3D Printed speakers - Unlock impossible designs compare to conventional

[jaakkopetteri]

I think that RDphysics build is pretty questionable. The cabinet looks good I guess, but the measurements they provide tell us basically nothing.

What do you mean they tell us nothing? I don't see anything questionable about the build, even if there's not a full Erin set of measurements

 

[kemmler3D]

What do you mean they tell us nothing? I don't see anything questionable about the build, even if there's not a full Erin set of measurements

They're super smoothed, in room, not gated... So we can't tell what's the room and what's the speaker. Can't say if the performance is good or bad from those graphs, but it looks bad. There's apparently some directivity error even despite the coax and rounded cabinet. And no distortion graphs, so who knows if it's straining or not.

And then they don't even provide real crossover settings, they just suggest 2khz as a starting point?

At that point it's not even a real DIY plan, just an example to use for inspiration.

 

[jaakkopetteri]

At that point it's not even a real DIY plan, just an example to use for inspiration.

Hardly "super smoothed" and the directivity measurements are gated, and gating isn't strictly required either. For a coaxial used off-axis the directivity seems very decent. SEAS coaxials (and the SB acoustics ones) are pretty proven anyway. Yes, it's not a complete plan, but I don't see how that makes it questionable. Poor measurements or other obvious flaws would make it questionable.

 

[kemmler3D]

Hardly "super smoothed" and the directivity measurements are gated, and gating isn't strictly required either. For a coaxial used off-axis the directivity seems very decent. SEAS coaxials (and the SB acoustics ones) are pretty proven anyway. Yes, it's not a complete plan, but I don't see how that makes it questionable. Poor measurements or other obvious flaws would make it questionable.

I think you may be reading my use of the word "questionable" as worse than I intend it. I just mean that it is not obviously a good choice, it might be a bad choice to build it.

However, there are some clear flaws visible in what they share.

Some of the measurements use 1/3 octave smoothing, which is more than you usually see in these contexts.

Aside from that, there's a 5dB on-axis boost from 5-8Khz and directivity errors in the same region. This suggests a lack of effort on the crossover but also the driver itself probably is to blame here. So why did they choose this driver over other coax drivers out there? There may be good reasons for this trade-off but it's not explained. The bass also drops like a rock below 100hz which suggests the driver isn't capable or the enclosure was designed without regard for bass response. Either way... why? It's only a 5" driver but given that they're specifying a DSP crossover you'd think they could try for more.


I also think their use of a casual trademark for "minimal edge diffraction enclosure" is cringe. It's not like they came up with the idea of roundovers.

Just compare this project description to something like Mechano 23 or Purifi SPK16. There is much more explanation of the design and the crossovers are at least designed for the speaker. This feels like a speaker project aimed at 3D printing enthusiasts, not a 3D printing project aimed at speaker enthusiasts, if that makes sense. I stand by my assessment of it being questionable.

All that said I guess my point is, if you know something about speakers you can probably do better than this. It's a good example of a printable enclosure but not a great example of a DIY speaker project.

 

[somebodyelse]

Just compare this project description to something like Mechano 23 or Purifi SPK16. There is much more explanation of the design and the crossovers are at least designed for the speaker. This feels like a speaker project aimed at 3D printing enthusiasts, not a 3D printing project aimed at speaker enthusiasts, if that makes sense. I stand by my assessment of it being questionable.

Or even compared to some of the 3D printed ones like the LCM that Erin measured - fairly conventional looking outside, but with the printing helping the internals (moreso on the Evo1 iteration) and the port, and an unconventional driver choice. The sides can even be conventional sheet material to save on some filament and printing time. Or these and their earlier iteration, both with integrated augerpro waveguides, large radii, some degree of modularity, and an experiment with resin fill. None of them are perfect, but it's a lot easier to come to a reasonable judgement on their imperfections.

Edit: I meant to mention the others in #99 too, many of which had more detail.

 

[Paul7777x]

I’m as ignorant as it gets when it comes to anything 3D printing. But I’m certain the possibilities of it for remarkably good and inexpensive speakers has not even been touched.

As it becomes cheaper to buy larger machines and new filament tech is developed, I expect to see some wonderful looking creations.

 

[kemmler3D]

@somebodyelse agree, lots of good examples out there that at least justify the design more.

As an aside and thinking more about 3D printed coaxial speakers... what would be really interesting from a 3D printing perspective is "make your own coax"... you would need a tweeter that can cross pretty low, but I think it should be possible to stick the tweeter in front of the woofer with a printed part. Something that is extremely hard to do with MDF, but at least feasible with printed parts.

 

[a4eaudio]

...but I think it should be possible to stick the tweeter in front of the woofer with a printed part...

I have seen some DIY coaxials with wooders employing a phase plug....the phase plug is removed and a tweeter installed. Since the tweeter-to-woofer transition may not be good enough, they created a 3D printed adapter for a smoother transition.

 

[jaakkopetteri]

Aside from that, there's a 5dB on-axis boost from 5-8Khz and directivity errors in the same region. This suggests a lack of effort on the crossover but also the driver itself probably is to blame here. So why did they choose this driver over other coax drivers out there? There may be good reasons for this trade-off but it's not explained. The bass also drops like a rock below 100hz which suggests the driver isn't capable or the enclosure was designed without regard for bass response. Either way... why? It's only a 5" driver but given that they're specifying a DSP crossover you'd think they could try for more.

The on-axis boost seems to disappear quickly, as is typical with coaxials, so it might not be much of an issue, and it's above the usual more problematic midrange anyway. I don't see an issue with the bass dropping off either, it's pretty reasonable for a small enclosure with coaxials to prevent modulation from the cone excursion (as I'm sure you know)

I also think their use of a casual trademark for "minimal edge diffraction enclosure" is cringe. It's not like they came up with the idea of roundovers.

I feel like you're taking that a bit too seriously, and the shape is definitely more complex than the typical prismatic enclosures with straight fillets that most people print

Just compare this project description to something like Mechano 23 or Purifi SPK16. There is much more explanation of the design and the crossovers are at least designed for the speaker. This feels like a speaker project aimed at 3D printing enthusiasts, not a 3D printing project aimed at speaker enthusiasts, if that makes sense. I stand by my assessment of it being questionable.

Yeah, this is getting semantic, but I feel like the project is more of an inspiration to showcase 3D printing and criticizing it for not being a complete project isn't fair. Every single 3D printed build out there is "questionable" if we set the bar at the SPK16

 

[Reydepancakes]

https://www.reddit.com/r/diyaudio/comments/1tkzz40/202605_continuation_of_my_solo_bifide_project

Nice interesting build.

 

[CJH]

Add a Pair of Halcro amps and you'll have an awesome looking set. CJH

 

[JohnnyNG]

As an aside and thinking more about 3D printed coaxial speakers... what would be really interesting from a 3D printing perspective is "make your own coax"... you would need a tweeter that can cross pretty low, but I think it should be possible to stick the tweeter in front of the woofer with a printed part. Something that is extremely hard to do with MDF, but at least feasible with printed parts.

Would be fun to see a DIY all-Purifi coax!

 

[kemmler3D]

Would be fun to see a DIY all-Purifi coax!

Could be kind of problematic with the large waveguide, but otherwise would be really interesting...

 

[aschen]

Hexibase latest project 5ft tall printed line source look really cool. I am not sure how I feel about line source, but I always like his no fluff videos and sense of humor. A lot of PLA was harmed in the process.

 

[Paul7777x]

Hexibase latest project 5ft tall printed line source look really cool. I am not sure how I feel about line source, but I always like his no fluff videos and sense of humor. A lot of PLA was harmed in the process.

I’ve not seen this guy before.

Quite extraordinary. Cheers for that