Small vented subs tuned low

We all know that for any given tuning frequency, the smaller the box gets, the longer the port needs to be.

What happens when the port volume approaches a significant fraction of the box volume, whatever that is? Could you keep increasing the port volume without negative effect? It would obviously start looking like a transmission line...
 
What happens when the port volume approaches a significant fraction of the box volume, whatever that is? Could you keep increasing the port volume without negative effect? It would obviously start looking like a transmission line...

Yes, though of course the vent will have increasingly strong resonances 'coloring' the driver's mids output if not damped, so at some point it's best overall to morph the cab, vent into an inverse tapered TL.

GM
 
What happens when the port volume approaches a significant fraction of the box volume, whatever that is? Could you keep increasing the port volume without negative effect? It would obviously start looking like a transmission line...

Hi 454Casull,

Simulating in Hornresp using the Loudspeaker Wizard will give you the answers.

The following related post may also be of interest:

http://www.diyaudio.com/forums/subwoofers/119854-hornresp-309.html#post3312998

Kind regards,

David
 

turbodawg

Member
2004-02-13 12:51 am
We all know that for any given tuning frequency, the smaller the box gets, the longer the port needs to be.

What happens when the port volume approaches a significant fraction of the box volume, whatever that is? Could you keep increasing the port volume without negative effect? It would obviously start looking like a transmission line...

I recently designed and built a box that has port volume at about 20% of net box volume. Seems to work fine, however it has a fairly high tune (35hz) and not all woofers will work well configured like this. Roughly 2.5 cubes net and effectively a 15"x2.5" by 27.5" long port.

If you were to attempt to tune low, say 20hz, in a 2-3 ft3 box, you need to be very careful that your port cross sectional area can keep up with your woofer's max output. Air speed needs to preferably be around 25 m/s or below at max output at tuning. With more powerful woofers the area needed increases, making the port length too long for a small box.

I've found that the best way to cram the most port you can into a small box is with a slot port on the longest dimension of the box with 90 degree turn on the end up the back wall, this gives you a bit more effective port length without extra complexity. Use a roundover and 45 deg slat to improve air flow on the inside, and roundover the outside end. Assume about 2" effective extra port beyond the actual measured length. See here for good discussion on end correction:

http://www.diyaudio.com/forums/subwoofers/219923-ultra-compact-ported-15-dayton-ref-ho-4.html

Inside end of port:

308156d1351129721-ultra-compact-ported-15-dayton-ref-ho-img_3154sm.jpg


Looking in from the outside:

308157d1351129721-ultra-compact-ported-15-dayton-ref-ho-img_3155sm.jpg
 
Last edited:

OscarS

Member
2011-01-02 10:44 pm
turbodawg, did you actually measure the true Fb? Reason I'm asking is because I've been doing subs like this for a while (not recently though), but none the less, when I use a slot port like yours where the slot ends up using 3 enclosure walls for its walls, I end up with an effective port length of ~66% longer than what it actually is (roughly 40% reduction in required port length, equivalently stated). This I verified with a WT3/DATS.
 

turbodawg

Member
2004-02-13 12:51 am
turbodawg, did you actually measure the true Fb? Reason I'm asking is because I've been doing subs like this for a while (not recently though), but none the less, when I use a slot port like yours where the slot ends up using 3 enclosure walls for its walls, I end up with an effective port length of ~66% longer than what it actually is (roughly 40% reduction in required port length, equivalently stated). This I verified with a WT3/DATS.

Hi, I ran some test tones at fairly high output and the frequency with the least cone movement was about 35hz or a hair above that. My original model has a 24" port length and 37.5hz tune. The box dimension on the port side is 22", so the actual port length with the bend is around 23-24". Also, my port is not exactly 2.5" tall, it's around 2-15/32", so that drives down the tune a hair.

Modeling with a 35hz tune and a not particularly accurate net volume around 2.5-2.75 cubes I get around a 2.5x15x27.5" port. So figure an additional 2" of end correction, vs. my actual, is about right.

I was expecting to end up with something tuned lower than the model, so I am pleased.
 

OscarS

Member
2011-01-02 10:44 pm
Hi, I ran some test tones at fairly high output and the frequency with the least cone movement was about 35hz or a hair above that. My original model has a 24" port length and 37.5hz tune. The box dimension on the port side is 22", so the actual port length with the bend is around 23-24". Also, my port is not exactly 2.5" tall, it's around 2-15/32", so that drives down the tune a hair.

Modeling with a 35hz tune and a not particularly accurate net volume around 2.5-2.75 cubes I get around a 2.5x15x27.5" port. So figure an additional 2" of end correction, vs. my actual, is about right.

I was expecting to end up with something tuned lower than the model, so I am pleased.

That's interesting indeed, because with a worst-case 2.5 ft³ and a 2.5"*15"*24" physical port gives me a standard tuning of 35hz. 27.5" gives me 33 Hz tuning. I am using Akabak to model. I would actually expect to see your box with no higher than 30Hz Fb. Would be interesting to see an actual impedance sweep of your box.
 

turbodawg

Member
2004-02-13 12:51 am
That's interesting indeed, because with a worst-case 2.5 ft³ and a 2.5"*15"*24" physical port gives me a standard tuning of 35hz. 27.5" gives me 33 Hz tuning. I am using Akabak to model. I would actually expect to see your box with no higher than 30Hz Fb. Would be interesting to see an actual impedance sweep of your box.

I'm using hornresp and winisd. If I get around to it in the next day or so, I can pull up the models and double check the numbers.
 
One solution for when ports become too large for the enclosure:

Long port options for subwoofers
Too long ports might have their first port resonance in the band you want to use them.
I aim at a 1st port resonance of at least double the highest frequency I want to use the sub.
So if a sub has to do 18hz - 80hz, I design the first port reonance to be at least 160hz.