small extra price.
Things like progressive spiders are not much more expensive to make.
Those are just stamped out of a mold, until the mold is just worn out.
Demodulation rings can be found in drivers or like 15-30 bucks as well.
If you have experience with OEM custom orders, almost everything can be done for just a very small extra price.
Right, but we have to admit that -still apart from the very clever voicecoil winding- the cone body design and cone-surround combi are among of the very best in currently available 6.5" format.
Much can be done by spec oriented ordering, but proper customized cone design is still a formidable stumbling block. The Chinese have not yet mastered the tricks involved in that in full, although FEM/BEM software is widespread nowadays. The Danes are still the current Masters.
- Home
- Loudspeakers
- Multi-Way
- The importance of Kms(X) and BL(x) for mid-ranges