Metal simple and powerful detector
Metal simple and powerful detector
My daughter got into her head to want to use a metal detector to search for ancient artifacts, where I live there are many areas to explore, the possibility of discovery are very low but, with a little luck you can find exhibits ranging from Etruscan times up to World War II. Not wanting to spend on a desire that could last less than a month before, find myself with a metal detector in the house that takes dust I decided to build one I. I appreciate this hobby, because it relaxes and leaves the mind free from everyday problems. Currently, the commercial metal detectors are equipped with microprocessors, This facilitates the processing of information, making them faster and more accurate identification and localization of metals.
But there are several detectors unprocessed who hit among those who want to realize their metal detector, including the Surfmaster PI and this http://blog.novaeletronica.com.br/pi-polones-construa-seu-detector-de-metal/
all of PI technology (Pulse Induced). But even these detectors are considered by many difficult to achieve, This is why I decided to design a version metal easiest detector, but with PI technology.
Although many engineers do not agree, a PI detector can discriminate if well designed, different metals, It depends on the sampling frequency.
Metal Detector Amilcar PI
My metal detector with false modesty that I called Amilcar Pi is easy to build it is already pre-adjusted for maximum efficiency and does not contain components difficult to find, and what not least is also very low cost.
The metal detector is composed of two main units, a transmitter that uses the generator of narrow pulses to be 150us, retrieved from the literature on the network is the most efficient solution and to lower consumption. These pulses amplified by a mosfet IRF740 piloting an emitter coil. The same coil also acts as a receiver and its main component is the LM324, a quadruple operational amplifier.
The starting point of this handset is inspired by a metal detector originally developed in Russia. In my there is a reworking of the same ideas with European components and, while I was there optimize the oscillator part to save mA always useful for battery-powered devices. The result was a very similar to Surfmaster PI detector in its operating principle, but much simpler and “quasi” the same efficiency.
The electronic circuit works so: It starts with a square wave oscillator which rotates on an oscillating operational at a frequency of about 130Hz with on-off ratio very unbalanced while 7,3mS Ton = Toff = 0,15mS.
The output is inverted by Q2 which serves to drive the mosfet, the scelda of R6 and R7 is designed to compensate for the fact that the operational amplifier output never reaches the supply voltage, They are typically 1.5V “eaten” by the operational, the divider serves precisely to ensure that when the output is at one of the transistor Vbe is below the conduction threshold to zero while when it is able to take it into saturation. The resistors R8 and R9 ensure that the mosfet has rising edges and very rapid descent to avoid unnecessary heat loss and give the maximum power to the transmitter coil. R11 together with D3 and D4 act with C4 ago detector for the extra voltage induced when stopping driving the mosfet. It appears obvious that esa will be much higher as much as a metallic element will be found in the magnetic circuit of the coil. R12, R13 and R14 securing the comparator threshold levels for signal detection so as not to have anything in headphones if not in the presence of a metal object. R15, r16, The potentiometer R17 and C3 serve to filter out high frequencies and simultaneously regulate the volume of the headphones. C7, R18 and the last transistor amplifies the signal picked up at an optimum level for headphones.
The consumption at idle settles on 20mA, these expedients guarantee an acceptable duration of 9V battery, But nothing prohibits to use a lithium battery pack 2S or 3S for a full day of uninterrupted research.
The metal detector coil
The construction of the metal detector coil is simple, the coil is composed of 25 turns of wire from 0,5 mm with a diameter of 19 cm.Con this setting the sensitivity to a coin 25 mm is 20 centimeters. Other diameters may be tested for an optimal result, the table below by some indications regarding the number depending on the diameter coils to always have an inductance of approximately 400uH
| coil Diameter | Number of turns | inductance |
| 100 mm | 42 | 402uH |
| 120mm | 36 | 405uH |
| 150 mm | 31 | 394uH |
| 175 mm | 28 | 387uH |
| 190 mm | 25 | 401uH |
| 200 mm | 26 | 406uH |
| 250 mm | 22 | 380uH |
| 300 mm | 20 | 390uH |
| 400 mm | 17 | 396uH |
| 500 mm | 15 | 400uH |
The base dual electronics face is 44mm x 58mm but it is not critical and you can achieve it however you like
I by the time I left everything in the hands of my daughter, at least until it will not be fed up of the new “toy” and shall put on a mobile gathering dust.
Amilcare Greetings
Salve, I'm going to build the circuit. However, there are a couple of things that don't come back to me. In practice, the last opamp is configured as a comparator but has both variable voltage references because in the end they both refer to the output of the previous operational. The latter is instead configured as a shunt and not properly as an amplifier. Else, I think the C5 capacitor must be connected with the negative pin to the pin 5 of the LM324 instead of ground as it should keep the Vref taken from the divider R12-R13 stable.
Lo schema presenta un piccolo errore nel disegno a colori, il negativo di C5 va effettivamente al pin 5 dell’integrato ma, chi lo ha realizzato anche in questa versione lo ha usato senza che ne venisse compromessa in alcun modo la sensibilità. Per tale motivo ho lasciato così lo schema anziché ridisegnarlo nuovamente a colori. in fondo in quel punto l’assorbimento da parte dell’integrato è irrilevante e la capacità per stabilizzare la tensione ha poca importanza. Il resto del circuito è esatto. Mi scuso per non aver risposto subito ma ho avuto qualche serio problema personale ed do tardato un po nel rispondere.
Good morning Amilcare ,
I have a question to ask if I don't disturb.
I own a metal pi-land craft-
It comes with two plates,one of 360mm, which from about 600uH ,and a 1200×1200 which from about 400uH . I don't understand why this difference as metal works well with both equally. I built a 200x180mm one with 413uH inductance ma ,as I balance the cable that goes to connect to the metal, measuring at the end of the cable,the inductance rises to 480…. even here I don't understand why. It is not possible to have an elucidation on this problem? Waiting for an answer I thank everyone in advance thanks
This kind of metal detector is very tolerant with inductances even with displacement of 50% of face value . As for the value that varies in the circuit, I don't know how to give you a certain explanation. It could also be the construction geometry that cures mutual inductance with the connecting cable
Thanks for your answer…so in the end it is just the field test to see which inductance is better from what I understand. I'll try to prepare more wraps ,with the same diameter, varying on the turns of the coils to understand which is the best. But I could run into circuit breaks ? Wonder why I am a welder,mold maker,I restore cars and modify engines but electronics for me is a novelty…except some modifications to the car electronic control units but nothing special. Thank you
I can't tell you many others it worked on the first try, try to check the coil and the various connections of the components.
Hello and thank you.
I have checked everything over and over.
Coil is 200mm diameter 0.5mmm enameled wire with 405uH industance.
Pcd made with Gerber files you provide.
I used irf740 mosfet, 1 bc557 transistor pnp, 1 bc337 npn, lm324 as opamp, 1 22k potensiometer and a beeper. Connections are like the schematic in your post. Is it possible to post photos in order you to check?
I don't know if those Gerber files can be corrected I made many variations. Following the future of the PC where I had placed the files, I have nothing left. Later I found files in a stick but I don't know if they are correct. Given the simplicity of the scheme that is certainly correct, I recommend you to do it again starting from the scheme. Or by following the diagram check that there are no differences with the pcb.
Hello I am trying to make this detector but have problem to make it work.
I made 2 pcbs based on schematic but they won’t work.
Is there any change in schematic?
Please help.
Hi I tried to build in breadbord the circuit according to the diagram but, the only difference being the use of two LM358N instead of’ LM324 may interfere with smooth? Also like transistors for headphones I used a BC547. Best regards.
Ciao!
R10 and R11 are 1 / 4w?
you can use IRF840 instead of IRF740?
it is possible to power the circuit 12V?
It certainly works well at 12V
For the mosfet is not critical and you can use the proposed model.
All resistors are 1/4 in W
Overall a nice job, but the schema publication of printed PCB would make excellent. It's possible? Ciao
As I mentioned in the comments all project data is lost due to premature death hdd on which were stored, for the scheme are the life-size if you click on the diagram is open the image in a new tab.
Missing only the indication of the transistors of the headphones which can be chosen at will just who is at least 100 mA and an operating voltage at least higher than that of the battery which is fed in my case I used a BC338 for the printout can not help you I should get back to do it again from scratch.
Good evening, thanks for sharing this nice circuit. I read that she used a BC328 to boost headphones, being a PNP transistor she connected to the opposite direction with respect to the schema say well? That is, with the emitter at the top and the collector to ground? Then I wanted to ask the inductance of the coil is critical? that is needed to force 400 uH or you can create other values? I'm building this circuit and soon will test it. Thank you.
No the transistor used is a BC338 but evidently having written the answer on the phone the corrector has changed in the BC328 I apologize for the inadvertent oversight but any NPN transistor with a few hundred mA of collector can be used. For the coil can groped other geometries as long as the value of inductance approaches 400MHz