[michael hurd]

I recently built a pair of sonosubs and here's the thread on them. From the outside, I was bent on designing a pair of subs to be used for music only, with the possibility of watching a movie once in a while. ( for second room )

Crossover <60 hz, closer to 50hz or less, depending on how the actual response was in room. Maximum enclosure volume possible with a large port for a few reasons. Large diameter port for low port velocity as well as linearity. I decided on 18 inch sonotube, and originally was set to build just under 9 cu ft net, 10cu ft before port and driver.

I estimated the outer diameter of a port, without having it on hand. DOH... as it was, I had to drive 75 miles to pick up sono and MDF, and a second trip was for a port and a few other odds and ends.

I had already cut the 18" sono ( actual inside measurement ) into two pieces 70 5/8" long, painted the outside of them with acrylic latex paint and cut endcaps before getting a chunk of 8" OD sono for the port. Never trust a label, measure it for yourself, lesson learned. And to think I was an aisle away from the tape measures... :sarcastic:

It actually worked out maybe a tad better, the port velocity was a bit higher but the port resonance was pushed up a bit more than 2 octaves above the intended crossover point.... 250hz.

 

[michael hurd]

Having issues in the past with T-nuts backing out at the most inappropriate time, I used baltic birch ply to mount the drivers, and MDF for the other endcap for the ports. Fasteners for the driver are the TC supplied hex head wood screws.

The inside port flare is two layers of baltic birch ply glued together ( the driver cut out at 13" and another 11" piece ) turned on a wood lathe. The outer flares are not yet attatched, I have to glue them on as well as figure out a base and legs for these.

 

[michael hurd]

I absolutely despise the smell of MDF outgassing from a vented enclosure during operation. I coated all of the MDF with Minwax urethane many times to seal it from outgassing, as well as the outside of the port. PL Premium construction adhesive was used in securing the port to the flare, as well as the endcaps. There are NO mechanical fasteners in the endcaps at all.

I don't have any pics of me turning the flares on the lathe or routering the endcaps for this build though. Note the big '8" OD' printing on the outside of the port tube, NOT. More like 7 3/8" ID and 7 5/8" OD. How is that close to 8" OD... home depot for ya...:explode:

 

[michael hurd]

A few days ago, I threw in a driver and tested one while the adhesive was curing on the other. I drove it off the amplifier in the trunk of my car, and played some music through it, noting some resonances on higher frequency bass. I did not use any high or low pass filters at all, and it was plain there was issues.

OK, I pulled the driver out and it was obvious that speaking or shouting into the enclosure produced a resonance. Not good.:doh:

 

[michael hurd]

I went down to my local store and bought 4 $ 6.00 pillows... returned, and installed them in the enclosure between the port flange and the endcap, it was a tight squeeze with a 13" OD flange and a 18" ID tube... about 2.5" to force the pillows down in there. I put all 4 pillows in the enclosure, re-installed the driver and did some rudimentary testing again. MUCH better, I was unable to get it to resonate, as well as when I tried the speak / shout test into it without the driver it was almost totally dead.

 

[michael hurd]

In the interest of actually producing some good measurement data, I decided to test the unstuffed enclousre against the stuffed one. The enclosures were placed side by side, while testing one I would short the terminals of the other, to keep it from drastically affecting the measurement.

I tested GP outdoors, mic 1m from driver, and this is what I came up with. The unstuffed trace is red, the stuffed is blue. Tests were back to back, as close as possible, nothing was moved during the sweeps.

 

[michael hurd]

I measured tuning by watching for lowest excursion on each of the enclosures. This is where it got interesting. First I set REW up for a 15 hz sine wave at a low level, and increased it slowly until the driver was very close to maximum excursion, then moving the frequency up in 1/2 hz increments.

Unstuffed sonotube came in at 18.5 hz. The results of the stuffed enclosure were quite interesting to say the least. With the unstuffed enclosure, the tuning was easy to verify.... cone motion was almost stopped.

With the stuffed sonotube, it was much harder to determine the tuning point, but I was able to determine it was 17 hz. There was a lot more cone excursion at tuning on the stuffed enclosure vs the unstuffed one.

In conversation with a friend, he mentioned that PWK ( Paul Klipsch ) always talked about the rear volume of a sealed box being a capacitance. Adding ESR to the cap he mentioned changes the "Q" of the cap, widening it.

So, the tuning point effectively gets wider with a bit less boost at the tuning frequency, hence the excursion goes up at tuning, with a bit less above and below.
__________________

 

[michael hurd]

Here's a graph of the two enclosures overlaid, I changed the scale a bit though. Interestingly enough HornResp predicted the resonances, when Mike Bentz modelled them, he sent me the picture through IM when we were talking about the enclosures.

 

[michael hurd]

Rain cut my measurement time short, I had to hurry to get everything all packed up before it got soaked. :thumbsdown:

I had wanted to get a measurement at the port exit as well, however I might have to wait for better weather, we still have some snow here... and it's raining currently still.

Some graphs of the model:

Impedance is high in the 30-35 hz range, where *most* of the low bass in music is, of course there are exceptions to the rule. This means that power across the coil is low, contributing to less thermal compression. At around 35 hz, it's about 14 ohms, going up to 24 ohms at 30 hz.

I modelled cone excursion with a high pass filter at 17 hz, and 500 watts, with a bit of heat added to to the voice coil temp rise box.

 

[ISLAND1000]

Mike,
I consider your experiments and build an important revelation. Your enclosure, the usual sonotube, proves to have identifiable resonance issues with the ( open space surrounding the PORT ). Just as interesting is that the cure for the resonance is to stuff the open area surrounding the port with poly pillows. Simple.
But wait, does this resonance issue apply to ALL sonotube enclosures? If so, will the poly pillows cure any resonance issues in any sonotube?
It makes me wonder if square and/or rectangular ported enclosures also have a similar resonance issue surrounding the port tube as in the sonotube design?
I had always assumed that any resonances created within the enclosures would usually be below the hearing threshold. Obviously they are not as your charting shows and you also noticed immediately upon firing the units up.
This is interesting stuff.
Oh, by the way the tubes and drivers look neat. Are you gonna use them as rolls on the floor or install them vertically?

 

[michael hurd]

Mike,
I consider your experiments and build an important revelation. Your enclosure, the usual sonotube, proves to have identifiable resonance issues with the ( open space surrounding the PORT ). Just as interesting is that the cure for the resonance is to stuff the open area surrounding the port with poly pillows. Simple.
But wait, does this resonance issue apply to ALL sonotube enclosures? If so, will the poly pillows cure any resonance issues in any sonotube?
It makes me wonder if square and/or rectangular ported enclosures also have a similar resonance issue surrounding the port tube as in the sonotube design?
I had always assumed that any resonances created within the enclosures would usually be below the hearing threshold. Obviously they are not as your charting shows and you also noticed immediately upon firing the units up.
This is interesting stuff.
Oh, by the way the tubes and drivers look neat. Are you gonna use them as rolls on the floor or install them vertically?

The main resonance is an end to end resonance, and is determined by the length of the sonotube. If you model a box with Boxnotes ( collo's program ) you can see where the resonances stack up, it's a handy program to have even when building a rectangular box.

All boxes have resonances, I try to make a rectangular box so that the resonances are spread out and not all multiples of one another, just like room modes. Absorption will help in every box. Non-parallel walls will help to keep a standing wave from forming,

The tubes will most likely be used vertically, but I will do some measurements as well in room eventually when I get them set up. I have just ran wire through the ports, but eventually will be putting a speakon connector in the endcap.

 

[ISLAND1000]

Well then, is it possible what you've accomplished by the poly pillows is created another but shorter reflective acoustic surface including the rounded mouth of the port? A shorter reflective length would raise the HZ past the pass band ending up somewhere in the 175hz range?

 

[SteveCallas]

ISLAND, no, that shouldn't be what is happening. The pillows aren't dense enough to reflect a wave, rather, they are significantly reducing the magnitude of the wave from endcap to endcap through absorption. This is why we line the walls and ports of ported enclosures with absorption.

Michael, excellent job on that port flare :T Even if the port OD came in a little small, you should have plenty of cross sectional area for a TC-1000 15" tuned ~17-18hz......plenty.

Looks like you should be able to go with an 80hz crossover as opposed to 60hz if you wanted. Dual 15s should be much more capable from 60-80hz than any sane set of mains - I'd attempt to get as much out of them as I could.

 

[ISLAND1000]

ISLAND, no, that shouldn't be what is happening. The pillows aren't dense enough to reflect a wave, rather, they are significantly reducing the magnitude of the wave from endcap to endcap through absorption. This is why we line the walls and ports of ported enclosures with absorption.

Normally, if the end cap was completely exposed without the intrusion of the port and flare that's what I would expect. But in Mike's project there is so little space between edge of flare and wall of the sonotube as to be a possible toss up as to what is actually reflecting a wave, the end cap or the port end and pillows or even a combination of all three? Thereby the shorter wave length causing any reflected wave to be above the pass band and not noticeable on the chart.

 

[michael hurd]

Sound is like water, it will move through the smallest hole. There is about 2 1/2 inches from the flange to the inside of the sonotube. The pillows as Steve mentioned are killing an end to end reflection, in the enclosure through absorption, as the wave has to make it's way through the pillows twice, once on the way to the far endcap, and once on the way back. As mentioned, they are not dense enough to cause an acoustic reflection.

I posted some more information on the AVS forum, and I will post it here as well. I measured the output of just the port, as well as did some experiments as to the number of pillows and their orientation in the enclosure. A single pilow in the center of the enclosure was enough to completely kill the end to end reflection, and was MORE effective than 4 pillows in the one endcap.

 

[ISLAND1000]

did some experiments as to the number of pillows and their orientation in the enclosure. A single pilow in the center of the enclosure was enough to completely kill the end to end reflection, and was MORE effective than 4 pillows in the one endcap.

Hmmm . . . . . . . very interesting.
Mike, you're pillows may give me nightmares. :foottap:

Now I wonder if centering (ALL) sound absorbent material in an enclosure rather than applying it to the outside wall, is a more efficient way to do this "interior reflection cancellation" stuff.

I'll go read the other stuff on AVS.

 

[SteveCallas]

Based on my own measurements (not outdoors unfortunately), lining the tube ID with a couple layers of poly batting and then wrapping the port in several layers of poly batting does the trick as well - no offensive peaks.

I gotta say that I'm still a little weary that pillows in the middle of the sub between the driver and the port would potentially decrease port effectiveness.

 

[michael hurd]

Steve, measurements show otherwise. Keep in mind that the air in the port is a mass... and is controlled by the 'spring' in the enclosure. You aren't really changing the air spring in the enclosure a lot.

 

[michael hurd]

Here's some more pictures and measurements. These are the pillows I used... 20" x 28" Jumbo, polyfill filled. Port output of the undamped enclosure in purple. Red is one pillow in the center of the enclosure, and the mustard color is with the 4 in the endcap. The microphone was placed inside the port, up to the XLR connector.

 

[michael hurd]

Here's some 'during' pictures. The enclosures wound up being right at 9.4 cu ft internally after driver and port, and with stuffing the tune ended up at 17hz.