Excursion limits in ported design

[pick]

[Sydney]

... I always include a high-pass filter slightly below the port frequency to limit excursion and prevent damage....

Note that Eminence provides designs ideas ( Jerry McNutt ) for their drivers ( especially for High Power applications ), they all make recommendations for using a High pass filter to prevent damage.

[Dirk]

Pragmatically speaking, what do you listen to, and at what level?

[pick]

Pragmatically speaking, what do you listen to, and at what level?

These days I listen to a mixture of country, rock, classic rock, and a little pop. No classical or pipe organ kind of stuff. I listen a at pretty high volume with my current system but that's not so much the point of this new project. Moderate listening levels would be OK. 

[Pete Schumacher]

20W listening levels will be just plain loud!!

[jhollander]

I pick a design SPL like 100 dB and design a port velocity below 26 m/s.  Once I get above my design values the port noise is an early warning.

[billfitzmaurice]

. I can assume most of the music happens above 30Hz

Most music content is above 50Hz, with very little below 40Hz. For music only there's no need for a lower than 40Hz f3. Move the Fb higher to get a 40Hz f3 and the maximum power above that will improve significantly.

 

[hardcorecap]

while most of the of the sound we interpret as music is above 50hz, isn't most of the power used to push the drivers in the lower frequencies? So, it'd take more and more power to push the same driver the lower in frequency you attempt. Just askin'
 

[dcibel]

The elephant in the room is the accuracy of the simulation itself. Cabinet simulations are using T/S parameters, which are small signal parameters by definition. The simulations do not take account the forces of the mechanical suspension over excursion, so they are not entirely accurate at high output. Rest assured your cone will not fly across the room at 20Hz like many simulations will show, however in a ported cabinet, below the tuning frequency the speaker driver is considered "unloaded" and the excursion will be similar to what you would get if there were no cabinet at all.

Scan-Speak claims to have improved the performance of these types of simulations, you may want to compare your results using the Scan Speak Toolbox:
http://www.scan-speak.dk/?page_id=496

[Sydney]

...Cabinet simulations are using T/S parameters, which are small signal parameters by definition....

A reality often overlooked.

[billfitzmaurice]

while most of the of the sound we interpret as music is above 50hz, isn't most of the power used to push the drivers in the lower frequencies?

Most of the power lies in the 50-80Hz pass band. While the fundamentals of LF instruments go down to perhaps 30Hz it's the 2nd and third harmonics that dominate. For instance, the benchmark electric bass speaker is the Ampeg SVT 8x10. It has an f3 of 58Hz, and that's all it needs, based on the spectrum of the signals it has to deal with.

 

[BOBinGA]

Most amps have a built in hi pass filter. You can add a 20hz 2nd order high pass to your Winisd model on the filters tab to see how it affects your design.

[pick]

Scan-Speak claims to have improved the performance of these types of simulations, you may want to compare your results using the Scan Speak Toolbox:
http://www.scan-speak.dk/?page_id=496

Thanks. I'll check out this tool and see how the simulation looks. 

[Brian Steele]

Thanks. I'll check out this tool and see how the simulation looks.

I don't think that Scan-Speak model takes into consideration the response of the driver when driven beyond Xmax. It seems to be some version of the "semi-inductance" model, which I believe also assumes linear behavior (not totally sure - I really haven't done a lot of research into it yet). Hornresp was recently updated to use that model as an option, and REW can derive the parameters from the driver's measured impedance.