Altec 2[.5]-way
mini horn setup
Altec 2-way system with 3000H super tweeter
Back in the late 90s, I took the challenge of building a "mini" horn based speaker system as an opportunity to learn more about the principles of integrating multiple drivers and getting them to sound as coherent to my ears as possible. This concept was inspired by tales I have read and heard about the ultra rare and evasive Western Electric WE757 studio monitor which uses a WE728B 12" for the lower frequencies and KS12027 horn attached to a WE713C compression driver for the high frequencies housed in a slant front 3 cu.ft. sealed cabinet.
Western Electric 757 monitors
I still believe that the classic WE/Altec 8" full range 755As in open baffles provide very natural and coherent midrange reproduction within it's 70-13,000 HZ bandwidth. However I am well aware of its inherent limitations in terms of maximum SPLs [sound pressure levels] and bass extension and I wanted to gain those qualities without compromising the midrange too much. Since I wrote the Sound Practices Homebrewer article, I have been fortunate to gather vintage Altec speaker/horn components to experiment with a high efficiency 2-way set-up.
The bass cabinets were built from Altec 614 plans using 3/4" birch plywood. Altec LF drivers were never meant to be used in sealed cabinets, so it had to be a ported enclosure [although my friends are happy mounting them in OBs]. These cabinets are loaded with a pair of Altec 419A (8 ohms), 12" Bi-Flexes. Ideally I should have used a pair of 414A or Z (16 ohms), but the only matched pair I found was quickly snatched up by my friends at SETUP, back in Manila. But I have no regrets with the 419As because its low frequency nature seems to be identical to the 414, plus I acquired them when they were still reasonably affordable.
Altec 614 cabinet plan
From the upper midrange up, I am using the early version of Altec's 811 sectoral horns attached to Altec 804A (16 ohms) compression drivers. After several listening comparisons with a borrowed pair of Altec N800D crossover, I settled on a simple first order slope [6dB/octave] crossover [woofer inductor not shown since it is inside the cabinet] hinged at around 900 hz with about 10 dB of padding for the 811/804A to match the sensitivity of the 419As.
Altec 419A, 804A/811horn lens 1st order crossover
I finally ended up with a speaker system that is closer in lineage to the Lansing Iconic or later Altec Valencias and can now enjoy greater bass extension and dynamic range while indulging in the capabilities of single ended amps putting out 1.5W or less.
Altec 2-way in the late 90s
One thing I learned through years of experimentation and investigation of several drive units is "ear testing". Initially I will put down a driver on my test bench and then hook them up to a small pocket FM radio tuned to a decent station [a trick from Walt Bender and readers of Audio Mart] with clip leads on one end and a mini phone plug for the earphone jack. I always listen for "articulate midrange" [especially vocals] whether it is a woofer, fullrange/midrange or tweeter. For example, a good tweeter should be able to render the overtones of a human voice so that one can still understand the text or lyrics. Try that with a modern dome tweeter and most likely all you will hear is spit, sizzle and tizz. A good low frequency driver should roll off smoothly without nasty peaks. If the driver sounds decent through a 1/4-1/2 W IC amplifier, it should even be better through DHT amps.
After passing these initial tests, I install the drivers in a cabinet, baffle board or horn lens and hook them up fullrange to my system using either my milliwatt SE171A [for compression drivers and tweeters] or an SE2A3 [for fullrange or LF drivers] amp and leave it in the system tuned to my favorite FM station all day just to familiarize myself with their inherent sonic quality. It is only after I have gone through this routine will I feel confident enough to try and build a system around the set of drivers in question.
Since I only use flea power amps the simplicity of a 1st order crossover [6dB slope/octave] is very much in line with my audio philosophy, because it allows the drivers to "breathe freely" and unencumbered. Complex crossovers function like negative feedback, they compensate for poorly designed drivers to make them look good in front of a spectrum analyzer. Crossover points are chosen based on what I heard during the "ear test phase" [with reference to their original specifications, if available] and then I refer to this crossover calculator site to figure out 1st order/6 dB per octave slope values for capacitors, inductors and fixed resistors for the L-pad. I find 'clip leads' handy for ease of changing component values or moving the crossover point around until I am satisfied with the integration of the drive units at the crossover point using a broad range of music and signal sources [digital, analog LP, tape, 78 rpm, etc.]. My musical training to recognize pitches quickly also comes in handy when fine tuning the crossover.
Revised Components and Crossover
414Z [replaced the 419A]
12" paper cone alnico magnet low frequency driver [same driver as a 414A less magnet cover]
16 ohms nominal
30-4000 hz freq. response
99db SPL/4 ft./1W
25W maximum input
I use "bi-wire/bi-amp" type speaker terminals for the LF wherein one pair has a 2.5mH choke in series [6db/octave roll-off @ around 900-1000hz] and the other pair is directly connected to the 414s allowing the unit to roll off mechanically. Fullrange operation of the 419A was not possible because it had a resonant peak starting around 2khz - 4khz that needed to be 'notched out' with a choke when paired with the 811/804A. This was my 'justification' for acquiring a pair of 414Zs in spite of the fact that the 419A did a good job as a LF driver. For a short while I got hold of a pair of mint 416 - 15" woofers as used in the Valencia or Voice of the Theater however I gave it up even if did slightly better in the bass because I preferred the ability of the 414, 12 incher to play better in the midrange running full range without being hampered by a choke.
802D [replaced the 804A]
16 ohm high frequency compression driver with aluminum diaphragm for 1" throat horn
500-22,000 hz frequency response
sensitivity - 111.7 dB SPL/4 ft./1W [with 811 or 511 horn installed]
30W [?] maximum input
A parallel set of paper in oil can type caps of 12uf adds the 802D/811 between 800-1000hz, which runs all the way up to its natural cutoff. Since the 802D is 2 dB more efficient compared to the 804A, I had to compensate [-12dB of padding] so that I can get the 811/802D down to the 99dB efficiency of the 414Z. I don't hear much difference between the 804A and the 802D, they both sound good owing to the superb transient ability of the original aluminum diaphragms. When I saw this absolutely MINT pair of 802D being offered by a friend [at a truly 'friendly price', thanks Sherwin!], I just had to have it!
3000H
8 or 16 ohms
3000-22,000 freq. response
20W [?] maximum input
Adding the .5
While attending a radio show a couple years ago, I acquired a single 3000A tweeter and was truly enamored by its ability to add shimmer and air above the capabilities of my then 804/811 combo. [It does the same job with any FR driver like 755A, 755C or E in an open baffle]. However I later gave up finding a suitable mate and traded it for an SME3012. A year past and I saw a NOS pair 3000H at eBay that made me lose sleep until I placed a winning bid..
Altec produced the 3000 series as the HF driver in the 601/602 Duplex, Melodist [756B + 3000A] and Seville. The N3000 2nd order [12dB/octave] crossover was specifically designed for this unit. However based on many blown units I've seen, this may have been set too low or the caps dried up unknown to the original owner who was driving the speakers at ear splitting levels with MC 275s! That's why I am a bit dubious about the 20W rating, likewise the 30W rating for the 802D from a 1960s Altec catalog.
The 3000H cuts in around 13-15khz using .7uf paper in oil cap with a 16ohm variable L-pad set around midpoint. It basically aids the falling response of the 811/802D combo due to its wider mouth flare. Thus, the 3000H is acting as a "super tweeter" adding that subtle but significant improvement in air and spaciousness. Technically it still is a 2-way speaker system. Here is the revised crossover.
*CAVEAT*
If your amp
has over 8W output you run the risk of frying the compression driver diaphragm
with a 1st order crossover!
Phasing
Phase may or may not be an issue to you but I did find that using terminal L2 as the [+] in the 802D and 804A presented a more focussed sound. Using L1 as [+] initially fools the ear to think there is more HF extension and air but prolonged listening convinced me that it was "phasey". Altec crossovers were mostly 12dB/octave devices which inverts phase in the HF output, therefore L1 was normally marked in red or [+] when 802s, 806s and 804s came from the factory.
Aligning the
sensitivity/efficiency of the horn/driver combination to the bass driver has a
significant effect on the overall character of the speaker system. There have
been many audio enthusiasts who have been turned off by the 'typical' honky and
forward sound ascribed to the sound of horns. This maybe due to the fact that
the HF horn is being driven at full tilt or with inadequate padding thereby
overpowering the bass driver. If you examine Altec crossovers, they were all
fitted with L-pads because none of their direct radiator low frequency drivers
ever equalled the efficiency of their compression drivers. In my set up I have
been satisfied with padding down the HF horn [802D] by 12 dB or 10dB [804A]
which brings it close to the manufacturer quoted sensitivity for the 414Z.
Fixed resistor values can easily be found by using the L-Pad
calculator
[bottom of the page].
I finally found a place that can resuscitate dead Altec 3000s:
Bill Hayes
MicroMike Labs
1537 West Cris Place
Anaheim, CA 92802
Tel: (714) 774-0342
Back to main page or Shop Talk
Updated: 10/01/06, JE Labs all rights reserved.