End Game DIY Loudspeakers

[Mr. Widget]

To be clear, this is not just something I've read about. I've done it. On two different speakers. The results, both subjectively and measured with REW, speak for themselves.

I am not saying it won't work. It gives you an alternative to look at which is possibly a good thing, but I would definitely want to compare the results against the original design.

On subjectivity and DIY... I have heard many DIY speakers that friends have built that pleased them to no end but I thought sounded atrocious. At the end of the day, if you build it and you like it, you are good to go.

Of course some DIY projects are pretty amazing too.

 

[Rick Sykora]

Depends what your end game is…

Am sure Linkwitz’s LX521 is for some. Dennis’s Philharmonic monitor would seem a good candidate as well. If going more traditionally, SB has some kits that are more premium like the Satori Rinjani. Should mention prolific designs by Troels and Heissmann Acoustics too.

 

[Marc v E]

Ok, here is my advice: don't do it.

The amount of expertise and measurements and rework that go into a well designed speaker is not something that can be matched with diy. Let alone the finish and enormous depreciation of a diy speaker.

By all means build the Directiva and enjoy it. Just don't spend more than that and end up with less than a Genelec, Neumann, Revel or Kef speaker.

 

[kemmler3D]

So far I think the conclusion is: Competitive-with-SOTA is basically possible with DIY and open plans, but not "obviously better than commercial SOTA".

Seems that most of us question the possibility of beating SOTA Commercial via DIY even with actual professional experience and an unlimited BOM cost.

Cosmetically, it is only possible to compete with SOTA if you're already 100% sure you can make nice furniture you'd be willing to put in your living room, because high-end speakers are nice furniture that make noise.

To circle back to OP's original aim -SOTA without the full cost... my ultimate advice is buy SOTA gear secondhand, unless you highly value the bragging rights of building your own speaker.

A point that has not been belabored much in this thread yet: DIY speakers have VERY LOW resale value, because very few of us have a bankable reputation as a speaker builder. KEF, B&W, Revel, JBL? Yeah, people have heard of them. Kemmler3D speakers do not command the same respect on the market and neither will yours.

The total cost of ownership of a DIY job is probably much higher than that of a secondhand Wilson or KEF or whatever. Unless you are certain that this is the absolute endgame, and that you will die with these speakers in your listening room... DIY may not be a wise investment.

I have had this experience in DIY furniture. I built a pretty cool desk, it was a clone of a commercial one, with upgrades. The finish was admittedly C+ compared to factory stuff but far from horrible. I spent >$800 building it and could hardly get $200 for it. A fair price for used commercial would have been $350-450. And again, speakers are just furniture that makes noise...

 

[Wolf]

My DIY speakers for about 2000 $.

Very nice! Care to spill? I see a 12MU mid, an 22W/1851, maybe a B&C woofer and I can't pinpoint the tweeter.

 

[TulseLuper]

My DIY speakers for about 2000 $.

These look really great. When did you build? The pair of mids alone is almost $700 with tax here. Plus Revelator woofers? Are they active or passive? Love the baffle material and grills.

 

[JPA]

Here's a suggestion if you really want to DIY. Contact Jim Salk at Salksound.com and ask him if he will sell you a couple of cabinets without drivers or crossovers, and with blank (uncut) front baffles. I have no idea what his answer would be, but it's worth a shot. Then, assuming he agrees, choose your drivers, route the driver holes in the baffles, mount the drivers, and start the measurement & tweaking process. Maybe you could even buy the drivers, send a set to Jim and have him route the baffle holes for you.

 

[Everett T]

Here's a suggestion if you really want to DIY. Contact Jim Salk at Salksound.com and ask him if he will sell you a couple of cabinets without drivers or crossovers, and with blank (uncut) front baffles. I have no idea what his answer would be, but it's worth a shot. Then, assuming he agrees, choose your drivers, route the driver holes in the baffles, mount the drivers, and start the measurement & tweaking process. Maybe you could even buy the drivers, send a set to Jim and have him route the baffle holes for you.

He would, but that would be a long and expensive processs for Jim right now. Also, you kinda need to know what bass driver before hand to know what size cabinet to make. It's not hard to veneer cabinets or paint them, it's just the high gloss, deep finish that takes time and good technique. Probably better off to find a local cabinet maker/woodsmith then have them done by Jim, avoid the shipping and they can do the baffles cutouts too.

 

[JPA]

He would, but that would be a long and expensive processs for Jim right now. Also, you kinda need to know what bass driver before hand to know what size cabinet to make. It's not hard to veneer cabinets or paint them, it's just the high gloss, deep finish that takes time and good technique. Probably better off to find a local cabinet maker/woodsmith then have them done by Jim, avoid the shipping and they can do the baffles cutouts too.

Are you a spokesperson for Jim?

 

[Everett T]

Are you a spokesperson for Jim?

No, I just know what the leads times are currently and a have had quotes for custom work in the past, in addition to purchasing from him.

 

[bothu]

Very nice! Care to spill? I see a 12MU mid, an 22W/1851, maybe a B&C woofer and I can't pinpoint the tweeter.

Here is the components :


Speakers :
Tweeter - Scan-Speak D2904 710003 , https://www.hifikit.se/media/attachment/file/d/2/d2904-710003.pdf
Mid range - Scan-Speak 12MU 8731T00 , https://www.hifikit.se/media/attachment/file/1/2/12mu-8731t00.pdf
Bass - Scan-Speak 26W 8861T100 , https://www.hifikit.se/media/attachment/file/2/6/26w-8861t00.pdf
Sub. - PA Audio C-18 ELF , http://p-audio.co.uk/pdf/C_18ELF.pdf

Bass and Sub. are sealed speakers.
Sub. are compensated with Linkwitz Transform to 30 Hz.
X-overs are 80Hz - 265Hz - 3,9kHz, 24 dB/octave LR filters.
Amplifiers for each speaker 150 W/8 ohm with clipping indikators.
Preamp is based on OP, LM 4562. External power supply.

Listening and measuring with REW have helped me to find out levels for each driver.

 

[bothu]

These look really great. When did you build? The pair of mids alone is almost $700 with tax here. Plus Revelator woofers? Are they active or passive? Love the baffle material and grills.

Here are components :


Speakers :
Tweeter - Scan-Speak D2904 710003 , https://www.hifikit.se/media/attachment/file/d/2/d2904-710003.pdf
Mid range - Scan-Speak 12MU 8731T00 , https://www.hifikit.se/media/attachment/file/1/2/12mu-8731t00.pdf
Bass - Scan-Speak 26W 8861T100 , https://www.hifikit.se/media/attachment/file/2/6/26w-8861t00.pdf
Sub. - PA Audio C-18 ELF , http://p-audio.co.uk/pdf/C_18ELF.pdf

Bass and Sub. are sealed speakers.
Sub. are compensated with Linkwitz Transform to 30 Hz.
X-overs are 80Hz - 265Hz - 3,9kHz, 24 dB/octave LR filters.
Amplifiers for each speaker 150 W/8 ohm with clipping indikators.
Preamp is based on OP, LM 4562. External power supply.

Listening and measuring with REW have helped me to find out levels for each driver.

/ Bo Thunér.

 

[voodooless]

To me it is ethically dubious to travel all around demoing speakers, taking up various dealers’ time in what they think is good faith exchange, for the purpose of duplicating a design to save money.

My guess is, these poor schmucks will also sell overpriced cables.. that should fix the ethical dilemma right there

 

[DrFG51]

I did a set of Wayne Parham's 4Pi speakers, with a pair of 3Pi flanking subs some years ago. I wouldn't necessarily suggest "end game", but they were really REALLY good, and pretty simple to make (disclosure - I'm a furniture maker, but even not being one, I think they're within the scope of a competent DIYer with "standard" power tools). 15" JBL mid driver, B&C horn. Incredibly sensitive and sounded really great on a small SET amp. I never once in the years I owned them considered they were lacking anything. They cost me around £2500 in parts/raw materials (including the subs), and imho are comparable with speakers with a 5 figure price tag.

 

[Jukka]

So far I think the conclusion is: Competitive-with-SOTA is basically possible with DIY and open plans, but not "obviously better than commercial SOTA".

Seems that most of us question the possibility of beating SOTA Commercial via DIY even with actual professional experience and an unlimited BOM cost.

Cosmetically, it is only possible to compete with SOTA if you're already 100% sure you can make nice furniture you'd be willing to put in your living room, because high-end speakers are nice furniture that make noise.

To circle back to OP's original aim -SOTA without the full cost... my ultimate advice is buy SOTA gear secondhand, unless you highly value the bragging rights of building your own speaker.

A point that has not been belabored much in this thread yet: DIY speakers have VERY LOW resale value, because very few of us have a bankable reputation as a speaker builder. KEF, B&W, Revel, JBL? Yeah, people have heard of them. Kemmler3D speakers do not command the same respect on the market and neither will yours.

The total cost of ownership of a DIY job is probably much higher than that of a secondhand Wilson or KEF or whatever. Unless you are certain that this is the absolute endgame, and that you will die with these speakers in your listening room... DIY may not be a wise investment.

I have had this experience in DIY furniture. I built a pretty cool desk, it was a clone of a commercial one, with upgrades. The finish was admittedly C+ compared to factory stuff but far from horrible. I spent >$800 building it and could hardly get $200 for it. A fair price for used commercial would have been $350-450. And again, speakers are just furniture that makes noise...

The first part is not entirely true. Cabinet and crossover design is not black magic. Nearly everything can be modeled and simulated to high accuracy, if you know what you are doing, before breaking the first glue bottle. The difficulty level depends on what you expect the loudspeaker to be, size in general, specific width, ease of use, all-in-one vs. 10 separate boxes, aesthetics, SPL levels and so on.

A classic 3-way cabinet from the 70's has many things right, if you bear the looks. That's a good starting point. A narrow baffle is more of a challenge, but rounded/beveled corners can be made with normal tools. DIY can easily give you enough so you don't want to spend 1M into a Magico for an upgrade.

 

[Plcamp]

If I had that budget I would consider the systems at Josephcrowe.com, particularly the onken bass cabinet combined with his 290 hz horn driven by a coaxial compression driver. He provides, for a very reasonable fee, full construction documentation.

3D CAD plans for ES-290 Biradial Horn -- Horn No.1670

By purchasing these plans you'll automatically receive the 3D CAD files for the ES-290 Biradial Horn No.1670 via email. Zipped Folder Includes: .STL files for 3D Printing (Assembly and part files) .STEP, .iges, & .x_t files for CNC Machine CAM programming Does not include CNC programming...

josephcrowe.com

 

[Rick Sykora]

With the demise of Selah Audio, there are fewer options for a DIY speaker that looks nice too. Does not mean they are extinct entirely. SB Acoustics is doing a few. The top of the line Sasandu looks nicely done…

At $5K USD, the main question is value. Would compare to Salk BePure at $6500. Do they compete well with a KEF R7? That seems a stretch (notably since you still have to build the Sasantu). On the other hand, if you are a good cabinet builder, the parts are only $2700.

 

[Burning Sounds]

Unfortunately these were all by text or word of mouth.

Ah, OK, thank you.

I did find a couple of comments on the midwest audio club forum thread on the 2022 show which were both positive regarding the LX521. One positive commenter mentioned that he heard later that some listeners did not like them. You yourself said you heard that they were not impressive. Hmm, not exactly reviews, eh?

 

[Wolf]

Here are components :


Speakers :
Tweeter - Scan-Speak D2904 710003 , https://www.hifikit.se/media/attachment/file/d/2/d2904-710003.pdf
Mid range - Scan-Speak 12MU 8731T00 , https://www.hifikit.se/media/attachment/file/1/2/12mu-8731t00.pdf
Bass - Scan-Speak 26W 8861T100 , https://www.hifikit.se/media/attachment/file/2/6/26w-8861t00.pdf
Sub. - PA Audio C-18 ELF , http://p-audio.co.uk/pdf/C_18ELF.pdf

Bass and Sub. are sealed speakers.
Sub. are compensated with Linkwitz Transform to 30 Hz.
X-overs are 80Hz - 265Hz - 3,9kHz, 24 dB/octave LR filters.
Amplifiers for each speaker 150 W/8 ohm with clipping indikators.
Preamp is based on OP, LM 4562. External power supply.

Listening and measuring with REW have helped me to find out levels for each driver.


View attachment 253949


View attachment 253950View attachment 253951


/ Bo Thunér.

Oh, it's the 26W, I was close. Good tweeter for sure.
So you have made 2 of those 4ch amps then? What an undertaking...

 

[fpitas]

I am not saying it won't work. It gives you an alternative to look at which is possibly a good thing, but I would definitely want to compare the results against the original design.

On subjectivity and DIY... I have heard many DIY speakers that friends have built that pleased them to no end but I thought sounded atrocious. At the end of the day, if you build it and you like it, you are good to go.

Of course some DIY projects are pretty amazing too.

It has been remarked that if one selects his own components, builds his own enclosure, and is convinced he has made a wise choice of design, then his own loudspeaker sounds better to him than does anyone else's loudspeaker.

-Beranek's Law, from his book on acoustics https://en.wikipedia.org/wiki/Leo_Beranek