aiming the subwoofer box

Well, as many of you know, the argument of whether the direction the subs are aimed in the trunk changes the decibel intensity or not is quite a heated argument. The guy over at installer.com did an excellent writeup on the topic (link)

Nonetheless, being a senior in high school, I had to do a science fair project and thus chose to complete a similar experiment. Although I'm sure to many of you my results aren't anything ground breaking, after winning Best in Fair out of over 500 entrants I just figured I'd share my results with the community. I'm more than open to comments and criticism as to how to improve the experiment / results...

Setup: Two 12inch 4ohm DVC Kicker CompVR speakers wired in parallel, mounted in a 3.68243 cubic foot ported box (sorry, don't have port dimensions atm). I recorded a 65hz sine wave at 0db and played it through the head unit.

Amplifier: Orion 300D

Results (db(c) vs. frequency):

Trunk-facing:



Cabin-facing:


Merged:



Analysis:
Cabin-Facing: Reflected Wave causes destructive interference because of its out-of-phase reflection



Trunk-Facing: Reflected Wave is nearly in-phase and thus causes constructive interference and a higher intensity wave

Edit: Apparently (as of this moment) the larger-size pictures are not working...I'll fix them tomorrow after school (around 2pm est)...sorry for the delay

If a moderator could edit the original post for me that would be awesome...here are the updated pics:

Nothing's working for you quite right HERE, but congratulations on winning best in show at your fair. Please tell the forum how you accomplished the testing.

-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.

oooohh, the forums were filtering the url of my website (which hosts the pictures) because it has the "s" word in it

here comes another try:

Trunk-aimed trial:


Cabin-aimed trial:


Merged results:


CAD drawing of cabin-aimed setup wave (blue = initial wave, red = reflected wave)


CAD drawing of trunk-aimed setup wave (blue = initial wave, red = reflected wave)


Procedure:
First, I folded down the back seats so that there would be the least resistance between the microphone and the speakers. I then moved the subs so that they were located 4.5 inches from the back wall of the trunk, and aiming backwards (away from the interior of the car). Then, I sat in the drivers seat with a laptop and taped the microphone to the center console (so that the location would remain constant). On the laptop I had "DSSF3" software, which has a built in oscilloscope so I could see the wave as it was picked up by the microphone. Then, after burning the 65hz sine wave I generated onto a cd, I played the cd through the head-unit on "Level 8." While yes, level 8 is an arbitrary number, by using it I knew it would remain at a constant output level and be the same for all of the tests. I then took a screenshot of the oscilloscope, and saved it for reference. Following that, I rotated the speakers 180 degrees so that they were facing the exact opposite direction (i made sure the location of the box remained at 4.5 inches from the back wall of the trunk) and once again, played the cd and recorded the results. I then did the same exact tests, except this time with the head-unit on "Level 15." These results, however, I did not include because the mic was overloaded by the volume and thus gave skewed results.

I then decided to draw the actual wave so I could provide a physical image as to what was happening. Knowing that the frequency was 65hz, and the speed of sound is ~1087ft/sec, I could conclude that the wavelength was 200.6769in (velocity=(frequency*wavelength)). With this information, I loaded up AutoCAD and drew the exact wave as it was projected out of the box for both the cabin-facing and trunk-facing trials. Seeing the wave in the cabin-facing trial, I noticed how out-of-phase the reflected wave was when it bounced off the back wall in the cabin-facing trial. Contrastingly, I also saw how much more in-phase the initial and reflected waves were in the rear-facing trial, and could therefore conclude that the waves were causing constructive interference and a resulting higher intensity output.

Conclusion: Aiming your subwoofers towards the trunk and placing them as close to the back end of the trunk as possible produces the highest intensity wave.

I'm still skeptical. I know that I have heard the "aimed front/rear" differences, but I am not terribly certain why. I see it in peoples posts, all over the 'net, but until I do it for myself, I'll just remain jaded.

I'm seriously going to have to figure this out for myself, someday. If it stops raining here in the valley by the weekend, I might just go out and try some of these experiments myself... Maybe saturday'll be my "someday". I've got my TrueRTA, my ECM8000, my USBPre, my Philips Aurilium, my laptop, my latest subwoofer build, my oscilloscope, and my 100A milspec power supply, with adjustable voltage outputs.

I'll get to the bottom of this! LOL More details to follow...

-------------
It all reminds me of something that Molière once said to Guy de Maupassant at a café in Vienna: "That's nice. You should write it down."

I need a better feel of what I'm looking at in those two diagrams with the black background.  In each one there is a green arrow.  What does that represent, exactly, and what should I look at that is representative of the 4.5" space?

The other question that came to mind is:  what is the depth of the speaker box, and does that have an impact at all with this test?  When you rotated the box 180 degrees, the driver was moved to a position relative to the box depth.

-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.

In the two diagrams with the black background, the "green arrow" is actually just a poor representation of the speaker inside the box. I figured it would be more clear showing a make-shift speaker inside the box than just having sine waves shooting out.

As for the 4.5" space, that's just showing that the distance between the box and the back wall of the trunk remained constant throughout the tests. Knowing that, I believe that also answers your last question. The box was always 4.5" from the back wall of the trunk, even after rotating it. Looking back, I'm wondering if it would have been a better idea to leave the driver in a constant location, rather than just the distance from the back of the trunk. I might have realized that earlier on, except the box barely fits in my trunk as it is, so I know that keeping the drivers in a constant position is out of the question for me.

EDIT: If the drivers were in the same position, then I am 99% sure the decibel output would be the same. HOWEVER, in the trunk of a car, it's impossible to get the driver as close to the back wall of the trunk in a cabin-facing setup than in a trunk-facing setup. The main idea is to try and get the speaker as close to the back of the trunk so that the reflected wave is reflected as soon as possible and bounces into an in-phase pattern.

I'm looking at your black images again, and you show that the wavelength is 200 inches, which you stated was 65Hz. (It's a little off, as the speed of sound at STP is 1127 ft/s - so the actual wavelength is 208 in. or 17.3 feet) At a wavelength like that, the interior of a car CAN'T cause a reinforcement... I honestly don't see how it could. ANY time the cabin/venue becomes smaller than the wavelength, you are now dealing with pressure, not sound (if that makes any sense...). Not only that, but if your box is not large enough to move the woofer 1/8 of one wavelength, I can't see WHY it does what everybody SAYS it does...

Like I said... experimentation will be in order this weekend...

-------------
It all reminds me of something that Molière once said to Guy de Maupassant at a café in Vienna: "That's nice. You should write it down."

Gotcha, inflames989.  Now that I see that green symbol as a woofer instead of an arrow, it becomes clear.  But the box depth should be defined so that it can be added to the to-rear distance measurement of 4.5" that you defined as a control in the experiment.

(Wonder what rear porting would do to that nicer set of soundwaves??  Looks like a good case for ports on the face of the box, by this demonstration.)

Here's one up for grabs:  a polyfiber-filled sealed enclosure actually (apparently, if you wish) slows down the backwaves, making the box air volume act as if it's somewhat larger.  As far as I know, there is no science (proven with math) to demonstrate to what degree this "slowing" action occurs.  It is estimated at somewhere between 10% and 30% greater box volume if polyfill is used, from what I've read.

Now, the graphs with the black background in this experiment are projections of soundwaves based on math, not actual recorded results, if I understood the above explanation correctly.  Consider this:  a scenario where the box, like many enclosures, is filled with polyfill or Acousta-Stuf fibers.  And, consider that there is no math to explain a difference in backwave emanation from the box (if indeed there IS a difference) because of the fill.  If the box were filled with fiber, would this projected result (which is based on math) be accurate? 

Also looking forward to your real-world experimenting this weekend, haemphyst.  This is a good thread... and thanks, inflames989, for sharing your project with the forum.

-------------
Build the box so that it performs well in the worst case scenario and, in return, it will reward you at all times.

the box I used actually has that...stuff...in it...I'm not sure if it's the polyfill or Acousta-Stuf fibers, but it has the filler material in it. (I know that sounds vague, but I really really really REALLY can't think of what it's called at the moment, and I'm not even sure if it's the same stuff you're talking about)

stevdart wrote:

---

Now, the graphs with the black background in this experiment are projections of soundwaves based on math, not actual recorded results, if I understood the above explanation correctly

---

This is actually a good point to keep in mind. The diagram is a picture of what theoretically should be happening.

Also, another thing overlooked is the passing of the wave through the 3/4" mdf. When a wave passes through a more dense material, it refracts towards the normal line drawn at the intersection point. I did not take that into consideration when drawing the diagram. However, since the same material and thickness is used throughout the experiment, it should not alter the results I got because each diagram would be altered equally.

Haemphyst, I am also looking forward to your tests this weekend