The Receiver Band Pass Filter Stage
The receiver and transmitter band pass filters are identical insofar as component values are concerned. Some clever illuminati from the EMRFD or uBitx reflector will undoubtedly comment "if they are the same why not just use one BPF" and switch it between functions". That is a possibility but adds to the switching complexity and raises some issues about circuit isolation. Thus the addition of two coils, five trimmer caps and two fixed caps is a viable solution and resolves the switching issues and enables greater circuit isolation. We are always looking to minimize unwanted oscillations.
Now again back to our picky illuminati, the two coils are exactly the same inductance but different cores were used because that is what I had in the bins. There may be some advantage to using the larger cores and that is why they are on the transmit side. Again both are the same inductance! Now there is a subtelty to this design (found accidentally). If you remove the two 150 PF NPO caps and retune the trimmer caps this network will tune 20 Meters! Four trimmer caps are 9 to 50 PF and the section coupling cap is 1-10 PF. So while the networks schematic shows fixed values these are peaked for max smoke. The filter on the left hand side is the Transmit Filter and the one on the right hand side is the receive filter.
Before you move on too quickly, below is essentially the IF Module and represents the heart of the Sudden Transceiver.
Note I used yellow cores as that is what I had available. Yellow cores turns were adjusted to match the inductance of 2.28 uH. This is a good exercise for the builder to learn how to wind toroids! The two network coupling capacitor affects the dip in the middle of the pass band. If you shift the 20M BPF coupling cap from 1 pF to 2 pF there is a dip in the curve. Wes Hayward W7ZOI, in his seminal publication Solid State Design for the Radio Amateur, in the appendix thereis a section covering Band Pass Filters whihc shows you how to construct small value coupling capacitors. Before my using LT Spice, I hand calcualted Band Pass Filter Constants using the method outlined by W7ZOI in SSDRA. Just my opinion but the EMRFD makes a great bookend while the SSDRA is a far more useful publication for those who like to homebrew.
An optional building approach is to use small shielded IF transformers wired back to back with a suitable small value coupling capacitor connecting the two transformers. Formerly Mouser (and others) sold a small 10.7 MHz IF transformer (green core) that was ideal for band pass filters. These transformers were easily padded to 9 MHz or even 7 MHzand a few connections and you were there. For higher frequencies, the internal coupling caps were easily removed and then these could be fitted to higher frequency networks. Here is a photo of such a bank of Band Pass Filters that I contsructed for my KWM4 Transceiver. The only appliaction were I saw some limitations was on 75/80 Meters where there just wasn't enough bandwidth to cover the entire band. For the KWM-4 I designed a BPF using the frmuals found in SSDRA (I keep telling you to toss out that EMRFD).
Regrettably these transformers are no longer stocked by Mouser. That said a friend in the UK put me on to transformer cores that are like the 42IF123 and can be used in this project. Model 2u6L.