KARLSON - Page 10 - diyAudio
Go Back   Home > Forums > Loudspeakers > Full Range

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 10th June 2013, 03:25 PM   #91
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
IG,
The response at 60 inches is a lot better than I imagined it could look. The bass extension seems to only go to 75 Hz in this case - the 40 Hz vent output seems to have disapperared. So you threw this away in the trash after you were done with it? Kind of a waste... would have made a good starting point for drilling holes to make the Flute Pipe TL.
  Reply With Quote
Old 10th June 2013, 03:41 PM   #92
IG81 is offline IG81  Canada
diyAudio Member
 
IG81's Avatar
 
Join Date: Feb 2008
The FC prototype had served its immediate purpose, which was a quick and dirty proof of concept. Wasn't thinking about flute holes then.

IG
  Reply With Quote
Old 11th June 2013, 12:51 AM   #93
freddi is offline freddi  United States
diyAudio Member
 
freddi's Avatar
 
Join Date: Aug 2005
it would be nice to have a way to set the response of a regular Karlson Ultra-Fidelity enclosure type to limit peaking with a given driver -- IG81 do you think a BP sim set for certain Q is reliable enough to forge ahead or will front chamber height play a significant role? -- for a new thread think about how the front chamber should look including any reflector.
  Reply With Quote
Old 17th June 2013, 02:43 AM   #94
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Default AkAbak sim of 1955 K15 Speaker

I put together a model of the original 1955 John Karlson K15 (vented no slots) http://home.planet.nl/~ulfman/images/Plans/original.gif. I modeled the back chamber as a waveguide of varying cross sectional area prescribed by the plan (see drawings below). I modeled the front chamber as a waveguide of varying cross sectional area as well but with a series of 9 rectangular ports with CSA prescribed by a quadratic function that describes the K-slot. The two waveguides are connected by a common vent (the 165 mm x 85 mm wide port). The driver is an Eminence Kappa Pro 12A (http://www.eminence.com/pdf/Kappa_Pro_12A.pdf). Measurements are at 3 meters away with speaker set on stand so that driver centerline is 24 in above floor and wall is 5 ft back. Even at 3 meters away, the SPL is 90 dB at 1 watt.

The curve fit of the K-slot is quite accurate when I compared it to the prescribed numbers. The fit is given by Y=0.974X^2+0.00874X+0.0207. This is for X=0,1 and Y=0,1. Simply scale function for any slot based on max height and width at Y(X=1).

The frequency response shows a very strong bass response down to 60 Hz where it starts to decrease but has a decent SPL at 50 Hz. This is a very efficient speaker (the Eminence provides this). There are several dips and peaks as the frequency goes higher. I see a peculiar looking "W" shape from 150 to 200 Hz. Also shown are the impedance and cone displacement curves.

For anyone familiar with Karlson speaker measurements, does this look anything like it is supposed to? I may be completely missing the crucial points by modeling it as two waveguides connected by a vent with a leaky front waveguide. I think I may be missing something because this driver has an fs of 37 Hz and I am only getting 60 Hz peak.

By far, this was the most difficult and complex model I have done to date in AkAbak. It used about 30 nodes total.

This model has a universal scaling constant that will allow me to resize it en masse to shrink the cabinet and pop in a smaller driver. That immediately will impact the lower tuning freq as it depends on the path length.
Attached Images
File Type: png K15-Freq-resp-3m.png (26.6 KB, 634 views)
File Type: png K15-Freq-Imped.png (22.2 KB, 612 views)
File Type: png K15-Freq-Displ.png (22.5 KB, 600 views)
File Type: png K15-akabak-1.png (561.3 KB, 1753 views)
File Type: png K15-akabak-2.png (508.9 KB, 1735 views)

Last edited by xrk971; 17th June 2013 at 02:51 AM.
  Reply With Quote
Old 17th June 2013, 03:50 AM   #95
diyAudio Member
 
grindstone's Avatar
 
Join Date: Jun 2007
Location: GMT-6
Nice work, X. Thanks for sharing. Knee-jerk here is to wonder about the second WG as || combinations or series/parallel. Awaiting weigh-in by the K-noscenti
  Reply With Quote
Old 17th June 2013, 04:11 AM   #96
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Default K15 with Betsy-K

I tried putting a Betsy K driver in the K15 box (full size box) and the sim results are surprisingly good - I would say better looking than the more expensive Kappa 12A. In particular, the impedance doesn't go into a wild high value.
Attached Images
File Type: png K15-Freq-resp-3m-BetsyK.png (26.6 KB, 808 views)
File Type: png K15-Freq-Imped-BetsyK.png (22.1 KB, 87 views)
File Type: png K15-Freq-Displ-BetsyK.png (22.2 KB, 40 views)
  Reply With Quote
Old 17th June 2013, 06:20 AM   #97
diyAudio Member
 
fringy75's Avatar
 
Join Date: Feb 2012
Location: eyrewell forest, west eyreton, NZ
Hi X.

How small do you think this would dim down to? As an FYI, weather has been shocking where I live so have not had been able to build the double BLH for the Peerless that you sim'd for me
__________________

One of the advantages of being disorderly is that one is constantly making exciting discoveries.
  Reply With Quote
Old 17th June 2013, 10:01 AM   #98
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Fringy,
Hope you are ok with the weather not being like a monsoon or something. The K15 scales terribly the moment you shrink it as the tuning freq goes up very high and you get no bass when scaled to a 10 in box the freq is 100 Hz. I think I need a two dimensional scale so that length can be preserved while width is reduced to accommodate driver size and volume.
  Reply With Quote
Old 17th June 2013, 11:52 AM   #99
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Default K15 with FF225WK

I added the ability to scale the length (height) and the width/depth independently. Doing this, I tried putting a Fostex FF225WK in a K15 box scaled in all dimensions to a smaller cabinet with the scaling factor of 8/15=0.53. This will give a 447 mm high cabinet, 242 mm deep, and 302 mm wide. In other words, a proportionally sized cabinet compared to the original K15. The frequency response is just awful - it ruined the speaker as you can see. So obviously, one cannot just shrink the cabinet all around for a smaller driver and expect it to sound good. However, I shrunk it in just the width and depth, but kept the height the same ("Skinny" version). This produced a nice flat bass shelf down to 60 Hz as can be seen in second plot. So it seems that if one wants to shrink the cabinet size, more path folding will need to be done to preserve the internal TL length - and this is apparent in the multitude of smaller cabinet variants of the Karlson that are available out there. Also shown are plots of the impedance and cone displacement for the "Skinny" version of the K15 with FF225WK. For grins, I added the VERTICAL polar SPL plot - notice the effect of the K-aperture ports on the low vs high frequencies. The bass is aimed downwards while the higher frequencies are lobed as you go from -90 deg to + 90 deg.
Attached Images
File Type: png K15-Freq-resp-3m-FF225WK-Small.png (26.6 KB, 50 views)
File Type: png K15-Freq-resp-3m-FF225WK-Skinny.png (25.9 KB, 43 views)
File Type: png K15-FF225WK-Skinny-Imped.png (22.2 KB, 37 views)
File Type: png K15-FF225WK-Skinny-Displ.png (22.2 KB, 21 views)
File Type: png K15-FF225WK-Skinny-Vert-Polar.png (20.4 KB, 32 views)

Last edited by xrk971; 17th June 2013 at 11:58 AM.
  Reply With Quote
Old 17th June 2013, 12:39 PM   #100
IG81 is offline IG81  Canada
diyAudio Member
 
IG81's Avatar
 
Join Date: Feb 2008
Quote:
Originally Posted by xrk971 View Post
I put together a model of the original 1955 John Karlson K15 (vented no slots) http://home.planet.nl/~ulfman/images/Plans/original.gif. I modeled the back chamber as a waveguide of varying cross sectional area prescribed by the plan (see drawings below). I modeled the front chamber as a waveguide of varying cross sectional area as well but with a series of 9 rectangular ports with CSA prescribed by a quadratic function that describes the K-slot. The two waveguides are connected by a common vent (the 165 mm x 85 mm wide port). The driver is an Eminence Kappa Pro 12A (http://www.eminence.com/pdf/Kappa_Pro_12A.pdf). Measurements are at 3 meters away with speaker set on stand so that driver centerline is 24 in above floor and wall is 5 ft back. Even at 3 meters away, the SPL is 90 dB at 1 watt.

The curve fit of the K-slot is quite accurate when I compared it to the prescribed numbers. The fit is given by Y=0.974X^2+0.00874X+0.0207. This is for X=0,1 and Y=0,1. Simply scale function for any slot based on max height and width at Y(X=1).

The frequency response shows a very strong bass response down to 60 Hz where it starts to decrease but has a decent SPL at 50 Hz. This is a very efficient speaker (the Eminence provides this). There are several dips and peaks as the frequency goes higher. I see a peculiar looking "W" shape from 150 to 200 Hz. Also shown are the impedance and cone displacement curves.

For anyone familiar with Karlson speaker measurements, does this look anything like it is supposed to? I may be completely missing the crucial points by modeling it as two waveguides connected by a vent with a leaky front waveguide. I think I may be missing something because this driver has an fs of 37 Hz and I am only getting 60 Hz peak.

By far, this was the most difficult and complex model I have done to date in AkAbak. It used about 30 nodes total.

This model has a universal scaling constant that will allow me to resize it en masse to shrink the cabinet and pop in a smaller driver. That immediately will impact the lower tuning freq as it depends on the path length.
Kool! That's some good work. I've never had a K15 myself, but from many of freddi's measurements plus what I measured from my K12 and SK8, I think you're fairly close. The "W" effect has been known to me ever since I started to measure Karlsons. For me, it marks the upper limit of the functional gain BW. This is where one should cross-over if wanting to use the Karlson as a bass enclosure only. There may be ways to minimize such artefacts for fullrange use. For simulating their bass performance as a BP6 enclosure, which is often accurate enough, the rear chamber tunes as a bass-reflex, but at a lower frequency than predicted by a BR simulation since th vent is mass-loaded by the air "trapped" in the front chamber. The front chamber probably has a few elements going on that determine its tuning, but for classic aspect-ration Karlsons, I think that the vetical dimension's QW resonance dominates. This frequency will usually be around 1.05 to 1.1 times the frequency dictated by the internal height. In any case, it can be set at Fb in the BP6 simulation as if it were a Helmholtz resonator, and Ql=3 usually approximates the de-Q'ing effect well enough.

Your findings about scaling-down are in line with the experimentation that has taken place through the years. Initially, Karlson was not recommending scaling down his K15, despite stating that it could be done if absolutely needed. Smaller models can become a bit problematic and I think it goes to prove that there is some QW action going on. He eventually came out with three different K12's and a re-configured, somewhat smaller K15, named X15, with an internal HF waveguide. This was initially a wooden "Rocket/Klam" driven by a cone tweeter and subsequently a metal (?) K-Tube with a University (I think) compression driver. There may have been pressure on him from the market since this was around the time Acoustic Suspension and Stereo became popluar and nobody wanted two hulking monsters in their living rooms anymore. Still, K12's can be pretty decent overall, but are probably more picky about the driver than K15. 8" drivers of medium motor strength can work fairly well in there. There was also a factory K8, with a few variations I believe. Freddi has some of these. The Super-K8, or SK8, actually of K10-size if going by scale-down based on K15, is about the smallest size practical with decent performance IMO, if keeping the classic form-factor. It was designed by a few members of the Karlson forum.

Smaller Karlson have been built with apparent good performance, but deviated in form-factor. Among these was the HAK6.5 (High-Aspect-Karlson) built for Fostex FF166En. Retaining some longer pathlength worked there. Even smaller units with folded TL back chambers have been said to be good. One was the Skip/Lee K-TL.

Getting low bass out of a Karlson can be somewhat hard since there is heavy cancellation below the rear chamber tuning. It usually has to be muscled-out with a very low Fs strong-motor driver (think Altec 416) and some EQ preferably. Karlsons seem to retain some cone-control below Fb still, so can probably take more EQ than a simple reflex.

I fudged a simulation of K12 as a tapped horn in HR a year or two back. I really cannot do a K-slot in there, so made a somewhat equivalent horn for the front chamber. The equivalency is not physical really, but more about matching the impedance and SPL curve. Driver is Philips AD9710, which I have used in K12. It was a very sensitive system, but not as much as shown below, subtract 3-5dB perhaps.


Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

Click the image to open in full size.

IG

Last edited by IG81; 17th June 2013 at 12:58 PM.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Karlson Enclosure Javaman Multi-Way 123 1st September 2011 09:00 PM
Karlson tuning dlhughes Full Range 18 21st August 2010 07:16 PM
when are you going to try Karlson? freddi Full Range 2 20th November 2007 08:45 PM
Karlson Kolumn PeteMcK Multi-Way 0 16th November 2007 08:08 PM
Karlson Slot jadybug Full Range 4 28th March 2007 04:16 AM


New To Site? Need Help?

All times are GMT. The time now is 04:10 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2