# Sideband Rejection

#### msseng

Joined Jan 19, 2009
11
Hi,
I have a problem to reject the lower side band of a signal that exists very close to the carrier!
I have a 10MHz signal with a few hertz deviation. I mix it with a 10MHz signal and do some work on remained deviation signal.
But when I mix it again to the 1MHz I have an image that is undesirable.
because of very close distance between 10MHz and that deviation, I can't use any filter to reject the lower signal.

#### Wendy

Joined Mar 24, 2008
23,062
Lets see, you are mixing 10Mhz to 10Mhz. Why, for a reference (as in calibration)? Two 10Mhz signals = a few hertz + 20Mhz (assuming the two base frequencies are surpressed, a bad assumtion). I think we'll need a bit more background before we can be of much help.

If this is a calibration then why not use a Lessajous curve setup?

#### msseng

Joined Jan 19, 2009
11
Lets see, you are mixing 10Mhz to 10Mhz. Why, for a reference (as in calibration)? Two 10Mhz signals = a few hertz + 20Mhz (assuming the two base frequencies are surpressed, a bad assumtion). I think we'll need a bit more background before we can be of much help.

If this is a calibration then why not use a Lessajous curve setup?
I am mixing a 10+delat MHz signal to 10MHz,to get "delta". because only delta must be changed (up or downconvert). and then I mix it with 10MHz to have a 10+(delta)' signal without any image (it means that 10-(delta)' is undesirable)
because of small delta, I can't use any filter to filter undesirale signal.
I hope to say my purpose!

#### msseng

Joined Jan 19, 2009
11
You can build a crystal filter with a quite narrow passband.

Have a look here:
Dear
I can't use any filter.because the delta is about a few hertz.
and the filter that can filter this, must have a very very high Q (about 10^6)

if you have the other solution, I would be greateful to hear it!
thanks.

#### bertus

Joined Apr 5, 2008
22,118
Hello,

When you mix 10 Mhz and 10 Mhz + few hz, you will get a few hz and 20 + few hz.
This principe looks like a direct conversion reciever.
A direct conversion reciever uses a double balanced mixer like a SL440 or NE602.
To get the few hz they use a lowpass filter after the mixer (in your case to get rid of the 20 + few hz).

Greetings,
Bertus

#### KL7AJ

Joined Nov 4, 2008
2,229
You want to use the phasing method. You can get the bandpass right down to 0 Hz that way. It takes some careful adjustment, but it can be done.

Another variation is the Weaver modulator. Both are referenced fully on the internet .

eric

#### msseng

Joined Jan 19, 2009
11
thanks a lot
but I had visited that link previously, and there is a question: Is there any mixer can mix a 10 MHz with a few Hz signal?!

#### Wendy

Joined Mar 24, 2008
23,062
How about the old fashioned diode mixer? I have to step out, but it is a common enough amature radio circuit.

#### Wendy

Joined Mar 24, 2008
23,062
thanks a lot
but I had visited that link previously, and there is a question: Is there any mixer can mix a 10 MHz with a few Hz signal?!
Sure, that is basic AM. If I understand you correctly you want to reject either the Upper Side Band or Lower Side Band.

#### bertus

Joined Apr 5, 2008
22,118
Hello,

Did you also scroll down the page ?

There a schematic can be found for SSB.

Here phase canceling is used to suppress the sideband.

Greetings,
Bertus

#### Attachments

• 20.2 KB Views: 97

#### Wendy

Joined Mar 24, 2008
23,062
OK, I studied balanced mixers in school, but this is new stuff for me. How do you get a consistant 45° shift across the band? Answers some questions I used to have though about SSB.

#### bertus

Joined Apr 5, 2008
22,118

#### msseng

Joined Jan 19, 2009
11
dears,

I report my problem again:
i have a 10MHz carrier with a 1-10Hz message.
I need a 10MHz+(1-10)Hz signal, and so know I can use a thing such as SSB Modulator.
But my question is does it exist any SSB Modulator Module or Mixer that can mix 2 mentioned signal?!

please help me if you can, but please don't propose some thing such as AM and etc. if you don't understand my problem completely!

thanks

#### bertus

Joined Apr 5, 2008
22,118
Hello,

The main problem is the low LF frequencies.
You can use the schematic in post 16.
For the phase shifter circuit use 300 times higher capacitors to use it from 1 to 10 hz in stead of 300 to 3000 hz.

Greetings,
Bertus

#### KL7AJ

Joined Nov 4, 2008
2,229
Hello,

Did you also scroll down the page ?

There a schematic can be found for SSB.

Here phase canceling is used to suppress the sideband.

Greetings,
Bertus

HE HE! Eat your heart out, Bertus! I have a Central Electronics 100V...the first full power PHASING sideband rig ever commercially made. A real collector's item.

(I picked mine up at the estate of a Silent Key....I've seen them on Ebay for as much as \$6600! Still not selling mine, though!)

Eric the avaricious.

#### msseng

Joined Jan 19, 2009
11
Hello,

The main problem is the low LF frequencies.
You can use the schematic in post 16.
For the phase shifter circuit use 300 times higher capacitors to use it from 1 to 10 hz in stead of 300 to 3000 hz.

Greetings,
Bertus

Thanks dear!