help with damping resonances in aluminium?

[mondogenerator]

hi there. IM building some loudspeakers using Visatons AL130 8 midbass driver...i have experience with 'doping' paper drivers to reduce break up and cone resonances, and although I am not sure it is the ONLY way to damp them, I have used bog standard PVA wood glue to effectively do this. Now im wondering if there are any methods to do the same 'damping' with aluminium cone drivers, as im unsure that PVA will sick properly, due to the non porous nature of the material.

Anyone have any ideas??

[el`Ol]

I once used hardwax oil to dampen a paper cone, first layer to pour in and second layer on top. Worked very well, and this stuff is probably better suited to surface-only application than PVA.

[LineArray]

Hi,

i know the AL 130 quite well. in which frequency range is
the problem you want to solve ?

I also think that PVA will not work on AL for long.

Just keep away from the breakup range ?

0.8 ... 1.3 Khz is a nice XO frequency for that driver.


I used a dual slope filter to get the input voltage down in
the presence region, where the resonances may cause trouble.

Used in that way, it is a very good driver.

[mondogenerator]

well i am using a slight variation on the vib130tl x-over on the visaton projec t pages as the freq response looks fairly flat to me. To my knowledge the xover point is about 2.5-3kHz so it should be out of the breakup mode at 7kHz(around).

The only reason i ask is that ive read a few blogs about others MEASuring the peak and finding it lower in the spectrum, about 5.5-6kHz.

Oh, and ive thought that i could maybe improve things by using a notch filter and designing my own x-over, as i dont really like the use of resistors in the high pass circuit for the tweeter GC20, as i prefer a Lpad resistor circuit.

If you have any ideas on whether any of this is really necessary, i would like to know, as i am really just spitting out ideas, trying to brainstorm an improved crossover to the one found on the visaton project page or impact audio pages (as i stated for the vibtl kit using al130 and gc20)

Thanks for any help you are able to offer, as i have had trouble finding anyone here in the UK (or indeed US) using visaton drivers, and i rate them quite highly so i have difficulty in understanding why this is!!!

Thanks again!

[LineArray]

Hi,

i am not a friend of fighting breakups with notch filters for
it does not solve the problem IMO, which is audible.

If you do so, you should make your own measurements, to adjust
the notch properly. Cone breakups are IMO also due to exemplar
of the driver, not only to type.
This would serve as an explanation for different frequencies of
the breakup measured by different owners.

My own ranking of measures is:

- use a lower XO frequency and a tweeter which is able to work
from 2,1 to 2,3 Hz . I would not go any higher.

When crossed at 1Khz the driver deliveres a midrange which is
nearly "perfect".

(To get lower with the tweeter i sometimes used a combination of two
tweeters in series, one of them bridged by a capacitor or RC,
which leads to a 2 1/2 way design in fact )

- ensure that the lowpass rolloff >4 Khz is steep enough
- Voice coil inductivity compensated properly ?
- Use dual slope filter to tame breakup candidates in a 2-Way
rather than notches.

- damping of the resonance
maybe a ring from a damping sealant around the dustcap has
good effect and has little mass, though i never modified an
AL 130 myself, could be worth a try.
If XOed low enough, it is not necessary to modify this driver IMO.

- last measure would be a notch filter


Just my personal opinions

Cheers


P.S. Link to an old 2-way of mine using an AL 130.

Astonishing discovery

The ribbon is crossed at 1Khz.

[mondogenerator]

Originally Posted by LineArray

Hi,

i am not a friend of fighting breakups with notch filters for
it does not solve the problem IMO, which is audible.

If you do so, you should make your own measurements, to adjust
the notch properly. Cone breakups are IMO also due to exemplar
of the driver, not only to type.
This would serve as an explanation for different frequencies of
the breakup measured by different owners.

My own ranking of measures is:

- use a lower XO frequency and a tweeter which is able to work
from 2,1 to 2,3 Hz . I would not go any higher.

When crossed at 1Khz the driver deliveres a midrange which is
nearly "perfect".

(To get lower with the tweeter i sometimes used a combination of two
tweeters in series, one of them bridged by a capacitor or RC,
which leads to a 2 1/2 way design in fact )

- ensure that the lowpass rolloff >4 Khz is steep enough
- Voice coil inductivity compensated properly ?
- Use dual slope filter to tame breakup candidates in a 2-Way
rather than notches.

- damping of the resonance
maybe a ring from a damping sealant around the dustcap has
good effect and has little mass, though i never modified an
AL 130 myself, could be worth a try.
If XOed low enough, it is not necessary to modify this driver IMO.

- last measure would be a notch filter


Just my personal opinions

Cheers


P.S. Link to an old 2-way of mine using an AL 130.

Astonishing discovery

The ribbon is crossed at 1Khz.

OK all points taken in...personally i think im just trying to be a little over perfectionist about things...reading the dadt on the zover again it apperas that the -3dB point of the woofer is just over 2k, so its better than i thought-ONE question about your project: what volume and tuning regime have you employed...im a little comfused as the box looks like a cube to me and isnt that pretty much the worst possible shape?

[LineArray]

Hi,

that box is no cube inside. It is not meant to be
an example for "BR alignment for AL 130".

It has two cascaded helmoltz resonators, the first one
behind the driver is lossy and serves to add acoustic
mass to the driver.

It is a low sensitivity broadband concept, not recomended for
general use. The (long) port for the main chamber is on the rear,
it is a bended tube, which would not fit inside otherwise.

The ribbon has a wedge shaped own partition behind the baffle too.
So we have three partitions with mostly non orthogonal walls
inside. I gave up the speaker, it is ridicoulously complex and work
intensive.

But i have never been sorry for choosing the XO frequency as
low as possible. The speaker serves sometimes as a kind of
personal reference in "uncoloured midrange to presence performance".

Cheers

[mondogenerator]

ok that makes sense...i didnt think anyone with knowledge would build a cuboid reflex box...hahah

so i guess that design is slightly akin to an aperiodic design in some respects...interesting

i was considering making some kind of resistive reflex design but it would've been purely an exercise in experimentation as there isnt really alot of data on the subject.

I only wish i had a tweeter that i COULD design a very low xover for, as i think the design im using strecthed the tweeter about as far as id like to go without introducing some nasty effects, although the tweeter resonance seems fairly well damped from looking at the impedance curve

by the way id love to try a ribbon but ive read that they can be hard to integrate into a system, both electrically and sonically

[MisterTwister]

Aluminum resonance is very hard to tame. I've done some experiments before. Applying damping material to the cone does almost nothing. I think very good cone mod is to make it extruded. Just take large philips screw driver and draw extruded lines from back side of the cone. It will greatly increase cone's stiffness. I'll post pictures later.

[rdf]

This is an outside one, but have you considered adhering strips of thin aluminum foil to the back of the cone with polyurethane glue? The latter dries very light and stays slightly pliable. Experimenting with comparable thin aluminum sheet in advance is recommended.