Jump to content

Modeling throat - vented horns - When is that arrangement useful to the hobbyist? How should it be simulated?


karlson3

Recommended Posts

(Has this already been discussed here before? - if so then I won't waste the bandwidth)

 

To my limited knowledge it was first patented August 18 1959 by James Novak and is seen in Bill Fitzmaurice's "Jack" series of midbass enclosures. (also Dana Moore's patents)

 

I believe hornresp can handle most cases and also series venting through a throat, more involved than venting the back chamber to the outside world.

 

Klipsch has the brains and resources to optimize this type.  I assume for low frequencies its treated like a 6th order assisted reflex with up to 6dB peaking at system tuning.

 

It would be interesting to know the basic outside dimensions of the new Jubilee to see if a rough guesstimate sim comes within spec on the low end.

 

 

https://imgur.com/Voxe6nB

 

Voxe6nB.jpg

 

 

 

 

Link to comment
Share on other sites

Never looked until this instant.  What I'm seeing in 5 grueling minutes of sims:

- stupid amount of efficiency down low if you make it big (400L)
- drops like a stone (~6th?)
- unloads less-poorly than some other stuff
- ringing b*st!@^

No-offense to anyone, an apparently re-patentable hrrmph is what I got in 5 minutes.  (Perhaps it takes more than 5 min work to achieve patentability?  Rhetorical).  Be some work to tame one.  Interesting corner on the Great Efficiency-Bandwidth exchange?  Be nice if some gung-ho person plotted the transfer function but that's not gonna be me today.  Beats working the Saturday list, though.  Have to poke at it some more.  

What are you seeing?

(Remember the "Have you considered a career as a commercial artist?" pitches?  Nice artwork).

Link to comment
Share on other sites

(I've nearly zero artistic ability other than perhaps make something from paper mache - online FREE: Adobe Background Remover, Vertex Upscaler, Image Colorizer plus good ol' GIMP are the non-artist's friend)

 

Some throat vented horns are concentric - like James Novak's patent and the "R-J" enclosure.  This new Klipsch speaker has

a baffle which holds a 15 inch speaker (I think) with 3- large and efficient ports mounted just below the woofer.  Most existing FLH cannot be converted to a throat loaded design using the R-J inner baffle as there's not enough space nor ability to tune the back chamber about one octave below the front horn's cutoff.

 

Here's the general approach to see how well a throat vented horn may work using the brilliant "hornresp" program by David McBean (and fueled with HF Olson power)

 

This might be ballpark to the epic Jubilee size-wise.

 

The input window gets: the front horn and throat arrangement in the 4 segments part of this window, down below that are chamber volume, port size and port length entries and driver parameters.   FWIW I picked one K33 spec.

 

On the EQ box section I first applied a 2nd order highpass to offer some protection below reflex tuning, then applied a peak boost(6dB in this case)in the region around tuning.

 

To get an idea how well throat venting vs venting the back chamber to the outside, it would take comparison

to a bass reflex the size and tuning of the throat-vented horn and also to a horn with back chamber vented to

the outside world.  I did that with a bass reflex and yes, the throat horn showed some advantage. (It should as nearly 3 times as large)

 

Seeing this approach vs pure FLH packed into the new Jubilee's bulk tells me the hybrid with EQ approach works better at

very LF.  It also says the hobbyist might have a less complex cabinet to build.  (IF they get the design right in the 1st place  -lol)

 

 

 

 

keCkoRT.jpg

 

 

BTW - the earlier Jubilee is very much like the old rectangularbox University Classic

 

dKnYVEv.jpg

 

3zYorAA.jpg

Link to comment
Share on other sites

Allow me to say thanks for visiting & stirring-up something technical again (seems different around here, that way).  Sounds like you saw behavior similar to what I saw.  There's a happier spot if you go big and are aiming for low.  No clue what driver is employed, and uncomfortable reverse-engineering our host's design on their own board, so I'm staying off the K33 and Jube size on general principle.  May well be pointless to try to apply the configuration to other (ie less LF) applications.  

 

Could never greenlight the S-8 Classic build as 400L is up there for my spaces and current driver-subs are iffy.  Moray started a thread here somewhere for alternative drivers and I recall a couple rising to the top.  The C15W stands alone for that cab and it's the lynchpin of it playing smooth, high, and almost low-enough (but not low-enough for sheets and sheets and 400L to get a nice 50Hz, IMO).  This would seem a better use of 400L+ than that.  World is full of 50Hz boxes (that's the problem).  760L is a Whole Other Thing (tm).

Don't know xmech of k33's, but that seems way past banging even with 15Hz HPF (?).  Box losses will take some of the (bottom of the) LF off, but the room might put some back on so boosting remains opaque, here.  Good ballpark, though, and apparently faithful in principle.  Stipulated that something higher-travel would fly.  

 

Re the atmospheric-vented reflex, throat-ported, and normal fronthorns, seems not a fair fight to go size-for-size w/o changing drivers to be what's best for each design type, but I get the comparison and appreciate the work.  That should certainly get a look and be a bunch of work to do right.  They're just all so different (and I'm arguing probably have different goals.  This is the crux, as always.  Unsure anything else plays as low, as high, as loud (& surely Klipsch would add as clean) as this thing.  Thinking that's the point (?) underpinning size justification.

 

At LF, people routinely argue a lot of throat geometry differences away, but...  Thinking it'd take a 3-D look (not our 1-D) to really get after what's happening in R-J/Novak-ey throats.  Suppose a person could do a full lemon throat to more broad-band the port (?) but we can't quickly/easily see that stuff in lumped-element-land and it might be exactly wrong, anyway; just a thought I associate with "R-J" albeit only one shape of theirs.  The parallel (port-throat) walls are close-enough where it might matter though, if a person tried to throttle these style ports for more damping.  In the home, sure--probably not pushing it that hard.  Also thinking 2 turns isn't enough to kill that parallel port-throat resonance unless that cone can't break-up to excite them.  If there's truly port tubes in there, then keeping that trash out of band enters, too.  Thanks for the work!  

 

Dunno.  We can get 105+/W in a corner with a lot of easier/smaller stuff, it's going after the 115 that separates the chaff.  These things look to be able to do it and down-low, but they're not petite.

 

Haven't seen Dana around here in a long time, and it's too nice here to read patents right now.  My sense is still that these are only worth doing if you're going for low+loud.  It does seem to play higher than it has any business playing, though, with things I tried.  That's probably the point--low+loud+high--to meet that schmancy top driver.  Glad people still make statement products.  Hope they are a screaming success for the mothership, but I've definitely slept in smaller volumes than 760L :) Sims here say that going real small is pointless for LF and the thing can be just a beast down low when you don't.  We know the diff between 20's and 40's is non-trivial, and this config seems to get it done and save a magnet over the previous rev.  A very hard problem of a 2-way with crushing bass would appear to be solved.  BFM's MR applications might be the next sweet spot up higher for these geometries.  Unknown.  I think we might've arrived at the same spot, but the utility differs.  I'm still at hrmph for most people, but it seems a solid approach for those with large spaces.  My spaces are headed the other way these years.  Gotta be a pretty visceral and immersive experience, though. 

Link to comment
Share on other sites

there's no real drive to reverse engineer - just to get a feel for the technique and also see the smallest volumes where it may do relative good on the low end..

 

Would  a 12/15 scale horn roughly derived from the Jubilee's bulk offer any pros?  How much influence does the front horn's air load have upon the system's low end with overall bulk fixed?

 

Its certainly easier to take a Belle, make the back chamber larger, vent and apply 6th order reflex technique as did the late and great DJK with the "Vented La Scala"

 

I'm not good with hornresp and would like to get a better grasp and that includes adding a throat vented model.

 

K33 certainly seems to work for such application although a beefier version with more power handling probably is in the new Jubilee.

 

I'd really like to get a good little 10" K-coupler before I pass away -horns will have attention and fans for the foreseeable future, while the Karlson takes work just to keep if from disappearing altogether. and its interesting that only GregB's "Karlsonator" and XRK971's "XKi" get built these days.  (IMO there's nothing wrong with the original Karlson models)

 

For the low end I have great faith in hornresp.

 

For a given bulk - there's only so much that can be done - but many response outcomes.

 

For starters one could reconfigure a V-LS size as a throat vented horn and see how that that little horn affects what one can get out of a back chamber vs the atmospheric vented LS and then just a 6th order reflex.

Link to comment
Share on other sites

4 hours ago, karlson3 said:

the Karlson takes work just to keep if from disappearing altogether. and its interesting that only GregB's "Karlsonator" and XRK971's "XKi" get built these days.  (IMO there's nothing wrong with the original Karlson models)

 

In the '50s and early '60s I had a Karlson enclosure High Efficiency Speaker Asylumin one corner, and a JBL C 34 rear loaded hornClassic JBL; C34 Enclosure, D130 Full Range, 075 Ring Radiator, N2500 on  PopScreen Two-way speaker system, WAY horn-loaded | Page 7 | Audiokarma Home Audio  Stereo Discussion Forums 

enclosure in the other corner.   Each had a JBL 030 system installed, consisting of a 15" D130 "extended range" speaker images?q=tbn:ANd9GcRfHOXIv-dKazzqwyVOe_J

 pair JBL 075 tweeters Bullet alnico magnet super tweeter.  and an 075 "orange juice squeezer" tweeter.  Here's the deal:

  • The bass in the Karlson sounded "better" to my teen age and early '20s ears.  The cannon in 1812 had much more umph.
  • A JBL engineer at the Hi Fi fair said, "Don't put a D130 in a Karlson, the C34 will be much smoother.  The Karlson will have a big peak at about 100 Hz."  I was not equipped to run and measure a sweep, but I had ears.  The Karlson provided a big sexy peak. 

My Karlson sits empty now, because our Klipschorns sound much better, but the Karlson is in an honored place at the back of our music room, because my dad built it for me in the mono days.

  • Like 1
Link to comment
Share on other sites

Enjoyed the story and the pics very much, garyrc.  Thanks for posting.  Seems like all of us like what "exciting" sound has currency when we're young.


--------------


Fred:  Take your base model (David's type, here):
1.  Copy the driver data (to memory)
2.  Add New record
3.  Resize wizard -> scale.  Use the cube root of the scaling you want
4.  Paste the driver data back in
5.  Check the throat (S1-3) sizes vs. driver, edit your VRC & LRC, Ap, Lp etc.  
6.  Hop in LS Wiz, change the manuals on S2,S3 etc to auto (lock 'em together)
7.  Slider yourself bleery, take many screens & upload to World


I think that gets you cut loose on looking at smaller stuff.  Apology if I've misread/misunderstood.  


Been waiting for you to cover the web with K-sims since the PH geom was added.  You'll need to ask the paraflex guys to show you which PH models to use (guess there's activity on FB, but not a FB person).  Draw them a system model of the K or even a crude schem and ask them to implement you a baseline/framework.  The problem with that thing is always the aperture specifics.  Needs (IMO) really to go akabak or 3-D/FE/BE methods.  Doing a scalable akabak script for K's like X did is a pain in the shorts--we're all too old to be that gung-ho anymore; takes a kid like X.  Unless it's about naps (I can be gung-ho about naps).  It's just drawing pictures & thinking out parametric waveguide and radiator sections.  Use the HR akabak export as a start (BP6S? or even this thing).   Have to change pi/2 to 2pi first, then it'll export--you have to add corner reflectors in the script for that.  If you wanna hack one together, we can look at it offline.  Start with the picture, though--the picture is Everything before you edit all that script stuff (and you'll need x,y, & z coord for radiators, too).  Keeping subscripts and directions straight will only go easy with a good picture.  Even then, sometimes one needs spreadsheets.  Did I mention to draw a good picture?


Anway, thanks again for stimulating technical/design discussion.

 

Link to comment
Share on other sites

hi grindstone - here's a 0.8 (^3) resize.  It did not like the Dayton PA310 specs one bit.  I could not find K281 but did use Bob Crites (RIP) spec for a "K28" speaker

 

hey Gary  - I never got that 100Hz peak with K15 - my K18 is very good.  Stubs can be used if needed to control cavity peaking, K's can be made cubical like Metro Systems "T15" - I'd love to hear / have a C34 !

 

This tells nothing about how to make a cabinet - just what should happen for these specs

 

eC5uDiH.jpg

Link to comment
Share on other sites

If I'm tracking, it looks like you kept the "resized" driver params after that operation?  Don't do that :) HR will resize Everything.  See step 4 above -- copy data for the driver you want to use before the resize.  Need to paste the right driver in after resize & before calcs or it's junk.  If not, apologies and you are now officially dangerous and loose.  

 

Saw our benefactor Dave's model at diyA, too--his way makes sense to me.  Might look at that when there's a minute to sit down and look again.  

 

Tried 12" "horn" driver yesterday and yeah, that was my exp, too.  I hate working in the blind w/o a transfer function/model.  Go caveman with me a sec:  Add an inertance in series with the normal series and parallel resonance "reflex" system, itself a compliance resonating with an inertance and against a driver (motor+suspension system).  Is that what this is or am I missing it at the root level?  Because the outputs are summed in the throat and we're talking macro (LF) effects in our cave, isn't that legal?    (No, not reading patents, either--gotta run). 

Link to comment
Share on other sites

Thanks for the correction--just saw 6XX Sd and you know...

 

Among the more frustrating aspects of this modeling approach (TH + vanishing L12,L23) is that offsets aren't possible so the column resonance is always an undampable gut-punch (in woefully undersized designs).  After working on pipes for ages, the confinement is very much constraining and unwelcome.  IMO, the only way to get that tamed (again, when sorely undersized) is an offset.  Highest-altitude, the thing is an end-driven open air column so it does what those do and we've long known how to fix that.  

 

David's model, being another "pinch" enroute, adds its reflection, and that one is too low to damp (and have any LF left), too.  I can't argue with his approach but chafe against one waveguide segment for that whole region.  You try to smooth those out and you end up with a normal fronthorn.  

 

Reviewing your atmospheric-vented Belle model, it's flirting with 110 @ 50 in < 400L.  In a corner, functionally & size-wise, it's very Dean-like (if 1/8-spaced), but with a K33 instead of needing a C15, so that's something.  The only reason that makes sense to me is if you're really excusion(=distortion)-driven.  Otherwise, yunno, stick a 12pe32 in an Aristocrat in 1/4 the size, save the delay and go on with your life.

 

This really is a continuum of stuff with a lot of waypoints that meet different goals.  One one end, there's Wayne's driver+port pointed backwards into the trihedral.  On the other end, there's these new-rev Jubes.  Thinking my answer to when the diy-er applies such a config is only when it's necessary to hit your low-corner target and still mostly retain horn-ish displacement limits & save some delay.  You can get the radiation-resistance bennies with horns at the cost of the delays & LF.  To get low-low and still play high-enough for 2-way, not seeing any ways around Big but open to education from anywhere. 

 

davids_method.png

Link to comment
Share on other sites

Hi Grindstone

 

David McBean certainly clarified this model shown in your post above.  With known baffle size Sd Sv and the two horn segments, the simulation should be reasonably accurate.   So if one wants to see what this type does then they can now simulate to see tradeoffs.  I bet it would take some thinking and comparisons to see what's best for a given bulk, driver and goal.

 

Bill Fitzmaurice's "Jack" series uses the series vented throat.  I'm not as sure how to simulate it yet and assume Bill does empirical adjustments based on a lot of accumulated experience.

 

nOoU5uL.jpg

 

 

Link to comment
Share on other sites

Did PWK generally choose the bifurcated folded horn vs single path for response smoothness?

 

Even if the method requires a very large cabinet for best result over parallel vented, it could offer an aesthetic advantage as no port openings would be seen. (or hidden with grille cloth.)

Link to comment
Share on other sites

Unknown re bifurc--could only guess (mid)bass/turns.  I've picked at it three times now, but make zero claims as to owing all the boundaries of the type.  (And I did go expo right-off, btw).  All I know is that it pushes boundaries in every way to achieve a 2-way ~20-20k.  Guessing it also fits through a door (if in pieces?).  Are you seeing anything different to make it sane-sized and still worth doing (above what's extant)?

Link to comment
Share on other sites

I think I did my first realistic sim of that type as did not understand how to define the throat size for a horn made as that one video showed.

 

It is huge - goes very low and reasonably level w.o. EQ  (I used a K28 spec as the correct woofer doesn't seem to have posted parameters)  Efficiency in a corner at 31Hz is around 10% w. K28 spec,

 

I might have made it too large (?)  876 liters internal volume !

 

If is officially released?  - I think it might go through a 36" wide doorway when turned sideways.

 

Mouth ~52x30 inches (???)  1.6 watts input at 21Hz for 107dB

 

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...