High output subwoofer enclosure matched to your car
How to make a high output subwoofer enclosure
The whole point of a high output subwoofer enclosure is that it sounds loud. We don’t care about linearity, transient response or anything else. The goal is to make some noise with the little that we’ve got, by making clever use of sound mechanics. I will take the subwoofer which I use in my car, and put it under test. It’s in a closed fiber glass box in my current setup. After that, we will make a high output subwoofer enclosure for it, and we shall see how much we can squeeze out of it.
Audio setup :
- JL 10W6v3 subwoofer (10″ 600W).
- Audison SR 1DK amplifier (900W @ 2 ohms).
- BMW 5 series F10 sedan.
- Stock head unit with Audison bit ten signal processor.
There are several steps we need to take before we get to the end result. I’ll make a list with what we are about to do, so you can get an idea of what’s to come :
- Measure the T/S parameters of the speaker. After I measured the parameters, they were pretty much matching with what the manufacturer declared in the tech sheet. Good job JL! If you don’t have means of measuring, just use the parameters quoted by the manufacturer.
- Measure the car’s transfer function and find which frequency gets the biggest boost courtesy of cabin gain.
- Design an enclosure with a Cebyshev frequency response (high peak near resonance).
- Tune the enclosure according to the number found at step 2.
- Measure the high output subwoofer enclosure after it’s complete.
- Tweak the tuning frequency if necessary.
Before we begin, I am going to measure the SPL of my sealed enclosure. I know it’s not fair to compare sealed with bass reflex, but under normal circumstances, the difference should be +3 dB. So anything above +3 dB is the “high output” part. Plus, having the same setup with the same volume settings, makes a very good comparison between enclosures alone. The enclosure is around 20 L of volume with no stuffing inside.
IMPORTANT! How I measured the SPL
Before I show you the actual numbers, I want to be clear on how I made the measurements. Excessive volume of any sort is not used, sine tones or any other speaker abuse. I’m just playing the subwoofer loud, within its Xmax rating. I’m not risking the life of my perfectly good subwoofer, just to get a high score. So don’t judge the absolute values of the measurements. If you do so, I’m sure many of you will say that you got a much better score. The point is to compare the scores between the two enclosures. The measurements were done playing Tyga – Glitta, because it hits hard on the frequency I am interested in.
Here is the measurement of the closed enclosure :
The audio system as a whole is oriented towards sound quality, with little regard to sound pressure level. While 124 dB is underwhelming, it’s pretty respectful.
Measure the car’s transfer function
I made a whole separate article on how to do this measurement and you can check it out here. But if you don’t feel like reading it, I’m going to make a brief description of the process.
- Make / borrow a sealed subwoofer with very linear response (Qtc of 0.7).
- Take a near-field measurement, so you have an anechoic response of the subwoofer.
- Place the subwoofer in the trunk and the measuring microphone on top of the dash.
- Take a frequency sweep measurement and scale the graph (depending how far you placed the microphone from the subwoofer).
- Compare the 2 graphs and see which frequency gets boosted the most.
On the last graph you can the see the anechoic response of the woofer (the yellow line) and the same woofer measured with the car response (the green line). You can clearly see a decent bump at 34 Hz. I like to call that the car resonant frequency, but it’s a little more complicated than that, since the car is made by a lot of individual objects with their own resonant frequencies. Now, all we have to do is to make our high output subwoofer enclosure peak at 34 Hz.
Designing the box
The design of a high output subwoofer enclosure is pretty straight forward. You make a box as big as you can, and then tune it to your specified frequency. The tuning frequency is given by the size and length of the port. The over-voluming of the box is to ensure we get a Chebyshev response. This means that we will get a high peak at resonance. The bigger the box, the bigger the peak. On one hand, this will have disastrous effects on linearity (which we don’t care for), on the other hand, it will have a beneficial effect on efficiency.
Since the car resonant frequency is at 34 Hz, probably your first assumption is that the enclosure should be tuned at 34 Hz as well. That assumption is incorrect. If you look at the response chart of a particular speaker-box system, you will notice that the resonant frequency is lower than the point where response starts to roll-off. People consider that the resonant frequency coincides with the point where the response starts to roll off. This is not true, especially for non linear responses, which we are discussing.
So, to have the maximum magnitude at 34 Hz we need to tune the box about half an octave lower than that. 26 Hz to be exact.
Modeling the response of our box
The box will have 98 liters of net volume. I subtracted from the overall volume 3 liters for each port and 2 liters for the driver displacement. Bracing has negligible volume. You can be very strict with exact volumes, but it’s not that serious with these high output boxes. The ports are 100 mm in diameter and 400 mm in length (flared at both ends). This will give me a tuned box frequency of 32 Hz. This tuning frequency is true for QL = 7 (box losses). But I know that I like to abuse the silicone sealant, and I always get a box with low losses. Lower losses will pull down my resonant frequency and hopefully we will get around the 26 Hz mark. You can read more about box losses here. Box losses are unpredictable, but I’m feeling confident about my decision.
Above we have the modeled frequency response of the high output subwoofer enclosure. Notice the +7 dB peak.
For our project, I’m using 22 mm (7/8″) thick MDF, double baffle, 1 horizontal brace, and 2 ports 100 mm (4″) in diameter. As for volume, i didn’t go as big as my trunk allowed, because i didn’t want to have any surprises that it doesn’t fit in the boot. Pay careful attention at the height of the box. Make it somewhat shorter than the trunk height. Otherwise, when you slide the box in, you might get it stuck.
Here are some pictures with the build :
Since the baffle is mostly cut out (2 ports and 1 speaker), the structural integrity is severely affected. The double baffle is to add more strength and the horizontal brace is to reduce flex. Don’t ignore this aspect, as it’s especially important in a high output subwoofer enclosure. Make sure the back of the ports are not to close to the back wall. In my case there is a 15 cm (6″) clearance, which is more than decent.
Measuring of the box
Now that the box is finished, let’s start measuring it. After I did an impedance sweep, I could tell the resonant frequency of the box and could measure the box losses. As I predicted, the box had a low loss figure. QL = 16 (higher number means lower losses) indicates that the resonant frequency should be lower than 32 Hz.
Resonant frequency is 28 Hz, but we were aiming for 26 Hz. Let’s see the frequency response chart, and then we can make a decision if we tweak the resonant frequency or not.
Turns out that the resonant frequency of 28 Hz is spot on. We get the highest magnitude at 34 and 35 Hz. This means there is no need for additional tweaking, since 34 Hz is our goal. If you need to raise the resonant frequency, you can reduce the length of the ports by cutting them. If you need to lower the resonant frequency, remove the rear flares, add 90 degree elbow joints and continue with PVC pipe (depending how long you need to make the ports). Now let’s get to measuring the sound pressure level :
We can see an increase of 10 dB compared to the sealed enclosure. As perceived loudness it’s about twice as loud. You can clearly observe in the spectrum analyzer below the decibel rating, the peak at 34 Hz. The peak is also present for the closed enclosure, because the cabin gain is dominating at that frequency. But by matching the enclosure’s tuning frequency with the car’s cabin gain, we can get an impressive boost at that particular frequency.
The improvement in sound pressure was considerable. However, all things being equal, a bass reflex enclosure will output an additional +3 dB, because it’s a more efficient design. But, the extra +7 dB are present because we chose a non-linear design with a high peak near resonance. Also, this peak was matched with the car’s transfer function to give even more output at that frequency. Bear in mind that the sealed enclosure benefits from cabin gain as well. While the enclosure will sound loud at 34 Hz, the linearity is severely affected and it’s not recommended for someone who is interested in high fidelity audio. But we all know bass-heads are not interested in linearity and all they want is to see their car shake.
Special thanks to Accuwave for providing the sound pressure meter, which I used for obtaining the decibel readings. Dead-easy to use : stick it on your windshield, connect it to your phone, and you’re good to go. If you need one, have a look here. A must-have gadget for all of you bass-heads out there.
- Image source : link.