Modeling bass for small speakers

[NdG1977]

Hi guys,
Probably a stupid question, but I'm not sure how to move forward:
When modelling small woofers do you model the woofer with the high-pass filter on or do you model the woofer unfiltered?

My take is that if it is supposed to play alone (no sub) it should be modeled with the filter and look for the best/lowest acceptable response, but if crossed over to a sub should be modeled without filter or better, this is what I'm doing:
Modeling volume and tuning without filter and port velocity and XMax with filter anyway this speaker is never going to play without filter...

Am I doing it right?

[DeZZar]

Without.

Pairing with a sub is more of an overall design decision. If your looking to make speakers that will never ever be played on their own and are intended as a package with a subwoofer then you might perhaps target a smaller box, maybe a sealed enclosure and you wont be too concerned with extended bass response. It means you can probably work with smaller woofers etc. Even so, you would still model without the filter.

[NdG1977]

OK thanks!

But what about port air velocity and xmax? If I model them without filter I end up with big ports and very low wattage in order to have an acceptable displacement... Is it OK to verify with filter and define port size that way? in the end they will never play full range, they cannot do it, just too small (2in) will always be cut at 150/200hz (active or cap still TBD) and fitting a big port looks like a waste... Sealed is not really an option because I get -3db at 200hz...
Odd project I know... But I refuse to buy a cheap logitech system when I can have some fun

[billfitzmaurice]

What the port velocity does below the crossover frequency to the sub doesn't matter, so ignore that. Size the port for acceptable port velocity above the crossover frequency. For that matter size the cabinet and set the tuning frequency for best results within the intended pass band. However, you may well find that a sealed enclosure with a larger driver may work better anyway and still be no larger a box.

[DeZZar]

Which driver are you using and how many?

[NdG1977]

Sorry I didn't specify drivers as I wanted to avoid double posting, anyway I'm doing 1 driver per channel NE65W in 0.6l as FR crossed at 150/200Hz (Active 200hz current assumption) and TCP115-4 as bass in about 3l. Objective is conference calls and casual background music in near field.
I'm going to build a test enclosure and post the results before the actual build anyway...
I'll also post the sim this evening so you can tell me if I'm setting it up right...

[DeZZar]

What Bill said.

Your peak excursion and port velocities are going to be outside the final operating frequencies of the driver so it's really not going to present an issue. Especially for the stated purpose.

[DrewsBrews]

The sealed box effectively creates a mechanical low pass filter. Saving the $ and complication of the large value components needed for high passing a woofer.

[billfitzmaurice]

Logically one would think that would be the result of the cone compressing the air in a sealed box, but where cone excursion control is concerned ported is better down to roughly F10. After factoring in the higher sensitivity of ported it tends to have the advantage in displacement limited output down to roughly F15. Besides, passive high passing a woofer, or in this case a midbass, has been replaced by active. Even very inexpensive amps are available as 2.1.

[DrewsBrews]

Yep, electronics have become the fix-all in day-to-day products. With the bulk pricing of custom or OTP chips that becomes the cheapest way to do it... not that there is anything necessarily wrong with that. Though, for us DIYers the same parts may not be so easily sourced.

[NdG1977]

What about passive cap on the input to the amp? Found a few articles about it and it looks pretty simple. Sent from my JDN2-W09 using Tapatalk

[billfitzmaurice]

It's better than nothing, but as it's a first order filter it's not brickwall by any means. You need to know the amp input impedance to calculate the cap value.