we know from Julian Hirsch's rag that the Karlson 12 made its debut in the fall of 1954 - I'm guessing from that and Wayne Green saying Karlson took off a year to develop the small cabinets, that K8 was ready by 1955. The only cool finish I've seen was on the little KR5 Rocket - it was basically white, kinda textured and had gold sparkles. Karlson was a skilled cabinet maker to do such things.
well - this one is bluish - maybe my memory is off - or the Rockets came in several hue
could this be the "pebble" finish?
well - this one is bluish - maybe my memory is off - or the Rockets came in several hue
could this be the "pebble" finish?
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Near-field measurement of 0.53X Karlsonator
Here is a quick measurement (and sim) at about 1 m away (I couldn't go farther due to short cable from pc to amp). The speaker is on floor about 4 inches from back wall. The stuffing appears to have damped the initial bass peak overshoot and much of the smaller peaks but the general shape of where the bass shelf starts and drops off looks good, as does the 450 Hz broad 'bread loaf' shaped peak in the middle and some of the subsequent peaks. My pc soundcard has a limitation of a lowpass at 10 kHz so the signal falls off like a brick. The stuff below 50 Hz is probably ambient noise from furnace/AC nearby. In general, I think the model does a surprisingly good job and confirms my ears' that the speaker has a punchy and relatively balanced response - although a bit coloured (by +7 dB or so) in the 400 to 500 Hz region.
Here is a quick measurement (and sim) at about 1 m away (I couldn't go farther due to short cable from pc to amp). The speaker is on floor about 4 inches from back wall. The stuffing appears to have damped the initial bass peak overshoot and much of the smaller peaks but the general shape of where the bass shelf starts and drops off looks good, as does the 450 Hz broad 'bread loaf' shaped peak in the middle and some of the subsequent peaks. My pc soundcard has a limitation of a lowpass at 10 kHz so the signal falls off like a brick. The stuff below 50 Hz is probably ambient noise from furnace/AC nearby. In general, I think the model does a surprisingly good job and confirms my ears' that the speaker has a punchy and relatively balanced response - although a bit coloured (by +7 dB or so) in the 400 to 500 Hz region.
Attachments
1.24X Karlsonator with Dual FF225WK Drivers
I got a request from Jamikl to run a sim of the K'nator with dual FF225WK's - I guess he was encouraged from the results that I got with the dual Vifa's. These drivers have absolutely nothing in common... But... the fact that they both do really well in a K'nator appropriately sized. The results really astonished me and make me wish I had 4 FF225WK's lying around to build this box because it is just a bass killer monster as you will see.
First of all, the design details: the cabinet was scaled to 1.24X in order to make the front driver mounting plate large enough to accommodate two FF225WK's mounted one on top of another (2 x 8.2 in = 16.4 in, plus added 1 more in for clearance making it 17.4 in, 17.4 in/14 in = 1.24X). The rest of the K'nator cabinet is simply scaled up. The plans can be found here:
The simulation results are as follows: 1st plot is the frequency response at 3 m, note the incredibly deep and flat bass shelf from 30 Hz to about 125 Hz with good efficiency of 90 dB. There is a room-suckout from 125 Hz to 200 Hz. Placement of speaker from wall will affect this and can be used to adjust optimal listening position. Second plot is frequency response at 5 m away, still respectable solid bass shelf 'punch'. Third plot is impedance, fourth is driver cone displacement, fifth is the transient response (impulse function) which looks really good as all the main content is within 8 ms for minimal time smear. Finally, the sixth plot is the polar response which looks very good over a 60 deg cone in front of speaker. The results are all for 2.83V into two 8 ohm loads in parallel so it really is 2 watts input. The speaker is placed on floor with back 3 in away from wall (this distance is critical as it affects first dip tremendously). This speaker needs the back wall though - move too far away and the bass response falls apart.
Well, there it is... another speaker that really kicks butt.
I got a request from Jamikl to run a sim of the K'nator with dual FF225WK's - I guess he was encouraged from the results that I got with the dual Vifa's. These drivers have absolutely nothing in common... But... the fact that they both do really well in a K'nator appropriately sized. The results really astonished me and make me wish I had 4 FF225WK's lying around to build this box because it is just a bass killer monster as you will see.
First of all, the design details: the cabinet was scaled to 1.24X in order to make the front driver mounting plate large enough to accommodate two FF225WK's mounted one on top of another (2 x 8.2 in = 16.4 in, plus added 1 more in for clearance making it 17.4 in, 17.4 in/14 in = 1.24X). The rest of the K'nator cabinet is simply scaled up. The plans can be found here:
The simulation results are as follows: 1st plot is the frequency response at 3 m, note the incredibly deep and flat bass shelf from 30 Hz to about 125 Hz with good efficiency of 90 dB. There is a room-suckout from 125 Hz to 200 Hz. Placement of speaker from wall will affect this and can be used to adjust optimal listening position. Second plot is frequency response at 5 m away, still respectable solid bass shelf 'punch'. Third plot is impedance, fourth is driver cone displacement, fifth is the transient response (impulse function) which looks really good as all the main content is within 8 ms for minimal time smear. Finally, the sixth plot is the polar response which looks very good over a 60 deg cone in front of speaker. The results are all for 2.83V into two 8 ohm loads in parallel so it really is 2 watts input. The speaker is placed on floor with back 3 in away from wall (this distance is critical as it affects first dip tremendously). This speaker needs the back wall though - move too far away and the bass response falls apart.
Well, there it is... another speaker that really kicks butt.
Attachments
I'm not sure I'd want to use two drivers in 4"+ sizes, but I bet it's fun for your two little Vifa. A single FF225WK might go just as deep (other than more room for EQ) in a properly sized Karlsonator. Like I said before, FF225WK is fair in SK8, but a stronger motor is probably a good idea for a klassic Karlson, so think FE206En, TB W8-1772. The TL-load Karlsonator might be better for the FF drivers. Maybe I should do foamcore box for FF85WK, though I'm buried in other projects right now.
It's a pretty good response you have there on your enclosure. Bass extension is as good as the SK8 and it's smooth overal - on Karlson terms at least. What smoothing value did you use? (looks like 1/6-octave to me) I'd like for a sub-10" Karlson to have less than 10dB of suckout where that first dip occurs, I know some of freddi's larger K's can do it, more problematic the smaller you go it seems.
IG
It's a pretty good response you have there on your enclosure. Bass extension is as good as the SK8 and it's smooth overal - on Karlson terms at least.
What smoothing value did you use? (looks like 1/6-octave to me) I'd like for a sub-10" Karlson to have less than 10dB of suckout where that first dip occurs, I know some of freddi's larger K's can do it, more problematic the smaller you go it seems.IG
Last edited:
Here is a sim of a single FF225WK in a K'nator. Scaled to 0.88X (any larger or smaller and it doesn't look as flat). The response is not bad but not as flat as a dual driver setup.
3 m and 5 m frequency response.
\
3 m and 5 m frequency response.
\
Attachments
Last edited:
Not bad, but I guess what makes the 2x FF225WK sim look so good is how much longer the TL rear is and this is not an aspect that'll scale in our very case (1x Vs 2x drivers), so the 0.88x model suffers right there. Making it larger overall for a single driver will probably be too much Vb and get peaky, which is why the two drivers work well there.
IG
Hi XRK971 - how would a single Fostex 225, FE206EN or a Tang W1772 look in Greg B's "Karlsonator Eight" which is 28" tall by 15.5" deep by 13.5" wide?
Which 8" fullrange best fits his Karlsonator 8 for overall performance?
Which 8" fullrange best fits his Karlsonator 8 for overall performance?
An externally hosted image should be here but it was not working when we last tested it.
I have plans for the 0.53X Karlsonator especially adapted for use with standard 20 in x 30 in x 3/16 in thick foam core boards in the Foam Core thread here: http://www.diyaudio.com/forums/full-range/223313-foam-core-board-speaker-enclosures-191.html#post3562928
xrk971, would you happen to have a WT3 or other means of measuring driver impedance? It could be interesting to individually measure each TC9F in your Karlsonator seeing how they probably are seeing a different airload on their frontside. An ideal test would be free-air (or baffled) versus the position inside the Karlson.
IG
IG
Mini Karlsonator (0.53X) with Dual TC9FD's Thread
For those of you interested in building the Mini Karlsonator with dual Vifas, I started a new thread here: http://www.diyaudio.com/forums/full-range/239338-mini-karlsonator-0-53x-dual-tc9fds.html#post3564364
For those of you interested in building the Mini Karlsonator with dual Vifas, I started a new thread here: http://www.diyaudio.com/forums/full-range/239338-mini-karlsonator-0-53x-dual-tc9fds.html#post3564364
Regarding scaling the Karlsonator,
As mentionned before, straight-up scaling won't work in all cases as the rear chamber is meant to have quarter-wave action. Looking at the Karlsonator, the length suggests ~60Hz resonance. The fact that it is an expanding line might make this frequency a bit higher, but then it is mass-loaded, so perhaps it is brought back down to this neighbourhood. Actual measurements of a build would be more revealing ultimately. We could then know by what factor to scale each dimension in order to retain sensible dimensions for the driver in question.
I'm not certain what happens with the front chamber in this case and maybe an impedance plot could tell. If I go by my usual method, it might tune around 120Hz, but that did not always work with HAK enclosures.
20" x 30" FC could actually be used to build a full-size Karlsonator prototype, but I'd suggest triple-thick panels all over and a real plywood baffle-board for the driver along with lots of bracing, especially for the wings. It might work OK small-signal.
IG
As mentionned before, straight-up scaling won't work in all cases as the rear chamber is meant to have quarter-wave action. Looking at the Karlsonator, the length suggests ~60Hz resonance. The fact that it is an expanding line might make this frequency a bit higher, but then it is mass-loaded, so perhaps it is brought back down to this neighbourhood. Actual measurements of a build would be more revealing ultimately. We could then know by what factor to scale each dimension in order to retain sensible dimensions for the driver in question.
I'm not certain what happens with the front chamber in this case and maybe an impedance plot could tell. If I go by my usual method, it might tune around 120Hz, but that did not always work with HAK enclosures.
20" x 30" FC could actually be used to build a full-size Karlsonator prototype, but I'd suggest triple-thick panels all over and a real plywood baffle-board for the driver along with lots of bracing, especially for the wings. It might work OK small-signal.
IG
GregB - if you're reading this thread - what tuning(s) did you have in mind for the Karlsonator?
Doing one of my sketchy Hornresp kludgalations, it looks like the TL's >~60Hz tuning might get mass-loaded down to ~40Hz once the top-choke and front chamber "horn" are added. I sim'ed this as an OD horn. I added a ~20L/120Hz reflex to the other side of the driver to somewhat approximate the front chamber's effect (assuming my prediction is OK, which I have doubts, but still...) and the whole thing looks decent enough for a single FF225WK.
Argh, HR tried to "divide by zero", crashed and lost my input data!
Anyway, when scaling the K'ator, adjustment to the top-choke might give some leeway to setting a reasonable tuning for the rear TL section.
IG
Doing one of my sketchy Hornresp kludgalations, it looks like the TL's >~60Hz tuning might get mass-loaded down to ~40Hz once the top-choke and front chamber "horn" are added. I sim'ed this as an OD horn. I added a ~20L/120Hz reflex to the other side of the driver to somewhat approximate the front chamber's effect (assuming my prediction is OK, which I have doubts, but still...) and the whole thing looks decent enough for a single FF225WK.
Argh, HR tried to "divide by zero", crashed and lost my input data!
Anyway, when scaling the K'ator, adjustment to the top-choke might give some leeway to setting a reasonable tuning for the rear TL section.
IG
I will try choking the top channel to get some MLTL action going to see if it will drive the response down. I know that when I scaled it to 1.24X, the combination of the volume growing at x^3 but the top channel area growing at x^2, somehow must have formed the perfect balance for the MLTL to kick in as the tuning was like 30 Hz. Choking a standard Karlson does not achieve this - so the TL-Karlson hybrid gives you this nice flexibility to tune easily.
Edit: OK, here are the results for choking the top channel, starting at 1.0X (no choke), 0.75X, 0.50X, 0.35X. You get down to 30 Hz at 90 dB with full choke. Not bad... Next plots are impedance and cone displacement with full choke. These are for FF225Wk in full size Karlsonator.
Edit: OK, here are the results for choking the top channel, starting at 1.0X (no choke), 0.75X, 0.50X, 0.35X. You get down to 30 Hz at 90 dB with full choke. Not bad... Next plots are impedance and cone displacement with full choke. These are for FF225Wk in full size Karlsonator.
Attachments
Last edited:
Agree.
We're already getting some good ML-choke by going from CSA of ~885cm^2 to 160cm^2 in that top section. Then we must remember this does not vent into free-space just yet.
IG
Thanks for the simulations!
Stock K'ator shows cone null at ~40Hz, so in line with my HR sim, good news there. The second null is ~150Hz on your sim, but mine cannot predict it, I did so at 120Hz, which might indeed be wrong. It looks normal IMO that the second cone null does not change frequency with greater choke, but its amplitude does, which might be an indicator of less de-Q'ing of the front chamber QW mode by a larger vent at its apex.
IG
Stock K'ator shows cone null at ~40Hz, so in line with my HR sim, good news there. The second null is ~150Hz on your sim, but mine cannot predict it, I did so at 120Hz, which might indeed be wrong. It looks normal IMO that the second cone null does not change frequency with greater choke, but its amplitude does, which might be an indicator of less de-Q'ing of the front chamber QW mode by a larger vent at its apex.
IG
Last edited:
Here is the 'stock' impedance with FF225WK. First null is 40 Hz, spot on. Second remains at 150 Hz, and as you say is not affected by vent choke.
Attachments
Matches up to the cone motion null.
In the event that a scaled-down box needs so much choke as to take away too much useful de-Q'ing from a larger vent, the aperture starting width can likely be widened a bit to compensate, assuming it all truly works out this way.
IG