AMT sub - using dynamic woofer

[mige0]

Originally Posted by Rudolf

I am still "in clothes" - at least below the 300 Hz belt : 15" in H frames 40x40x40 cm

Rudolf

get undressed - at least up from the 100Hz belt !

[theAnonymous1]

Anyone ever thought of a DIY LAT type driver made out of passive radiators and one or two subs with with really strong motors; connected together with carbon fiber rods?

[djk]

The LAT driver is just a rip of the ESS Transar, updated for newer materials.

Why not just stack some cone drivers as in the thumbnail sketch I posted a while back?

Having the slots at an angle (instead of perpendicular) will improve the HF bandwidth.

[WithTarragon]

Originally Posted by djk

The LAT driver is just a rip of the ESS Transar, updated for newer materials.

Why not just stack some cone drivers as in the thumbnail sketch I posted a while back?

Having the slots at an angle (instead of perpendicular) will improve the HF bandwidth.

I am trying to understand your last point about the HF bandwidth.
Is this because the plenum (which effectively acts as a low pass filter) is smaller and is now filtering less?

[mige0]

Upper BW limit of an AMT sub - like my Deep Frame Dipole arrangement or the LAT / Transar is set by two factors - as easily can be seen by comparing my measurements above.

One limiting factor is the separation distance of the OB / framed OB - the second limiting factor is the pipe behaviour of a "pleat".
These are two completely different underlaying mechanisms we have to stay clear of.

Both mechanism superpose the well known comb filter effect above first peak.
As we usually would like to avoid the area above the first peak, this frequencies actually are the limiting factor (plus some margin).

As I have shown with making the "pleat" of conical shape we can up-shift at least the first peaking of the pipe at the same footprint - and thus extending the useful bandwidth.

Another way to up-shift first peak (due to dipole operation) is to lower separation distance - meaning - making a physically smaller speaker.
The trade off of small footprint AMT bass (like the LAT without additional baffle for example) is the loss of low frequency SPL, as also quite easily seen in my measurements above.


Michael

Originally Posted by WithTarragon

I am trying to understand your last point about the HF bandwidth.
Is this because the plenum (which effectively acts as a low pass filter) is smaller and is now filtering less?

[theAnonymous1]

Originally Posted by djk

Why not just stack some cone drivers as in the thumbnail sketch I posted a while back?

Because the magnets on the drivers limit how small you can make the "slots".

With a DIY LAT style setup you could fit quite a bit more moving surface in the same space. It would require a beast of a motor to drive it though.

[djk]

"I am trying to understand your last point about the HF bandwidth.
Is this because the plenum (which effectively acts as a low pass filter) is smaller and is now filtering less? "

"Because the magnets on the drivers limit how small you can make the "slots". "

The dimensions of the plenum determine the corner frequency of the low-pass filtering, primarily the depth. I think the tilt allows more on-axis SPL from the driver whose cone you can see. Electro-Voice made a speaker with one pair of 15s that was able to cross at 800hz to the HF horn. With some good 6-1/2s it should be possible to push it higher than this.

Madisound has some surplus Aurasound 6-1/2s for around $11 or so that look good. The T/S look suitable. They have paper cones and a neo magnet with a 1/2" thick top-plate with an underhung coil. The x-max at the 70% BL point looks to be about 4mm.

[theAnonymous1]

I was thinking more along the line of narrow plenums for mass loading à la "ripoles".

[djk]

Then I would probably just do perpendicular.

[mige0]

The "mass loading" saga of Ripols is mainly just that.

What looks as to be a down-shift in low cut is *mainly* a illusion caused by the SPL suck out towards higher frequencies.
Or in other words - we sadly do not gain anything - SPL wise. Maybe 1-2 dB at around 20Hz at best.

Ripole might be interesting for all-passive guys though, as it needs less EQing - despite the peak.

For optimization of SPL versus volume sadly its the wrong direction.

Michael