qrp SSB transceiver filter problem

[neazoi]

Hello,
I consider the next simple transceiver circuit.
It includes an SSB crystal filter in the receiver section (x5, x6, x7, x8).

I do not understand what is the purpose of X3 ?
It seems to me like a CW narrow filter, but if so, what is it's purpose bwfore the SSB filter?

 

[vfone]

Ideally the mixer should be followed by a diplexer and by the IF filter next.
In this way provide the right impedance for the mixer output, minimize the mixer spurs, make easier the job of the first IF amplifier minimizing the bandwidth at its input.
Generally good superheterodyne receivers use a wide IF filter just after the mixer, and a narrow IF filter after the first or second stage.
The small amount of increase in system noise figure (because the filter is placed in front of the IF amplifier) is insignificant compared to the other benefits.
The posted schematic tries to implement somehow the ideas above.

 

[neazoi]

Ideally the mixer should be followed by a diplexer and by the IF filter next.
In this way provide the right impedance for the mixer output, minimize the mixer spurs, make easier the job of the first IF amplifier minimizing the bandwidth at its input.
Generally good superheterodyne receivers use a wide IF filter just after the mixer, and a narrow IF filter after the first or second stage.
The small amount of increase in system noise figure (because the filter is placed in front of the IF amplifier) is insignificant compared to the other benefits.
The posted schematic tries to implement somehow the ideas above.

I thought that it could be to minimize unwanted signals to the ssb filter preamplifier, but I am confused about the bandwidth.
A single crystal should have a bandwidth much lower than an ssb crystal ladder. It seems to me that this single crystal is killing the voice signal there.
Am I wrong?

 

[vfone]

Yes, you are right. A single series crystal could have a narrow bandwidth (few hundred of Hertz, and asymmetrical) in case if load impedances are high.
In the posted schematic it is mounted in a one pole ladder filter with about 1k at both ends, making somehow the impedances lower, so wider bandwidth.

 

[neazoi]

Yes, you are right. A single series crystal could have a narrow bandwidth (few hundred of Hertz, and asymmetrical) in case if load impedances are high.
In the posted schematic it is mounted in a one pole ladder filter with about 1k at both ends, making somehow the impedances lower, so wider bandwidth.

So you suggest to leave this crystal there as it is?
What if I use a 10MHz SSB filter and thus a 10 MHz crystal at this point? Do I need to change the values of capacitors in this single pole filter?

 

[vfone]

You can let the crystal there, and see after how behaves the SSB reception with or without it.
I don't think using a single 10 MHz crystal will need values changed.

 

[neazoi]

You can let the crystal there, and see after how behaves the SSB reception with or without it.
I don't think using a single 10 MHz crystal will need values changed.

Another thought is to put out the single pole crystal filter and the fet after it and use just the ssb filter for rejection of unwanted signals. I think the dual gate mosfet is a mixer with gain, so maybe it will be able to compensate for the filter loss.