JBL GT4-12 DIY Ported Home Theater Subwoofer


Active Member
About a year ago I had a go at designing a ported DIY subwoofer using a JBL GT4-12 subwoofer. Using WinISD I settled upon a 140L box tuned to 23Hz using a single 100mm round port measuring 264mm long.
Below are some details I created when designing the box:


After using the sub for a while, while reading some forums I discovered that ideally the port air velocity needs keeping below 20m/s (from memory). However at tuning this sub reaches 27m/s. I don't really hear any port chuffing until I watch movies with extreme low bass such as EOT, but it annoys me that this happens.

I started to look into what I could do to the sub to modify it, but keeping the current box dimensions. My thinking was to add a second port. This drastically reduces the port velocity down to 13.5m/s.

I then got to thinking (as I have a second matching driver laying around) that I could remove the current port and add a second driver in its place. I could then add a slotted port toward the bottom of the box. WinISD calculates a 270mm x 102mm port measuring 1199mm long which tunes the box to 21Hz. The max air velocity of this is 14.7m/s which would be fine.

To experiment, I also modelled the sub with 2 drivers in a sealed box. This is a viable but not preferred option as the low end drops off too much.

Below is a graph showing the 3 modelled configurations
Green = 1 driver and a second port
Blue = 2 drivers and a slotted port
Red = 2 drivers sealed

I've read about group delay of ported speakers and am led to believe that at very low frequencies, due to the wavelength, it doesnt really impact the sound, but I've included it below anyway:

For anyone who is interested the Thiele parameters for the GT4-12 are below

I would like to make my decision going forward based on the optimum modelled arrangement, but wanted to put the information out there for anyone who knows more about this to give me their view. I don't want to miss anything again like I did last time with the port velocity.


Distinguished Member
A 1.2m long port is certainly quite the pipe, the 1st port resonance will be quite low so you should consider where that is and whether it's in the intended pass band. Can you actually build such a thing into the current box? It's not immediately apparent how that fits without multiple folds. Finally look at excursion as well at realistic power levels (and with any DSP you need/intend to add applied)


Active Member
Hi Matt. Thanks for the reply. I believe I can achieve building the port in the current box by running it along the bottom and up the back of the box.

The 1st port resonance is at 143Hz which is well above my crossover.

The cone excursion at 400W is the blue line below:

If I add a Butterworths 2nd order Highpass at 17.5Hz, the excursion is as follows:

Is there anything else I may have missed, or could cause a problem?


Active Member
I'm probably going to be looking at implementing this mod to my sub in the next few weeks, so in the mean time, please could someone put my mind at ease that I haven't missed anything? I know these arent going to be ground breaking subs, but I've modelled them next to a Marty cube with a B+C 18TBW100 (which I'm considering doing) and they hold up pretty well.

The latest video from AVForums

Samsung S95B Update, B&W Panorama 3 & Q Acoustics Reviews, HDR PQ EOTF/Gamma Calibration Discussion
Subscribe to our YouTube channel
Support AVForums with Patreon

Top Bottom