I killed two odd Phillips from 90's minisystem that never make make sense and sound and go to karlsonators.
A recycling issue. Kkk
I liked them a lot more, the sound is very diferent and cozzy, if is possible. I took the originals (tks greg) and resize for a 51/2" woofers, buyed a two way crossover in china and used the olds ones only for the mids. Only one modification, the curves are Tactrix.
i just wonder why the hi frequencies aren't so goog.
if someone want, i do have the blueprints in SketchUp.;)


  • Screenshot_2020-12-03-17-45-17-790_com.julian.apps.AudioTool.jpg
    431 KB · Views: 292
  • IMG_20201125_182520.jpg
    508.7 KB · Views: 298
  • IMG_20201127_125259.jpg
    448 KB · Views: 293
  • IMG_20201127_122503.jpg
    751.3 KB · Views: 296
  • IMG_20201125_174042.jpg
    616.3 KB · Views: 286
  • IMG_20201109_183951.jpg
    429.8 KB · Views: 152
  • IMG_20201113_135857.jpg
    987.4 KB · Views: 131
  • Like
Reactions: 1 user
Joined 2005
Paid Member
- That's a pretty amazing transformation - - looks nice, measures nice.

It would be good to see the Sketchup files. At Job Ulfman's old Karlson Speaker forum, there were two users of tractrix curve apertures for high aspect Karlson and also trax for a klam. (see attachments)

How did you generate the tractrix curve? What is that coating or paint which looks like a damping compound?

How would you describe the sound in the Karlsonator vs the original box? How about the way instruments and vocals sound now vs former "home" ?


  • KLAM12. CARL.jpg
    KLAM12. CARL.jpg
    23.4 KB · Views: 103
  • KEN L K8.jpg
    KEN L K8.jpg
    37.1 KB · Views: 97
Joined 2005
Paid Member
Hey XRK971 - -might you eyeball the drawing of this guy's small Karlsonator for Kappa12A and see what it would really do vs his hornresp input? He asked me for help some time ago but I don't even have hornresp anymore to see what his input says for volume, tuning, etc.

I like its small size and wonder if it would play well and perhaps reach 70Hz solid.

Well, if memory serves it's either a 1/4 or 1/2 WL of the slot height, so if 150 Hz then ~13543/4/150 = ~22.57"/45.14", so what's the actual slot height?
I'm sorry, have patience with me but I didn't understand how you got the result.
what number is 13543?
then you divided this number by 4 (13543/4). Does 4 mean 1/4 wavelength?
and then you divided by 150 Hz again. would the frequency be where the decline begins?
what does this 22.57 "/ 45.14" division mean?
I long assumed the classic K15 (and its scale-downs) front chamber was a QW resonator, but it might just be that the volume and slot geometry simply tune as a Helmholtz resonator and the frequency coincides with the vertical QW. Certain modern high-aspect designs have thrown-off my assumption a bit though, so I'm not certain what to think now, but it does not matter a whole lot as the tuning is adjustable via the slot gap as well as how high it begins.
I'm sorry, have patience with me but I didn't understand how you got the result.
what number is 13543?
then you divided this number by 4 (13543/4). Does 4 mean 1/4 wavelength?
and then you divided by 150 Hz again. would the frequency be where the decline begins?
what does this 22.57 "/ 45.14" division mean?

Speed of sound in air in inches used by Hornresp (34400 cm); yes; no, where the notch is. Don't recall the math to figure its slope order (Q), hence its bandwidth (BW). Means 1/4 WL = ~22.57", 1/2 WL = ~45.14".
if I increase the verticalization of the exponential curve, can I move the drop peak after 1KHz?
Correct. Think of the slot as a panel where its base = LF cutoff and the other end its HF with the vertical length (L) needing to be long enough to span these two frequencies with a curvature (flare frequency) = 1/4 (L).

Note though that you don't need to increase (L) just to get to 1 kHz though since its 1/4 WL diameter = ~13543/4/pi/1000 = ~1.078" wide. In short, the factory K-slots already goes much higher, so would need to shorten it.

.....and from the previous posts it appears that any angled baffles reduces these to a variable approximation.
Hi everybody! Referring particularly to post #323, I own some Visaton BG-20 speakers, and I'm planning to build a couple of Karlsonato8, or maybe a couple of TQWT boxes, as dr.Scottmoose suggested. Starting from the original plans I found on the posts, I prepared some Sketchup3D drawings (in mm), hoping this could be useful to somebody. The thickness of the wood is supposed to be 19mm (3/4in, more or less), and the hole for the loudspeaker is bigger than needed, please carefully check before going further. Thanks, everybody, for your HUGE effort of spreading and sharing precious info about the audio world: I've learnt more in this last year, reading and lurking on this site, than in many years of magazines and I'd like to be useful somehow. Finally, last but not least, please forgive my rough english and greetings from Roma, Italy!


  • KarlSonator8&TQWT.zip
    430.1 KB · Views: 63
Joined 2007
Paid Member
I think you need a big XKi. Like a Deltalite II 2515 15in or something like 100dB sensitivity. The XKi controls cone excursion very well so SPL should be able to go pretty high. Of course use with K tube tweeter crossed around 1.2kHz.

Tune box to 41Hz and hit the low bass notes.

99.5dB should be something special - price seems very reasonable for a neo magnet driver with this kind of sensitivity. I think it will have a jack hammer punch and is articulate with fairly light cone (72g) and smooth breakup free response out to 3kHz.

Let me know and I will work up box and vent dimensions for you.


I'm late to the party, but count me as interested! How big of an XKi are we talking? I'm looking at bass guitar use like freddi is describing and it seems like there's been a lot of discussion in the varied "K" threads but no settled or agreed upon solution. If there's already something out there and I missed it, please show me the way :)

This is all a fun experiment, so I'll try XPS foam (1.5"?) with some sort of facing.

The Deltalite II 2515 is out of stock at PE for the moment. I have some Beta 15A's that could go in for now until they restock. Worth trying or is the Vas too high on the beta (11.8 vs 7.2)? Are there better candidates that have come out since this post in 2016?