Aside from the OpenRC F1 car that I am printing I am working on a pair of Nautilus Shell Speaker. As a fan of the Bowers & Wilkins Nautilus speakers the minute I saw these on Thingiverse I knew I had to print them. Not only that, but the designer, Alex Mordue put them together specifically for printing on his BigBox. http://www.thingiverse.com/thing:1328923 The 'hot off the press' photo after a mammoth 32 hour print: An in progress photo: The above was printed with a 0.4mm nozzle at 0.2mm layer height at 40mm/s in E3D Everyday PLA. My plans are to sand the speakers and apply a primer, a further sanding and primer if required and then spraying with black car paint followed by a gloss lacquer. They will be fitted with 4" drivers and connected to a Class-T amplifier. I will post more photos as I print the second speaker cabinet, process them and then wire them up.
Any information yet on the enclosure volume? I've tried to find time to simulate the speaker but I usually forget or get distracted by other issues... (Like playing settlers of Catan with my son)...
A decent 3D-tool should allow you to isolate the inner volume using some editing and boolean operators, then you can get the enclosed volume very precisely. I assume the speaker does not need any damping material? Nevertheless, it might be beneficial for the acoustic quality to put it in an outer box and fill that with quartz sand. There is just one good measure to minimize wall oscillations: Mass. If that is not enough, get more mass.
I assume this is a closed speaker so recommended is to fill it at least 75 to 85%. Mass is of course the best way but that is unlikely since the BB build volume has set the max size. Unless the enclosure is way to big for the speaker there will be no more mass... Or... Print it with brass filament... Stiffness is also a solution to the problem. Move the resonances above the upper limit of the ear and the problem can be ignored... Not the best way. But one way...
That's why I proposed an outer sand mantle: Stuffing that shell volume (now I know why slicers are asking for the shell count!) with damping material could become a bit of a challenge. Print a damping structure?
sorry, as a non-native english-speaker, I am not quite sure what you mean by this question. If you aim at keeping the shell shape visible - hmm, yep, that would get lost in a box. But do you want to look at your speakers or listen to them? If both, build one dummy set for the eyes..
Rectangular loudspeakers create a natural resonant panel that needs careful management to avoid resonance but with tapered horn loaded loudspeakers with circular cross section there is little or no resonance provided the structure is stiff and this looks very stiff. Any filled filament should provide reasonable resonance control. Filling the inter wall gap may improve things.
With a quick wash in the bath I can see they're going to look great in black gloss. Even with a clear coat of lacquer I think they'd look amazing. Not sure if the camera does it justice: This is just the first sand in 320 grit. I will give it another sand and then a sand in 1200 grit. Then primer and a further sand.
Yes you understood perfectly. Dummy set would solve the problem... But somehow I do not think that is a solution that is acceptable... There is a few problems with speakers... One is resonances due to the cavity then the enclosure/material will introduce their... Let's just say that I'm not surprised that most commercially successful speakers are square boxes... Since this a broadband speaker then many of the simple strategies for handling resonances cannot be used... I hope to simulate today... Looks good. while you rinse, measure the volume. Since it is a closed speaker it do not have to be extremely precise. Just so I can put that volume in and se how the speaker will react. And what response you can expect.
Filling a multi-threaded shell with water without enclosing any air bubbles may end in some contortionist's exercises . Getting the water out again is a step simpler. Or use Archimedes' concept, measuring the displacement. Subtract the shell's wall. Or do it empirically based on what I suggested above. You could also integrate the cross sectional area over the length of the deconvoluted shell. I think the usual simulation programs like boxsim etc. are not very precise for non-square housings. When I designed an optical membrane position feedback for a speaker some 45 years ago (not unsimilar to the IR sensor for the bed zeroing), I was surprised that the speaker had no resonances any more - without any filling - and that was a box. Disabling the feedback loop led to the expectable non-linearities.
Could you not use CAD. You could subtract the model from a block and get a fairly accurate idea of volume that way?
I've started spraying some primer on the cabinet. Aiming to print the second one this Friday/Saturday.
Yes the camo vibe is cool. I need to work out how to fill some holes where there was too much overhang. Some epoxy resin or something maybe? I'm not sure what's best to use. Also about to set the second one printing!
