RELIC HAWK

[scs]

DD coil 27 cm. diameter.R coil=T coil =120 turn 0.3mm.Frequency is not critical -5-12 khz.R coil and T coil must be in resonance by C 5 .Shield-aluminum foil or graphite.C3/R9-fase shift.

[plboy]

You speak for C3 R4 and C4 R5?
What about the coil? if you use standatr dd27 Tx must to be about 48 turns/ 0.50-0,60 for 1mH and Rx must to be 180 turns/ 0.20-0.30 for 16mH. Capacitor of Rx must to be about 68nF what you make. Which type of shield you use

[scs]

This detector is mod in RH-more sensitive in small pieces of metal.This detector work with receiver and transmitter coil in resonance.When the coil is made to seek a minimum receiver Signal no with elements C1,R1,R6 and C7.Fix coils with epoxy resin .With C1,R1-phase and R6 C7-amplitude make the received signal minimal.Is no problem use standard DD coil and standard generator.
Regards

[plboy]

I think showed you enough how to make a good detector. I'm to here.

Ich denke, dass sie genug wie man einen guten Detektor zeigten machen. Ich bin bis hier.Fiele gruse!

This schematic is NOT for beginners!Work perfect with coil with good compensation receiver signal.

Hi to all which make this project,
I make Bulgarian version 'Coinseeker 2' 3-4 times and Musketar which is like Minelab RH....
SCS what means good compensation of receiver signal? What type of coil and preamplifier you use? Are you use standart DD coil and standart generator

[scs]

I think showed you enough how to make a good detector. I'm to here.

Ich denke, dass sie genug wie man einen guten Detektor zeigten machen. Ich bin bis hier.Fiele gruse!

This schematic is NOT for beginners!Work perfect with coil with good compensation receiver signal.

[bernte_one]

hello scs,

have seen that you have boar layout from relic hawk for sprint layout
would you please kindly share the (maybe upload as zip file)

many thanks

[scs]

R54 and C39 connections.

[scs]

I found only two schematic errors -z54 and C39 connections.
Corrected schematic-

[Thomas]

Sacho, Ok?

[scs]

pcb-

[scs]

In reality-

[scs]

Home made.Work.

[Davor]

Please note that this gleaner thing is in effect exactly the same thing as a fine implementation of Costas loop in a metal detector. Actually brilliant solution, even for a IB detector without two box Rx and Tx separation. See the costas loop as a PSK demodulator, it draws only Im signal out at "Demodulated binary output":


To extract real part you'd need a lower branch output as well.

There is an interesting twist to a Costas loop used for PSK/QPSK demodulation that makes it a real 4-quadrant MD Rx candidate that would sort out coil phases by itself, hence being quite happy with differential coils, and it is configured as follows:


See the switch for choosing between PSK and QPSK? As in 2-quadrant/4-quadrant choice. This implementation has limiters, while gleaner has AGC ... same function.

I've noted the benefits of this approach some time ago, see http://www.geotech1.com/forums/showt...883#post144883 and of course there is an implementation already that may be radically rejuvenated. I think some soft limiters would provide much better performance than AGC over much larger signal level span. A simple anti parallel diode limiter in a feedback of an op amp nicely compresses over, say, 6 orders of magnitude.

Otherwise, yeah, nice catch

[mikebg]

Here is the reinvented block diagram of most sensitive metal detector.

[mikebg]

REVISION AND REDESIGN OF RX in Relic Howk

Before this, we should know how to make revision and redesign of RX used in an arbitrary metal detector.

Part 1. THE MOST SUITABLE BLOCK DIAGRAM for RX

Every design of an electronic appliance should start with study and analysis what is the best block diagram.
When an amateur designer starts to revise and redesign the circuit diagram of an arbitrary metal detector, he should know the best block diagram, to find and to repair the differences.

The sensitivity of a metal detector depends on gain of RFA (radio frequency amplifier). In a competent designed RX, the gain of RFA is limited by interference and internal generated noise.

Because of incompetent designed block diagram, the sensitivity of conventional metal detectors is limited by a large AIR&GND signal existing in RX input. Target signal appears as a very small modulation index of existing signal. To avoid saturation of RFA with AIR&GND signal, the incompetent designer should reduce its gain. For example, the gain of RFA used in metal detector C-scope 1220 is only 9 times.

A competent designed metal detector should have ABC (Automatic Balance Control) to increase modulation index of TGT signal by suppressing AIR&GND signal in RFA input. Then the RFA can operate with maximal possible gain maintained by AGC (Automatic Gain Control) . The AGC increases gain until output starts to saturate with noise generated by resistors and semiconductors. I posted in the forum an image how seems the RFA output on an oscope when the ABC eliminates AIR&GND signal in input and when the AGC maintains maximal possible gain (when noise peaks start to saturate RFA output).

(R)EMI group showed the most suitable block diagram of RX for narrow band metal detectors. The project named GLEANER is a reinvention because removes drawbacks of an invention made by Vaino Ronka and described in US patent 3,614,600. Attached is the invented by Ronka block diagram where several blocks are renamed by (R)EMI group according terms used in amateur radio. Here is a short explanation of block diagram and idea for its improvements:

DESCRIPTION for Fig. 5
1. Radio Frequency Amplification
The RFA is formed by ferrite rod antenna 12, opamps 68 , 72 and subtracting amplifiers 100. All received signals are compensated in the input of RFA by an ABC (Automatic Balance Control). This allows the RFA to operate with extremely high gain because there is no AIR&GND signal to saturate RFA . The gain is limited only by input interference and noise generated in resistors and semiconductors. However there is no AGC (Automatic Gain Control) to maintain the maximal possible gain of RFA (when its output starts to saturate).
2. Automatic Balance Control
The ABC uses two sine waves in phase quadrature to compensate all input signals with TX frequency. Both compensating waves are generated by carrier recovering section. For this purpose, an additional ferrite rod antenna 14 receives AIR signal. Opamps 80, 84, 88 and 90 produce also two square waves in phase quadrature. They are used as Re and Im reference voltages for synchronous demodulation. Two P-I controllers use demodulated by 106 and 108 signals to control electronic attenuators (potentiometers) M1 and M2 until outputs of 106 and 108 become zero.
3. Target signals
Outputs 116 (11 and 117 (119) of controllers are used to obtain information for GND signal (because the apparatus is used for prospecting). This not suitable for metal detecting. The TGT signal appears as output of demodulators 106 and 108 (see the explanation below - point 2).

IDEA FOR IMPROVEMENTS: 1). AGC. Without AGC, the operator should adjust manually the maximal possible RFA gain according internal noise and environmental EMI in the region. An AGC can maintain the maximal possible RFA gain (near to saturation of its output).
2). Low noise RFA circuit. Ronka describes other method for signal subtracting in Fig. 4 with text in column 4 row 37. A 10 ohm resistor pos. 33 connects the cold lead of RX coil to ground. This allows to design low noise preamp connecting RX coil as bandpass filter formed with capacitor 32 to inverting input of opamp 30.
3). TGT signal should be taken from outputs of demodulators. Outputs 116 and 117 of controllers are not suitable to obtain TGT signal. This will be a motion type metal detector because P-I controllers maintain zero output of demodulators 106 and 108 and TGT signal exists in outputs of demodulators during setting time of controllers only. To avoid suppression of TGT signal, the controllers should operate slowly (with enough large setting time).
4). A twin loop RX coil is necessary to suppress GND signal and EMI.
5). No need of antenna 14 for receiving AIR signal because it is not the true carrier signal. The carrier wave is AIR&GND signal received by antenna 12 and it is phase shifted by its LC tank circuit. The TGT signal is shifted by LC tank with the same angle and this eliminates the need of phase stability attributable to conventional metal detectors.
6). No need of subtracting amplifiers 100 to eliminate AIR signal because P-I controllers eliminate it in RFA input.
7). No need of limiters 104 because this reduces sensitivity to small and deep targets. The limitation of received signal makes ABC and AGC useless.
Read also US patent 3,500,175.

My next post will contain GLEANER block diagram.