Jump to content

A K-402-Based Full-Range Multiple-Entry Horn


Chris A

Recommended Posts

It's from my father's woodworking shop, among many other tools he accumulated.  It's taken me a couple of years to be able to think about moving them.  They do need to be in use, however. 

 

Chris

  • Like 1
Link to comment
Share on other sites

  • 1 month later...

Just an update to the performance of the K-402-MEH prototype--its phase/group delay and step response using a first order Bessel filter and the BMS 4592ND driver, which is good down to an advertised 300 Hz crossover frequency.  In the below plots, the crossover point is still 450 Hz, but the shallow crossover filter uses that lower frequency potential.  Note the 1 dB/division vertical resolution on the SPL and the closeness of the total phase curve to the calculated minimum phase curve:

 

1596604082_K-402-MEHw-BMS4592ND--1storderBesselSPLandphase(inclminphase).thumb.jpg.b26e21bac280ac10162513b18545fa8e.jpg

 

Here is an impulse spectrogram:

 

567214982_K-402-MEHw-BMS4592ND--1storderBesselimpulsespectrogram.thumb.jpg.590629d22159498d094c11b392c80f59.jpg

 

and a step response plot:

 

1528907297_K-402-MEHw-BMS4592ND--1storderBesselStepResponse.thumb.jpg.400e2c53f4bac49b2a5aaac59282760f.jpg

 

For those of you that have some background with these type of plots, I think that when you compare these plots to any other horn-loaded loudspeaker model that has also been measured, these are fairly spectacular. 

 

The listening performance has also taken a step up since the BMS 4592ND dual-diaphragm driver has been in dialed in.

 

Chris

Link to comment
Share on other sites

11 minutes ago, Chris A said:

For those of you that have some background with these type of plots, I think that when you compare these plots to any other horn-loaded loudspeaker model that has also been measured, these are fairly spectacular. 

 

The listening performance has also taken a step up since the BMS 4592ND dual-diaphragm driver has been in dialed in.

I'm fixin to try a pair of those on my Jubes sometime this year. How does it compare to TAD drivers?

Link to comment
Share on other sites

I've made that shift already.  The listening sound quality improvement is noticeable by flattening the phase curve (and its resulting group delay).

 

...but that's a subject for another thread. (Perhaps I'll start one on that subject soon.)

 

Chris

  • Like 2
Link to comment
Share on other sites

19 hours ago, ClaudeJ1 said:

How does it compare to TAD drivers?

The TAD is of course cleaner on impulse plots, but how that translates into differences in listening performance is perhaps not as pronounced:

 

TAD TD-4002  on K-402 Left Jub Spectrogram.jpg

 

When I put on a really good recording of jazz or other acoustically performed  music that wasn't built from multi-tracked recordings, I can just barely hear the differences in the smoothness of the upper harmonics and the definition of the cymbals, etc.  It's a pretty small difference though--enough so that I can recommend either the BMS 4592ND or the TADs for highest performance reproduction.  Someone with 20-year-old ears (especially women) might hear a little more difference, and particularly classically trained musicians that have spent some time training their ears to hear very small differences in timbre and attacks/releases.

 

Chris

  • Like 1
Link to comment
Share on other sites

4 minutes ago, Chris A said:

When I put on a really good recording of jazz or other acoustically performed  music that wasn't built from multi-tracked recordings, I can just barely hear the differences in the smoothness of the upper harmonics and the definition of the cymbals, etc.  It's a pretty small difference though--enough so that I can recommend either the BMS 4592ND or the TADs for highest performance reproduction.  Someone with 20-year-old ears (especially women) might hear a little more difference, and particularly classically trained musicians that have spent some time training their ears to hear very small differences in timbre and attacks/releases.

According to my ears, when I had older version K1133's on K402's (the ones with the long snout) I was thrilled to be able to enjoy a 320 Hz. crossover point and letting the mass rolloff take it's course at about 4-5 Khz. on that combo. As you know, Danley uses a 300 Hz. crossover point as well. I'm crossing my TADs at 400, but I think I could do better as a 3-way when done right, but the 300 Hz. point is what I'm after, whereby that "lower midrange" comes out of a straight axis horn instead of the Jube's labyrinth.

Link to comment
Share on other sites

14 minutes ago, ClaudeJ1 said:

I'm crossing my TADs at 400, but I think I could do better as a 3-way when done right, but the 300 Hz. point is what I'm after, whereby that "lower midrange" comes out of a straight axis horn instead of the Jube's labyrinth.

Yes, the anechoic (full space) response of the Jub bass bins is pretty narrow above 300 Hz due to the dual horn mouths side-by-side spaced apart 21.25" and using a truncated mouth like a Belle (i.e., not like a La Scala). I'm not sure what the Jub bass bin polar coverage actually is in-room with floor and wall boundaries since I've not hauled them outside to do polar tests to date. 

 

Crossing at 300 Hz with a pair of BMS 4592NDs would be interesting to hear.  If I can spring for another 4592ND driver in the near future, then I might be able to hear the difference crossing them over at a lower frequency.  The 400-450 Hz crossover frequency presently used on Jubs is almost at tuning fork (A440) pitch--which is right in the middle of the midrange band.  Not having to put those frequencies through the folded horn of the Jub bass bin would be an advantage in sound quality, one would guess. 

 

The K-402-MEH avoids all that and you can hear the difference in the midrange clarity and presence quite clearly.  I believe that this is the source of the real advantage of the K-402-MEH over the Jubilee in terms of sound quality.  The MEH has good directivity down to just below 100 Hz--just like the Jubilee and Khorn bass bins--before they both lose directivity control in the horizontal axis.  The transition to room boundary support at 100-130 Hz or lower is much less audible than using a direct radiating reflex ported woofer box and crossing at 500-800 Hz (like the KPT-942 or a Cornwall). 

 

The crossover of drivers within an MEH is seamless--and you can see and hear that seamlessness with Danley Synergies which use large amounts of driver overlap in their MEHs with their fractional order filters.  So crossing at a lower frequency point doesn't make a lot of sense in the case of an MEH.  It's better to cross where the drivers sound better and have good acoustic power output capability.

 

Chris

  • Like 1
Link to comment
Share on other sites

1 hour ago, ClaudeJ1 said:

BTW, Chris, you plus/minus 1.5 db magnitude response is the best I've ever seen! Good work.

To give others an idea of the differences in total phase growth using higher order Linkwitz-Riley filters (48 dB/octave) vs. lower order Bessel (18 dB/octave), here are two SPL and phase plots. First the Linkwitz-Riley 48.  Notice the bottom phase curve with respect to the minimum phase curve:

 

260099975_JubileeOn-AxiswithLR48dB-octxoverSPLandphase.thumb.jpg.0e00b5dbd52335ca98cc8b4b9a9d5d8c.jpg

 

and the same thing but using Bessel 18 filters instead:

1110150726_JubileeOn-AxiswithBessel18dB-octxoverSPLandphase.thumb.jpg.0f6d75c20f8d05a8f747fbb2520daf32.jpg

 

I didn't use lower order than 18 dB/octave due to the already low crossover frequency for the TAD TD-4002s in both cases (450 Hz crossover frequencies in both cases).

 

Chris

Link to comment
Share on other sites

12 minutes ago, Shiva said:

Are you listening to a stereo pair, or is this a single converted 402 you are testing?

I always use one loudspeaker at a time, i.e., mono, when listening for sound quality differences in loudspeakers, including the case of the K-402-MEH prototype.  I learned that lesson from Roy using mono speakers in listening tests in Hope years ago when listening critically for differences in loudspeakers and their setup.  Stereo always disguises sound quality issues.  This is why an array of five loudspeakers positioned around a listener always seems to sound better--even if using much poorer performing loudspeakers.

 

The K-402-MEH occupies the center position in my array.  But I've also listened to it in the corner, both on the floor and elevated to the height of the Jubilee K-402.  The comments I made above were based on all those listening positions.  The K-402-MEH picks up almost 12 dB of bass response at 40 Hz in a room corner relative to its performance in the elevated center position.  The corner loading case was EQed again to match performance as in the center case. 

 

All loudspeakers that are moved to a corner position need to be re-EQed to compensate for the pick up in bass response (which thus lowers bass distortion levels in the case of the corner loading).  This is well documented in PWK's writings.

 

Chris

  • Like 2
Link to comment
Share on other sites

3 hours ago, Chris A said:

When I put on a really good recording of jazz or other acoustically performed  music that wasn't built from multi-tracked recordings, I can just barely hear the differences in the smoothness of the upper harmonics and the definition of the cymbals, etc.  It's a pretty small difference though--enough so that I can recommend either the BMS 4592ND or the TADs for highest performance reproduction. 

 

The BMS 4592ND is a cool design. Pictures of its diaphragm and phase plug look interesting.

  • Like 1
Link to comment
Share on other sites

The ring dual diaphragm design makes a lot of sense especially when you consider the need for a phase plug and the need to avoid diaphragm break-up modes at high frequencies. 

 

The BMS dual-diaphragm drivers also control any FM distortion (Doppler) that might show up at extremely high SPL due to the large pass band that it covers.  Breaking that up into separate parts kills the growth of FMD above the crossover frequency (6.4 kHz).

 

The fact that they can do it with polyester diaphragms also tells a good story from the standpoint of material shape stability of the moving diaphragm which moves as a ring rather as a membrane with higher order modes--like a drum head.

  • Like 1
Link to comment
Share on other sites

The TAD is of course cleaner on impulse plots, but how that translates into differences in listening performance is perhaps not as pronounced:
 
638564980_TADTD-4002onK-402LeftJubSpectrogram.jpg.5ef779941ba6e07e6a8737d1ac833b83.jpg
 
When I put on a really good recording of jazz or other acoustically performed  music that wasn't built from multi-tracked recordings, I can just barely hear the differences in the smoothness of the upper harmonics and the definition of the cymbals, etc.  It's a pretty small difference though--enough so that I can recommend either the BMS 4592ND or the TADs for highest performance reproduction.  Someone with 20-year-old ears (especially women) might hear a little more difference, and particularly classically trained musicians that have spent some time training their ears to hear very small differences in timbre and attacks/releases.
 
Chris
Do you have any idea how either driver would compare to the EV DH1A?

Sent from my SM-T550 using Tapatalk

Link to comment
Share on other sites

Actually, I haven't had a DH1A to test or to hear.  I've heard good things about the driver but I can't say much about it due to my lack of experience.

  • Like 1
Link to comment
Share on other sites

I've updated some of the plots found at the beginning and middle sections of this thread (the various plots linked at the beginning of the thread) using the current Xilica settings--PEQs, crossover filters, delays--and a BMS 4592ND 2" dual diaphragm compression driver--which rivals the TAD TD-4002 drivers in the Jubs on either side. 

 

Note that I've yet to update the polar measurements that I did 2 1/2 years ago--which will be significantly improved from those old ones posted here, I believe.  It takes a little while to understand how to do the crossover filters in an MEH, I've found. 

 

Chris

Link to comment
Share on other sites

46 minutes ago, Chris A said:

I've updated some of the plots found at the beginning and middle sections of this thread (the various plots linked at the beginning of the thread) using the current Xilica settings--PEQs, crossover filters, delays--and a BMS 4592ND 2" dual diaphragm compression driver--which rivals the TAD TD-4002 drivers in the Jubs on either side. 

 

Note that I've yet to update the polar measurements that I did 2 1/2 years ago--which will be significantly improved from those old ones posted here, I believe.  It takes a little while to understand how to do the crossover filters in an MEH, I've found. 

 

Chris

A "little while" is an understatement. I spent a week on Jubes alone.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...