I Tested 12 Different Designs So You Don't Have To (Gyroid Comparison Inside)

Ajouté : 2026-05-22

[00:00:00]Last time, I said the future speakers might not be a box, but there's a big difference between theory and proof. In the world of 3D printing, there's a dangerous trap. Just because you can make something weird doesn't mean that you should. A sphere looks cooler than a cube, a pyramid more interesting than a rectangle.

[00:00:18]And a gyroid lattice looks like it's straight out of a sci-fi movie. But, does any of that actually make a speaker sound better? Or am I just designing the world's most complicated way to hold a driver? Today, [music] we find out. I'm putting it all to the test. Same driver, >> [music] >> using the same material with the same internal volume, but completely different shapes. We're going to build them, measure them, and with real-world data, find out how much enclosure geometry actually matters.

[00:00:52]So, what's the problem we're trying to solve? We need to start there. Speaker boxes are boxes for one main reason.

[00:00:59]They're easy to manufacture. But, that simple shape comes with acoustic compromises, internal reflections, standing waves, and sharp edge diffraction. Now, 3D printing lets us break free from the box, but that freedom is often a double-edged sword. A weird shape can introduce new problems just as easily as it has solved the old ones. And that's why this test is so important. I'm putting six designs to the test, all printed with the exact same internal volume, 0.4 L, using the same Dayton Audio RS175 driver.

[00:01:34]And by keeping the driver and the volume the same, we can isolate the one variable that matters, the shape itself.

[00:01:43]First up, our baseline, the cube. This is our control group with all of the classic problems, flat parallel walls and sharp edges. Every other design has to prove it is better than this. Next, the sphere. Acoustically, this should be a cheat code. No parallel walls inside and no sharp edges outside to diffract sound waves. Spheres are a nightmare to make with wood, but fairly trivial for a 3D printer. Now, our third contender is the pyramid. Its angled walls aren't parallel, which should break up internal standing waves, and this is our weird, but still geometric option. And then we have the cylinder. It gets rid of the vertical edges, but still has flat parallel ends, and its curved inner wall is interesting [music] because it could actually focus sound internally, potentially trading one problem for another. Next, the hexagon. This is a practical 3D printed shape. It's made of flat faces, and so that makes it easier to print than a sphere, but the angled sides should prevent the simple reflections found in a cube. And our sixth contender, the organic shape. Now, this was my what if the whole box is a diffuser experiment. A complex flowing form with no obvious flat walls. Now, finally, the wild card, the lattice interior. [music] This isn't about the outside shape, but what's inside.

[00:03:06]A gyroid lattice can act as both bracing and a diffuser, and for this test, every enclosure will be tested [music] twice, once hollow and once with the lattice structure inside. Now, designing these in Fusion was a challenge, especially getting all of the internal volumes the same.

[00:03:23]Printing them was a marathon.

[00:03:25]I used PLA with the same wall counts for a rigid, airtight enclosure, and this was days of printing. And while PLA is great for prototypes, for more serious materials, our sponsor, PCBWay, >> [music] >> offers on-demand 3D printing in nylon, resin, aluminum, and even titanium. So, if you need a stronger enclosure or custom CNC part for your own project, they can turn a design into a real part.

[00:03:50]You can check them out at the link in the description. But, for this test, I controlled every variable. Now, how do we measure the difference?

[00:04:02]Saying something sounds good is subjective. I want cold, hard data, and my setup is built exactly for that. The microphone is a calibrated UMIK-1 made for accurate acoustic [music] measurements. It feeds into a powerful piece of free software called Room EQ Wizard or REW. Now, REW generates test tones and creates graphs that reveal the truth about our speakers. I'm also using the DATS V3 from Dayton Audio, which measures a speaker's impedance. That's its electrical resistance that changes with frequency. Now, this reveals resonances, tuning issues, and even air leaks. And we're going to look at four key measurements. First, frequency response. REW sweeps a tone across the hearing range to measure the speaker's loudness at each frequency. A perfectly flat line is the goal, and we will measure center, 15°, and 45°. Second, impedance response. A smooth curve is a happy curve. Now, sharp spikes tell us the enclosure itself is vibrating or resonating.

[00:05:06]Third, total harmonic [music] distortion or THD for short.

[00:05:11]This measures the tiny unwanted notes a speaker accidentally [music] creates.

[00:05:15]Lower is always better. Finally, the waterfall plot. This 3D graph shows us how sound dies [music] down over time.

[00:05:22]We want sound to vanish instantly, but long ridges on this graph mean audio will sound smeared and unclear. And with this suite of tests, we can get a scientific look at how much shape really matters.

[00:05:38]Okay, our contenders are built and the test bench is ready. If you're enjoying this deep dive into 3D printing and audio, [music] consider subscribing. It's a huge help, and you won't miss any of the other crazy experiments I have planned. Now, let's get to the results. I tested every enclosure in the same spot in my room.

[00:05:57]Now, it's not an anechoic chamber, but what matters is the comparison. If they're all tested in the same imperfect room, the difference between them will still tell us a story.

[00:06:08]Also, remember the gyroid infill takes [music] up space, making an enclosure act smaller.

[00:06:14]For it to be a success, it has to help more than it hurts. Starting with the cube, the results immediately showed this was not going to be simple.

[00:06:23]The empty cube had a smoother frequency response, but the gyroid filled one had a lower broader impedance peak, meaning it was better damped. For the cube, the gyroid was a clear trade-off. It controlled resonance, but at the cost of our frequency accuracy. Next, the hexagon, [music] and this is where the gyroid started to make sense. The frequency response was slightly [music] smoother with the gyroid.

[00:06:49]Distortion was basically a tie and the impedance peak was knocked way down.

[00:06:53]This was one of the first [music] shapes where the gyroid looked useful without a major downside. Then came the pyramid.

[00:07:01]This one was weird. The empty version had a smoother frequency response and distortion was a toss-up. The empty version was cleaner in the bass, but the gyroid was cleaner in the treble.

[00:07:12]And the pyramid was a clear case of choose your compromise. The cylinder was another oddball.

[00:07:19]The empty version had a smooth response head-on, but the gyroid version became more consistent as I moved the mic off-axis.

[00:07:28]And the results were mixed. The gyroid helped a little, but it wasn't an >> [music] >> astounding breakthrough. Then, we get to the sphere. The empty sphere was already one of our best performers with a smooth response right out of the gate.

[00:07:43]The gyroid did not offer much improvement. The takeaway, the sphere shape is already so acoustically sound that adding a complex internal structure didn't provide a major benefit. And finally, the organic shape. In frequency [music] response, the empty version was better, but the impedance measurement told a completely different story. The gyroid-filled organic enclosure had a much [music] lower impedance peak.

[00:08:10]Electrically, it was one of the clearest signs that the gyroid was damping the system. Now, looking across all six shapes, the biggest surprise was that the best enclosure depends on what measurement you care about. If I only look at the frequency response, the sphere and the pyramid were some of the strongest performers, especially without the gyroid.

[00:08:34]But, if I look at the impedance damping, the gyroid-filled hexagon and organic shape stand out at the top. And that's the important part. The smoothest response was not always the most damped enclosure.

[00:08:50]Here's the verdict. The gyroid, it was not a magic filter. It did not automatically flatten the frequency response, and in fact, some of the smoothest responses came from shapes that were already doing the acoustic work on their own.

[00:09:06]But, the gyroid often added damping and changed how the energy was stored inside the enclosure, which showed up most clearly in the impedance measurements.

[00:09:16]So, if you want the smoothest response, shape matters. If you want better damping, the gyroid can help, but you can't just stuff it into a random box and expect magic.

[00:09:30]So, the conclusion, do weird speakers sound better? The answer is a resounding sometimes. Shape for the sake of being weird is not a magic bullet, but what 3D printing gives us is the power to use shape and internal structure as a deliberate engineering tool. We proved that a sphere can sound better than a box because its shape is acoustically superior, and we proved that a complex lattice structure can have a real measurable effect on how the speaker behaves. Project started with a simple question and ended with a clear answer.

[00:10:08]The enclosure is not just the container for the speaker. It is the speaker, and it's a testament to what's possible when you break [music] free from the box and start designing with real acoustic goals in mind. But, thanks for joining me on this adventure, and be sure to [music] like and subscribe if you want to see more. Let me know down below what you'd like to see next time.

[00:10:32]And until then, thanks.