Double superhet, filters considerations.
Especially the design marked as "Fixed frequency first oscillator" at the middle of the page, because of it's simplicity.
The way I see it working is this, and please correct me if I am wrong:
The first filter (rf frontend) must be a LPF. For example 30MHz.
This will reject all frequencies above 30MHz (image rejection)
The second filter must be a BPF centered at the first IF frequency that would allow each amateur band to pass through (eg 200KHz bandwidth)
The crystal LO must be switched for each band. for example if RF=10MHz, and first IF is 40MHz, then the LO crystal must be 30MHz.
Do I see this right?
But why is the second IF amplifier and filter needed? Can't it be directly converted to AF?
This is a tuneable IF set. You have a fixed down convertor followed by a tuneable IF, which is actually another receiver with a tuneable down convertor and second fixed IF.
If you do not use the second narrow band IF and filter, then you will demodulate all the signals in the pass band of the first IF at the same time.
The second tuneable receiver is often 5 -> 5.5 MHZ, so to receive a signal on 3.5001 MHZ, you mix the incoming signal with 9.0 MHZ and the 9- 3.5001 sideband then is selected by the second receiver ( 5.4999 MHZ). if you did not have the second narrow band IF, all the frequencies in the range 4 - 3.5 MHZ would be received.
The linearity of the whole first receiver up to the narrow band second IF filter must be good, else a strong signal up to .5 MHZ away can block your weak wanted signal.
Frank
I think that the second tuneable receiver shall be tuned to the high IF range, not 5-5.5MHz, say for example somewhere above 30MHz for a 1-30MHz receiver. Is that true?
Else what would be the meaning of the high first IF?
For a continously tuned HF receiver, the first IF must be above the reception ranges, i.e. for 1-30 MHZ, first If could be 38 MHZ. But for an amateur band only receiver, the IF needs not to be on an amateur band, hence 5 MHZ. In about 1968, G2 DAF (?) designed and published such a receiver in the RSGB hand book. Also local oscillator stability for the reception of SSB was (is?) more problematic with a continously tuned oscillator at Fsig + 38 MHZ, bit easier now with PLL oscillators.
Frank
Ah, you mean like this? http://serwis.avt.pl/manuals/AVT2960.pdf
This used an 8.6MHz If but It can cut the image since there is an LPF at the rf input and the band of coverage is 3.5MHz
Am I right?
Many amateur band only transceivers use an IF of around 9Mhz because very effective and low cost narrow band crystal filters are available, or can be readily home made.
The only disadvantage is that it can be difficult to dodge some of the spurious mixer products and oscillator harmonics, and you can get "birdies" appearing at some frequencies on particular amateur bands, which can be a real nuisance if your choice of frequencies turns out to be a bit unfortunate.
Another disadvantage of doing it this way, is that some bands may of necessity have to tune backwards on your frequency tuning dial.
The more HF amateur bands you decide to include in your rig, the more difficult all this becomes.
If you want to build a full coverage HF receiver (or transceiver) with something like thirty 1Mhz bands, and have all of them tune in the same direction, AND be free of birdies and spurious products, the only way is to convert straight up to an IF much higher than 30 Mhz, then convert back down through one or more lower IF frequencies, which may be either tunable or fixed.
This also lends itself to using a frequency synthesiser for the first mixer that can tune through the entire 30 Mhz wide span.
I am in the process of building something exactly like this myself right now.