strong point might be a lot of "hit" and low distortion in the midbass
one 8 cubic foot coupler started out as a 15" box but worked fairly well with a lower mass old Eminence 18" and had much less distortion than a Yorkville UCS1 This coupler was a one 4'x8' sheet project. There's no rear shelf in this 32"x21.5"x20" coupler. The rear shelf doesn't usually have much effect upon response but does help tune the transient kick. Using a rear shelf with tight gap puts a lotta of bend on the rear panel. A front shelf deflects some highs away from the upper part of the front chamber and in K15 adds to the vent boundary.
subjectively, adding a front shelf to this coupler made it sound more like K15 - subjective power was good indoors and out and the bass could be felt.
omitting a front shelf might be a bit better for some horns on top if the xover is high
pulling the baffle forwards might help eliminate the first dip and also would allow an internal slotted Karlson waveguide tube
JEK's Fig 6 reflector from patent 3540544
Effect of reducing rear chamber volume with a half-sack of dry dog food (guessing around -1 cubic foot)
Effect of front shelf upon ~ground plane measurement - (its tuning the two vent holes a bit lower)
K vs USC1 horn small signal 34Hz sine
K vs an 11 cubic foot commercial pipehorn 20vrms/50Hz
Yorkville which I used for comparison
- impedance of another coupler with 20 liter air mass stub added to its top
rough sketch showing the 20 liter stub I added to the front chamber
stubby-K ground-plane - perhaps its 80 liter rear chamber could have been a bit larger-?
Metro approached a coupler/BP as follows with its T-15
one 8 cubic foot coupler started out as a 15" box but worked fairly well with a lower mass old Eminence 18" and had much less distortion than a Yorkville UCS1 This coupler was a one 4'x8' sheet project. There's no rear shelf in this 32"x21.5"x20" coupler. The rear shelf doesn't usually have much effect upon response but does help tune the transient kick. Using a rear shelf with tight gap puts a lotta of bend on the rear panel. A front shelf deflects some highs away from the upper part of the front chamber and in K15 adds to the vent boundary.
subjectively, adding a front shelf to this coupler made it sound more like K15 - subjective power was good indoors and out and the bass could be felt.
omitting a front shelf might be a bit better for some horns on top if the xover is high
pulling the baffle forwards might help eliminate the first dip and also would allow an internal slotted Karlson waveguide tube
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.
JEK's Fig 6 reflector from patent 3540544
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.
An externally hosted image should be here but it was not working when we last tested it.
Effect of reducing rear chamber volume with a half-sack of dry dog food (guessing around -1 cubic foot)
An externally hosted image should be here but it was not working when we last tested it.
Effect of front shelf upon ~ground plane measurement - (its tuning the two vent holes a bit lower)
An externally hosted image should be here but it was not working when we last tested it.
K vs USC1 horn small signal 34Hz sine
An externally hosted image should be here but it was not working when we last tested it.
K vs an 11 cubic foot commercial pipehorn 20vrms/50Hz
An externally hosted image should be here but it was not working when we last tested it.
Yorkville which I used for comparison
An externally hosted image should be here but it was not working when we last tested it.
- impedance of another coupler with 20 liter air mass stub added to its top
An externally hosted image should be here but it was not working when we last tested it.
rough sketch showing the 20 liter stub I added to the front chamber
An externally hosted image should be here but it was not working when we last tested it.
stubby-K ground-plane - perhaps its 80 liter rear chamber could have been a bit larger-?
An externally hosted image should be here but it was not working when we last tested it.
Metro approached a coupler/BP as follows with its T-15
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.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Very interesting. thanks.
I couldnt quite understand your "rough sketch". Is that a separate chamber on top?
With that "fig 6" type curve, have you ever tried to curve it the other way? Convex rather than concave in relation to the driver?
Just a crazy thought but has anyone ever tried the rear chamber sealed?
Any ideas as to why the shelves effect transient "kick". I thought they were just to act as a sort of low pass.
What is the source of that massive dip around 250?
MrKramer
I couldnt quite understand your "rough sketch". Is that a separate chamber on top?
With that "fig 6" type curve, have you ever tried to curve it the other way? Convex rather than concave in relation to the driver?
Just a crazy thought but has anyone ever tried the rear chamber sealed?
Any ideas as to why the shelves effect transient "kick". I thought they were just to act as a sort of low pass.
What is the source of that massive dip around 250?
MrKramer
yeah - the top of that coupler was removable so I added a 20 liter chamber which connected to the front via a 1.4" x ~20" slot to see what happened.
I never tried the other way but Walter Zintz's KHYBOE and Acoustic Control's 115BK have an inverse transflex path.
internal construction of 115BK - it has a 3"x9" vent cut in the center of its reflector
the rear shelf along with damping forms a lowpass filter before the main vent - impedance can be juggled a bit with the lowpass gap - maybe the extra rear chamber is damping something which occurs during a strong bass passage.
here's an old outdoors graph of the regular Karlson with EV15L vs a reflex the size of its rear chamber with Peavey 1502-8 woofer . Although both tuned around 50Hz by Z-minimum, 37Hz and both cabinets loaded with M151, the Karlson had much less cone excursion than the reflex with two 4" ID ducts
K15 could do well on two tone sine tests compared to reflex. K12 with the distributed slit vent looks poor on sine
rear chambers have been run sealed - here's a groundplane of the little 115BK with cardboard over its vent - not a good seal to play but you
can see the trend and response from that perspective with vent open vs blocked
the front geometry must affect the bounce and dip(s) but perhaps the rear chamber shape has some role too plus there can be a cavity peak
then fall which seems to be filled in (?) by the rear wave
I never tried the other way but Walter Zintz's KHYBOE and Acoustic Control's 115BK have an inverse transflex path.
An externally hosted image should be here but it was not working when we last tested it.
internal construction of 115BK - it has a 3"x9" vent cut in the center of its reflector
An externally hosted image should be here but it was not working when we last tested it.
the rear shelf along with damping forms a lowpass filter before the main vent - impedance can be juggled a bit with the lowpass gap - maybe the extra rear chamber is damping something which occurs during a strong bass passage.
here's an old outdoors graph of the regular Karlson with EV15L vs a reflex the size of its rear chamber with Peavey 1502-8 woofer . Although both tuned around 50Hz by Z-minimum, 37Hz and both cabinets loaded with M151, the Karlson had much less cone excursion than the reflex with two 4" ID ducts
An externally hosted image should be here but it was not working when we last tested it.
K15 could do well on two tone sine tests compared to reflex. K12 with the distributed slit vent looks poor on sine
rear chambers have been run sealed - here's a groundplane of the little 115BK with cardboard over its vent - not a good seal to play but you
can see the trend and response from that perspective with vent open vs blocked
An externally hosted image should be here but it was not working when we last tested it.
the front geometry must affect the bounce and dip(s) but perhaps the rear chamber shape has some role too plus there can be a cavity peak
then fall which seems to be filled in (?) by the rear wave
Last edited:
the distributed vent in this case is the Karlson 12 as presented in July 1958 Popular Mechanics and sold by Karlson probably from 1956 onwards - look at the lower right cabinet with 6 - saw blade slits
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Has anyone still on this forum tried K wings on an H or maybe even U frame open baffle? I can imagine them either flat and really close to the cone or angled out at the base to make a truncated pyramid kind of thing.
MrKramer
MrKramer
Magnetar experimented with a dipole klam with 21" woofer but didn't post picture. I don't think it would help an existing design like MJK's GW-1858 H-baffle. "Icebear" had a flat panel version of Forsman's VSS (Vertical Split System) which features an inverted slot on the rear side something happend to the pic at imageshack so all I can find is this 7kb thumbnail
here's a patent which fails to mention Karlson and uses a dual driver "klam"
http://news.google.com/patents/about?id=O1M7AAAAEBAJ&dq=4325454
An externally hosted image should be here but it was not working when we last tested it.
here's a patent which fails to mention Karlson and uses a dual driver "klam"
http://news.google.com/patents/about?id=O1M7AAAAEBAJ&dq=4325454
Last edited:
Freddi, Thanks for all the plots.
So it looks like something that will do okay < 150Hz. +5dB in the bass reflex comparison chart is good.
Any comparisons vs a TL design?
So it looks like something that will do okay < 150Hz. +5dB in the bass reflex comparison chart is good.
Any comparisons vs a TL design?
Ben, you still have not explained anything about how you ADD an extra hundred watts?? How much did you start with to be able to ADD a second hundred watts??
If it was 100 watts , then you "added" exactly 3dB of headroom, eh?
And what tweeter can handle "100 watts"??
Please explain??
I'm not saying anything about if ur right or wrong, just want to know what ur talking about??
Ben, I am not sure what ur talking about exactly WRT "coconuts", since I never said anything about using them??
But you may be mistaking added harmonic distortion for "loudness" of certain instruments or portions of the spectrum. Many people think that a speaker sounds "loud" when they are actually hearing added distortion from the tweeter!
It is surprising to hear a system that does not particularly increase (above a threshold - but that's a more complex discussion...) in added harmonic distortion(s) being played loudly - it just does not sound loud - it sounds BIGGER. Most systems don't do that trick at all.
If you think that "horns" do "brass" well then you have only heard horns that are adding harmonics - ie. not really good horns.
My point is that as DIYers we can accomplish outstanding speaker reproduction and signal chain quality at a fraction of the cost that it would take to buy something that actually did the job as well.
-------------------
On an aside, since this is on K-slots, not K-horns, I know quite a bit about the K-horn. If ur still using the stock EV diffraction tweeter, mounted stock, then you have a tweeter that can't handle much power and has monster harmonic distortion. The older larger format mid horn is actually not too bad, although it does have some issues.. imho.
_-_-bear
freddi,
That "Metro" speaker looks interesting - got a link?
Would like to know more technical details and if the front slot is for show or if it actually plays a role...
Have you ever seen a hyperbolic folded bass horn where the rear enclosure of the driver is ported to the front??
Ben, that story about the hi-fi show down on "radio row" was cool... too bad no one thought to take pictures!
_-_-bear
That "Metro" speaker looks interesting - got a link?
Would like to know more technical details and if the front slot is for show or if it actually plays a role...
Have you ever seen a hyperbolic folded bass horn where the rear enclosure of the driver is ported to the front??
Ben, that story about the hi-fi show down on "radio row" was cool... too bad no one thought to take pictures!
_-_-bear
I'm new to the Karlson discussion, but it seems to me that the warts on the Karlson bass cab measurements--specifically the deep notches--aren't due to any failure of the fundamental concept, just due to problems with execution (parallel surfaces in the kavity causing resonant cancellations).
To avoid tossing the baby with the bathwater, it seems a good idea to look more closely at designs that are simpler, purer applications of the concept (like well-built sealed klams with no parallel walls, or look at rockets or, most of all, at k-tubes).
To me, these would offer the most interesting insights into the Karlson concept because they aren't obscured behind overlays of reflex ports, internal reflections, etc.
I'm staying tuned.
To avoid tossing the baby with the bathwater, it seems a good idea to look more closely at designs that are simpler, purer applications of the concept (like well-built sealed klams with no parallel walls, or look at rockets or, most of all, at k-tubes).
To me, these would offer the most interesting insights into the Karlson concept because they aren't obscured behind overlays of reflex ports, internal reflections, etc.
I'm staying tuned.
Last edited:
Bill F - the dips are inherent to the design afaik.
The Karlson models like a horn that is undersized for the job - you get the same dips - it's a comb filter effect.
The one like the Metro incarnation reduce to a "bad bass horn" enhancing the directly coupled output of a 15" speaker. The area that is unclear is IF they managed to put the effect of the "bad horn" (the Karlson slot) at a frequency where it is "good" and there is some additional output from the rear (acoustically lowpassed) short horn, then you do augment the response in a potentially good way for PA work, in a smallish box.
Found the 2007 discussion over at the asylum on this topic, with the pdf for Metro. They seem to think that the output of a port is out of phase with the woofer... which leads me to think that they are guessing 100% about everything. It is called an "acoustical phase inverter" for a reason??
_-_-bear
The Karlson models like a horn that is undersized for the job - you get the same dips - it's a comb filter effect.
The one like the Metro incarnation reduce to a "bad bass horn" enhancing the directly coupled output of a 15" speaker. The area that is unclear is IF they managed to put the effect of the "bad horn" (the Karlson slot) at a frequency where it is "good" and there is some additional output from the rear (acoustically lowpassed) short horn, then you do augment the response in a potentially good way for PA work, in a smallish box.
Found the 2007 discussion over at the asylum on this topic, with the pdf for Metro. They seem to think that the output of a port is out of phase with the woofer... which leads me to think that they are guessing 100% about everything. It is called an "acoustical phase inverter" for a reason??
_-_-bear
Let's do a little thought exercise about this design concept.
Let's say we had a ported cabinet, direct radiator.
Now, we take some volume, and place it in front of the front baffle.
Next we cut a hole in that new front box.
How big?
Dunno.
What does that look like?
Maybe a "bandpass woofer" of some sort?
Problem is that the hole we cut will try to resonate at one singe frequency best.
So, now we want to "spread" the resonances - turn the hole into a triangle?
Alright, make it look like an exponential curve...
Make it BIGGER and the "Q" goes down too...
Next we change the internal shape of the front volume so that it slopes, and that gives some compression and variation in distance, and again some more spreading of resonances...
Viola! Karlson Slotted Enclosure??
Kind of a low Q horn, but with not enough mouth size to avoid the comb filter effects... whaddya think folks?
_-_-bear
Let's say we had a ported cabinet, direct radiator.
Now, we take some volume, and place it in front of the front baffle.
Next we cut a hole in that new front box.
How big?
Dunno.
What does that look like?
Maybe a "bandpass woofer" of some sort?
Problem is that the hole we cut will try to resonate at one singe frequency best.
So, now we want to "spread" the resonances - turn the hole into a triangle?
Alright, make it look like an exponential curve...
Make it BIGGER and the "Q" goes down too...
Next we change the internal shape of the front volume so that it slopes, and that gives some compression and variation in distance, and again some more spreading of resonances...
Viola! Karlson Slotted Enclosure??
Kind of a low Q horn, but with not enough mouth size to avoid the comb filter effects... whaddya think folks?
_-_-bear
Are you sure? Others maintain it is dimensional cancellations. Measurements I've seen of Karlson tubes -- the purest application of the concept -- always had cancellation notches that looked to me like measurement artifacts, probably due to the fact that the tube diameter is acoustically small through most of the passband, and the aperture is far forward of the baffle, so bounce from rearward radiation is hard to avoid.
I would be very interested to see a horizontal polar measurement.
Look at a graph of a horn that is too small and short... same graph.
It is somewhat irrelevant where the "bounce" comes from - be it a reflection from the mouth of a horn or another physical or acoustical boundary. It is inherent in the design.
_-_-
PS. if one utilizes the effect by rolling off the speaker with a LOW pass, it reverts to being a type of "bandpass" design (forget that all speakers are in effect a "bandpass") of an octave or two width...
It is somewhat irrelevant where the "bounce" comes from - be it a reflection from the mouth of a horn or another physical or acoustical boundary. It is inherent in the design.
_-_-
PS. if one utilizes the effect by rolling off the speaker with a LOW pass, it reverts to being a type of "bandpass" design (forget that all speakers are in effect a "bandpass") of an octave or two width...
Last edited:
Ken Lehman made an 8" coupler having a large curved reflector with what I'd assume was a W0838R Eclipse poly woofer - KenL said it had dips and echoes, he attached a tapered stub to the top of its front chamber (see removable plate in pic) and it smoothed out. There was a duct right above the woofer.
KenL's K8
if an inverse taper transflex scheme is used rather than a port, will the result be moving away from coupled cavity and the result more "mellow"?
Here's Metro's E-Array manual in English which discusses the T-15 Karlson hybrid
http://www.metroaudiosystems.gr/pages/manuals/E_Array_Manual_en.pdf
KenL's K8
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.
if an inverse taper transflex scheme is used rather than a port, will the result be moving away from coupled cavity and the result more "mellow"?
Here's Metro's E-Array manual in English which discusses the T-15 Karlson hybrid
http://www.metroaudiosystems.gr/pages/manuals/E_Array_Manual_en.pdf
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:
Maybe I'm a bit thick, but I'm still not convinced. Let's take the measurements in the previously linked Acoustical Tube study for example.
The K-tube FR graph shows a big notch at 2.1kHz and some light ripple above that. The 7cm k-tube is very short and, factoring in the compression driver's throat length, I'm pretty sure it violates the design guideline I recall that at least 3/4 of the total tube length should be slotted (I wish he'd tested a more standard design along the lines of the Transylvania Tube). Perhaps that explains the ripple, but I'm not sure it explains the 2.1kHz notch.
Because look now at the FR graph for the 14cm angle-cut "elliptic profil" tube. It's twice as long and a completely different profile, but it has the *exact* same 2.1kHz notch. Clearly a measurement artifact, IMHO.
Also notice that that angle-cut tube, with its higher proportion of "slotted" length, displays far less treble ripple.
So, I suggest that with the long cut you no longer have much loading of any sort and the "Q" of the system is greatly reduced, thus the ripples are also reduced.
With the shorter cuts the inherent ripples are quite evident.
As is the case with the standard Karlson Enclosure.
Perhaps if someone scaled it up to a large size, one might see the ripples reduce? But then, is it any better than a horn of equivalent size, or a direct radiator box??
The notch at 2.1kHz? I didn't even take a close look at that... but it is not likely a measurement artifact... a bigger issue is the impedance curves. The impedance curve of my somewhat larger compression driver on a multii-cell horn that is not optimal at all looks very much like a flat line for the most part... not a dipsy doodle affair... the dipsy doodle indicates that the load on the diaphragm is varying all over the place...
Seems to me that Western Electric did a tweeter with a similar shape in the 30's but perhaps I am imagining I read this...
these issues are somewhat immutable and not up to my say-so...
_-_-bear
With the shorter cuts the inherent ripples are quite evident.
As is the case with the standard Karlson Enclosure.
Perhaps if someone scaled it up to a large size, one might see the ripples reduce? But then, is it any better than a horn of equivalent size, or a direct radiator box??
The notch at 2.1kHz? I didn't even take a close look at that... but it is not likely a measurement artifact... a bigger issue is the impedance curves. The impedance curve of my somewhat larger compression driver on a multii-cell horn that is not optimal at all looks very much like a flat line for the most part... not a dipsy doodle affair... the dipsy doodle indicates that the load on the diaphragm is varying all over the place...
Seems to me that Western Electric did a tweeter with a similar shape in the 30's but perhaps I am imagining I read this...
these issues are somewhat immutable and not up to my say-so...
_-_-bear
"The Tube"
Ti205 in-room
An externally hosted image should be here but it was not working when we last tested it.
Ti205 in-room
An externally hosted image should be here but it was not working when we last tested it.
- Status
- Not open for further replies.