I'm hoping the folds would roll-off content above 300Hz even further considering TF1020 extends far into the midrange. It's pretty well behaved though, meaning no huge peak before its own roll-off like lots of comparable pro drivers, so I'd hope that a very simple network would do it, maybe none at all?
that might well do it - my baby Classic's throat and overall geometry seems to allow a lot of un-desirable stuff above the passband. My other midbass horn, by RCA -Fan cuts off pretty fast above a point. Your horn could be flipped over, a triangular stand/mouth brace to hold it upright, a tweeter mounted in the woofer back chamber, then a midhorn placed on top.
my RCA-Fan horn in blue
RCA-Fan horn general fold - its wider and has a somewhat different
expansion than the version below
Look at my baby Classic's response with the low Q SWR (Eminence) 10 - the mic was pretty close - I think it might look a bit different if the mic were not peeking directly at the final turn. Maybe inverting the horn and placing the mouth right at the floor would give a better close graph.
You can see the horn is passing a 2K peak strong. I had to put a notch filter on that peak.
I wonder what lowpass filter would make the most sense with this horn/driver combo?
my RCA-Fan horn in blue
RCA-Fan horn general fold - its wider and has a somewhat different
expansion than the version below
Look at my baby Classic's response with the low Q SWR (Eminence) 10 - the mic was pretty close - I think it might look a bit different if the mic were not peeking directly at the final turn. Maybe inverting the horn and placing the mouth right at the floor would give a better close graph.
You can see the horn is passing a 2K peak strong. I had to put a notch filter on that peak.
I wonder what lowpass filter would make the most sense with this horn/driver combo?
Last edited:
you can see even it with the Community 12" is appearing to have a rising (or elevated) response area. I asked RCA-Fan whether a 12pe32 would go higher and the response IIRC went something like - no - it will be tilted up. I wish he would write a book on folding horns, tricks and "trade secrets" 
I'm not one bit unhappy with my baby Classic but need to find a good midhorn solution. Any ideas ?
I'm not one bit unhappy with my baby Classic but need to find a good midhorn solution. Any ideas ?
back to the Karlson for awhile - what do you feel might be a best cavity aspect and shape? What is the influence of any upper reflector? Can a curved reflector exacerbate issues? Should one add a resonator stub to the front chamber at times to limit peaking and if so, can it be scaled from one size K t- another?
here's a Chinese pig in a poke 15Cx - from the rear it almost looks like an old Peavey 15Cx
Wharfedale Pro Twin D-243B 400 Watt 15" Cast Frame Coaxial Speaker & Horn 8 ohm | eBay
if the motor is strong enough then may be ok. Those P-Audio 15cx, normally $280 ea. which closed out at $99.95 ea were a bargain. I tried to recommend them but most rather have little FR. In K15 those BM15CXA would have kicked.
Wharfedale Pro Twin D-243B 400 Watt 15" Cast Frame Coaxial Speaker & Horn 8 ohm | eBay
if the motor is strong enough then may be ok. Those P-Audio 15cx, normally $280 ea. which closed out at $99.95 ea were a bargain. I tried to recommend them but most rather have little FR. In K15 those BM15CXA would have kicked.
another old patent with K (and ~keystone) slot is this - can't remember if I've mentioned it in recent years
US3138220.pdf
US3138220.pdf
Last edited:
Freddi your faith in the Karlson is not misplaced. When I was reading Martin Kings paper on TQWT i noticed the line length was at the resonant frequency.
This reminded me of the Tabaq which was originally designed for 3inch drivers and seemed to do well for a variety of drivers. The line length is 30 inches (110 hz) which is where most 3 inch drivers have their resonant freq.
Both these designs had 2 forms of resonance. Quarterwave at the driver resonant freq (which reduces harmonic distortion?) and helmholtz resonance for the system. The Karlsonator is a ML-TQWT with a K-coupler so i worked on a few models(based on X models) and liked the fact that according to the model excursion was reduced(in some cases halved)
Unfortunately results have been disappointing. As X noted double chamber reflex sound better (are cleaner) in bass region. the regular K is of course a DCAAV (a form of double chamber reflex)with a k-coupler.
You have also noted when testing a mix of two freq a K outdoes a bass reflex. the only thing is in a model there isnt reduced excursion in a K compared to the karlsonator but to my ears the K sounds better.
You are also looking for kick. I think this is represented by energy. in the speaker models energy at resonance is stored as pressure in the helmholtz resonator. the problem then is that at the same pressure difference large speakers store more energy than small speakers. properly designed smaller speakers may sound the same but they will not have the kick (thump in the chest?) you desire. so looking for kick and wanting a small speaker may be mutually exclusive.
This reminded me of the Tabaq which was originally designed for 3inch drivers and seemed to do well for a variety of drivers. The line length is 30 inches (110 hz) which is where most 3 inch drivers have their resonant freq.
Both these designs had 2 forms of resonance. Quarterwave at the driver resonant freq (which reduces harmonic distortion?) and helmholtz resonance for the system. The Karlsonator is a ML-TQWT with a K-coupler so i worked on a few models(based on X models) and liked the fact that according to the model excursion was reduced(in some cases halved)
Unfortunately results have been disappointing. As X noted double chamber reflex sound better (are cleaner) in bass region. the regular K is of course a DCAAV (a form of double chamber reflex)with a k-coupler.
You have also noted when testing a mix of two freq a K outdoes a bass reflex. the only thing is in a model there isnt reduced excursion in a K compared to the karlsonator but to my ears the K sounds better.
You are also looking for kick. I think this is represented by energy. in the speaker models energy at resonance is stored as pressure in the helmholtz resonator. the problem then is that at the same pressure difference large speakers store more energy than small speakers. properly designed smaller speakers may sound the same but they will not have the kick (thump in the chest?) you desire. so looking for kick and wanting a small speaker may be mutually exclusive.
for small, I'll take anything up to where FedEx oversize pricing kicks in . Subjectively, and from what little experience I have, it seems to take at least a lower mass 10" speaker to have adequate air movement to get the "kick" - if the upper mids are weak, then balance will be negatively affected leaving the tweeter harmonics and midbass hump as the signature. If the motor on a regular/small K is too strong, then other balance problems arise. (inductor DCR and/or amp output Z could help)
for dual chamber reflex, do you think the classic 2:1 chamber ratio with 3 identical ports is the way to go? (hornresp sometimes seems to indicate better performance with different ports)
what sensible moves can one make with the Karlson for a new build that's worthwhile? (I have no problem with classic K15 nor K12 - with appropriate drivers)
if a K produces too rough of graph, then it will likely be rejected by prospective builders. For graph readers, other than floor bounce, we want ideally to minimize the dips and the height of the first peak.
. Subjectively, and from what little experience I have, it seems to take at least a lower mass 10" speaker to have adequate air movement to get the "kick" - if the upper mids are weak, then balance will be negatively affected leaving the tweeter harmonics and midbass hump as the signature. If the motor on a regular/small K is too strong, then other balance problems arise. (inductor DCR and/or amp output Z could help)for dual chamber reflex, do you think the classic 2:1 chamber ratio with 3 identical ports is the way to go? (hornresp sometimes seems to indicate better performance with different ports)
what sensible moves can one make with the Karlson for a new build that's worthwhile? (I have no problem with classic K15 nor K12 - with appropriate drivers)
if a K produces too rough of graph, then it will likely be rejected by prospective builders. For graph readers, other than floor bounce, we want ideally to minimize the dips and the height of the first peak.
Last edited:
For kick out of a helmholtz it needs to be as short as possible.
Also the speaker cone should be in close proximity of the inner port opening.
Last edited:
speaking of vent placement - look at the graphs I just took of a 6-slit vent style Karlson 12 (maybe late 1956-57 - published in July 1958) with Eminence's first B102 vs the first Karlson 12 from fall 1954 which had a single horizontal vent, loaded with Dayton's PA310-8. There isn't a great difference between their front chambers other than the later model had a 10 degree forwards slant.
I think the distributive vent is making the difference. What do you say?
These are very close measurements - like 6" from the aperture - its probably more meaningful (?) to take plots at a greater range than the aperture's height.
I think the distributive vent is making the difference. What do you say?
These are very close measurements - like 6" from the aperture - its probably more meaningful (?) to take plots at a greater range than the aperture's height.
Last edited:
Hi Freddi,
It looks like the distributed slot has sharper dips, which are less audible. However, they may cause more harmonic distortion in frequency space. Floor measurements remove floor bounce (but not room wall bounce) so needs to be done in an open field not inside a small room. I usually measure at least 0.5m away with speaker on a stand to avoid room and floor bounce. I know you use RTA and Omni Mic. Try using REW and set frequency dependent window (FDW) option to 6 cycles. This gates measurement and makes sure what you see is intrinsic to the speaker and not result of floor or room bounce. It’s really useful to have that feature.
It looks like the distributed slot has sharper dips, which are less audible. However, they may cause more harmonic distortion in frequency space. Floor measurements remove floor bounce (but not room wall bounce) so needs to be done in an open field not inside a small room. I usually measure at least 0.5m away with speaker on a stand to avoid room and floor bounce. I know you use RTA and Omni Mic. Try using REW and set frequency dependent window (FDW) option to 6 cycles. This gates measurement and makes sure what you see is intrinsic to the speaker and not result of floor or room bounce. It’s really useful to have that feature.
here's my first attempt at REW for two diy Karlsonette (1st K12 cabinets) - neither has substantial damping materials. The cabinet loaded with PA310-8 has its stock ~14" wide by 15/16" wide slot port. The other cabinet, with Pyle PYM1298 (~like Kappa 12A) has the stock port blocked and a rough ~K15 style port of maybe 13 sq. inch. cut into its upper board.
subjectively, although the cabinets were tried in different parts of the room, I thought the Pyle 12 sounded much clearer, faster, and more agile when coupled to a K-tube and compression driver. The Dayton has a good motor, low Q but double the moving mass of pym1298.
(Brain Steele will probably laugh or be aghast at the LF extension compared to TH POC3 and Boom Unit - K12 have about 35l back chamber and tuned around 62Hz)
the cabinets were away from walls a fair amount, and elevated about 33"
subjectively, although the cabinets were tried in different parts of the room, I thought the Pyle 12 sounded much clearer, faster, and more agile when coupled to a K-tube and compression driver. The Dayton has a good motor, low Q but double the moving mass of pym1298.
(Brain Steele will probably laugh or be aghast at the LF extension compared to TH POC3 and Boom Unit - K12 have about 35l back chamber and tuned around 62Hz)
the cabinets were away from walls a fair amount, and elevated about 33"
Welcome to the world of REW and nice looking measurements! The FDW 6 cycle shows what is intrinsically happening with the speaker independent of the room. You will see how handy it is to not have to hand type legends in a paint program anymore. You can overlay dozens of plots and click the enable button next to the legend to show or hide - very handy.
THANKS ! - it sure will help make decisions vs 18 million squiggles - that pc must be 15 years old, and has 768MB ram - program seems to run fine on xp.
I need to take a factory Karlsonette, remove its port panel, then cut something close to the "Dutch K12" 3 - slot port for comparisons.
- that pc must be 15 years old, and has 768MB ram - program seems to run fine on xp.I need to take a factory Karlsonette, remove its port panel, then cut something close to the "Dutch K12" 3 - slot port for comparisons.
Last edited:
here's a few more from the K-pile
Low qts SWR 10" bass guitar speaker from Eminence in a 0.62 scale K15 vs PYM1298 in a K12
K12 with PYM1298 again vs a high aspect K8 with 3" port at the top and Nirvana Super 8 vs Art Welter's Dad's "K8" with a Visaton BG20.
there is something screwy with the tall K8 and port near the top
Low qts SWR 10" bass guitar speaker from Eminence in a 0.62 scale K15 vs PYM1298 in a K12
K12 with PYM1298 again vs a high aspect K8 with 3" port at the top and Nirvana Super 8 vs Art Welter's Dad's "K8" with a Visaton BG20.
there is something screwy with the tall K8 and port near the top