David Limmer GQRP 5668


The design is for 80m (3.5 to 4.0 Mhz to allow the use of HF converters), although constructors should be able to modify for 160m without difficulty. 40m may be possible, but should be regarded as the upper limit due to the low IF


The specified filter (Murata FM455 J1) is the most expensive single component, but has a respectable performance for such a compact unit. (2.68Khz@ 6/50db). Other 455Khz units could be tried. An alternative tried was the cheap CFU455 1T (Khz) which was suprisingly good.


T3 was chosen as it offers a centre tapped primary for IC1 and a high impedance secondary for TR2


Uncritical - any of the more common MOSFETS and FETS can be used for TR1,2,3,6 & 7, and similarly any common bipolar for TR4 & 5. Diode types are also uncritical. IC2 was chosen for low noise, but LM324 will work well. IC3 was an LM386N1 and was run from 8 volts, it does not like 12 volts. If the LM386N4 is available, the regulator REG1 can be omitted.


This is a little more complex than is usual in receivers of this type, using a phase splitter and full wave diode bridge, but is worth the effort. Very strong signals will cause overshoot, in which case VR1 should be backed off and some attenuation switched in. Delay time can be modified by changing C33/R40. Use a tantalum capacitor for C33. R36 should be chosen to give 5 volts at TR5 collector. (between 820 ohm and 2K2 approx.)


Runs on the low side of the signal. C24 is 180pf and C25 is 470pf with a 350pf air-spaced variable to cover the full 500Khz. Use good quality polystyrene capacitors and mount L2 vertical.


The VFO was set using a general coverage receiver. With the AGC off, T4 was adjusted to zero beat with VR3 approx at mid travel. T3 is simply peaked for maximum noise (keep the AGC off) and T1/T2 peaked at mid-band.





This article appeared in SPRAT 72 Autumn 1992


Chevin circuit