Karlson Slot

[jadybug]

Hi all, this is my first post of too many questions to hold in my head. I have
been floating around the internet and have seen many different types of full
range speakers. My main interests tend to lean between TL's and Horns. I have
looked at some of the designs put forth by Planet 10. The iBIBk looks like a
contender but I have a question about the Karlson Slot. Would the speaker react
the same way if it had a 27 in. sq. area opening/mouth verses the Karlson Slot?
A paper posted here

http://www.snippets.org/pipermail/di...ry/004478.html

states that there shouldn't be a difference.
Has anyone tried both on the same speaker to verify an improvement?
Thanks
Jady

[Scottmoose]

www.zillaspeak.com/bib.asp

Dave's iBIB is based upon the work a few of us have done with 1/2 wavelength tuned TQWT -it just happens to have a Karlson slot sliced into it.

People have hammered K-slots for a long time. Granted, some implementations have been decidedly dodgy, but the basic principle behind them is perfectly sound. From what I gather, intitial builds of the iBIB indicate that the Karlson sliced into them do indeed help reduce ripple in the passband.

[freddi]

regular Karlson coupler feature downwardly tapering wedge-shaped structure whose aperture is flaring postively at the same time. The hope was to achieve better impedance match to room by the bottom of the slot and the originals fired into the V aperture to widen dispersion while adding a type of "reverb" (Karlson's invention list of 1950 mentions this speaker)

K15 and similar bulk couplers can play extremely strong transients without much cone movement IMO.

Karlson's Rocket must have appeared in the late 1950's. Some years after Mr. Karlson's death, Walter Zintz built small 15" coupler and a cast-metal tube under the name "Transylvania Power". I like that tube better than DDS 90.


KR-5

[ww_je]

Originally posted by jadybug
Hi all, this is my first post of too many questions to hold in my head. I have
been floating around the internet and have seen many different types of full
range speakers. My main interests tend to lean between TL's and Horns. I have
looked at some of the designs put forth by Planet 10. The iBIBk looks like a
contender but I have a question about the Karlson Slot. Would the speaker react
the same way if it had a 27 in. sq. area opening/mouth verses the Karlson Slot?
A paper posted here

http://www.snippets.org/pipermail/di...ry/004478.html

states that there shouldn't be a difference.
Has anyone tried both on the same speaker to verify an improvement?
Thanks
Jady

Karlson's original patent is concerned with two issues -- controlled diffusion of the emitted sound (the sort of thing the Karlson sound reinforcement boxes (eg, the Rocket IIRC) were intended for), and with impedance matching of driver and air, especially at low frequencies. Poppe's AES paper is about impedance matching but only in a very much simpler version of the boxes. Does all that theory transfer right off onto the K coupler boxes? There's lots of argument both ways; and lots of clever sarcastic derision about it all too. Not so much from karlson fans, so I guess lots of folks have some troubles understanding just how those complex boxes work.

Experiemental reports of observed or measured happenings are confused by at least two issues. What driver was being used? And is it really a Karlson design? Those boxes are so complex internally that anyone building them is mighty tempted to fiddle with this or that. And the unspoken question is, of course, did Karlson himself get it right and is deviation from his practice inevitably sliding down a hill away from the optimum peak? Karlson was a radar engineer in WWII and I suppose knew enough sums and physical theory to get some (all?) of it rightish. But all in advance of Thiele/Small and their application of electrical filter theory to box design, well simple box design, anyway.

If that impedance matching is well done across several octaves, the driver/enclosure combination should be quite efficient in that range, rather like a properly matched horn/driver combination which also heavily loads the driver while matching acoustic impedance. Or a bass reflex enclosure at system resonance when cone motion goes to almost nothing. I find that the K-coupler designs (for 15", 12", and 8" drivers -- the one with the tulup curved openings on the front surface of a box, and the complex baffling behind/above a slant mounted driver -- to give me reflexive headaches when I try to make sense of what's happening, so I can't definitively anwer the question from theory.

But I can recount an experiment from years ago. Big Dynaco power amp (the 400?) with meters (might not have been a Dyna then...) and a homebuilt signal generator (based on a Exar chip) feeding directly into the Dynaco, driving a Karlson 15" enclosure (I was assured it had been built from measurements taken from a factory box) with an Altec 15" driver installed. The Exar chip was/is reasonable, but not really low distortion (sine wave approximation approach), so we ignored distortion -- though there wasn't any audible to any of the ears involved at any frequency or power level. Not even when the fundamental tone was inaudible to everyone. One was a cellist, one an MIT engineering graduate student, one an audio nut. The Karlsons were serving as a kind of woofer/subwoofer in a three way stereo system driven by a TEAC (?) electronic crossover. Pretty well too. Perhaps at 100Hz? Anyway, we swept the signal generator down from about 300 Hz to around 20Hz with the crossover out of the circuit so its settings didn't make any difference, at pretty high levels. We were looking for evidence of acoustic loading on the driver, so we went to high levels after a while so we could see any motion at all. At low levels there wasn't any. As I recall, we were indicating something like 75-100 watts output; we were real nervous about burning out the driver. So we gave the voice coil plenaty of time to cool off. We watched the cone as closely as we could, peering through the tulip slot with a flashlight, and never -- at any frequency or power level -- saw or felt anything more than perhaps a 1/16" or 1/32" cone motion, if detectable at all. For those who use a more rational measurement system, that's in the range of 1mm. Motion was not periodically visible and then invisible as we swept the generator down. Loud though, really loud, until we got below audibility. All of us found that interesting, as we each dropped out at different points, though everyone could feel the pressure in chest and abdomen. Very odd.

This implies to me, as to us all, very heavy loading of the driver by the enclosure, just the sort of thing Karlson's claims about the impedance matching he was achieving would produce. Paul Klipsch is supposed to have measured them and gotten a really seasick response curve. But then he was a competitor, with large horns. What was he measuring (box and driver)? In the 70s, we didn't have the equipment necessary to take the measurements. Should have measured impedance across the frequency range, though...

Anyway, if Karlson's claims about acoustic impedance matching are correct, and I can't dismiss them based on the data from my little experiment, then substituting a (square, round, or whatever) port wouldn't be equivalent. Something's going on with that tulip shaped thing. And it almost certainly wouldn't have similar directional properties either. Which maybe you wanted, or maybe not.

[freddi]

(sorry to catch your post late) I'm impressed by the loading and output for so little cone movement - something seems to be happening in and below the passband

if higher cotoff for midbass use can be accepted, perhaps 12" coupler with say PYM1298 could do work in a home system to reach fairly high levels as the 12 would be close to the same level as 15-18 from 90Hz up. To make pretty graphs would take a bit of work and I'd expect a hinged baffle and RTA could make that go fast.

I wonder how Karlson would have built them if he had lived another 30 years?

even 18" did pretty well in K15 size (~8 cubic foot) couplers

this one used one 4'x8' sheet and although didn't go lower than K15, had good dynamics with a shockwave rolling across the room


this variant used 23 degrees baffle and taller pipe and its also a good player in several areas