[ProCast99]

Hello all:

In my ongoing research to find a nice pair of high-efficiency speakers that I can mate with an NS-SEP20, I have come across a rare Kenwood knock-off of the Altec 604 design. There is a local guy willing to sell for a good price, and strangely, has two pair. According to the scant info about them available on the web, they are about 93-94 dB efficient into 8 ohms.

Here is a link to a photo if anyone is interested:
http://www.hifiengine.com/gallery/images/kenwood-ls-770.shtml

I have heard them and they sound v. nice. I may buy them. But when the two pairs are stacked, they sound REALLY good. So this is the root of my question.

I understand efficiency rating is a ratio. But I am thinking speakers put an absolute load on a system - i.e. if I have two speakers of the same efficiency, that presents twice the load of one speaker - if so, and if I am choosing this speaker so that it will play well with a lower-wattage amp, then it would be better to just stay with one speaker per channel -- is this correct logic?

My second related question is -
if I wanted to go the diy route and create cabinets that would house two of the speakers instead of one (i.e. same effect as stacking, but instead of two separate speakers per side, only one), would that change things? I am guessing it would somewhat, depending on the cabinet's design, but I am not sure if it would be significant, as I am assuming most of the load comes from the power needed to move the drivers.

As is perhaps clear, I am no electronics whiz, but I can understand basic mathematical relationships - so simpler responses will probably be more meaningful to me -- thanks in advance for any thoughts / advice you may have on this topic.

Cheers!

[steve f]

It may work. Try stacking the speakers on top of each other. Place the top pair upside down. You end up with a WTTW set up. I have some concern over the spacing between the tweeter pair, but I'd try it out.

[James Romeyn]

Link leads only to Google, at least in my Firefox. 

What is your amp's power rating into the Alec 604's "20 Ohm" input? 

I'm highly suspicious of grotesque amount of dreaded comb filter effect (especially in the vertical domain) resulting from 19" vertical distance between two HF horn centers covering 2k-16 kHz (15" mid bass radius = 7.5" + 2" cabinet edge = 9.5" x 2 speakers = 19").  General theory is distance between two drivers < 1/2 highest shared wavelength.  1100fps speed of sound / 16kHz = .06875' x 12 = .825"/ 2 = .4125" maximum theoretical tweeter distance, beyond which comb filter effects audible.  19" / .4125" = 46 x theoretical maximum.  Granted these are theoretical limits, but 46x is pretty far from any potential gray zone.     

If your ear height was exactly mid-point between both tweeters I suppose it would work at that one location, but as soon as you moved up/down there'd be gross audible phase effect.  Maybe turning off one tweeter (on the lower 604) is an option. 

You might audition THIS ARRAY with two regular stand mounts, which should indicate whether or not you'd like it with two Alec 604 in custom stand mount cabinets.  The above array gives up about 1/2 the gain vs. both speakers firing forward, but in every other measure performance should be multiples better.  You can easily test for yourself, though.

The above driver distance limits don't apply to the above array because the vertical dipole output is delayed at least 10ms.  The ear perceives same-spectrum output delayed 10ms as a secondary source, hence all the wonderful resulting spatial and density effects.  Personal experience is the effects are superior to all other known bipole/dipole effects.   

Regardless what array you employ, you can make one crossover for both speakers rather than separate xo for each speaker (per channel).     

Re. the amp/load electrical interface: Whether one speaker sounds worse or better than two depends entirely on the specific load and the specific amp.  I read the input impedance of the Altec 604 crossover is 20 Ohms.  This being the case, with any amp except possibly OTL, it's best to parallel two 604, not series.  The high load impedance of 10 Ohms requires only moderate speaker wire AWG, but as always, keep it as short as possible.

[Russell Dawkins]

link leads to Google for me, too.
James, although theory says this vertical spacing of two identical sources should not work, nevertheless I have heard this sort of arrangement work spectacularly well!

[James Romeyn]

link leads to Google for me, too.
James, although theory says this vertical spacing of two identical sources should not work, nevertheless I have heard this sort of arrangement work spectacularly well!

I stand corrected and am absolutely not surprised to hear your report, which I trust 100% accurate (IOW, I trust your opinion very highly).  In fact, I tried best to word shape my reply to allow for your interesting and insightful reply.  Can you explain your experience?  I sure can't.

I'll relate a similar experience.  Jeff Hedback mentioned that 1/4 wavelength is the theoretical limit for maximum distance between subs and mains at the crossover frequency.  My room is about 25.5' x 16.5', speakers on the short wall.  One of my four subs sites on the back wall near the L wall.  The largest distance between sub and main is about 24'.  The sub LP is 4th order @ 67 Hz increasing to 6th order @ 80 Hz (main HP is 2nd order @ 80 Hz).

Jeff's theoretical maximum spacing: 1100fps / 67 Hz = 16.4' / 4 = 4.1'.  24' actual / Jeff's 4.1' limit = 5.9 x Jeff's theoretical limit. 

Admittedly, the entire philosophy of the distributed sub array counters all other sub philosophies except its cousin the Dual Bass Array.  The distributed array creates four entirely new "bass modes" to damp the effects of the room's permanently fixed modes.  Each sub sites at a null point, easily found with an SPL meter while playing bass tones.  One of the sub's polarity is inverted.

Anyway, not only is the crossover absolutely seamless, it's impossible to discern the presence of any sub unless one physically contacts their ear to a sub cabinet.  Yes, 3" away from the cabinet, with SPL not to far below live, with significant sub output, the sub's presence is undetectable, even with a meter.  IOW, the meter reads the same 3' from a sub vs. 3" from a sub.  That's how thorough is the mode damping effect.

My experience with regular sub applications is that Jeff's rule may be correct.   



       

[Russell Dawkins]

James - I'm not correcting you at all, just making an observation. 
I tend to be non-technical by nature, partly because math never was my strong suit.

I can't explain my experience, but I can describe it (at the risk of boring those who have already heard reference to it in another thread).

(cue old-time western music)
It was back in 1968 in Montreal when a DIYer brought a pair of his Wharfedale speakers over to stack on the identical pair my friend owned. We listened through a Heathkit receiver. When Janis Joplin sang it was one of two times in my life where, with my back turned to the speakers it sounded like the voice was materializing in the room. The effect was completely missing with the single pair. I don't know the model of Wharfedale, but I do remember they had 12" paper cone woofers which were probably run full range, as was the practice in those days, when manufacturers felt they could get away with it. Even JBL did this.

[James Romeyn]

In the 70s (19, not 18) my then-roomates and I stacked Larger Advents head-to-head (in two corners of the living room in a Victorian duplex on 8th Avenue one block S of San Francisco's GG Park) with a high-powered receiver.  Thanks for reminding me of a long-forgotten audio high point!  Those things rocked!   

[steve f]

Double Advents, oh yeah!  I graduated high school in '70. My buddies and I did the same thing, except in Chicago. Great sound and at a time when rock & blues were fantastic music. Lots of great memories.

[ProCast99]

Ha - ! Thanks, guys, for sharing the memories along with the theory ! 
there definitely was a little "something extra" in the room when the two sets were stacked.

Here is a link to an article I was directed to by an electronics friend of mine - seems v. germane to the amp-speaker relationship
http://komkris2000.webs.com/The%20story%20of%20Tube%20Friendly%20Speakers.pdf

So far my decision will be that, in absence of an impedance curve, and the fact I'm in a relatively small space and don't need the extra volume (nor do my neighbours!), I'll stick with the one pair for now. They are a really nice speaker, I'm happy with their presentation and "studio" like sound. Very clean.

Here's their photo - (sans grills)

[James Romeyn]

Digg. 

So the bottom mid-bass clones the co-ax driver above, minus the horn tweeter?

What are impedance ratings of output taps on your amp?  Are not modern tube amp output taps rated more like 4 Ohms and 8 Ohm, lacking the 16 Ohm tap?  An 8 Ohm tap into 20 Ohms is, I suppose, about 40% 8 Ohm power.  But still, 8 Watts @ 100 dB is no slouch!         

[ProCast99]

Yeah, the bottom is a passive radiator, which extends the bass response - the specs claim it goes down to 40 Hz.

The amp has both 8- and 4-ohm taps. My concern with stacking is that the 8 ohm is a nominal value, and I don't know if there are significant dips in its profile.

If the two speakers were running in parallel, (4 Ohm impedance), and there were big dips in the curve, i understand that this could cause problems. So I just want to play it safe. There is not a huge plus-side to stacking (although admittedly, it would be cool-sounding), and I don't want to screw up an amp...

I've just happened upon a historical photo gallery of the Kenwood-Trio speaker offerings - and this design is the only co-axial design they offered - it would be interesting to hear the story behind it...

[James Romeyn]

The driver is Japanese-made Kenwood clone of Alec 604, right?  Internet states Altex 604 crossover input is 20 Ohms.  What is Kenwood input impedance? 

Could you re-post an info link, or state the Kenwood model?  Looks like a cool speaker! 

[ProCast99]

Hi James:

Finding detailed info on this relatively rare speaker is a challenge - here is what I've been able to glean -

Closest thing to a spec's sheet I've seen :
http://audio-heritage.jp/TRIO-KENWOOD/speaker/ls-77.html

historical Kenwood / Trio speaker gallery (it is called LS-77 in this gallery)
http://audio-heritage.jp/TRIO-KENWOOD/speaker/index.html

The fellow who sold it to me re-capped it, but says he does not know what the cross-over impedance is. Interestingly, the cross-over is at 4KHz - which is higher than the 604 (tends to be around 1500 Hz), so I'm not sure what to say or guess...

Thanks - and I'd be interested in any further thoughts .. !

Stuart

[ProCast99]

Oops - sorry - the model is alternatively - Kenwood LS-770 or Trio LS-77 (or some combination of the two)...

[James Romeyn]

So many speakers, so little time....

Still remember my old Kenwood L-O7M 200W bipolar mono blocks...thought I died and went to amplifier heaven at the time...still probably decent bass amps.