[abd1]

So I’ve pretty much settled on the regular song3’s. I spoke to Jim and while the song3-A is more accurate I just don’t listen to music to analyze it. I listen to relax and I think the base driver will be more than adequate.

[Brian D]

Thank you for that update.  I have a set of Song3's being built and this thread was starting to give me second thoughts.  Enjoyable sound is my endgame, not the never ending upgrade cycle, and your reply hit me perfectly.  I'm really looking forward to mine!

[abd1]

Thank you for that update.  I have a set of Song3's being built and this thread was starting to give me second thoughts.  Enjoyable sound is my endgame, not the never ending upgrade cycle, and your reply hit me perfectly.  I'm really looking forward to mine!

Yeah, I was tempted to go to the Song-3A's too, but after speaking with Jim he said the detail in the regular model is similar to what I'd hear in the SCST's which use the magnesium Seas driver. Not that the Song3 has more detail just similar, and that accuton takes it up even more. I've listened to speakers using those drivers and they're as detailed as I prefer. I think if the sound gets too detailed it can take some of the enjoyment out of the music, at least for me. Plus I'm going to use the money saved to upgrade to the Primaluna Dialogue Premium HP integrated when a good deal on one come along.

[Paul K.]

In addressing your point #2, I wondered the same thing many years ago when first venturing into TL design.  So, on an MTM design like in so many Salk speakers I've done for Jim, I modeled first the full version with the tweeter's center as the TL's design center (the proper way to do it) as if there were a single midwoofer located at that point with twice the Vas and Sd.  Once I had optimized that design, I then reduced the cabinet (the line) volume (area) by half and separately modeled each midwoofer's response where they were actually located.  The results showed that each midwoofer had virtually the same response below about 200 Hz, but somewhat different response above 200 Hz (the peaks and dips in their responses were different).  But, most of those peaks and dips occurred such that they cancelled each other out for the most part, leading to a nicely smooth overall response like that modeled for both midwoofers in the MTM.  In conclusion, two drivers located close together will work quite well in a TL.
Paul

Late to the party, but it allows me to steer it back onto topic...

Jim and his wife are great people and they're not too far from me, but I'm not a big Salk fan as:
1.) The drivers in the lower priced speakers don't blend to my (and others) ear (perhaps due to using design by committee);
2.) I'm a long time TL fan (Fried), but can't imagine how 2 drivers at different points along the line can work correctly;
3.) I'm not a MTM fan either because I don't want to be locked in height wise to avoid phase issues (as I've heard before);
4.) His Fried tribute speaker would have made Bud roll over in his grave (some design aspects were downright anti-Fried);
5.) His finishes would put my solid cherry furniture to shame.   

You guys need to spell out abbreviations!  Along with nested drop-downs on the Salk site it took me half an hour to figure out that SCST stood for SuperChargedSongTower. 

I'd vote the SCST off the island (MTM using small misplaced woofers in a TL cabinet for in-room use) and speak with Jim regarding Song 3 vs. 3A, either of which should integrate drivers better and therefore image better.  And as you say, one larger dedicated woofer should be able to fill your overall space better.  And the quality of drivers in the Song 3/3A would be hard to beat (impressive content value).

[DMurphy]

Late to the party, but it allows me to steer it back onto topic...

Jim and his wife are great people and they're not too far from me, but I'm not a big Salk fan as:
1.) The drivers in the lower priced speakers don't blend to my (and others) ear (perhaps due to using design by committee);
2.) I'm a long time TL fan (Fried), but can't imagine how 2 drivers at different points along the line can work correctly;
3.) I'm not a MTM fan either because I don't want to be locked in height wise to avoid phase issues (as I've heard before);
4.) His Fried tribute speaker would have made Bud roll over in his grave (some design aspects were downright anti-Fried);
5.) His finishes would put my solid cherry furniture to shame.   

I'm not sure which "lower end" models you're referring to, but I can assure you the crossover integration is just as good on the less expensive speakers.  That's something Jim never compromises on.   Paul has already answered your second comment.  I think Bud is something of a false god when it comes to transmission line design.  He didn't have the necessary theory and modeling resources to implement them correctly.  I don't think there are any "anti-Fried" elements in the tribute speaker.  I worked with his original notes to develop the series crossover, and we used the same tweeter he championed.  The transmission line implementation does differ from Bud's, but I think to the better. 

[R Swerdlow]

I'm glad to see Paul Kittinger and Dennis Murphy both responded to an earlier post.  I'd like to add one more observation about transmission line speaker design.  Earlier TL designs, such as Bud Fried's, involved a significant amount of trial & error (guesswork) that produced reasonably good results, but were not optimal.

Martin King and George Augspurger, two researchers and designers, independently succeeded in creating the first successful mathematical models of TL speaker behavior in the early 21st century.  Their work realistically predicts TL speaker behavior without trial & error with lumber & sawdust, and allowed for improved TL designs.

See this Wikipedia article on Transmission Line Loudspeakers for more.  https://en.wikipedia.org/wiki/Transmission_line_loudspeaker

"However from the 21st century, Martin King and George Augspurger (both separately and referencing each other's works), produced models which show these to be "generally less than optimal" designs which "did a good job of approaching what was possible in their day".  Audio engineer Augspurger had modeled TL using an electrical analogy, and found it to agree closely with King's existing work, based on a mechanical analogy.[18]  D'Lugos concluded in his overview of TL modeling and design theory: "I think that using modern drivers & tools such as King's software you can build a better TL easier today".[18]"

All the Salk TL speaker cabinets were designed by Paul Kittinger who uses Martin King's software combined with his own expertise and experience.

Instead of spinning in his grave, I suspect Bud Fried might smile in approval.