On Sun, 14 Oct 2018 23:55:27 -0700 (PDT)
Jeefaw K Effkay wrote:

On Sunday, October 14, 2018 at 11:01:15 PM UTC+1, Ralph Mowery wrote:
In article , lid
says...

On Sun, 14 Oct 2018 12:04:52 -0700 (PDT)
Jeefaw K Effkay wrote:

On Sunday, October 14, 2018 at 7:33:10 PM UTC+1, Jeff Liebermann
wrote:

It might help to understand why some bands use LSB while
others USB. In the early daze of sideband radio, the common
IF frequency was 9MHz. The radios had only one sideband
filter. With one filter, it was cheaper and easier to mix
and up convert in the transmitter. So, to save the cost of
adding a second filter, the bands below 9MHz were designated
as LSB and the band above 9MHz became USB. Eventually, radios
were built with two sideband filters, and this was no longer
important. As usual, the legacy technology remained in place
to haunt the survivors to this day.

I've seen this explanation before, but it doesn't make sense.

A 9MHz USB signal mixed with a 5.0 to 5.5MHz VFO will produce
mixing products in the 80m and 20m bands - but both will be
upper sideband.

When the 9 MHz is mixed with the 5 mhz the 20 meter signal is upper
sideband. The 80 meter signal is inverted and becomes the lower
sideband NOT usb.

Consider a 2 tone signal at the 9MHz USB IF, comprising 900Hz and
1300Hz tones.

The components will be 9.0009 and 9.0013

Subtract the VFO at 5.5MHz:

9.0009 - 5.5 = 3.50009
9.0013 - 5.5 = 3.50013

Nothing has been inverted. The 80m signal is still upper sideband.

GB3BERNIE

Ralph is posting from rec.radio.amateur.antenna and google groups
strips the crosspost - without a repeater, he's not going to answer you.

In article ,
says...

Consider a 2 tone signal at the 9MHz USB IF, comprising 900Hz and
1300Hz tones.

The components will be 9.0009 and 9.0013

Subtract the VFO at 5.5MHz:

9.0009 - 5.5 = 3.50009
9.0013 - 5.5 = 3.50013

Nothing has been inverted. The 80m signal is still upper sideband.

GB3BERNIE

Ralph is posting from rec.radio.amateur.antenna and google groups
strips the crosspost - without a repeater, he's not going to answer you.





Try it the other way around and use a ssb generated at 5 mhz and the vfo
at 9 mhz. It is difficult for me to remember which was used for the vfo
and ssb generator.

On Mon, 15 Oct 2018 10:43:09 -0400, Ralph Mowery
wrote:

In article ,
says...

Consider a 2 tone signal at the 9MHz USB IF, comprising 900Hz and
1300Hz tones.

The components will be 9.0009 and 9.0013

Subtract the VFO at 5.5MHz:

9.0009 - 5.5 = 3.50009
9.0013 - 5.5 = 3.50013

Nothing has been inverted. The 80m signal is still upper sideband.

GB3BERNIE

Ralph is posting from rec.radio.amateur.antenna and google groups
strips the crosspost - without a repeater, he's not going to answer you.

Try it the other way around and use a ssb generated at 5 mhz and the vfo
at 9 mhz. It is difficult for me to remember which was used for the vfo
and ssb generator.

I'm fairly sure the SSB was generated at 9MHz. Googling for a
reminder, I find a large number of 9MHz sideband crystal filters
available, while nothing for 5MHz. Presumably, the 9MHz sideband
crystal filter is use for both the receiver IF filter and in the
exciter SSB generator to strip off the unwanted sideband.
https://www.google.com/search?q=9+mhz+crystal+filter
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 16/10/2018 08:08, Jeff wrote:

I'm fairly sure the SSB was generated at 9MHz. Googling for a
reminder, I find a large number of 9MHz sideband crystal filters
available, while nothing for 5MHz. Presumably, the 9MHz sideband
crystal filter is use for both the receiver IF filter and in the
exciter SSB generator to strip off the unwanted sideband.

You are correct 9MHzwas a common IF for both tx & rx.

A common way of generating both usb and lsb was to have 2 switched
crystals with frequencies just above and below 9MHz in the oscillator,
feeding a balanced mixer, before the xtal filter, and switch depending
on which sideband you required.

Is there a mathematician on here that can explain the maths of sideband
inversion/retention?


--
Spike

"Nearly all men can stand adversity,
but if you want to test a man's character,
give him an internet group to manage"

On 16/10/2018 09:14, Jeff wrote:

A common way of generating both usb and lsb was to have 2 switched
crystals with frequencies just above and below 9MHz in the oscillator,
feeding a balanced mixer, before the xtal filter, and switch depending
on which sideband you required.

Is there a mathematician on here that can explain the maths of sideband
inversion/retention?

No inversion is required with this method.

If you feed a ~9MHz signal and audio into a balanced mixer the output
will be both sidebands plus a suppressed carrier.

Your xtal filter is ~2.4kHz wide centred on 9MHz, so if you move the
frequency of the ~9Mhz signal (switch a crystal) going into the balanced
mixer either above or below 9MHz you can select which side band goes
through your filter.

Simples.

Wasn't a similar system used in the Yaesu FT-200 (9MHz IF, 5 MHz VFO)?

IIRC the set had a NORM/INV sideband switch.


--
Spike

"Nearly all men can stand adversity,
but if you want to test a man's character,
give him an internet group to manage"

On Tue, 16 Oct 2018, Spike wrote:

On 16/10/2018 09:14, Jeff wrote:

A common way of generating both usb and lsb was to have 2 switched
crystals with frequencies just above and below 9MHz in the oscillator,
feeding a balanced mixer, before the xtal filter, and switch depending
on which sideband you required.

Is there a mathematician on here that can explain the maths of sideband
inversion/retention?

No inversion is required with this method.

If you feed a ~9MHz signal and audio into a balanced mixer the output
will be both sidebands plus a suppressed carrier.

Your xtal filter is ~2.4kHz wide centred on 9MHz, so if you move the
frequency of the ~9Mhz signal (switch a crystal) going into the balanced
mixer either above or below 9MHz you can select which side band goes
through your filter.

Simples.

Wasn't a similar system used in the Yaesu FT-200 (9MHz IF, 5 MHz VFO)?

IIRC the set had a NORM/INV sideband switch.

That wasn't uncommon, the conversion scheme allowing for the "default"
sideband to be one switch position, so the only time you needed to switch
sidebands was if you needed the "wrong" sideband".

Michael

On Mon, 15 Oct 2018, Ralph Mowery wrote:

In article ,
says...

Consider a 2 tone signal at the 9MHz USB IF, comprising 900Hz and
1300Hz tones.

The components will be 9.0009 and 9.0013

Subtract the VFO at 5.5MHz:

9.0009 - 5.5 = 3.50009
9.0013 - 5.5 = 3.50013

Nothing has been inverted. The 80m signal is still upper sideband.

GB3BERNIE

Ralph is posting from rec.radio.amateur.antenna and google groups
strips the crosspost - without a repeater, he's not going to answer you.





Try it the other way around and use a ssb generated at 5 mhz and the vfo
at 9 mhz. It is difficult for me to remember which was used for the vfo
and ssb generator.

I think the origins are with a 5MHz IF. This has come up before, the same
explanation given, yet if I wasn't sick and did the figuring, I think it's
that the 9Mhz one wasn't it, but a 5MHz IF does do the inversion.

But I can't remember what rig had a 5MHz IF. THey existed, but the ones I
can think of came later. So maybe it was a phasing rig, but which did
conversion rather than generate the SSB signal on the signal frequency.
The Central Electronics 10 transmitter maybe, reinforced by their later
20, but I havent' checked.

There was a popular rig in QST that worked out the figures so the low IF
was converted up to an intermediate frequency with one crystal, one caused
no inversion, but if you multiplied the crystal frequency by three, it was
high side and inverted the sideband. But since it did both sidebands, it
wouldn't have been a standard for LSB below a certain frequency, There
were early ssb rigs that didn't have selectable sideband, they just picked
conversion frequency properly so below 10MHz, it was LSB, and above was
USB. Since nobody used the opposite sideband, no need for a switch.

Michael