JUST 142Vpp for +/-20mm excursion !

[ChocoHolic]

With slow baby steps I am going ahead with two sadhara-cubes.
Sealed design. 30 litres netto. Linkwitz transform down to 12Hz (natural resonance of the sealed design would be 42Hz).
Today I made some first measurements in order determine the
required power of the amp. I feel comfortable with +/-20mm excursion, or little more. The drivers are specified with a linear excursion of +/-27mm. (Well I measured bottoming at +/-25mm in one driver and +/- 28mm in the second driver)

For a 20Hz signal, I found that I will reach +/-20mm excursion already at 142V pp .
The required power was just around 390W. I had expected that I would need about double of the power.
Really nice! This will make the design of the class D amp and the SMPS easier.

Here a pic of one cube.

[ChocoHolic]

Further on, I did some thermal examinations.
To get reliable values for the thermal speaker facts, the best way is to examine the RDC at a reasonable power level. Increase of RDC will give a clear information about the temperature of the voice coil. Fxxx on all power ratings given in any data sheet,
no matter which national/international standard they apply.
A copper coil will increase its RDC by 38% with every 100°C temperature rise. I applied a 10Hz sine wave with 40Vpeak-peak and monitored the DC resistance.
The initial value at 20°C was 2.77 W, after some hours it stabilized at 3.73 W.
According the temp. coefficient of copper this means about 91K temp rise of the coil, ==> running at approx. 111°C, which is not critical.
The power under this conditions was 43W.

Below is the screen shot of the excursion test with 142Vpeak-peak and 390W. Under this conditions the voice coil heats up quite fast. Already after 10 sec. it exceeds 150°C .... well, I decided not to run it continuosly at this load...
White trace: Voltage across speaker, 20V/div.
Colored trace: Current through speaker (inverted polarity), 5A/div.

[Optical]

how do you measure driver excursion?

[ChocoHolic]

Hi Optical!

Well, the measurement is a simple mechanical set up with fixed
reference metal bar (must be quite solid to avoid bending) above the cone and a slide gauge.
No sophisticated optical system, sorry.

First I measure the distance from the reference bar to the
cone without signal.
Second, when the signal is applied, I measure again with the slide
gauge. CAREFULLY. You can hear it very clear, when the cone
starts touching the slide gauge. Difference of both readings is the excursion. Accuracy is limited of course, say about +/-0.5mm.
And furtheron I cannot see if the excursion towards the back
is identical with towards the front, as I only can measure the
outwards excursion.
But in order to get an idea about the required power, this
should be sufficient ....I guess....

Cheers Markus

[ChocoHolic]

...the unit for the RDC must be Ohms, not W !
The increase from 2.77 Ohms to 3.73 Ohms indicates an
temperature rise of littel more than 90°C....

How did the W get in that posting???
I took the text with C&P from a word doc... and there the unit is fine!!!

[simon5]

Well it's a bad caracter table conversion, most of the time if you copy/paste PDFs you'll get W instead of ohms symbol.

Usually, Adire rate drivers with thermal ratings, so I'm surprised to see you're recommending less than 400W on this coil rated at 800W. I won't argue with you because you seem to know what you are doing hehe!

[ShinOBIWAN]

Transform down to 12hz?

Why? Group delay would be terrible at that gain, it would also explain why your only able to use 400w to reach maximum excursion, in other words your seriously underusing the max SPL of these drivers but then again perhaps this doesn't matter.

I would have gone for transform to 20hz, you'd have higher power handling, low group delay and more SPL plus room gain would factor in and the natural 2nd order roll off of a sealed design would ensure a -3dB at 15/16hz. 12hz is useless unless played at high SPL which would then induce the whole room is moving feeling.

[richie00boy]

Markus, good stuff -- always like to read your posts.

Sninobiwan, surely the whole point of Markus transforming down to 12Hz is to get the group delay and phase change down really low in the audible band?

[mike.e]

Markus, good stuff -- always like to read your posts.

Sninobiwan, surely the whole point of Markus transforming down to 12Hz is to get the group delay and phase change down really low in the audible band?

What if its already inaudible??

What if all that filtering to fill in the response causes a similar response to a ported enclosure with its clearly known sharp cutoff near F3

[richie00boy]

By using the LT Markus is able to achieve a GD, phase and transient response that will be the envy of any vented box. The downside as mentioned is absolute SPL, but if it's loud enough for what he wants and his top criteria is ultimate accuracy... And don't forget just how tiny this sub is, this kind of performance is just not going to happen with a vented box even double the size.