So, this is how the set up was done:

150 watts to an ARC 8 in an enclosure slightly too small for it. True RTA with external USB preamp, and Behringer MC8000 mic. Ran pink noise, and made sure the sub was in the same location, the mic, the volume, etc.

Placed mic on 2" of foam to prevent vibrations from interfering.

Ran a test with a good amount of polyfill, and one without. This is how it came out:



As you can see - with polyfill, we lost a few DB from 40-50db. However, from 50-55 or so, its a little stronger. The peak shifted slightly lower and to the right with polyfill.


From what it looks like, the polyfill actually shifted the response to the right, marginally. You get the same effect when you make the box smaller.

So in conclusion, it looks like Polyfill simply 'takes up space' and shifts the response slightly forward, just as if making a box slightly smaller.

So as far as using polyfill to make hte box seem bigger? Which would increase low end response and overall spl?

Busted.

 

hmmm.....I'm interested to see exactly how much polyfill they used and how tightly it was packed. Supposedly, the density of the polyfill per airspace volume is key to getting the "larger box" effect. If the density is too much or too little, you lose SPL

on a complete sidenote: DAMN i missed that show

 

Since the box is .3, I tried a few varying amounts, without much result. What you see is the result of the most amount I tried. Much less, and the curve looked nearly identical to the 'without' polyfill graph.


I didn';t mean to imply Mythbusters ran the test - I just put that in the thread title because they do this kind of ****

fwiw, I tested my idmax response too, not with polyfill, just to see. I have the same volume at 20hz that I do at 80hz, with a 5db peak at 44hz. Somehow, I made the box almost perfectly by accident.

 

.75lb per 1cu^3 generally results in about 20% increase in enclosure size. With a box that small I would have to think a very small amount of polyfill would be needed. Something in the range of about 2.7 ounces of polyfill.

Since you made the effort try that weight to see if the normally regarded chart still makes sense.

http://www.moodym.com/audio/fiber.html

Fej

 

I'm not exactly sure how much i put in mine, but it's plenty loud and certainly beats my old alpine SWS 1222D's by a long shot.

 

I had always used fill to reduce the boominess sound of the sub, to "tighten" it up some.

I've used eggshell matress padding, polyfill materials and even pillow filling, most had the same effect.

Its cool to "see" what it actually does to the sound though.

I see you have a couple threads with your sound graphing stuff now... is that new?

 

I've been a site lurker for a few months since a friend turned me onto this forum. I have a lot of audio experience to share so I thought I might make my first post as helpful as possible...

Unfortunately, a simple RTA isn’t generally capable of demonstrating this well-documented phenomenon. It is ABSOLUTELY true that stuffing a box can/will increase its apparent size as well as increase output efficiency in certain areas of the curve.

The first place this is evident is in the impedance curves (200 data points per curve, 20-200 Hz, actual measurements):



The resonant frequency shift is obvious, as is the change in the Q factors. This particular driver has Qts=0.46 and Fs=29.3. When it’s loaded into a 1-cubic foot box, the Qcb=0.81 and Fb=55.3. When the box is stuffed properly, these numbers change to Qcb=0.53 and Fb=49.7. Getting this closed box Q without stuffing to be near 0.53 would require a much larger box.

If that isn’t enough evidence, the following response chart should convince anyone (again, 200 data points per curve, 20-200 Hz, actual measurements):



Note how much the fill has smoothed the response. Also, note the large increase in output below 50 Hz. The 6 dB increase is similar to multiplying amplifier power by 4 times or more.

I’ve built countless boxes over the 20 years I’ve been into this stuff, and this condition is always considered in the design and tuning process. While this driver provides an extreme example, this condition applies to all low-frequency drivers in boxes.

It’s not a myth.

 

Guys, just so you think Sprkboy there isn't some lurker looking to pick a fight, he's the "friend" I've often referred to in posts. He's been building car audio systems and speakers for a long, long time and really knows his stuff.

So feel free to give him a hard time about having to sell his '92 as a downpayment on his house, but don't think for a second he doesn't know what he's talking about on this stuff.

 

Welcome spkrboy. I am with you as well and I think that Kale simply overstuffed his box effectively reversing the benefits of polyfill. I also include the polyfill into my enclosure specs, although I try and get within 20% of my recommended "ideal" enclosure as I know the value of polyfill that nets me around a 20% volume gain. I am still waiting to see kale's retest with much less polyfill.

Fej

 

Thanks, fej... and thanks Komo, for the props.

The real problem Kale has is that an RTA isn't designed to measure sound in a way that easily shows these changes. It has to be done in a certain way where reflections and other contributors are eliminated. An RTA only shows you total sound energy at a given moment in time at a particular point in space. This is also complicated by the impedance changes the amp sees, which further muddies what the RTA measures. Even if he doesn't touch the amp volume control at all, the current and voltage will change simply for the change in impedance.

There's a ton more to this equation than an RTA can measure. I wouldn't be at all surprised if he had the same results no matter what level and type of stuffing he uses in this system.

 

welcome Spkrboy.

Apparently i have a lot to learn about audio....mostly testing/tuning equipment.

i feel like a noob for asking this.... but what the *^$* is "Q"?

(I'm so glad i don't have to ask the OT guys this or i'd get 100 replies that it's a "god-like" entity from star trek)

 

You'd get MRVETTE saying subwoofers can't be loud in cars, and its impossible to play under "80 hz doubled" in a car.

 

I've know him for nearly 15 years and when he starts talking in those terms my eyes STILL cross.

But he keeps repeating and I soak a little more in each time.

 

great to have such an addition to the forum. welcome

 

I am still confused... I have never understood exactly how polyfill could make the box 'bigger'...

 

Thanks, DPG...

Those of you who understand this stuff already… please don’t flame the lack of detail here. This is just a quick blurb to bring others into the fold.

“Q” is a term used to describe a factor of electro-mechanical systems (a loudspeaker driver is one example). Technically, it’s the energy stored per cycle divided by the energy dissipated per cycle.

In terms of loudspeaker drivers, Q is a description of the electrical and mechanical damping factors of a driver alone as well as in combination with boxes and/or other drivers and crossovers. When you look at a raw driver’s specs, the Qes is the electrical Q, the Qms is the mechanical Q, and Qts is a combination of the two called the “total system Q.”

The idea behind building a box for a driver is to make the box in such a way as to augment the driver’s damping ability. You build a closed box for a driver with a target Q in mind, called the Qcb (Q of the closed box). For a given driver, the smaller the box, the higher the Qcb will be. The larger the box, the lower the Qcb will be. There are a number of equations and simulators to help in this design process.

Ideally, you’d build for a target Qcb somewhere between 0.5 and 1.2 depending on your musical taste and the acoustic environment the speaker will play in. Think of a high Q as “fat,” “slow” or “boomy” and a lower Q as smooth, tight and fast.

Rap and dance fans will prefer a high Qcb, perhaps 1.2 to 1.5. Rock fans will prefer something like 0.7 to 1.0, jazz fans around 0.7, and classical 0.5 to 0.7. The higher the Qcb, the more attention the subwoofer will bring to itself, but the less low bass it will produce. A Qcb of 0.707 is called a Butterworth response. It has a maximally flat response and the lowest cutoff frequency. In most practical instances, this is the largest box you would want for a given application. Anything larger will generally cause a loss of efficiency and/or higher cutoff frequency. You can see from the charts in my first post that a higher Q causes a peak in the response and considerably less low bass. Most times that matters. Occasionally it doesn’t.

This is why you should not simply choose a driver for a system and run with it. Rather, the best method is to determine the size of the box you’re willing to make then choose a driver that performs well in that box. There are dramatic differences in driver performance in different boxes, and what sounds good in one application may be truly awful in another. There is no such thing as a perfect driver, no matter the application. There is no “best car audio subwoofer” or “best home subwoofer” in any meaningful sense.

***In relation to the thread here, box stuffing changes the apparent box size in part by lowering the Qcb.

Q is inescapable and rules everything in audio.

 

In th simplest terms i understand, the filler material causes the sound waves on the inside of the box to slow down as they move from the cone's underside to the edges of the box. This simulates having a larger box.

 

since I don't want to overstuff my box, what would you guys say is the right amount to put in a 1.9-2cu ft box (not counting driver displacement) with 2 12"s?

 

Gonna very much depend on WHICH 12's you're using.

 

...and the kind of fill you use, your intended goals for sound, your crossover point, the quality of the amp, the typical listening level, the equalization used, the environment it's used in........

I's entirely a trial-and-error process unless you have some basic test equipment to measure impedance (not at all hard to do), and even then it still requires adjustment by ear for taste.