Modula PSD- small system idea simmering on the back burner for months...

[JonMarsh]

OK, I'm calling it a Modula PSD at this point, but the initials have changed and evolved depending on the brands of parts- the "P" has been constant, because of the PTT6.5X04-NFA woofer.

What's with the other letters at this point?

"S" = Seas, for Seas H1499 DXT tweeter. I picked up samples and did baseline measurements in 2019, saw some positive stuff but also figured optimizing a crossover would take some effort- day job was too demanding then.

"D" = Dayton RSS210PR, the smaller PR I've been waiting and hoping for, for certain system applications.

The LF analysis with two PRs looks very promising...





This PuriFi driver has some "quirks" regarding response, as does the DXT. But it seems possible to get them dance together nicely...




With data just from a sealed cabinet, without the extended LF performance, the crossover design has been optimized "on paper" from driver measurements.




Care has been taken to insure that the load impedance is not demanding high current monoblocks that this author is far too fond of...





Components are mostly on hand for a test build, some crossover parts pending. It's in the queue... And Steve has local friends that want a build like this!

[Jonasz]

Wow, twenty components for each tweeter, must be some kind of record? 😁

What does the 22 ohm resistor to ground in the ultra sonic notch(?) do?

[JonMarsh]

It trims the HF level to be flatter and match up better with the woofer. No LPAD though.

I think this is what Zaph used to call a "perfectionist" crossover. It's an experiment to see just how much performance I can get out of the DXT sonically.

I did review quite a few other DXT crossover designs from the past, none of which address the ultrasonic breakup, which is a key issue for resonance amplification of distortion products and IMD.



And of course, once you start "fixing" a few things, some of us just don't know where to stop... IMO the DXT is suited for applications in the 83-85 dB sensitivity area, though, and I have several of those now, including the Kii inspired Ardent/Modula Xtreme. And this two way for Steve and his friends. Which could be a nice HT surround speaker, if a bit in the overkill area. And a possible HT CC build coupled with the Peerless FSL0512R01 mid. But then the targets vary from an LR3 at 2300-2400, to the top end of a Duelund concept for the Kii inspired Modula Xtreme. The Xover above for the two way has a VERY tight fit to the LR3 requirement.... but then, that's the easier one.

Off axis behavior is quite nice, much flatter at 30 degrees off axis than any conventional dome I've measured. So, I figure if Kii liked it, then exploring what it will do may be interesting. The first pair I have I bought in 2019, didn't have time to follow up.

And I have another tweaked version of the Kii inspired Duelund design with altered LF tuning, and 1-2 dB more overall sensitivity. If the build actually hits this, I won't complain- from 40Hz to 20kHz within a 2 dB window? That would be a crummy phono preamp, but for a speaker, respectable.




Of course, the "smart" thing for me to do would be to not post ANYTHING about this until it was all done and documented, and a known success, but I think some folks enjoy seeing how the sausage is ground...

But you know, this is DIY, and if some folks just want to use a series cap and resistor, well, that's their choice! It's a big wide world out there...

[Evil Twin]

Wow, twenty components for each tweeter, must be some kind of record? 😁

What does the 22 ohm resistor to ground in the ultra sonic notch(?) do?

Actually, I believe are over counting the "Tweeter crossover" parts, as L9, C12, and R12 are just impedance zobel networks to avoid having a large peak in the input impedance, which would provoke "misbehaving" with many higher output impedance amplifiers.

The Modula MT XE has a similar network, and one of Zaph's waveguide designs did this for the same reason. Most of the parts have arrived for the three way network version to test, and a few more on the way, should arrive at the Imperial Lab on Thursday.

This version is pending for some neighbors of Steve's... I have been brought in for oversight due to the interest of the Chancellor...