[R-390] R-392 help - Green Speaker-ology

[Barry Hauser]

Hi Mark & gang:

I coulda' told ya' ... the primary limiting factor with the LS-166 (and 
LS-454, etc.) is not necessarily the transformer, it's the basic 
speaker/enclosure design.  They are built to be waterproof and 
blast/concussion resistant.  The R-392 is waterproof, which was 
bathtub-verified by one of the list members a few years ago -- and actually 
floats, though face-down -- which is not particularly convenient. ;-)

The LS-166 and others of the series, can be called Accidental Acoustic 
Suspension design.  In addition, the cone is made of heavily 
varnished/sealed cloth and has two form-fitting grilles or baskets front and 
aft as part of the water-proofing and blast resistance, respectively.  The 
suspension is very stiff.  Also, the intention is to cover the code and 
voice frequency ranges, to hopefully improve intelligibility, as with other 
"communications" type speakers.  That's on the presumption that much of the 
signal content outside the range of, oh, 300-3,000 Hz is likely to be noise 
or not needed.

I used to home brew speaker systems years ago -- with hightly variable 
results -- so had studied up on it.  So here's some more background for what 
it's worth.

There are two basic types of speaker enclosure designs -- unsealed and 
sealed.  For the most part, until the late 50's or so, maybe mid 60's, the 
unsealed were the rule.  These ranged from simple open back designs -- like 
many popular acccessory speakers for communications gear, to rather 
elaborate bass-reflex designs.  It all has to do with the back-wave.  When a 
speaker driver physically oscillates, it produces both.  For HF transducers, 
it doesn't matter much as high frequency audio is directional.  However, the 
back wave of a regular or LF/woofer speaker cancels out much of the front 
wave.  If you run a woofer driver outside of an enclosure, sometimes you can 
barely hear it.  If you do the same with a full-range driver, it will lack 
bass and you'll mostly be hearing the higher frequencies.

So, a primary challenge in speaker design is to deal with the back wave. 
Simple open back speakers sort of deal with it -- providing side walls which 
suppress/redirect some of it.  Then there were the bass-reflex designs and 
variants which generally attempt to make use of the energy by physically 
reversing the phase of the back wave and putting it out the front of the 
enclosure through a tuned port.  Just how well in-phase it becomes as well 
as a bunch of other parameters would determine resulting frequency response 
and overall fidelity.  Bass reflex designs usually benefitted from size --  
the bigger the better -- but not always -- as the "monsters" I built proved 
out.

Along came the acoustic suspension design.  The basic idea was to bottle up 
the back wave -- but as with most things, there's a lot more to it.  This 
design is inherently less efficient, requiring more power, but allowed for a 
more compact enclosure.  (Remember the wattage wars of the old days? --  
Triggered by the introduction of lower efficiency speakers.)  Not only is 
the back wave not make use of, but sealing the enclosure puts much more 
physical resistance on the movement of the cone -- the driver is basically 
"trying to" compress and expand a given volume of air.  This begat the need 
for more compliant suspension parts - -surround and spider (corrugated disk 
that supports the voice coil), yet stiffer cone material.  Many drivers are 
specifically designed for either accoustic suspension or free-air 
enclosures.  In fact, some of the more extreme suspension drivers can 
self-destruct if operated at high volumes in free air because the thing is 
supposed to be impeded by the trapped air and there's nothing to restrict 
movement.

Anyway, you can buy a small metal speaker unit about the size of the LS-166 
that is acoustic suspension and will sound pretty good.  That's largely 
because the driver is high-compliance and acoustically matched to the 
enclosure.  They also sell a lot of small bass-reflex speakers of similar 
size -- they have small ports either front or back.

Which reminds me ....   If you take an old National, or similar, open back 
communications speaker and place it so the back is about 12 inches from a 
wall, it will improve the lower frequency end.  You can experiment with 
varying the distance -- effectively tuning the phase of the reflection of 
the back wave.  Also may benefit from being in a corner - as with many 
speakers - for that and other reasons.  There are a number of other relevant 
parameters re speaker systems, such as dealing with peak resonance of the 
drivers and enclosures, etc.  Fortunately, I don't remember the rest of it 
all that clearly. ;-)

Back to the LS-166.  Here's a simple experiment -- try running it with the 
back off, if you haven't already, and vary its position.  There may be some 
improvement.  However, the tinny sound is also due to its construction --  
the stiff, waterproofed cone and suspension materials, etc. which restrict 
movement.  In addition, the enclosure was not designed and "tuned" for best 
fidelity either.

The next step would be to replace the driver, however, I'm not sure what 
would be the best choice.  A unit made for acoustic suspension may require 
higher wattage than the R-392 can put out. Probably better to use a 
universal type and leave the back off.  There would still be the limitation 
of the transformer, but you could use a Hammond instead.

Or, leave the LS-166 for display and use a different speaker, or even 
amplified computer speakers and bypass the audio stage of the R-392.  (You 
can remove the 26A7 and reduce heat.)

Oddly though, the speaker in the "Angry-5" -- AN/GRR-5  R-174 "gas" 
receiver -- built into the power supply half, is of similar design --  
waterproof, concussion-resistant -- front and back screens, etc.  However, 
they sound a good deal better -- actually not all that bad.  I'm sure part 
of it is due to the larger enclosure space -- the power supply section --  
but the driver is somewhat different and, I suppose, other factors are 
involved.  They were from the same time frame as the LS-166's, though.

Probably more than you wanted to know about speakers, eh?

Barry














Mark wrote:

> Change 1;  Just got my hands on an old LS-166/U speaker.  I repeated my 
> experiment with the low frequency cutoff of the 600 ohm to 8 ohm 
> transformer.  Either I mis-remembered, or they changed the transformer 
> design.  It swept 3dB down at 4 Hz to 0dB down from 10 Hz to greater than 
> 200kHz!  Noted a few peaks and troughs less then 3 dB from the speaker 
> load. as I swept it.  I did note that it sounds quite as tinny as I 
> remembered.  But the speaker itself is entirely different from what I 
> remembered.  So changing the transformer won't work on the LS-166/U for 
> bettter lows.  At least not without changing the speaker itself.  Cabinet 
> size probibily has something to do with it, but the sound does not change 
> with the back off.  Can't do an audio sweep because the Sound Level Meter 
> brought the farm many moons ago.  Sorry about the bum steer.
>
> Mark Huss wrote:
>
>> That was my first thought, too.  That is why I tried Stand-By.  I think 
>> Barry is right.  Sounds like an AGC problem. I also own an R-390A from 
>> 1955, Origional Collins #2792.  And if anybody has a spare PTO they want 
>> to part with, I am buying. (the origional one has about a khz of 'slip' 
>> from wear).
>> As for the LS-166.  I actually need the UG-77 connector on it.  As a 
>> point of note about the tinny sound.  Researched this while stationed in 
>> Korea for the RATT Rig operators, who liked to tune in shortwave on the 
>> secondary receiver.  They couldn't use stereo speakers because 2nd LT 
>> 'Crash' Rothman objected to the 'Unauthorized Equipment'.  The little 600 
>> to 8 ohm transformer is the main culpret.  Manually swept it using an 
>> audio oscillator and there is a nice rolloff below 300 hz.  Cured the 
>> problem by replacing the transformer with a 70.7 volt one, and stuffing 
>> the metal cabinet with fiberglass.  The ops were appreciative.  Said it 
>> sounded a lot better.  And 2nd LT 'Crash' Rothman was none the wiser.
>> R-392/URR, Stewart-Warner, Sn# 2681R, Order nr. 11653-PH-52.  And from 
>> the paint on the case, belonged last to the Radio PLT, Co B, 198 Sig.
>
>
> _____________________________________________________________
> R-390 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/r-390
> Help: http://mailman.qth.net/faq.htm
> Post: mailto:R-390 at mailman.qth.net
> Unsubscribe: http://mailman.qth.net/mailman/options/r-390
>