pa0nhc ARDFRx2m 0141013-4 /
20170623 / 20171112
Make it yourself easy. Order a easy to solder professional PCB.
version 2m ARDF receiver is a combination of the best properties of my 80m
ARDFrx and the former version 2m ARDFrx. The purpose for redesigning it was : a for you easier to
finish project, using less components and readily wound coils.
Some components which could to be exchanged for adjustment purposes, now have a bit bigger holes.
Anyone who can solder, should be able to finish this project successfully.
Using the high quality factory made PCBs :
- you don't have to drill 250 0.8mm holes
- all components only have to be soldered on the bottom side
- have solder masks, with less chance of hidden short circuits
- top and bottom silks make placing of over 100 components easier
- shows a nice professional finish of your project
In this design purposely NO scarce
radio ICs are used. It is a single super
The three SMDs are especially selected for their availability and optimal properties for this critical application. All other components are wired or pinned for easy assembling. The coils are already wound. Together with the highQ 10.7MHz IF coils, the cheap 10.7MHz IF crystal filter gives very good selectivity. A cheap and light weight screening housing can be made from a piece of square 45 mm ALU pipe. See "Suggestions for housing".
The PCB is fixed in it with two M3 screws. A 9V battery makes 6 hours continues operation possible. The antenna input is 50 Ohms coaxial with a SMA bus for easy disconnection and service.
|Sensitivity ::||-120dBm (S4, 0.22uV / m=80%) is clearly readable.|
|RF blocking level :||-35dBm (equal to S9+60dB ! ).|
|Gain adjustment :||Over 90dB range. -35dBm
(blocking niveau of IC1) can still be
Max. and min. gain can be set.
|RF selectivity :||12 kHz -3dB, better than +/- 250 kHz -80dB.|
|Audio selectivity :||300Hz to 1.5 kHz - 3dB, slopes
Output max. 4Vpp. The max loudness produced by your headphone MUST be adjusted for max. 85dBspl to prevent ear damage.
|Battery :||9V 6LR61. Max. 17 mA for least 12 hours continues use.|
|Battery indication :||The LED battery indicator darkens when the tuning will become unstable due to low battery voltage.|
|Total mass :||500 grams.|
An effective and supple gain adjustment
is realized by varying the G2 voltages of Fets 1 and 4. The needed negative and
positive voltages (-0.67Vdc to abt. +0.63 Vdc)
are generated by the voltage drops at D1 and D2.
IC1 needs no gain adjustment. This enhances oscillator stability and VHF large signal capacity.
REM: as the full battery current runs through D2, this diode will be damaged in case of an accidental short circuit from B+ to a PCB ground surface.
The good sensitivity is accomplished by an optimal match and low insertion loss of the antenna input circuit.
A complicated optimal selective RF input 145 MHz filtering (like a 3-stage helix filter) is unpractical here. It needs more space, is heavier, costs more and has more insertion loss.
Mirror reception of air traffic around 124.1 MHz is less annoying than
constant reception of very strong pager transmitters around 165.9 MHz.
A free running 133.5 MHz oscillator is more stable than an nearly equal 156.2 MHz oscillator.
So i decided for "Under mixing", with a local oscillator frequency below the receiving frequency, around 133.5MHz.
The impedance transformation ratio
of the antenna circuit should be 50 : 1500. The voltage transformation should
therefore be 1 : 5.5.
As a consequence, a link or tap coupling to L1 could not be used. An 145 MHz antenna circuit with good match to 50 Ohms, was not achievable with a ring core coil.
In stead, an exactly dimension able
capacitive voltage divider is used, resulting in always perfect match.
If tuning to resonance should be done with a trimmer capacitor, the transformation ratio changes during tuning.
L1 is therefore tuned for resonance by its core.
The single coil L1 has low losses.
R1 bleeds static's, preventing damage to C1 and IC1.
The double sided, precisely designed PCB, results in no de-sensibilisation, nor ghost signal reception from both low power oscillator circuits. Thanks to the optimal separation between circuit blocks. Shown in the length-wise PCB layout. This also makes a cheap home made tube like housing possible.
De-coupling capacitor C3 (100pF, SRF 100 - 180 MHz) and choke L15 (1uH SRF 180MHz) are effective for VHF signals. The 22nF (SRF 8-17MHz) capacitors and 15uH (SRF 40MHz) chokes are optimal for filtering 10.7 MHz IF signals. But only if their connecting wires and traces are of minimal length.
Stabilizer Vr1 acts as a very good circuits separator. Audio IC2 is powered directly from the battery. All other circuits are powered from a stable +5V. Resulting in a very stable tuning (as for a free running 135 MHz oscillator without temperature compensation) and stable gain adjustment.
The LED is a simple but effective battery condition indicator. It darkens when the battery voltage runs down to 6V, warning that the tuning and gain adjustment can become unstable.
High Q IF coils L3 and L6 are not damped by the MOSfets, as these have very high input and output impedances. Resulting in a maximal possible amplification factor and selectivity. In fact, with the now available 5170 coils, which have a Q of 130 (!), the total gain became so high, that maximal possible gain is not needed. It can be restricted by enlarging the value of R25 to abt. 15 kOhms.
wideband selectivity from the crystal filter + IF coils is very good, and IC1 only starts blocking at a very high input level of S9+60dB.
The values of Rx and Ry depend on the type of the used crystal filter. Fb2 and Fb5 in the FETs G2 circuits discourage parasitic VHF/UHF oscillations.
In practice, a distant and very weak fox hunt transmitter can be received without problems, even standing in the vicinity if a strong finish fox transmitter.
The coupling track between F1 and F2 is designed relative wide on purpose. It acts as a coupling capacitor to ground surface for the filter. Depending on the filter crystals used, an extra capacitor (0.5-1.5pF), connected between the F1 to F2 connection, and "ground", could be placed to optimize the crystal filter band pass curve and/or the far away selectivity.
Direction finding receivers should NOT have automatic gain
control. But this can result in a danger :
BE AWARE: a suddenly active strong transmitter could be ear damaging loud when listening to a distant weak transmitter on the same frequency.
Even the "plop" while switching the receiver on-off can be far to loud.
You MUST reduce the
maximal loudness from YOUR used headphone
to 85dBspl, by adapting the resistive value of R20.
The value of R20 depends on the sensitivity properties of the individually used headphone.
The AM detector circuit around Fet5 is of the "Infinite Impedance" type. Its positive properties are : simple circuit, low distortion and good performance at very low input levels, high impedances and low current consumption. It is set to class B by means of R22. Fet5 should be a low IDss Jfet.
The 10.7 MHz BFO oscillator T1 injects a little signal for demodulation of CW and SSB signals.
SPECIAL : this injected BFO signal acts as a bias for FET5, vastly enhancing its sensitivity and lowering its distortion.
If you only want to detect AM signals, then order a 10.670 MHz crystal for Xt. It is still needed for enhancing the sensitivity of the Jfet detector, but will not cause whistling due to the fact that its frequency falls outside the total IF and audio pass band.
As even with very low input voltages the signal distortion in the detector circuit is very low, the following audio amplifier could be made very sensitive. A third IF amplifier could therefore be omitted, lowering power consumption. And reducing the number of components, and needed PCB space.
Audio amplifier IC2 has hiZ FET inputs, 5V capabilities and rail-to-rail output. Its two stage gain is over 60dB (!) still having negative feedback. Its surrounding components reduce the audio bandwidth from 150Hz to 2kHz with both 18dB/oct filter slopes. Enhancing the heard signal-to-noise-ratio.
REM: C32 MUST be a film capacitor. Ceramic capacitors show microphonic effects in this sensitive location.
Important detail :
1. The antenna is connected to the receiver by 50 Ohms coax. Sometimes there will be a mismatch between antenna and coax (due to nearby objects). If an electrically half wave long coax is used, it will behave neutral to mismatch, and will not worsen the mismatch.
The total mechanical coax length
of a half wave long piece RG169 Teflon coax must be : 300
/ f / 2 x v = 145 / 300 / 2 x 0,7 = 72,4 cm.
The total mechanical coax length of a half wave long piece RG174 coax must be : 300 / f / 2 x v = 145 / 300 / 2 x 0,66 = 68,3 cm.
REM: this is the total mechanical coax length between the receiver PCB and antenna dipole, including coax plugs and windings through ferrite cores.
2. As the a transmitter also generates RF currents onto the outside of this coax line, AND onto the headphone cable, both can influence the directional pattern of the antenna as experienced by the listener. These RF currents should be blocked.
At the dipole of the antenna, and near the coax connector at the outside of the receiver box, and inside the receiver box at the headphone connector, these lines have to be wound through a ferrite cores. They function as mantle current chokes, and help having an optimal directivity pattern.