PURIFI finally did a fully purified passive speaker design! The SPK 16 prototype is here - with a PTT tweeter

[Kaladin]

For the cardboard spools, you can print simple additional rings (which are reusable)—that way, this won't happen anymore. There are plenty of templates available on Makerworld.

For larger print projects, I often buy TINMORRY 2 kg CF PETG Rapid Filament on a large spool (black) when it's on sale for around €32. Of course, that doesn't fit in the AMS.

View attachment 530474

That's a good idea to save some money. Your prices look better but in the US it's about $8 less per roll for the cardboard rolls.

A picture, please?

I've attached a couple pictures plus a few older ones. The pieces are just stacked together for now. I'm not going to glue them up until after they've been filled.

 

[DatSpeakerTho]

Incredible. Almost makes me not want to cut wood anymore for speakers.

 

[Mattcc22]

That's pretty awesome! Once they are glued together and I assume you sand and maybe bondo the seams or something? I've never worked with 3d printed parts. Are you able to get rid of the seams entirely?

 

[Salt]

That's a good idea to save some money. Your prices look better but in the US it's about $8 less per roll for the cardboard rolls.



I've attached a couple pictures plus a few older ones. The pieces are just stacked together for now. I'm not going to glue them up until after they've been filled.

Appreciate the 'bed' on the back for former active enhancement

 

[kemmler3D]

That's pretty awesome! Once they are glued together and I assume you sand and maybe bondo the seams or something? I've never worked with 3d printed parts. Are you able to get rid of the seams entirely?

Pretty much this. Some people use wood filler, others use filling primer (spray), but it amounts to filling in the gaps and sanding down. After a few passes you can eliminate the visual gap pretty well. The trick for a speaker cabinet is to make sure the glue-up has a fighting chance at being airtight. So basically very even glue coverage before clamping, and hope.

Update from my end, I have what I think is a 95% ready-to-go design that's slightly modified, I have a couple glitches in the model I need to resolve and then send to Lars for approval... it's meant to be printable on a 220mm bed, so most printers will be able to do it, albeit in 8 pieces.

I will also share a 4-piece version that should be printable on Bambu P1S which is what I use.

 

[Paul7777x]

Pretty much this. Some people use wood filler, others use filling primer (spray), but it amounts to filling in the gaps and sanding down. After a few passes you can eliminate the visual gap pretty well. The trick for a speaker cabinet is to make sure the glue-up has a fighting chance at being airtight. So basically very even glue coverage before clamping, and hope.

Update from my end, I have what I think is a 95% ready-to-go design that's slightly modified, I have a couple glitches in the model I need to resolve and then send to Lars for approval... it's meant to be printable on a 220mm bed, so most printers will be able to do it, albeit in 8 pieces.

I will also share a 4-piece version that should be printable on Bambu P1S which is what I use.

It looks to me to be an extraordinary job.

Top class

 

[Kaladin]

Incredible. Almost makes me not want to cut wood anymore for speakers.

It feels like cheating. Especially the large roundovers.

That's pretty awesome! Once they are glued together and I assume you sand and maybe bondo the seams or something? I've never worked with 3d printed parts. Are you able to get rid of the seams entirely?

Yeah, bondo spot spotty works great for filling seams. From there, it's using a filler primer to help fill gaps and layer lines. I'll use dura-block sanding blocks that are used for autobody work to get all of the lines straight and avoid creating waves that will show up in the final paint.

For filler primer it depends on the paint job. I'm going to spray Evercoat Super Build over epoxy primer but I've also used Rustoleum sandable filler primer cans in the past with good success.

 

[DatSpeakerTho]

Especially the large roundovers.

I will gladly be labeled a cheater for getting round overs like that, after I've been trying to figure out how to do it on wood for awhile. I've hit up wood shops but no one near me has a shaper with the cutters to do it.

Yeah, pretty much filling and sanding is how you'll have to get a good finish. It's pretty easy.

I took these garbage looking waveguides and made them nice. Wood filler and flat black enamel spray can.

 

[Roland68]

That's a good idea to save some money. Your prices look better but in the US it's about $8 less per roll for the cardboard rolls.



I've attached a couple pictures plus a few older ones. The pieces are just stacked together for now. I'm not going to glue them up until after they've been filled.

Pretty much this. Some people use wood filler, others use filling primer (spray), but it amounts to filling in the gaps and sanding down. After a few passes you can eliminate the visual gap pretty well. The trick for a speaker cabinet is to make sure the glue-up has a fighting chance at being airtight. So basically very even glue coverage before clamping, and hope.

Update from my end, I have what I think is a 95% ready-to-go design that's slightly modified, I have a couple glitches in the model I need to resolve and then send to Lars for approval... it's meant to be printable on a 220mm bed, so most printers will be able to do it, albeit in 8 pieces.

I will also share a 4-piece version that should be printable on Bambu P1S which is what I use.

If the joints are designed with a continuous tongue-and-groove configuration, stability is significantly enhanced, the gluing surface area is increased by 50–100%, and airtightness improves markedly as well. Furthermore, the risk of parts shifting during gluing is significantly reduced—and can even be completely eliminated.

Ensuring that everything fits together properly requires a bit of practice and experience; however, the tongue and groove are permitted to have a small amount of play, which is then filled with adhesive.

 

[kemmler3D]

If the joints are designed with a continuous tongue-and-groove configuration, stability is significantly enhanced, the gluing surface area is increased by 50–100%, and airtightness improves markedly as well. Furthermore, the risk of parts shifting during gluing is significantly reduced—and can even be completely eliminated.

Ensuring that everything fits together properly requires a bit of practice and experience; however, the tongue and groove are permitted to have a small amount of play, which is then filled with adhesive.

View attachment 530607

Sensible and probably something I should add to the design. Would not be super hard to make. My hesitation is that on some printers you may have varying tolerances or slight warping, which makes fittings that span large areas iffy. It would also potentially increase the need for supports, but that's not the end of the world.

Right now I'm relying on dowels to line up the parts, which makes gluing pretty manageable, but for airtightness I have a feeling that might not be a great approach. Something for the 0.2 version...

 

[Roland68]

Incredible. Almost makes me not want to cut wood anymore for speakers.

That's pretty awesome! Once they are glued together and I assume you sand and maybe bondo the seams or something? I've never worked with 3d printed parts. Are you able to get rid of the seams entirely?

The long-term goal is to 3D-print speaker enclosures as single-piece units, featuring a screwed-on, airtight back panel or merely service openings. Furthermore, different materials could be printed simultaneously—such as foaming materials for internal filling and resonance damping.

 

[Roland68]

I will gladly be labeled a cheater for getting round overs like that, after I've been trying to figure out how to do it on wood for awhile. I've hit up wood shops but no one near me has a shaper with the cutters to do it.

Yeah, pretty much filling and sanding is how you'll have to get a good finish. It's pretty easy.

I took these garbage looking waveguides and made them nice. Wood filler and flat black enamel spray can.

View attachment 530541

View attachment 530542

You can get small gantry milling machines for this kind of wood and plastic processing for as little as €250–500.
With one of these, you can naturally mill parts like this out of wood or MDF, but you can also finely surface-mill printed parts. The result is then ready for painting immediately.

 

[DatSpeakerTho]

You can get small gantry milling machines for this kind of wood and plastic processing for as little as €250–500.
With one of these, you can naturally mill parts like this out of wood or MDF, but you can also finely surface-mill printed parts. The result is then ready for painting immediately.

Yeah, but I can print this purifi speaker for a lot less than that and don't have to store a big loud ass machine.

 

[Roland68]

Yeah, but I can print this purifi speaker for a lot less than that and don't have to store a big loud ass machine.

You misunderstood me there; I was referring solely to the waveguides—or rather, the post-processing required after printing. These devices are actually relatively compact; with a working area of 200–300 mm and a maximum height of 100 mm, they are no larger—and in fact, often smaller—than standard 3D printers, and notably flatter. Fundamentally, they are nothing more than a 3D printer equipped with a milling motor.

 

[DatSpeakerTho]

You misunderstood me there; I was referring solely to the waveguides—or rather, the post-processing required after printing. These devices are actually relatively compact; with a working area of 200–300 mm and a maximum height of 100 mm, they are no larger—and in fact, often smaller—than standard 3D printers, and notably flatter. Fundamentally, they are nothing more than a 3D printer equipped with a milling motor.

I just used a better setup printer. Pretty much no post work needed. Also when I say I, I mean my cool friends with 3d printers. One gifted me a ender 3 for free and it was a nightmare so I just ppl to print stuff for me.

 

[andrew]

Pretty much this. Some people use wood filler, others use filling primer (spray), but it amounts to filling in the gaps and sanding down. After a few passes you can eliminate the visual gap pretty well. The trick for a speaker cabinet is to make sure the glue-up has a fighting chance at being airtight. So basically very even glue coverage before clamping, and hope.

Update from my end, I have what I think is a 95% ready-to-go design that's slightly modified, I have a couple glitches in the model I need to resolve and then send to Lars for approval... it's meant to be printable on a 220mm bed, so most printers will be able to do it, albeit in 8 pieces.

I will also share a 4-piece version that should be printable on Bambu P1S which is what I use.

Stupid, novice, question but what are the 4 or 8 pieces?

The long-term goal is to 3D-print speaker enclosures as single-piece units, featuring a screwed-on, airtight back panel or merely service openings. Furthermore, different materials could be printed simultaneously—such as foaming materials for internal filling and resonance damping.

Yes, the thing that worries me, as a novice, is the idea of having to glue the various panels together in an airtight manner that looks good. So, the idea of a single piece unit with a screw on back panel is appealing. I know that this is beyond most home 3D printers but those of us without a printer could take such plans to a local 3D printer shop. (Here a local manufacturer seems to offer 3 prints using metal, fibre and various plastics https://www.sqpengineering.com.au/services/additive-manufacturing/)

 

[Roland68]

Stupid, novice, question but what are the 4 or 8 pieces?


Yes, the thing that worries me, as a novice, is the idea of having to glue the various panels together in an airtight manner that looks good. So, the idea of a single piece unit with a screw on back panel is appealing. I know that this is beyond most home 3D printers but those of us without a printer could take such plans to a local 3D printer shop. (Here a local manufacturer seems to offer 3 prints using metal, fibre and various plastics https://www.sqpengineering.com.au/services/additive-manufacturing/)

Surely you realize, though, that 99% of all loudspeakers on the market—regardless of price range—consist of at least six glued-together parts, if not more. This applies to the standard kit version of this Purifi loudspeaker as well.
Do you find yourself worrying about the adhesive joints on *every* single one of those speakers?

To print such enclosures effectively and affordably, you need printers specifically designed for the task. The Purifi enclosure is actually small enough that it can be printed as a single piece on relatively affordable, readily available printers costing around €1,000. However, doing so does require a certain amount of experience.

You can certainly request such a print from this service provider—but you’d better buckle up, because the price is going to absolutely knock your socks off .

 

[andrew]

Surely you realize, though, that 99% of all loudspeakers on the market—regardless of price range—consist of at least six glued-together parts, if not more. This applies to the standard kit version of this Purifi loudspeaker as well.
Do you find yourself worrying about the adhesive joints on *every* single one of those speakers?

To print such enclosures effectively and affordably, you need printers specifically designed for the task. The Purifi enclosure is actually small enough that it can be printed as a single piece on relatively affordable, readily available printers costing around €1,000. However, doing so does require a certain amount of experience.

You can certainly request such a print from this service provider—but you’d better buckle up, because the price is going to absolutely knock your socks off .

Yes, of course, it’s my ability to glue and seal the cabinet that is in question. I hadn’t realised that there was a standard kit version cabinet (in MDF or ply) of the SPK16. Can you provide a link?

You’re no doubt right about the manufacturer’s cost

 

[Roland68]

Yes, of course, it’s my ability to glue and seal the cabinet that is in question. I hadn’t realised that there was a standard kit version cabinet (in MDF or ply) of the SPK16. Can you provide a link?

You’re no doubt right about the manufacturer’s cost

What I meant was that the standard version calls for a conventional MDF enclosure. There are service providers where you can have such housing parts milled, but also have complete housings built, but for the housing, that won't be a special offer.

For experienced 3D printing enthusiasts, it is standard practice to print larger objects in multiple sections and then glue them together. However, achieving good results with this method requires a fair amount of experience. Speaker enclosures of this size, in particular, present an even greater challenge.

If you have no prior experience with this sort of project—or with 3D printing in general—I would recommend investing directly in a sufficiently large printer with a build volume of 500x500x500mm, or at least one capable of printing the object in just two sections. And, crucially, ensure it features an enclosed build chamber.

Nowadays, even small CNC portal mills and kits are affordable enough that you can acquire a machine capable of milling enclosure parts from MDF for under €500. The range of options expands significantly once you move into the €500 to €1,000 price bracket.

Alternatively, you can build the enclosure yourself using MDF or plywood, exactly as specified in the kit instructions.
For this, you simply need pre-cut wood from a hardware store and a router equipped with a circle-cutting attachment (which typically costs between €50 and €200). There is even a method that requires nothing more than a jigsaw.

However, regardless of which approach you choose, you will need at least a certain degree of DIY mentality—or enthusiasm—along with the drive to teach yourself the necessary skills.
If you lack that mindset—or if you view this merely as a means to an end, devoid of any genuine enthusiasm—then you are probably better off passing on the project; none of these options are "set it and forget it" endeavors—not by a long shot.

Even regarding the first option—ordering a pre-milled kit from a professional service provider—I would still recommend paying them to assemble the kit for you if you have no prior experience with such assembly work. That is actually the smallest cost item—and money very well spent.

 

[kemmler3D]

Stupid, novice, question but what are the 4 or 8 pieces?


Yes, the thing that worries me, as a novice, is the idea of having to glue the various panels together in an airtight manner that looks good. So, the idea of a single piece unit with a screw on back panel is appealing. I know that this is beyond most home 3D printers but those of us without a printer could take such plans to a local 3D printer shop. (Here a local manufacturer seems to offer 3 prints using metal, fibre and various plastics https://www.sqpengineering.com.au/services/additive-manufacturing/)

So basically the SPK16 cabinet measures roughly 250x330x450mm (IIRC) and a standard print bed on a consumer-grade printer is around 220x220x250 (LxWxH). As such the cabinet needs to be printed in pieces, it won't fit in the printer in one go.

@Roland68 is correct that these speakers are small enough to print in one go if you get the right printer. Printers that big are far less common but not hard to get if you want one. In that case, it can look pretty decent straight off the printer when things go right. Removing supports from the interior would be kind of a pain, but the removable back would help with that.

A Bambu P1S has a 256x256x256 print bed so you can split the cabinet into 4 pieces, but I'm trying to get it to fit on smaller printers, so it ends up making sense to split it into 8. Front left top, front right top, rear right top, rear left top, etc. etc.

Gluing the pieces together should not be TOO hard, depending on what glue you use. For this I will probably use a 2-part epoxy, (it's thick enough to make sure you get a nice even layer) and I plan to print a little spreader / scraper tool to help get an even coating on the cabinet walls. For tiny gaps, the filling and painting steps should (hopefully) help with airtightness.

The design I'm working on has little dowels that connect each piece to its neighbor, so you just need to put the glue on, slot the dowels in, then clamp it down. If I do that part of the design right, it will not demand much of your DIY skills.

Still, I will tell you right now, no 3D print that's been glued together looks good - just like MDF enclosures, you need to do some finishing work to make them look nice.


In the case of finishing 3D prints, you usually do a quick sand to knock down print or glue flaws that protrude, then slather the piece in some kind of filling compound (bondo, wood filler, etc to fill in any gaps.) From there, you typically use another filling agent that smooths out the layer lines (filling / building primer is popular), sand that, and probably repeat that process a few times to get it really smooth. From there you paint using whatever methods you prefer.

This probably sounds like a pain in the ass, and it is, but it's no worse than finishing wood pieces.

There are buttloads of Youtube videos out there that walk through this process, so if you have the motivation to try, it's doable.