What building my own speakers has taught me...

Interestingly, for my build, I also went three-way active. I did not try to do the cabinets myself. Woodworking is not already my hobby and I could not do a good enough (solid and robust and visually accurate) job myself without first putting in years of effort. It was still not cheap, as the Purifi drivers cost a pretty amount. In some ways I got the active crossovers for free. But first I had to pay for BACCH4Mac software, that happens to have an active crossover module built in. It is only fourth-order. But I have drivers that have pretty wide frequency overlaps. So it is fine. The 10-inch cross to the 6.5-inch around 200Hz. The 6.5-inch to the tweeters around 2500Hz. So the 6.5-inchers really cover the imaging.

I considered using broadbands with no tweeters. But when playing very loud in tests, I found broadbands to get stressed.

That said, there are some pieces that sound simply phenomenal with a broadband surrounded with a horn. With what I “saved”, maybe I can afford another system with horns for when they are better.

@carousel  - I think my next build, if I _ever_ do another would be a mid sized 2 way with compression driver and horn crossed to 12"  woofers.  Something inspired by the Altec Voice of the Theater or JBL speakers used behind the screen.  With a low and single crossover point and high efficiency I think those would also make really good DIY builds. 

@erik_squires 

Bravo…

Nice job on the speakers. I have also used Solen for custom work (crossover design, parts and board build) with excellent results.

Do you already have ideas on any “improvements “ or design changes (just in case you had the crazy idea of doing it all over again)?  

@erik_squires wrote:

By the way, @phusis - If you are thinking of the RAAL ribbons, I highly recommend them.  Never heard them but interacted with the designer via DIYaudio.  Super knowledgeable, very friendly and helped me learn a great deal.  I was lucky to get to learn from him.

I was particularly impressed by his command of his manufacturing.  He makes custom transformers for buyers at scale which allowing a variety of impedance and low frequency options. 

I’ve heard very good things about the RAAL’s. What I have coming up though is a Stage Accompany product, a dutch manufacturer, and their HF/MF driver is originally based on the Philips magnetostat RSQ8P which SA have then further developed into its current SA8535 state with a neodymium magnet.

The one sitting in the speakers I’m about to receive is an older version with a ferrite magnet weighing in at 20 pounds (SA8525). It’s a high sensitivity unit and horn loaded, which means the 1-4kHz range is boosted (which is of course compensated for in the built-in electronic XO). The essentially same, current version of the driver is also used non-horn loaded in their monitor series M57 and M59. A friend of mine is using a pair of M57’s outboard actively configured via DSP and sub augmented, whereas my upcoming pair of Stage Accompany speakers are actively configured with internal class A/B amps and an analogue, electronic crossover that is microprocessor controlled and protected. They will be sub-augmented as well with my tapped horns. 

Come to think of it I actually do know the Mundorf AMT’s; another friend of mine is using the larger (i.e.: longer) version for HF-augmentation in a horn-loaded speaker system. Very nice sounding, and by that I mean they integrate with and complement the horn well, not drawing attention to themselves but aiding resolution and extension. 

What is your impression of the sound from your speakers in their current, active configuration vs. their earlier, passive state? I’m assuming you only used the top section with the Scanspeak woofer/mid and Mundorf AMT unit passively (powered by the Luxman?) with the dual 10" Dayton woofer coming in later actively, but maybe you are still able to discern some overall differences here. 

@ticat 

Do you already have ideas on any “improvements “ or design changes (just in case you had the crazy idea of doing it all over again)?  

I have actually thought about this.  Honestly my current speakers are too good for my room and better than I can hear, so I have no practical reason to change a thing, but thinking as a tinkerer I do.   I'm not going to claim my speakers are better than any others, but that they make me extremely happy and if that for me, in my room, my own ears are probably not good enough to hear better.  If I were to move to a home with a bigger/better living room that may change. 

A Little History ... 

An active three-way speaker was not ever the original plan.  A high end two-way with subwoofer support was, so if I were to start with a blank piece of paper and re-do this speaker as a three way to start there are some changes I'd make.

First, I'd change the tweeter.  It is excellent but when I purchased it around 2017 I might have been on the leading edge of it's adoption, and it was half the price (I think).  Now, megabuck speaker makers use it, like this $85,000 Gryphon: 

And while it is stellar, it is not 3x more stellar than a number of high end domes.  So I'd probably give up on the $850 Mundorf, or even the smaller version which is ahem, only $450.  The equivalent Beymas are also about the same.   Per ChatGPT, Mundorf probably started AMT tweeter manufacturing around 2004.  Not sure when this particular model was first released. 

So, first thing that would go is the tweeter.  A dome or ring radiator, probably not Be.  Next, since I'd have to raise the crossover from 1.8 kHz to 2.5 kHz I'd consider downsizing the mid-woofer to a 5" to get the same horizontal dispersion.  Further, the large size of my current top cabinet can be cut in half by the change in driver and using a sealed cabinet. 

I think this would  give me equivalent performance, with a much smaller cabinet and about 1/3rd the part cost.  

The woofers are really great.  I could go with single, ported, but honestly I prefer output to extension, and for the money these are just really outstanding performers.  I could spend 2x as much and I'd probably never hear a difference. 

Lastly, if I was a truly rational man, and I'm not, I could probably also cut down on the plate amplifier power, and get the next size down.  In my modest 14' x 20' living room, 600 watts per channel, in a tri-amped set up is serious overkill. 

Of course, I just built these, in six months who knows how I would feel?