Re: Elecraft Digest, Vol 49, Issue 52

[alan alan]

Good Morning Ian,

This could be a good short piece in RadCom

73  Alan  G0HIQ

--- On Sun, 25/5/08, [hidden email] <[hidden email]> wrote:
Message: 1
Date: Sat, 24 May 2008 09:03:49 +0100
From: Ian White GM3SEK <[hidden email]>
Subject: Re: [Elecraft] USB on all bands ??
This is a classic detective story, with more than the usual share of red herrings!   .......



      __________________________________________________________
Sent from Yahoo! Mail.
A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[gm3sek]

Hello ALan

>Good Morning Ian,
>
>This could be a good short piece in RadCom
>
Been there, done that (or most of it) when the Editor asked for
"something historical" for the July 2003 RSGB Anniversary issue.

The information about 10MHz came along later, from amateurs who were
personally involved with the international ISB links. The UK had a
world-wide network of links between Commonwealth countries at that time,
using transmitters supplied by Marconi, and thus had a very large
influence on the international standards that emerged.

What amazes me is the persistence of the belief that it's all because a
9MHz SSB generator and a 5MHz VFO produce opposite sidebands on 14MHz
and 3.5MHz. Anyone who can rub two numbers together can see that isn't
true... but that makes no difference at all.


--

73 from Ian GM3SEK
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[gm3sek]

Ian White GM3SEK wrote:
>Hello ALan
>
>>Good Morning Ian,
>>
>>This could be a good short piece in RadCom
>>
>Been there, done that (or most of it)

Sorry, that wasn't intended for the whole list.


--

73 from Ian GM3SEK
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Julian, G4ILO]

GM3SEK wrote

What amazes me is the persistence of the belief that it's all because a
9MHz SSB generator and a 5MHz VFO produce opposite sidebands on 14MHz
and 3.5MHz. Anyone who can rub two numbers together can see that isn't
true... but that makes no difference at all.

9+5.0=14.0. 9+5.5=14.5.
9-5.0=4.0. 9-5.5=3.5.

Sorry, I don't see how that isn't true. I'm sure I can also recall some published transceiver designs that took advantage of this fact to provide two band coverage using a single VFO and 9MHz IF.

[gm3sek]

G4ILO wrote:

>
>
>GM3SEK wrote:
>>
>> What amazes me is the persistence of the belief that it's all because a
>> 9MHz SSB generator and a 5MHz VFO produce opposite sidebands on 14MHz
>> and 3.5MHz. Anyone who can rub two numbers together can see that isn't
>> true... but that makes no difference at all.
>>
>9+5.0=14.0. 9+5.5=14.5.
>9-5.0=4.0. 9-5.5=3.5.
>
>Sorry, I don't see how that isn't true. I'm sure I can also recall some
>published transceiver designs that took advantage of this fact to provide
>two band coverage using a single VFO and 9MHz IF.

This particular mixing process gives 3.5MHz and 14MHz very conveniently
- but it gives the SAME sideband on both bands.

Imagine a phasing-type SSB generator with a 9.000MHz suppressed carrier
frequency. When configured for USB, an audio tone at 1kHz gives output
in the upper sideband, at 9.001MHz. Then:

9.001 + 5.300 = 14.301 - that's 14.300MHz USB
9.001 - 5.300 = 3.701 - that's 3.700MHz USB

We have the SAME sideband on both 20m and 80m!

To swap sidebands in this example requires the SSB generator to be
switched to LSB. With a 1kHz tone, the output switches to 8.999MHz. Then
the same calculations give:
8.999 + 5.300 = 14.299 - that's 14.300MHz LSB
8.999 - 5.300 = 3.699 - that's 3.700MHz LSB

Once again, we get the SAME sideband on both 20m and 80m.

A 9MHz filter exciter is slightly different, because it's the filter
passband that is centred on 9.000MHz. Such exciters swap sidebands by
switching between carrier oscillators. For USB the CO is below the
filter passband at 8.9985MHz, or for LSB the CO is above the filter at
9.0015MHz. The arithmetic is a bit more messy, but the result is exactly
the same - generating SSB at 9MHz does NOT automatically swap sidebands
between 80m and 20m.

That is clear mathematical proof that the 9MHz SSB exciters were NOT
responsible for the ham USB/LSB convention. On the contrary, when
changing between 20m and 80m they were forced to switch sidebands at
9MHz in order to *follow* that convention!  There is also plenty of
historical proof that the USB/LSB convention had been in existence for
several years before the first published 9MHz design came along in 1956.




--

73 from Ian GM3SEK         'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[WILLIS COOKE]

It does not give the same sideband on 80 and 20.  You
also have to consider that the rotation of the tuning
capacitor of the VFO turns the opposite direction for
20 as it does  for 80.  You will find this on the dial
of many pre 1970 commercial offerings.  Everyone needs
to use one of the old SSB rigs for a few days to
appreciate how good even the worst of today's rigs
are.  Then use a crystal controlled one or two tube
transmitter and a 40s or 50s affordable receiver like
a S-38 or NC-53 and appreciate how good the 60s SSB
rigs were compared to the 40s and 50s.

Cookie, K5EWJ

--- Ian White GM3SEK <[hidden email]> wrote:

> G4ILO wrote:
> >
> >
> >GM3SEK wrote:
> >>
> >> What amazes me is the persistence of the belief
> that it's all because a
> >> 9MHz SSB generator and a 5MHz VFO produce
> opposite sidebands on 14MHz
> >> and 3.5MHz. Anyone who can rub two numbers
> together can see that isn't
> >> true... but that makes no difference at all.
> >>
> >9+5.0=14.0. 9+5.5=14.5.
> >9-5.0=4.0. 9-5.5=3.5.
> >
> >Sorry, I don't see how that isn't true. I'm sure I
> can also recall some
> >published transceiver designs that took advantage
> of this fact to provide
> >two band coverage using a single VFO and 9MHz IF.
>
> This particular mixing process gives 3.5MHz and
> 14MHz very conveniently
> - but it gives the SAME sideband on both bands.
>
> Imagine a phasing-type SSB generator with a 9.000MHz
> suppressed carrier
> frequency. When configured for USB, an audio tone at
> 1kHz gives output
> in the upper sideband, at 9.001MHz. Then:
>
> 9.001 + 5.300 = 14.301 - that's 14.300MHz USB
> 9.001 - 5.300 = 3.701 - that's 3.700MHz USB
>
> We have the SAME sideband on both 20m and 80m!
>
> To swap sidebands in this example requires the SSB
> generator to be
> switched to LSB. With a 1kHz tone, the output
> switches to 8.999MHz. Then
> the same calculations give:
> 8.999 + 5.300 = 14.299 - that's 14.300MHz LSB
> 8.999 - 5.300 = 3.699 - that's 3.700MHz LSB
>
> Once again, we get the SAME sideband on both 20m and
> 80m.
>
> A 9MHz filter exciter is slightly different, because
> it's the filter
> passband that is centred on 9.000MHz. Such exciters
> swap sidebands by
> switching between carrier oscillators. For USB the
> CO is below the
> filter passband at 8.9985MHz, or for LSB the CO is
> above the filter at
> 9.0015MHz. The arithmetic is a bit more messy, but
> the result is exactly
> the same - generating SSB at 9MHz does NOT
> automatically swap sidebands
> between 80m and 20m.
>
> That is clear mathematical proof that the 9MHz SSB
> exciters were NOT
> responsible for the ham USB/LSB convention. On the
> contrary, when
> changing between 20m and 80m they were forced to
> switch sidebands at
> 9MHz in order to *follow* that convention!  There is
> also plenty of
> historical proof that the USB/LSB convention had
> been in existence for
> several years before the first published 9MHz design
> came along in 1956.
>
>
>
>
> --
>
> 73 from Ian GM3SEK         'In Practice' columnist
> for RadCom (RSGB)
> http://www.ifwtech.co.uk/g3sek
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft   
>
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[gm3sek]

WILLIS COOKE wrote:
>It does not give the same sideband on 80 and 20.  You also have to
>consider that the rotation of the tuning capacitor of the VFO turns the
>opposite direction for 20 as it does  for 80.

The direction of tuning is not what inverts the sideband.

The only thing that inverts the sideband is if the SSB frequency is
subtracted in the mixing process. So:

9MHz USB + 5MHz VFO = 14MHz USB

9MHz USB - 5MHz VFO = 4MHz USB - sideband is not inverted.

But in Norgaard's 'New Approach to SBB Generation' in QST, June 1948,
the SSB is generated on 5MHz and the VFO is around 9MHz, so:

9MHz VFO + 5MHz USB = 14MHz USB

9MHz VFO minus 5MHz USB = 4MHz **LSB** - now the sideband is inverted.

If anyone is still in doubt about this, words aren't going to help any
more. Please take a few minutes with a pencil and a calculator, punch in
some numbers and verify it for yourself.


--

73 from Ian GM3SEK         'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On May 25, 2008, at 7:50 AM, WILLIS COOKE wrote:

> It does not give the same sideband on 80 and 20.

Actually, it does.

Imagine you are generating your sideband at a 9MHz IF. In the IF you  
are generating USB, i.e. you have a 9MHz carrier and 1KHz and 1.2KHz  
tones injected to the mic jack. This produced a spectrum at 9MHz of:

9000.0 KHz - carrier (suppressed)
9001.0 KHz - first tone
9001.2 KHz - second tone.

Now we mix subtractively with 5100 KHz to produce a result in the 80M  
band:

9000.0 - 5100 = 3900.0 KHz
9001.0 - 5100 = 3901.0 KHz
9001.2 - 5100 = 3901.2 KHz

Notice that the spectral lines are above the carrier in increasing  
frequency. That is USB.

Now let's repeat this using additive mixing to get 20M.

9000.0 + 5100 = 14100.0 KHz
9001.0 + 5100 = 14101.0 KHz
9001.2 + 5100 = 14101.2 KHz

The spectral lines are still above the carrier indicating USB.

Now if you repeat the process by setting the VFO to 5500.0 KHz you  
will see that the radio would tune to a lower frequency on 80M while  
tuning to a higher frequency on 20M thus indicating that the radio  
tunes "backwards" for 80M but in both cases it still produces an USB  
signal.

--

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com



_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[WILLIS COOKE]

I think that you have the answer to the riddle Ian.  9
megacycle filters were not available until about 1960
with the second generation of transceivers.  By the
time of Galaxy 3 and 5, NCX-5, Swan 350, etc the need
for the USB/LSB selection was passed, but we stayed
with the tradition.

I have seen some remarks that would indicate that the
Elecraft is unique in being able to select the
sideband, but I have not had a transceiver that would
not.  Even back to the Hammarlund HX-50, if I remember
correctly one could select the sideband.  I know on
the two transceivers I have in the shack, Kwd TS-850
and Yaesu FT-900 when you select SSB it puts you on
the preferred sideband, but if you press the button
again it puts you on the other sideband.

Cookie, K5EWJ

--- Ian White GM3SEK <[hidden email]> wrote:

> WILLIS COOKE wrote:
> >It does not give the same sideband on 80 and 20.
> You also have to
> >consider that the rotation of the tuning capacitor
> of the VFO turns the
> >opposite direction for 20 as it does  for 80.
>
> The direction of tuning is not what inverts the
> sideband.
>
> The only thing that inverts the sideband is if the
> SSB frequency is
> subtracted in the mixing process. So:
>
> 9MHz USB + 5MHz VFO = 14MHz USB
>
> 9MHz USB - 5MHz VFO = 4MHz USB - sideband is not
> inverted.
>
> But in Norgaard's 'New Approach to SBB Generation'
> in QST, June 1948,
> the SSB is generated on 5MHz and the VFO is around
> 9MHz, so:
>
> 9MHz VFO + 5MHz USB = 14MHz USB
>
> 9MHz VFO minus 5MHz USB = 4MHz **LSB** - now the
> sideband is inverted.
>
> If anyone is still in doubt about this, words aren't
> going to help any
> more. Please take a few minutes with a pencil and a
> calculator, punch in
> some numbers and verify it for yourself.
>
>
> --
>
> 73 from Ian GM3SEK         'In Practice' columnist
> for RadCom (RSGB)
> http://www.ifwtech.co.uk/g3sek
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft   
>
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>

Willis 'Cookie' Cooke
K5EWJ
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com