Port Related Issue: Thoughts on how I can resolve?

[Efalegalo]

I'm building an MTM with a slot port up on the front bottom of the enclosure. The port was made large enough so I wouldn't run into port-noise issue.

After taking far-field measurements, I observed the response of the woofers was a little more ragged than I had hoped.

Then I took measurements of the port and noted that there were resonant peaks that coincide (more or less) with the dips observed in the woofers far-field frequency response. Maybe there is some sort of cancellation going on?

Was there any way for me to know that I would have run into this issue? I used WinISD for modeling enclosure volume.

Any thoughts on how I can resolve this issue?

[Zvu]

WinISD and Unibox give predictions about port resonances up to 1000Hz. Unibox even gives rough assessment of how will it affect the frequency response. Relation between port noise and port resonances usually tells you if the driver is better suited for passive radiators or regular port (if the port resonances appear too low in frequency when you make the port large enough surface not to create chuffing for planned SPL). What is too low depends on design. In a two way you have to juggle with port noise and port resonances much more than in a three way. In your case i'd put port on the back side and try to reduce the port surface to move it at least over 700Hz if port noise doesn't get any worse. Front loaded port on two way speakers i used successfully only on hi-eff speakers.

There is no other way to mitigate it now other than make another cabinet (if you glued this one) or to try to hack the existing inner part of the tunnel and glue some flexible material to that wall - hoping to get something like flex port developed by Kef.

http://www.htguide.com/forum/showthread.php?44377-Great-Scott-Retro-Audio-Rulz-Back-to-the-Future-or-Forward-to-the-Past&p=633321&viewfull=1#post633321

[Paul K.]

Just a quick question; what was the predicted peak air velocity for the port as you made it? A general rule of thumb many people use it for that velocity to not exceed 5% of the speed of sound, or ~17 m/s. If you designed for a significantly lower velocity by using a larger port area, the port length required to achieve the appropriate system tuning frequency will, of course, be longer than with a smaller-area port, and the longer the port, the more likely you'll run into the very problem you're experiencing. Also keep in mind that how large the port air velocity is before it becomes objectionable also depends on how loudly the system is playing and at what frequency the peak velocity occurs relative to the musical content.
Paul

[Efalegalo]

Paul,

The enclosure was modeled so port velocity does not exceed 28m/s at full excursion (35 watts). The predicted SPL was 103db.

I just noted the "Use "transmission line"-model for port simulation" and noticed how the predicted peaks nearly perfectly match the measured port response.

What is also interesting is that the slop port has predicted resonances, but the circular port doesn't. Very interesting.

I might have to ditch the box :-(.

[Evil Twin]

Painful when that happens...

One way to avoid port resonances is to use a passive radiator instead. More complex design modeling, and higher order roll off, but no port resonances.

[Efalegalo]

Thanks, Evil Twin.

This is the second time I've run into a port-related issue :-(. The thing is, both times it was a "budget" build, and thus cheaped out by not going the passive radiator route. I might have "saved" some money, but in the end "wasted" my time. It's all good though - all part of learning :-)

[fbov]

There's simple way to damp port resonances... line the inside of the box with 1-2" of acoustic absorption. The 450Hz port resonance wavelength is ~10x shorter than the 45Hz box resonance wavelength, so the absorption will damp the port resonance much more than the box resonance you want. Very common practice.... you don't need new boxes.
Have fun,
Frank