Question about reflex duct shape

[ermes]

hi all!,

i have a question about reflex duct shape.
is better rectangular or triangular ?
help me to exam the two options

thank you

[enigmaticEntity.]

I assume you mean "port" as is commonly used around here... I'm no expert, but my guess is that neither of the two are necessarily "better" than one another. One thing you do have to keep in mind is to not make the opening too long/thin, otherwise you get into trouble with "fluid" properties of air and the inside surface of the port.

Also, it is usually better to round the edges of the opening, and this may be easier with a rectangular port. Circular ports are the easiest to round-over if you have a router.

Do you have a design where a triangular port is used? It would be interesting to see ;-)

[bjorno]

Originally Posted by ermes

...rectangular or triangular ?
help me to exam the two options

Rectangular is better than triangular but both are not optimal...
Optimal port shape: See the picture.

b

[moray james]

Bjorno: that is a very interesting attachment you have shown. Do you have a link to the article it came from? Or could you explain further how to design such a port? Thanks and best regards Moray James.

[AdamThorne]

I've made some rectangular ports. I have a router and roundover bits, but I haven't put a roundover on any of the ports I have made. I wasn't sure if it would change the tuning of the port. If anyone had thoughts on that I'd be interested to know.

[jerome69]

It is a french master thesis for a company heliaetus
A french presentation
http://www-cast3m.cea.fr/html/ClubCa..._GPellerin.pdf

The original website is here but not work:
http://old.lam.jussieu.fr/src/membres/pellerin/pub

The AES article is more interesting because you can calculate the profile. I made an excel spreadsheet to calculate them. Not so easy you need to use the excel solver.

Originally Posted by moray james

Bjorno: that is a very interesting attachment you have shown. Do you have a link to the article it came from? Or could you explain further how to design such a port? Thanks and best regards Moray James.

[speaker dave]

Originally Posted by AdamThorne

I've made some rectangular ports. I have a router and roundover bits, but I haven't put a roundover on any of the ports I have made. I wasn't sure if it would change the tuning of the port. If anyone had thoughts on that I'd be interested to know.

General flaring of the ports is good for reducing turbulance. If you start with a given length and then flare it, it will tend to make the effective length shorter (tuning higher) because a percentage of the port is having its diameter increased.

I'm talking about big end flares here. Roundover bits wil barely change the tuning, yet getting rid of the sharp edges will cut down turbulance a significant amount.

Regarding port cross section, we would like to have the maximum area with the minimum port wall surface. Best would be from a circular port. Worst would be a very thin and deep slot. I can't imagine an equilateral triangle would be much worse than a circle. If it fits the cabinet better, go for it.

I once tried a compact sub with an elliptical profile from end to end of the port, supposedly one of the best cross sections and patented by our friends at Bose. It was better but not by leaps and bounds. We ended up going with a PR.

David S.

[weltersys]

Originally Posted by AdamThorne

I've made some rectangular ports. I have a router and roundover bits, but I haven't put a roundover on any of the ports I have made. I wasn't sure if it would change the tuning of the port. If anyone had thoughts on that I'd be interested to know.

Rounding ports with a 3/4" round bit will reduce chuffing (turbulence noise) and won't appreciably change tuning.
Edit: Just like Speaker Dave said :^).

[AdamThorne]

thank you both!

[bjorno]

Thank you jerome69 and all

That's the right origin for my drawing, agree the AES paper is the one to study.

Most important is that the port becomes directional and have output lower than fb in a BR box besides a comparable smaller size than a straight cylindrical port.

b