that Transylvania tube is wonderful for home stereo. Here's my 15" klam made by John Lapaire with a diy K-tube crammed down the coaxial's HF horn throat. David A. Young was the first person I saw on forums to mention conversion of coaxial to koaxial
Karlson type klam playing a bit of Nokie Edwards and Adventure "Autumn Leaves" speaker demo - YouTube
I made an 18" klam. Carl Neuser has made them from small to 18" - a 12"-klam should be punchy and manageable (kinda) - could be series vented but that might need active highpassing when driven really loud.
I bet 4->8 little K12s in parallel with good drivers would make a punchy bass rig.
klam18
a coax or even 12lta could be fun in a 12 klam
Karlson type klam playing a bit of Nokie Edwards and Adventure "Autumn Leaves" speaker demo - YouTube
I made an 18" klam. Carl Neuser has made them from small to 18" - a 12"-klam should be punchy and manageable (kinda) - could be series vented but that might need active highpassing when driven really loud.
I bet 4->8 little K12s in parallel with good drivers would make a punchy bass rig.
klam18
An externally hosted image should be here but it was not working when we last tested it.
a coax or even 12lta could be fun in a 12 klam
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
E.G. Becks incarnations of Karlsons ideas:
Akustik, Lautsprecher, Boxen
In German, but there are interesting pics of both Karlsons and other concepts.
Most noteworthy some developments on Karlsons pronciples.
Greez
SigFire
Akustik, Lautsprecher, Boxen
In German, but there are interesting pics of both Karlsons and other concepts.
Most noteworthy some developments on Karlsons pronciples.
Greez
SigFire
Which company's brochure would that be, Transylvania Power Company or LEECH UK "Sound Lasers"? Would it be possible for you to scan and upload it if it's not something that's already around?
Did either Karlson or Green ever offer any insight as to what they were doing with the rear chambers. I know those doing modeling have treated the rear as a bandpass or BR, but what was in the mind of the designers? Was it just a baffled chamber for tuning and phasing? From the patent one might guess that it started as a single rear chamber vented at the top with further refinement from there. Nothing wrong with that approach, I knew an antenna feed horn designer at Comsat Labs that would get an idea and build a prototype for testing instead of first going to mathematicians.
Last edited:
Ron E. can you find an illustration of that type whistle?
the way K15's damping pads are arranged , the rear shelf appears partly a lowpass element to clean up things before the port cycles the rear wave into the front chamber. (The slit vent as in K12 would probably kill the cleaning action if driven by sine wave) Poppe seemed to think that John did work without resorting to math and has a lot of respect for Karlson.
Karlson's 1952 Audio Engineering Magazine article presented two simplified equivalent circuits
one of Wayne Green's ads when Karlson was only selling ~1400 enclosures/month. Julian Hirsch's "The Audio League Report" of 1955 covering the 15 inch and 12 inch Karlson enclosures is interesting.
http://img64.imageshack.us/img64/8420/wayne1400.gif
the way K15's damping pads are arranged , the rear shelf appears partly a lowpass element to clean up things before the port cycles the rear wave into the front chamber. (The slit vent as in K12 would probably kill the cleaning action if driven by sine wave) Poppe seemed to think that John did work without resorting to math and has a lot of respect for Karlson.
Karlson's 1952 Audio Engineering Magazine article presented two simplified equivalent circuits
An externally hosted image should be here but it was not working when we last tested it.
one of Wayne Green's ads when Karlson was only selling ~1400 enclosures/month. Julian Hirsch's "The Audio League Report" of 1955 covering the 15 inch and 12 inch Karlson enclosures is interesting.
http://img64.imageshack.us/img64/8420/wayne1400.gif
An externally hosted image should be here but it was not working when we last tested it.
JK was an accomplished horn/speaker designer among other things and he wanted to 'shrink' a horn, which like all cab alignments is a band-pass one, so the obvious way was to convert it into one or more series coupled resonant chambers, which today we define as 4th, 6th, 8th, etc., order BPs depending on how many there are.
4th, 6th order BP alignments were nothing new as these were what the pioneers chose for the original horn compression drivers back in the 1920s, so well known technology by those 'practiced in the art' of speaker design. Ditto the bass reflex, which dates from the early 1930s.
GM
IMO, much like an Altec VOTT A7 is a ~80Hz(?) front horn with a bass reflex supplementing the lower frequencies, the Karlson's front chamber will usually have gain in the 100Hz-200Hz region, depending on size, and the rear chamber's output vents in there to reach a bit lower. One could also vent it to the exterior I suppose. It'd look something like this, I drew it a few years back:
An externally hosted image should be here but it was not working when we last tested it.
What's the bandwidth (low and high) of a K sub with sub tuning (boost on the low end with PVC ports) ? Karlsons sound "funny" in the midrange, other merits apart.
not sure on the boosted - re-tuned K15 but in my rickety drywall house it was able to play a pretty clean 22Hz sine from my Altec 416 - it sounded pretty darn good with an 80Hz crossover to a K12 on theater organ CDs - a lot better quality and if makes any sense, "more adept" than an Eminence/Dayton 295070 in a reflex tuned and boosted similarly. Some of that impression may be the Altec being the better driver (?) Exemplar's K15 seems to get good reviews. Do you still have
a K15 for experiments?
a K15 for experiments?
Last edited:
Testimonies to the Karlson Faith
Steve Schell of Cogent loudspeaker fame
Re: Karlson -- why?
209.178.152.4
[ Follow Ups ] Thread: [ Display All Email ] [ High Efficiency Speaker Asylum ]
Posted by Steve Schell (M ) on September 27, 2004 at 17:21:15
In Reply to: Karlson -- why? posted by serenechaos on September 27, 2004 at 15:42:59:
I sense a bit of serene skepticism in your question. No problem, Karlsons do look pretty goofy at first glance. There is something exceptional in their performance, however, that audio writers and hobbyists have been struggling to explain for 50 years now.
It took me many years to appreciate accuracy in bass reproduction, which is quite different from sheer level or extension. Karlsons are exceptionally clean sounding as well as very extended for their size. It takes a bass horn of considerably greater bulk to outperform them. They impose no limitations on bandwidth- Karlson Associates used to recommend their use with the coaxial and triaxial drivers of the day. I much prefer them loaded with field coil theatre woofers, and big ol' midrange horns and compression drivers sitting on top.
*********************
John Lapaire
Hi Fred,
No scientific study here, but my RS spl meter measured a clear 8-10 db
difference between my old Theater 4s with Mike's Gauss/JBL components and
the Karlsons with either the cheap-*** Eminence or the Gauss. Guess which
had the most punch. These things really do have gain, with a visible
reduction in excursion. The big, full-sounding Theater 4s became suddenly
thin and weak.
We started by listening to his dual-10" 3way all-aluminum setup, which has
beautiful clear deep bass and good volume considering lower eff motors.
Then the Theater 4s with Gauss/JBL, then Eminence-loaded Ks, then swapped
top ends, including the K-tube (great sound, weak lower mids-no surprise).
Then all hell broke loose with crossover swaps, driver swaps etc. At one
point we stacked the Ks (wrong way, but we didn't have enough wire to flip
them!) Finally we wound up with Gauss in one K, Em in the other, 500 hz
crossover to JBL HF, cranked up the JBL lpads to match the Ks. This
arrangement actually shook dirt out of the ceiling/floor joists.
Mike's incredible OTL monoblocks each have 7 pairs of tubes running in
parallel, easily pushing the Ks up over 120 db at 1 meter, occasionally
pegging the meter over 126 db (I was wearing hearing protection). We
didn't do any sine wave tests, just music: rap, rock, jazz, but watching the
needle jump left no doubt about what frequency range was being energized.
One surprise was the difference between the $400 Gauss and $50 Em - or lack
of difference. Mike compared them in the Th 4 boxes, and we compared them
in the K boxes. Similar in tone and sensitivity, the Gauss slightly edged
out in control/detail IMO.
The Ktube did well enough to pursue further. Gotta get it down a little
lower without losing the top end.
I got some photos before running out of memory, they're on the hp site.
Don't want to publish Mike's equipment, suffice to say the quality of the
signal producing chain can probably not be improved on.
John Lapaire
*************************************************
Lars Moseholm
Subject: Some K15 observations - workings and principles
Dear Group
My sister’s vest is some useful equipment. Can I borrow some…. Hmm..
Take care out there and do not forget to enjoy the nice outdoor summer days and have some fun! Next week I go with my family to Lithuania of all places. It's a rather unic mix of old Northern and Baltic way of life behind all the late Eastern European Communist fixes. Wonder if John Karlson ever was in Europe?
Lars
A K15 BASS GUITAR KARLSON
Date: 07/18/02 11:13:19 AM
Name: Lars Moseholm
Email:
Subject: Some K15 obserþ£S‚. °T›
2 à=Ð/ LARS BUILDS A K15 BASS GUITAR KARLSON
Date: 07/18/02 11:13:19 AM
Name: Lars Moseholm
Email:
Subject: Some K15 observations - workings and principles
Dear Group
My sister’s boy wanted an enclosure for his bass guitar - a low cost and efficient one. I proposed to build a K15, and so we did, using the original x-15 dimensions. We used a 15” guitar speaker from Monacor we had in hand (the SP-385G: Fs=35Hz, Qts=0,4 modified, Vas=260 l, EBP=60).
First some observations without the tapered panels. The resulting back chamber vol. was 95 l without stuffing. We tuned it to 42 Hz using one 15-cm diameter port of length about 15 cm using nearly 65 vol-% stuffing. Unfortunately, we had to tweak the vent somewhat resistive in order to lower the box Q to get rid of some + 3 dB booming due to too small K15 back-chamber vol. for the given speaker. This turned the box into a hybrid between an aperiodic closed-box and a vented-box with an increased f0 near 60 Hz and a Qts of 0.7. It sounded really good. Dry bass and everything.
The speaker Z-plot revealed a major resonance at 61 Hz and a minor one at 315 Hz (the major front chamber parallel side walls resonance, there was no top-bottom resonance due to the titled speaker board, shelf and tilted top board). The port gave useful output between 40 – 120 Hz (-9 dB down on both sides). The speaker output was fairly flat (+/- 2 dB) down to about 60 Hz followed by the usual 4th order vented drop off.
The above results without the panels were all expected and fit with the usual TS box calculations.
With the tapered panels in place the front chamber was 55 l (behind the panels) and tuned to 150 Hz by an acoustic efficient tapered opening of 1320 cm2 (less than the nominal of the tapered opening of 1560 cm2 – not all area takes part in the tuning probably due to the tilted speaker board). With the tapered panels in place the usual K15 speaker terminal impedance top at 150 Hz showed up (other measurements have show 160 Hz), and other changes took place:
a) The back chamber port really took over. At 15-25 Hz the port output increased from 8 to 2 dB. The 2 dB increased output continued up to 40 Hz, and further increasing to 3 dB at 50 Hz. The vented port output dropped 9 dB from 50 to 120 Hz as mentioned above, however, the total K15 output was constant up to the 120 Hz (showed no drop). The K15 port further revealed a +4 dB top between 120 and 170 Hz, thereafter sharply dropping to meet the vented output at 250 Hz. So, the speaker drove the front chamber through the back chamber port!
b) The speaker output dipped at 200 Hz. The speaker output increased about 2 dB from 25 Hz to 80 Hz, and was the same as for the aperodic box in the range 80 – 100 Hz. That was followed by +1-3 dB up to 140 Hz. Between 140 to 300 Hz the speaker output suddenly dipped 10 dB at the near field, but that dip was somewhat compensated (but not fully) by the increased port output in that range.
We observed another constant 8-10 dB dip at 500 Hz (between 350-750 Hz), but that could be due to the guitar speaker itself. However, Fred also have measured the dips at 200 and 500 Hz recently using the Eminence and the JBL 2220H in his posting of 29th of June, and others as well. I also remember some curves showing a big K15 dip at 400 Hz.
c) Overall SPL increased. The nominated speaker SPL was 96dB. The measured K15 SPL was larger than 100 dB. Estimated total output was close to 120 dB. That speaker box really could punch out dry bass with practically no speaker movements.
How to make some sense out of this?
I modelled our K15 using standard 6th order band-pass set-up. This is a combination of a 4th order high-pass vented function, and a 2nd order front chamber low-pass function. This assumes output from the two ports only. The results clearly showed the observed increase in SPL, and the 200 Hz high-pass cut off.
However, the K15 also had output form the speaker itself at frequencies above the front chamber low-pass cut-off due to direct radiation. That could evidently explain the 200 Hz “dip”.
In a usual band-pass the two ports radiate into the environment. In the K15 the port from the back chamber radiates directly into the front chamber, but the two filter functions would not necessary significantly interfere with each other. Any additional mass loading from the front chamber on the back chamber port in our ported design would be small compared to the amount of port mass set into motion. This may not be the case in the original design.
The area between the shelf and the back wall is larger (and in our case definitely shorter, but that is not the case in the original design) than the port, and the same can be said about the opening between the front shelf and the tapered panels. In accordance, we saw no significant interference from the shelf on box resonance’s.
I think the K15 is what Freddy calls a series band-pass box. A more precise term could be that the K15 is a direct radiator 6th order series band-pass.
The coupler seems not to do any ¼ wave mass loading (no high-pass filter function). The right experiment to do would be to replace the tapered panels with a panel with a circular hole of similar effective size, and measure the acoustic filter response. If the same response was seeing – the tapered design means nothing for the filter function (but still a lot for the off-axis dispersion and some front chamber resonance's – I know). Unfortunately, the boy drove off with the box before I could do that small experiment. But now my own son got interested into this design after hearing the K15 box given full blown rap-music for the benefit of our neighbouring area. So, I think time will come for that panel shape experiment also.
The high efficiency is inherent to the band-pass and is – among other things – determined by the difference between the two Helmholtz frequencies. The Helmholtz resonance’s also explains the small diaphragm movements even for the high outputs.
The sound “colour” of the K15 was clearly determined by the balance between the direct radiated sound and the front chamber resonance’s and high-pass filter function. Also the diffraction of waves from the narrow part of the tapered opening was important.
It is not possible to specify a usual alignment for a band-pass. However, using the K15 dimensions and a front chamber tuning of about 150 Hz, the optimal driver parameters seem to be within the following ranges: Qts near 0.3, fs around 40-45 Hz, and Vas 400 – 450 l for a fairly flat frequency response. Does that fit with the early fifties good quality speakers?
A flat response is not so important in the bas range, so a large degree of flexibility should be allowed in practise (“the K15 fit many speakers”). It should also be possible to come up with some “cookbook” design rules for given drivers – or to justify the use of specific drivers. A lower front chamber resonance (140 Hz ?) may also help to smooth out the response (a smaller flare rate, tractrix curves)
Changing the diameter of the speaker – all other things equal – will only change the two “port” sizes in order to keep the proper tuning. However, Vas will also change as a function of sq.(Sd), so no simple scaling of dimensions exist.
In summary:
1. That a speaker box shows a tapered opening does not mean that the effect of that opening is the same for any box. The tapered opening in a ¼ wave pipe works in a very different way compared to the opening in the K15 (high-pass filter versus a low-pass filter, respectively!).
2. The K15 is a viable design, and it can basically be understood in usual engineering terms. However, in particular it’s on- and off-axis output cannot be easily predicted at present from TS-parameters.
3. The cabinet seems fairly robust to a range of speakers.
4. No simple scaling of the dimensions of the K-cabinet according to the speaker diameter can be recommended.
If one focuses at the need for total absence of resonance’s and a flat on-axis response in a speaker-box people should go for other designs. If one goes for high efficiency, low frequency punch, good dispersion and some work in the area of the unknown, so choose a Karlson. In fact, the K15 seems quite well behaved and a very interesting design.
I would like personally to thank the group for blowing new interest into this bass-box. After I build a dissapointing K12 35 years ago, I now understand that the K15 is different. I also think I understand why. But there is no way around the fact, that in order to build a broad recognition of the value of the design among the speaker community, things have to be justified in terms of TS-parameters and basic acoustic principles.
Further, I’m not sure I understand the seemly disagreement between the claim of usual degraded transient performance from a band-pass and my own - and certainly others at this forum observations of the “dry punch” from a K15. Can anyone enlighten me here without referring to some Karlson magic.
Best regards,
Lars
************************
Willem Norloos
Sorry Fred it took so long to answer, been busy rearanging and cleaning out my workplace (>20 bags of once "must have" garbidge went out).
As i said before, i like to use the K. as it was designed originally, however, i use it only for low up to 500 Hz. I build my first K's (12") when i was 14 years old and i got stuck with them ever since. On the other hand i spent most of my working years in the professional audio business (i took a job as a graphic designer in the 90's because of back-problems). I never used any K's in pro-audio, not because they are no good, but because other types of enclosures are usually better suited for the jobs on hand. In the end it is quality AND SPL's that counts.
For the big venue systems (mostly 4 to 6-way) we had a choice of 2x15" (raelly big) RCA bins, Cord-bins, several types of Cerwin Vega (L36, sensurround, etc) JBL and Martin-bins. The K is a rather broadbanded enclosure, it will work from 25 to whatever the woofer can do, where the other (horn systems) give more SPL in their narrow band, wich is a good fit to the 4 and more-way systems. In our studio i could never convince anyone to try anything else than their trusted JBL's (4343) and Altec's (don't remember the type but is was 2-way with a funny 90 deg. bent horn).
Last year i decided to make an active 2-way system for home use.
Before i made any choices i wrote down the design parameters and worked along those lines. I wanted reasonable flat response from 30-20kHz, constant dispersion of no less than 90 deg., and max. SPL of 115 dB over the entire band without ANY sign of stress or compression.
The upper part was really very easy, i have heard enough makes of drivers to vote for JBL or Altec without looking further. There are only a few types capable of starting at 500hZ and giving acceptable high's. JBL 2441 (Diamond pattern) would do fine. I found a pair of 2445 wich are also oke, just need a bit more EQ to lift 16kHz. The JBL 2380 horn was the obvious choice for this driver.
For the low end i looked at a few horn designs (Klipsch, Altec) but they were either too big or needed a corner to stand in, i do not have suitable corners.
I had my doubts about cone-control at high output levels with the normal vented boxes, so in the end there really was no competition left for the Karlsons.
I am happy to say, they do all that i expect from them!
At the moment i use the Cetec-Gauss 4583XR. The specs are:
Fs 31Hz
Qms 2.5
Qes 0.33
Qts 0.30
Vas 10.00
Highest Xo 800Hz
Xmax (lin) + & - 0.250"
Xmax (max) + & - 0.375"
RMS 400W
Flux 12000 Gauss
Eff. 97dB 1W@1m. (swept from 30 - 800Hz)
XR stands for:
- Kevlar reinforced cone
- High temp. VC
About the dips (400-500 region) you mentioned, they do not show up with the impedance measurement! There is only one (small) resonance at 140Hz. The peaks i found with reponse measurement were at 125 and 350Hz but they are probably room related.
About measuring from the slot:
I expect you measure with warble tones within one octave at the time. D.B. Keele (Electro Voice) has researched measuring nearby pressure fields (within the loudspeaker piston-range!) and found the results to be in good agreement with more traditional methodes (AES Anthology nr. 1). I would place the mike very close to the slot (1"), warble one octave and move the mike up and down for max. readout, and use this maximum. The max should be the same for each octave.
I never tested Karlson's claim for 120 deg. dispersion up to 10kHz because i do not use them that high. Maybe someone using a coax can verify the claim?
Well, i hope i have been of some help
Regards,
Willem
**********************************
Paul Searce builds a Karlson Ultra-Fidelity
Date: 09/27/02 11:18:59 PM
Name: Paul S.
Email: nscearce@citlink.net
Subject: University Karlson
I have just completed my first Karlson enclosure and loaded it
with a University UXC-123 driver. To these ears it sounds great.
I used the plans listed in the Oct. 1955 issue of Radio-Electronics,
slightly modified to allow a thicker back and tapers.
I listened and took some measurements with, and without the tapers in place.
Without the tapers there was a noticable peak in the mid bass as one would expect
from this driver in a 3.5 cu ft ported box.
With the tapers, bass is less peaked, and seems deeper. Hasty measurements show 3-6 dB
increase from 30Hz to 70Hz.
I wasn't sure how well the University would work in a Karlson, but it seems to be a decent match.
From what I have been reading here, I expected good things from the Karlson, and I'm not dissapointed.
Thanks everyone.
Paul
« Last Edit: May 12, 2008, 06:25:47 AM by freddyi »
Online
Steve Schell of Cogent loudspeaker fame
Re: Karlson -- why?
209.178.152.4
[ Follow Ups ] Thread: [ Display All Email ] [ High Efficiency Speaker Asylum ]
Posted by Steve Schell (M ) on September 27, 2004 at 17:21:15
In Reply to: Karlson -- why? posted by serenechaos on September 27, 2004 at 15:42:59:
I sense a bit of serene skepticism in your question. No problem, Karlsons do look pretty goofy at first glance. There is something exceptional in their performance, however, that audio writers and hobbyists have been struggling to explain for 50 years now.
It took me many years to appreciate accuracy in bass reproduction, which is quite different from sheer level or extension. Karlsons are exceptionally clean sounding as well as very extended for their size. It takes a bass horn of considerably greater bulk to outperform them. They impose no limitations on bandwidth- Karlson Associates used to recommend their use with the coaxial and triaxial drivers of the day. I much prefer them loaded with field coil theatre woofers, and big ol' midrange horns and compression drivers sitting on top.
*********************
John Lapaire
Hi Fred,
No scientific study here, but my RS spl meter measured a clear 8-10 db
difference between my old Theater 4s with Mike's Gauss/JBL components and
the Karlsons with either the cheap-*** Eminence or the Gauss. Guess which
had the most punch. These things really do have gain, with a visible
reduction in excursion. The big, full-sounding Theater 4s became suddenly
thin and weak.
We started by listening to his dual-10" 3way all-aluminum setup, which has
beautiful clear deep bass and good volume considering lower eff motors.
Then the Theater 4s with Gauss/JBL, then Eminence-loaded Ks, then swapped
top ends, including the K-tube (great sound, weak lower mids-no surprise).
Then all hell broke loose with crossover swaps, driver swaps etc. At one
point we stacked the Ks (wrong way, but we didn't have enough wire to flip
them!) Finally we wound up with Gauss in one K, Em in the other, 500 hz
crossover to JBL HF, cranked up the JBL lpads to match the Ks. This
arrangement actually shook dirt out of the ceiling/floor joists.
Mike's incredible OTL monoblocks each have 7 pairs of tubes running in
parallel, easily pushing the Ks up over 120 db at 1 meter, occasionally
pegging the meter over 126 db (I was wearing hearing protection). We
didn't do any sine wave tests, just music: rap, rock, jazz, but watching the
needle jump left no doubt about what frequency range was being energized.
One surprise was the difference between the $400 Gauss and $50 Em - or lack
of difference. Mike compared them in the Th 4 boxes, and we compared them
in the K boxes. Similar in tone and sensitivity, the Gauss slightly edged
out in control/detail IMO.
The Ktube did well enough to pursue further. Gotta get it down a little
lower without losing the top end.
I got some photos before running out of memory, they're on the hp site.
Don't want to publish Mike's equipment, suffice to say the quality of the
signal producing chain can probably not be improved on.
John Lapaire
*************************************************
Lars Moseholm
Subject: Some K15 observations - workings and principles
Dear Group
My sister’s vest is some useful equipment. Can I borrow some…. Hmm..
Take care out there and do not forget to enjoy the nice outdoor summer days and have some fun! Next week I go with my family to Lithuania of all places. It's a rather unic mix of old Northern and Baltic way of life behind all the late Eastern European Communist fixes. Wonder if John Karlson ever was in Europe?
Lars
A K15 BASS GUITAR KARLSON
Date: 07/18/02 11:13:19 AM
Name: Lars Moseholm
Email:
Subject: Some K15 obserþ£S‚. °T›
2 à=Ð/ LARS BUILDS A K15 BASS GUITAR KARLSON
Date: 07/18/02 11:13:19 AM
Name: Lars Moseholm
Email:
Subject: Some K15 observations - workings and principles
Dear Group
My sister’s boy wanted an enclosure for his bass guitar - a low cost and efficient one. I proposed to build a K15, and so we did, using the original x-15 dimensions. We used a 15” guitar speaker from Monacor we had in hand (the SP-385G: Fs=35Hz, Qts=0,4 modified, Vas=260 l, EBP=60).
First some observations without the tapered panels. The resulting back chamber vol. was 95 l without stuffing. We tuned it to 42 Hz using one 15-cm diameter port of length about 15 cm using nearly 65 vol-% stuffing. Unfortunately, we had to tweak the vent somewhat resistive in order to lower the box Q to get rid of some + 3 dB booming due to too small K15 back-chamber vol. for the given speaker. This turned the box into a hybrid between an aperiodic closed-box and a vented-box with an increased f0 near 60 Hz and a Qts of 0.7. It sounded really good. Dry bass and everything.
The speaker Z-plot revealed a major resonance at 61 Hz and a minor one at 315 Hz (the major front chamber parallel side walls resonance, there was no top-bottom resonance due to the titled speaker board, shelf and tilted top board). The port gave useful output between 40 – 120 Hz (-9 dB down on both sides). The speaker output was fairly flat (+/- 2 dB) down to about 60 Hz followed by the usual 4th order vented drop off.
The above results without the panels were all expected and fit with the usual TS box calculations.
With the tapered panels in place the front chamber was 55 l (behind the panels) and tuned to 150 Hz by an acoustic efficient tapered opening of 1320 cm2 (less than the nominal of the tapered opening of 1560 cm2 – not all area takes part in the tuning probably due to the tilted speaker board). With the tapered panels in place the usual K15 speaker terminal impedance top at 150 Hz showed up (other measurements have show 160 Hz), and other changes took place:
a) The back chamber port really took over. At 15-25 Hz the port output increased from 8 to 2 dB. The 2 dB increased output continued up to 40 Hz, and further increasing to 3 dB at 50 Hz. The vented port output dropped 9 dB from 50 to 120 Hz as mentioned above, however, the total K15 output was constant up to the 120 Hz (showed no drop). The K15 port further revealed a +4 dB top between 120 and 170 Hz, thereafter sharply dropping to meet the vented output at 250 Hz. So, the speaker drove the front chamber through the back chamber port!
b) The speaker output dipped at 200 Hz. The speaker output increased about 2 dB from 25 Hz to 80 Hz, and was the same as for the aperodic box in the range 80 – 100 Hz. That was followed by +1-3 dB up to 140 Hz. Between 140 to 300 Hz the speaker output suddenly dipped 10 dB at the near field, but that dip was somewhat compensated (but not fully) by the increased port output in that range.
We observed another constant 8-10 dB dip at 500 Hz (between 350-750 Hz), but that could be due to the guitar speaker itself. However, Fred also have measured the dips at 200 and 500 Hz recently using the Eminence and the JBL 2220H in his posting of 29th of June, and others as well. I also remember some curves showing a big K15 dip at 400 Hz.
c) Overall SPL increased. The nominated speaker SPL was 96dB. The measured K15 SPL was larger than 100 dB. Estimated total output was close to 120 dB. That speaker box really could punch out dry bass with practically no speaker movements.
How to make some sense out of this?
I modelled our K15 using standard 6th order band-pass set-up. This is a combination of a 4th order high-pass vented function, and a 2nd order front chamber low-pass function. This assumes output from the two ports only. The results clearly showed the observed increase in SPL, and the 200 Hz high-pass cut off.
However, the K15 also had output form the speaker itself at frequencies above the front chamber low-pass cut-off due to direct radiation. That could evidently explain the 200 Hz “dip”.
In a usual band-pass the two ports radiate into the environment. In the K15 the port from the back chamber radiates directly into the front chamber, but the two filter functions would not necessary significantly interfere with each other. Any additional mass loading from the front chamber on the back chamber port in our ported design would be small compared to the amount of port mass set into motion. This may not be the case in the original design.
The area between the shelf and the back wall is larger (and in our case definitely shorter, but that is not the case in the original design) than the port, and the same can be said about the opening between the front shelf and the tapered panels. In accordance, we saw no significant interference from the shelf on box resonance’s.
I think the K15 is what Freddy calls a series band-pass box. A more precise term could be that the K15 is a direct radiator 6th order series band-pass.
The coupler seems not to do any ¼ wave mass loading (no high-pass filter function). The right experiment to do would be to replace the tapered panels with a panel with a circular hole of similar effective size, and measure the acoustic filter response. If the same response was seeing – the tapered design means nothing for the filter function (but still a lot for the off-axis dispersion and some front chamber resonance's – I know). Unfortunately, the boy drove off with the box before I could do that small experiment. But now my own son got interested into this design after hearing the K15 box given full blown rap-music for the benefit of our neighbouring area. So, I think time will come for that panel shape experiment also.
The high efficiency is inherent to the band-pass and is – among other things – determined by the difference between the two Helmholtz frequencies. The Helmholtz resonance’s also explains the small diaphragm movements even for the high outputs.
The sound “colour” of the K15 was clearly determined by the balance between the direct radiated sound and the front chamber resonance’s and high-pass filter function. Also the diffraction of waves from the narrow part of the tapered opening was important.
It is not possible to specify a usual alignment for a band-pass. However, using the K15 dimensions and a front chamber tuning of about 150 Hz, the optimal driver parameters seem to be within the following ranges: Qts near 0.3, fs around 40-45 Hz, and Vas 400 – 450 l for a fairly flat frequency response. Does that fit with the early fifties good quality speakers?
A flat response is not so important in the bas range, so a large degree of flexibility should be allowed in practise (“the K15 fit many speakers”). It should also be possible to come up with some “cookbook” design rules for given drivers – or to justify the use of specific drivers. A lower front chamber resonance (140 Hz ?) may also help to smooth out the response (a smaller flare rate, tractrix curves)
Changing the diameter of the speaker – all other things equal – will only change the two “port” sizes in order to keep the proper tuning. However, Vas will also change as a function of sq.(Sd), so no simple scaling of dimensions exist.
In summary:
1. That a speaker box shows a tapered opening does not mean that the effect of that opening is the same for any box. The tapered opening in a ¼ wave pipe works in a very different way compared to the opening in the K15 (high-pass filter versus a low-pass filter, respectively!).
2. The K15 is a viable design, and it can basically be understood in usual engineering terms. However, in particular it’s on- and off-axis output cannot be easily predicted at present from TS-parameters.
3. The cabinet seems fairly robust to a range of speakers.
4. No simple scaling of the dimensions of the K-cabinet according to the speaker diameter can be recommended.
If one focuses at the need for total absence of resonance’s and a flat on-axis response in a speaker-box people should go for other designs. If one goes for high efficiency, low frequency punch, good dispersion and some work in the area of the unknown, so choose a Karlson. In fact, the K15 seems quite well behaved and a very interesting design.
I would like personally to thank the group for blowing new interest into this bass-box. After I build a dissapointing K12 35 years ago, I now understand that the K15 is different. I also think I understand why. But there is no way around the fact, that in order to build a broad recognition of the value of the design among the speaker community, things have to be justified in terms of TS-parameters and basic acoustic principles.
Further, I’m not sure I understand the seemly disagreement between the claim of usual degraded transient performance from a band-pass and my own - and certainly others at this forum observations of the “dry punch” from a K15. Can anyone enlighten me here without referring to some Karlson magic.
Best regards,
Lars
************************
Willem Norloos
Sorry Fred it took so long to answer, been busy rearanging and cleaning out my workplace (>20 bags of once "must have" garbidge went out).
As i said before, i like to use the K. as it was designed originally, however, i use it only for low up to 500 Hz. I build my first K's (12") when i was 14 years old and i got stuck with them ever since. On the other hand i spent most of my working years in the professional audio business (i took a job as a graphic designer in the 90's because of back-problems). I never used any K's in pro-audio, not because they are no good, but because other types of enclosures are usually better suited for the jobs on hand. In the end it is quality AND SPL's that counts.
For the big venue systems (mostly 4 to 6-way) we had a choice of 2x15" (raelly big) RCA bins, Cord-bins, several types of Cerwin Vega (L36, sensurround, etc) JBL and Martin-bins. The K is a rather broadbanded enclosure, it will work from 25 to whatever the woofer can do, where the other (horn systems) give more SPL in their narrow band, wich is a good fit to the 4 and more-way systems. In our studio i could never convince anyone to try anything else than their trusted JBL's (4343) and Altec's (don't remember the type but is was 2-way with a funny 90 deg. bent horn).
Last year i decided to make an active 2-way system for home use.
Before i made any choices i wrote down the design parameters and worked along those lines. I wanted reasonable flat response from 30-20kHz, constant dispersion of no less than 90 deg., and max. SPL of 115 dB over the entire band without ANY sign of stress or compression.
The upper part was really very easy, i have heard enough makes of drivers to vote for JBL or Altec without looking further. There are only a few types capable of starting at 500hZ and giving acceptable high's. JBL 2441 (Diamond pattern) would do fine. I found a pair of 2445 wich are also oke, just need a bit more EQ to lift 16kHz. The JBL 2380 horn was the obvious choice for this driver.
For the low end i looked at a few horn designs (Klipsch, Altec) but they were either too big or needed a corner to stand in, i do not have suitable corners.
I had my doubts about cone-control at high output levels with the normal vented boxes, so in the end there really was no competition left for the Karlsons.
I am happy to say, they do all that i expect from them!
At the moment i use the Cetec-Gauss 4583XR. The specs are:
Fs 31Hz
Qms 2.5
Qes 0.33
Qts 0.30
Vas 10.00
Highest Xo 800Hz
Xmax (lin) + & - 0.250"
Xmax (max) + & - 0.375"
RMS 400W
Flux 12000 Gauss
Eff. 97dB 1W@1m. (swept from 30 - 800Hz)
XR stands for:
- Kevlar reinforced cone
- High temp. VC
About the dips (400-500 region) you mentioned, they do not show up with the impedance measurement! There is only one (small) resonance at 140Hz. The peaks i found with reponse measurement were at 125 and 350Hz but they are probably room related.
About measuring from the slot:
I expect you measure with warble tones within one octave at the time. D.B. Keele (Electro Voice) has researched measuring nearby pressure fields (within the loudspeaker piston-range!) and found the results to be in good agreement with more traditional methodes (AES Anthology nr. 1). I would place the mike very close to the slot (1"), warble one octave and move the mike up and down for max. readout, and use this maximum. The max should be the same for each octave.
I never tested Karlson's claim for 120 deg. dispersion up to 10kHz because i do not use them that high. Maybe someone using a coax can verify the claim?
Well, i hope i have been of some help
Regards,
Willem
**********************************
Paul Searce builds a Karlson Ultra-Fidelity
Date: 09/27/02 11:18:59 PM
Name: Paul S.
Email: nscearce@citlink.net
Subject: University Karlson
I have just completed my first Karlson enclosure and loaded it
with a University UXC-123 driver. To these ears it sounds great.
I used the plans listed in the Oct. 1955 issue of Radio-Electronics,
slightly modified to allow a thicker back and tapers.
I listened and took some measurements with, and without the tapers in place.
Without the tapers there was a noticable peak in the mid bass as one would expect
from this driver in a 3.5 cu ft ported box.
With the tapers, bass is less peaked, and seems deeper. Hasty measurements show 3-6 dB
increase from 30Hz to 70Hz.
I wasn't sure how well the University would work in a Karlson, but it seems to be a decent match.
From what I have been reading here, I expected good things from the Karlson, and I'm not dissapointed.
Thanks everyone.
Paul
« Last Edit: May 12, 2008, 06:25:47 AM by freddyi »
Online
Gave it away (and it's still in use to this day) when I moved a few years ago although kenpeter has one I could possibly borrow if needed. My interest is on something to play a clean 30 Hz to 300 Hz, thereabouts.
I guess a decade is a bit of a stretch for almost anything with some kind of usable efficiency. The better measuring Karlson types seem to manage 2-2.5 octaves of flat response between the bass roll-off and the first dip. The efficiency is maybe not quite as high as a full-size horn, but still up there IMO.
Here are some near-slot measurements I once took for the 10" Celestion TF1020 mounted in my DIY Karlsonette. IIRC, the mic was pretty much at the slot plane, probably somewhere above driver axis, where the slot is getting narrower.
An externally hosted image should be here but it was not working when we last tested it.
I'd like to know why some off my near-cone measurements showed a null 1/2 octave below system tuning. I got derailed on TF1020 when one arrived with frame bent bad enough to crease the cone - what would you build for a dedicated K10 using TF1020?
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Too bad about that busted TF1020. You got a pair right? Maybe stick the good one in an orphan K you may have, worth it IMO.
I figure it may actually have been not bad in SK8, which pretty much is a K10. Karlsonator 8 or perhaps even 12 would be a safe bet IMO. It probably has enough BW to allow 1st order Xo to a K-Tube; it nearly hits 5khz, fairly clean to boot, on-axis.
I figure it may actually have been not bad in SK8, which pretty much is a K10. Karlsonator 8 or perhaps even 12 would be a safe bet IMO. It probably has enough BW to allow 1st order Xo to a K-Tube; it nearly hits 5khz, fairly clean to boot, on-axis.
With regards to the first page mention of the Karlon slot antenna's....they are still around and in use!
Thought these 2 pictures would make Freddi's day better!
CJ
One of the first things I noticed when I began working for a defense contractor manufacturing microwave antennas, were the similarities between microwave horn feeds and audio horns. An engineer friend of mine who was into audio a bit told me microwaves and audio waves have similar problems, but in different frequency ranges.
For example. High frequency audio waves are inhibited by objects which are larger than a wave length. Microwaves have difficulty propogating around rain drops.
We had many "Karlson" type horn feeds of different shapes scattered about the lab for testing, but I don't recall actually installing them in a system for deployment, save for one which was about a 4" long solid square aluminum tube, with a Karlson quad arrangement. If I remember right I think that horn was designed to receive and transmit a signal between 18 to 40 Gigahertz. At the time I was there we were developing a 60 Gig horn feed which was nothing more than a rectangular waveguide.
Last edited:
I'm re-posting a comment of mine from another thread.
I built a 15" version of the box using an EVM 15L for a bass player friend of mine, and the vocalist in his band told me the bass sounded out of tune, then in tune. Not only I, but two friends familiar with enclosure design thought the guy was nutz. He was after all, prone to indulging in the cannabis on an hourly basis.
Do not click in the link. I don't know where it came from, and I don't know how to get rid of it.
About two years later in 1985 on a gig using a K12, where during the second set I had to share a piece of music with the keyboard player on the other side of the stage, I noticed I was a bit flat, so I tuned up. After the song was over I returned to my side of the stage I noticed I was sharp.
It hit me like a ton of bricks! It appeared that off axis higher frequencies suffered from, lack of a better term, a doppler effect. It explained why the singer told me the bass sounded out of tune, then in tune. The singer noticed this because he was moving about the stage.
It was when I discovered the problem with the box I quit producing Karlsons. However, I built a new prototype two months ago using an EVM 12L with the intent to going back in production for use as electric bass rigs.
I have to say, I'm a bit tickled my thread has taken on a life of it's own. Thank you all for responding with thoughtful and intelligent reply's.
H.F.
I built a 15" version of the box using an EVM 15L for a bass player friend of mine, and the vocalist in his band told me the bass sounded out of tune, then in tune. Not only I, but two friends familiar with enclosure design thought the guy was nutz. He was after all, prone to indulging in the cannabis on an hourly basis.
About two years later in 1985 on a gig using a K12, where during the second set I had to share a piece of music with the keyboard player on the other side of the stage, I noticed I was a bit flat, so I tuned up. After the song was over I returned to my side of the stage I noticed I was sharp.
It hit me like a ton of bricks! It appeared that off axis higher frequencies suffered from, lack of a better term, a doppler effect. It explained why the singer told me the bass sounded out of tune, then in tune. The singer noticed this because he was moving about the stage.
It was when I discovered the problem with the box I quit producing Karlsons. However, I built a new prototype two months ago using an EVM 12L with the intent to going back in production for use as electric bass rigs.
I have to say, I'm a bit tickled my thread has taken on a life of it's own. Thank you all for responding with thoughtful and intelligent reply's.
H.F.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.