XKi - X's ab initio Karlson 6th Order Bandpass

Founder of XSA-Labs
Joined 2012
Paid Member
Hello XRK. I love the work you do. It Might be a stupid request, but I have one subwoofer laying around. And I thought it would be cool to build it in something good. Sadly the woofer is made for car audio, but I think it has nice specs. Maybe you can tell me, if It is a good mach please?

Focal 27A
cone 20,5 cm
Sd 330,06 cm2
Xmax 6 mm
Fs 41,29 Hz
Qes 0,540
Qms 4,350
Qts 0,480
Vas 21,22 litres
Res 23,36 ohms
Mms 107,12 g
Bl 12,22 N/A
Le 20,70 mH
Re 2,9 ohms

Impédance 4
Puissance maximum 500 W
Puissance nominale 250 W
Sensibilité (2,83 V / 1 m) 90 dB
Diamètre d'encastrement 232 mm
Profondeur d'encastrement 130 mm

This looks like it can work as a subwoofer, which is what the driver was designed for. The large Le causes a bump in response near 70Hz, which actually will give this woofer a psychoacoustic impression that it makes more bass than it does due to 2nd harmonic of 41Hz bass note being at 82Hz.

So make the XKi box 44 liters total volume with a 12in wide box x 18in tall x 12in deep box that has a 12in wide x 0.75in high x 9in long vent. Make the setback of the vertex of front chamber 4.75in from the front with a baffle board that is at least 12in tall to accommodate a 10 in driver and magnet when angled back.

Here is predicted SPL at 2.83vrms with 36Hz -12dB/oct high pass filter:

attachment.php


Here is predicted cone displacement at 20vrms just about hitting xmax for both peaks:

attachment.php


Here is predicted SPL at xmax requiring 20vrms input. Can achieve about 109dB with F3 of 38Hz:

attachment.php


If you will use it as a subwoofer from 40Hz to 100Hz, I think this is fine. Just be aware of the hump at 70-80Hz. Hope this helps you.
 

Attachments

  • XKi-Focal27A-Freq-2.83v.jpg
    XKi-Focal27A-Freq-2.83v.jpg
    73.9 KB · Views: 795
  • XKi-Focal27A-Displ-20v.jpg
    XKi-Focal27A-Displ-20v.jpg
    74.8 KB · Views: 789
  • XKi-Focal27A-Freq-20v.jpg
    XKi-Focal27A-Freq-20v.jpg
    73.5 KB · Views: 828
If running a tweeter, maybe a small one could be mounted on one of the wings? Would get c-t-c closer than mounting right at the top and avoid worrying about compromising the start of the slot.

I don't know if XRK or Bonjonno has experienced this, but when manually evaluating Karlson polars by walking 120deg in an arc around the speaker (keeping both ears close to on axis), and also when I turn the k-slot sideways and eval vertical and horizontal, I've found that the most intense HF diffraction seems to happen at the top of the K-slot making it a strong re-emission source for HF. This apex point is pretty close to the tweeter. Time alignment would probably help tighten up the impulse but other than that I don't see a glaring frequency response impact to having a 3K cross up 1-2" from the top of the K-slot. Measurements might show my intuition is incorrect but I have played with *a lot* of k-slot type configurations, specifically evaluating polar dispersion.
 
Actually not the apex, but the top 1/3rd of the slot seems to have the most HF emission going on. That's one thing about the K that is kinda interesting. Real world situations and instruments both have diffraction and surface emission happening at various points on their surfaces. With a guitar these resonances that have different decay characteristics define the shape and size of the instrument. The Karlson waveguide and speaker configuration provides some not too unnatural diffusion of a signal creating a larger than the speaker psychoacoustic "size" that is probably closer in reality to a real sound source and it's "recording scene". Lately, I've been re-reading the thread "beyond the Ariel" and have noted how often Lynn talks about ambient room cues blending properly with the environment (rather than not being affected by the environment) as an important factor in "you are there" reproduction. I think the Karlson approach is an interesting middle-ground between open baffle / 360deg dispersion systems and sealed box configs. The K offers broadband temporal cues that suggest an apparent size of scene while not relying on the room to pull this off. And then there are other documented advantages.
 
its good at times to use the X15 method (introduced 1965 with a wooden asymmetric projector) with internal K-tube. Also I think the upper reflector can have some effect in focusing HF energy towards that part of the slot.

with really tall K and beamy driver, perhaps not a lot of HF energy gets towards the top (?) - its still enough to be heard.

I've got a lot of respect for Karlson's work - and hope it can be explored further.
 
So make the XKi box 44 liters total volume with a 12in wide box x 18in tall x 12in deep box that has a 12in wide x 0.75in high x 9in long vent. Make the setback of the vertex of front chamber 4.75in from the front with a baffle board that is at least 12in tall to accommodate a 10 in driver and magnet when angled back.

I have sketched some measurements, I want to ask If I got them right. And also there is one measurement missing, that I don´t know how to figure out.

Thank you!

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
what are the best K's for PA310-8? - fwiw a 0.89 scale version of TB46's 15" Karlflex w. PA310 vs Brian Steele's cool Boom Unit

I'll mount a PA310 in a copy of the old 1955 K12 and give it a whirl in a day or so


goofed typing the K-flex dims - its about 17" deep
e2NgBWc.jpg
 
any guesses how a regular vertically stacked pair 10" (Delta10a) or 12 (Kappa12a) would fare as a midbass module vs typical small midbass horn?

of course, the nominal 4ohm impedance isn't as friendly to some tube amps as higher but with K its good to have significant input power for peaks. (for class AB - I've 150wpc tube and 1KW SS)

I'd be happy with a solid 70Hz as long as it would play auditorium levels with little cone excursion and work well with K-tube or horn/WG

one of my old K10 has a Delta10A and although doesn't go "low", is quite impressive. It might be made better with an extra chamber at the top. (?)

Kappa12a based on an earlier sim of pym1298 could require 50l each for the rear chamber & 12.5l for front so a low Q 10" could help keep overall bulk down yet move enough air to be fun.


An externally hosted image should be here but it was not working when we last tested it.



An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Questions on your lovely new MTM XRK

New XKi MTM design with Beta 8cx and aluminum dome tweeter. 94dB at 2.83v sensitivity, BIG sound and big size. Thanks to member, Korny for the inspiration as it was a custom design for him.

First sound test here powered by the ALPHA 52w Class A Big Boy (BB) amp.

X, OK I'll bite :) is the Al dome crossed high as a super-tweet as the 8CX is a co-axial ?(can't quite work out what's going on here) Could this arrangement work as a 'regular' MTM with the recessed tweeter placement giving a degree of time alignment w.r.t regular mid-basses? I'm thinking something like Faital pro 6Fe200, though tweeters of matching sensitivity would be an issue if mid-basses in parallel...