[ARC5] Some info on the S-38

Fri Nov 20 01:02:51 EST 2015

    A note on "sensitivity". The specs given are really for gain, i.e. 
audio output for a given RF input. The input signal should also have a 
spec for modulation frequency and percentage.  Such specs are sometimes 
useful for trouble shooting but are not a measure of the sets ability to 
recover weak signals.  A better, although not really satisfactory 
measure is the signal to noise ratio. This is a very common spec on all 
sorts of better quality receivers. It is the level of a  modulated 
signal plus noise over unmodulated signal plus noise for a stated ratio. 
The ratio is sometimes 10db (i.e. 3 to 1 voltage) and sometimes for 6db 
(2 to 1 voltage). A receiver like the S-38B, with a noisy converter 
working directly from an antenna, that is, without an RF stage, will 
have anything from around 10uV to 100uV to obtain the 10db ratio.  It 
will vary with the tube characteristics and the Q of the RF coils.  
Typical 1940s communication receivers will have something like 4 to 10uV 
again depending on the tubes and RF coils but will usually be better 
than the cheap receiver because of having two or more IF stages and 
narrower bandwidth.
     More modern receivers, post WW-2, with higher Gm miniature tubes, 
will have something on the order of 2uV for 10db  SNR.  Of course the 
SNR depends on the bandwidth and to some degree on noise from spurious 
responses which add to the overall noise.
     A much better measure is Noise Factor or Noise Figure but it is 
seldom encountered in lower frequency equipment although common for VHF 
through microwave gear.  NF is the ratio of the noise power output from 
the receiver to the theoretical noise power output from a "perfect" i.e. 
noise-free system.  It is measured by measuring the output of the 
receiver from a calibrated noise source and comparing that to the output 
with no input but with the input terminated in a "perfect" resistor of 
the design value for the antenna.  Since the test signal is subject to 
the same bandwidth limits and spurious responses as an actual signal it 
is fairly valid measure of the noise performance of the receiver or 
amplifier being tested.  Typical "modern" vacuum tube receivers have NF 
on the order of perhaps 4 to 10db in the HF band. Because atmospheric 
noise is high in this region this is quite good performance. However at 
VHF and higher frequencies the noise of the antenna system is mostly 
resistive (i.e. Johson noise) with perhaps some galictical noise so much 
better NF is desired, hense the use of super-cooled devices at the input.
     It is common to state the performance of an IF filter as its 
bandwidth ratio from 6db down to the bandwidth at 60db down (1000 to 
1).  For a single tuned 455 Khz IF transformer a ratio of perhaps 10 to 
1 is common but can be a little more if the coils are high Q or if the 
frequency is lower.  The skirt width is pretty much a function of the 
type. The selectivity is also a function of the mutual coupling of the 
two coils.  When more IF stages are added the selectivity becomes 
greater (meaning sharper) and the combination of critically tuned and 
over coupled stages can yield good skirt selectivity.  An example is the 
Hammarlund HQ-120-X and 129-X.  Complex filters using crystals or high-Q 
magnetic resonators (mechanical filters of the Collins type) can have 
much smaller 6 to 60db ratios. Around 2 to 1 is typical for the 6 or 8 
pole filters found in today's equipment. Obtaining greater ratios can be 
done but the problem with eliminating ripple across the pass band and 
obtaining reasonably good phase and time delay response become difficult 
and expensive.  Bell System used crystal lattice filters up to 20 poles 
(maybe more) in its carrier and microwave systems but one does not 
usually encounter more than 8 poles in ham or commercial communications 
equipment.
     The S-38B does OK for a simple receiver.

On 11/19/2015 9:29 PM, K5MYJ wrote:
> Here's what a dB is:
> https://en.wikipedia.org/wiki/Decibel
>
> Bob Macklin
> K5MYJ
> Seattle, Wa.
> "Real Radios Glow In The Dark"
> ----- Original Message ----- From: "Leslie Smith" <vk2bcu at operamail.com>
> To: <arc5 at mailman.qth.net>
> Sent: Thursday, November 19, 2015 7:22 PM
> Subject: Re: [ARC5] Some info on the S-38
>
>
>> OMG!  OMG!  OMG!  How I hate it when I'm wrong!
>> Worse still - how I hate it when I was wrong and I should have known I
>> was wrong!
>> Worse than that - how I hate it when EVERYONE sees the mistake and
>> thinks "YOU WERE WRONG!"
>>
>> Sob, sob, sob.  I'm off to eat spaghetti!  (Better than eating worms!)
>> Tnx everyone.  I feel better now!
>>
>> 73 de Les Smith
>>  vk2bcu at operamail.com
>>
>> On Fri, Nov 20, 2015, at 11:26, Leslie Smith wrote:
>>> Correction:  I wrote that 60dB "down" amounts to one-millionth of the
>>> signal (or something like that).
>>>
>>> I should have written 60dB "down" amounts to an attenuation of the
>>> signal by one-million.  The signal is attenuated 10x 
>>> log(one-million).  Anyway, I figure that every-one on this list 
>>> knows that, and made the
>>> correction for me already.  -  Les
>>>
>>> -- 
>>> http://www.fastmail.com - Access your email from home and the web
>>>
>>> ______________________________________________________________
>>> ARC5 mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:ARC5 at mailman.qth.net
>>>
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>>
>> -- 
>> http://www.fastmail.com - The professional email service
>>
>> ______________________________________________________________
>> ARC5 mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:ARC5 at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

-- 
Richard Knoppow
1oldlens1 at ix.netcom.com
WB6KBL