OpenBCM V1.06 (Linux)

Packet Radio Mailbox

ED1ZAC

[C.D.G. OpenBCM]

 Login: GUEST





  
G8MNY  > TECH     12.08.18 08:30l 131 Lines 5666 Bytes #6 (0) @ WW
BID : 33040_GB7CIP
Read: GUEST
Subj: 4 Aerial Doppler RDF
Path: ED1ZAC<ED1ZAC<GB7CIP
Sent: 180812/0629Z @:GB7CIP.#32.GBR.EURO #:33040 [Caterham Surrey GBR]
From: G8MNY@GB7CIP.#32.GBR.EURO
To  : TECH@WW

By G8MNY                            (Updated Jan 15)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)

THE DOPPLER RDF PRINCIPLE
By moving a Rx aerial towards the Tx (rotating it around a center), the Rx
frequency increases, & when going away it decreases. So on a FM Rx a tone will
be heard at the frequency who's phase compared to the aerial rotation position
is the direction.

 + Freq          ->
                                           Tx
No change    v ROTATNG ^                    *

 - Freq          <-       plan view

To mimic a spinning aerial a least 4 identical aerials are needed. And each of
these needs to be faded in & out in pseudo 1/2 sinewave manner. This assumes
the 4 aerials don't interact!

EARLY KIT RDF
I have a 1981 commercial USA kit by Doppler Systems Arizona, model 3003, it
uses a crystal timebase to clock 4 phases of an identical waveform stored in
Proms.
       .-.
       Ţ1Ů               They give the 4 phases of
      Ů   Ţ              smooth gain changing bias
.-~-./  0° \.-~-,        for gate 2 of the 4 dual
.         .-.            gate mosfet preamps mimicing
 Ţ        Ţ2Ů            a spinning aerial rotating
 Ů       Ů   Ţ           300 times per second.
  \.-~-./+90° \.-~-.
 .-.         .-.
 Ţ Ů         Ţ3Ů
Ů   Ţ       Ů   Ţ
     \.-~-./+180°\.-~-.
    .-.         .-.
    Ţ Ů         Ţ4Ů
   Ů   Ţ       Ů   Ţ
-./     \.-~-./+270°\

SCHEMATIC 

Ant 1  Ant 2
   │     │
 ┌─┴─────┴─┐ 6 way    ┌─────────┬────────┬──────────┬───────────┬─────────────┐
 │Four Gain├==========┤Waveform │ Crystal│  Phase   │Calibration│Heading Deg &│
 │Modulated│          │Generator│Timebase│Comparitor│   Delay   │Polar Display│
 │ Preamps ├=======┐  └─────────┴────────┴────┬─────┴───────────┴──────┬──────┘
 └─┬─────┬─┘ coax  │                          │300Hz                   │Kill
   │     │        ┌┴──────┐            ┌──────┴─────┐ No 300Hz Squelch │Display
Ant 4  Ant 3      │Scanner├────────────┤  AF Signal ├──────────────────┘
                  │  Rx   │ 300Hz + AF │Conditioning├─Overload LED
                  └───────┘            └────────────┘
AERIALS
These must be identical with same length feeder etc. I used 4 magmounts with
short 1m leads to BNC plugs. Into these I can plug in aerials, telescopic ones
give 100-432MHz, for lower frequencies "coat hanger" wire can be cut & put into
PL259 plugs. It is important they are wired up in order correctly!

 -------------      -------------      -------------
|    '    '   |    |     .  .    |    |             |
|      []     |    |      []     |    |     [] ::   |
|    .    .   |    |     '  '    |    |             |
 -------------      -------------      -------------      ˙ = Magmount
 50MHz spacing          144MHz             432MHz        [] = Preamp

The aerial spacing needs to be about 1/4 wave for the doppler tone to develope,
the distance & the shape of the vehicle roof all affect the bearing
calibration!

RX SETUP
I use the simular sized AOR2002 Scanner for the Rx, modified with a Pre Vol AF
output (see below). The aerial arrangement with external preamp is optional,
but less cables. Mine preamp came unmounted outside the unit, so I put it in a
diecast with magnetic underside & 4 BNCs, but I also modified it with 4 step up
broad band RF transformers for more gain & terminated them on the PCB with 470R
instead of the 47Rs & added 2 clipping diodes as protection in case I Tx! With
the preamp output coil removed, it now gives useful gain 50-432MHz.

AF MODIFICATIONS
The original circuit had very little pre signal conditioning, only a few dBs
down at 200Hz & 1kHz, so any modulation kills the system. So I desigend a very
narrow wayne filter on a couple of the existing initial opamps in circuit...

             ┌───────100K─┐         dB │
    10n      │    │\      │          0 ┤             .˙.
In──┤├─100K─┬┴────┤- \    │            │             Ţ Ů
            │     │    >┬─┴─┬─      -10┤            Ů   Ţ  
        u1 ===  ┌─┤+ /  │   o          │            |   |
            │   │ │/    │ Test      -20┤           /     \
            ├───)────┤├─┘ Point        │          /       \
           220  │     u1            -30┤        .'         '. 
            │   │                      │      .'             '.   
          100R  │                   -40┤ _ .˙'                 '˙. _     
         Preset │                      └────┬────┬────┬────┬────┬────┬──
  ──────────┴───┴────                      100  200  300  400  500  600Hz

The exact centre frequency is set on the multiturn preset by setting identical
losses at 290 & 310Hz, on the 1st op amp's test point & then the 2nd. The
overload detector pick of point is moved to an earlier stage & made it more
sensitive, so it operated before clipping occured.

DISPLAY
Polar display uses 16 LEDs in a circle for ease of navigating & 1 in the middle
to indicate power. As well as that there is a 0-360° bearing indication.
┌───────────┐
│ 360 .˙˙˙. │
│    :  .  :│
│ ! .'˙...˙'│
└───────────┘
An overload LED indicates if there is too much AF drive.

IN PRACTICE
Anoyingly my scanner gives different 300Hz delays (phases?) between wideband &
narrow band FM modes, so the barring changes if the bandwidth is changed. But
I found on narrow band FM mode, the heavily clipped wideband FM AF, RDF 300Hz
signal is fine. So I added a dedicated unsquelched narrow band FM AF O/P for
RFD feed, & I can still listen to the AM & WBFM modes.

The main problem is RF multipath, if any of the 4 aerials sees an upset RF
field (nulling) the result is usless.



Why Don't U send an interesting bul?

73 de John, G8MNY @ GB7CIP


Read previous mail | Read next mail


 16.08.2018 15:38:45lGo Back Go up