Modernization of my HF transceiver: making VFO to work on lower frequencies
What is the problem?
The communication transceiver, that i have described on my transceiver page, has one disadvantage: on "higher" bands (10 and 15m) the required VFO frequency becomes rather high, and absolute stability decreases dramatically when common components are used. This is felt very well especially on 10 meters, where VFO generates frequencies beyond 20 MHz. The higher frequency is being generated, the lower capacities and inductances have to be used, the higher is influence of many external factors over the VFO. So, the problem can be formulated by following way: to make VFO working on lower frequencies on 10 (and on 15 if possible) meters band. Here i give two ways to solve the problem.
First, but not the best variant to solve this problem.
If you retain the main schematics of the transceiver, there is only one way to decrease required frequency of VFO on 10 meters band - to increase the first IF. But the second IF at 500 KHz does not allow to do it - if you increase the first IF, say up to 10 MHz, the mirror selectivity at second IF becomes too poor. The solution is:to introduce the THIRD IF, and the third bi-directional frequency converter. The good value that i used, is 15.5 MHz. In this case, VFO frequency at 10 m band becomes shifted down from the original value for 10 MHz and becomes 12.500 MHz at 28 MHz reception, which is quite acceptable from all viewpoints. Additional frequency converter uses 10 MHz LO signal to convert 15.5 MHz to 5.5 MHz, which is first IF in original unit. This LO frequency is obtained from a stable quartz-oscillator. When divided by factor of 2, this signal (5 MHz) can be used for the next mixer. This does not add any additional oscillator to the unit, which is quite good from a viewpoint of internal interference.
The implementation of this idea is rather simple. you make the bi-directional frequency converter on two JFETs according to the schematics of transceiver's mixers. Between it and the former first mixer you set the BPF tuned to 15.5 MHz. This converter works only on 10 meters and on others bands is bypassed. You also replace the original 5 MHz oscillator to 10 MHz and add a 2:1 divider with subsequent emitter repeater. The signals of LO's are appropriately directed on all bands by additional relays. The block schematics is here:
So, we have double frequency conversion transceiver on 160-15 m, and triple frequency conversion transceiver on 10 meters. The construction i implemented here does not cause signifcal gain change by introducing additional conversion stage, and the rig works as good as before on 10 meters, but with much improved frequency stability on this band.
To make the VFO more stable at 10 m the following should be done:
First of all, the corresponding inductor and two capacitors in VFO, which are responsible for the 10 m band, should be changed to the values providing frequency coverage 12.500 - 13.500 MHz.
Then, quartz resonator in 5 MHz/500 KHz oscillator should be changed to 10 MHz. The feedback capacitors should be changed to 68 pF. The small inductor (~5-15 uH) should be used in series with a crystal to set the frequency exactly to 10 MHz. The former 5 MHz output becomes now 10 MHz output.
10 MHz signal should be connected through half-sinewave former (on diode and 1k resistor) to the 'C' input of a flip-flop clock SN7474, which works in counting mode (inverted output and 'D' input are connected to each other). The IC is fed like counter in original block - trough resistor from a main source (about 510 Ohm). The output of clock is directly connected to the input of original counter, as well as to conventional emitter-repeater with LC-low pass filter. After all changes are done, this block supplies 10 MHz, 5 MHz and 500 KHz signals.
The main one. You make additional bi-directional frequency converter with the design similar to the original two ones. As a BPF for 15.5 MHz serve three resonant circuits with inductive and capacitive connection. Inductors are wound on 6 mm polystyrene core with a ferrite slug and have 75 pF capacitors in parallel, and 6.8 pF capacitors for the connection between circuits. The physical distance between coils - about 25-27 mm. Additional relays should also be introduced, which provide bypassing of the signal on 160-15m, as well as proper LO's signal switching (see scheme above).
After all is done and BPF's are tuned, the rig on 160-15m should work without any changes, and on 10 meters the same gain as before should be provided. But the frequency must be much stable.
This variant gives slightly higher stability due to lower VFO frequency, but it suffers from following disadvantages: receiveing becomes too 'dirty' due to many points of heterodyne interference, because the digital divider 10 -> 5 MHz, used in LO gives a lot of harmonics at the output. Also, in TX mode on 10 meters, as it appeared, parasite emission exists on 31 MHz. This is the second harmonic of last intermediate frequency, which is not suppressed adequately by the BPFs, because it is too close to the passband. I could not decrease its intensity to appropriate limits.
My second solution, which i made several weeks ago, is free from these disadvantages.
Another variant of modification.
The problem of choosing additional IF for 10 m band is very tricky. First of all, it is very good not to use any additional oscillators, except those, which already exist in the transceiver. The more oscillators you have, the more internal interference you get. Also, the problem of outband emissions should be regarded. It should be noted, that mixers used in this construction are not-balanced. It means, that output frequency of any mixer should be far away from input frequencies (both heterodyne and signal) as well as from their lower harmonics.
I have chosen the following frequency allocation scheme: 5.5 MHz (first IF in original receiver) is mixed with existing 5 MHz oscillator, giving 10.5 MHz additional IF. Then this IF is mixed with VFO signal, which lies in range 17.5-18.5 MHz, giving target frequency (28-29 MHz). In such scheme 2-nd harmonics of signals are far outside the passband of the filters, (for the last mixer, they are 35-37 MHz and 21 MHz). There is still hazard, that 2-nd harmonic of LO (10 MHz) will pass trough the additional BPF without necessary suppression, but as it appeared, it does not, when filter is made properly (do not try to increase gain by making the connection between resonant circuits in filter too strong - otherwise, you will lose in selectivity).
In everything else, the second solution is similar to the first one. But you do not have to change the resonator in 5 MHz/500 KHz oscillator and do not have to make additional divider. The VFO frequency is higher, that in 1-st variant, but generally the present design is easier to make and do not suffer from outband emissions on 10 meter band (which tends to get open more frequently, by the way!).
I hope, that explanations about the modification are clear here, and no additional coments are requeired.
73!, Valentin Gvozdev , RU3AEP.