Metal detector for medium-large objects

[eduardo1979]

Hi Eduardo1979,

What is actually the operating pps of the GS5, GS5lite and GS6?

Hi Geomax. The GS5 Lite's frequency is adjustable from 2.6 to 3 kHz. I set mine to 2.8 kHz and the TX pulse width to 100 µs.

I assume the GS5 has the same frequency, or very similar. The GS6 has the same frequency as the Surf Master Pi, around 615 Hz.

[GeoMax]

Hi Eduardo1979,

thanks for the reply.

[David222]

Hello friends. I have a strange problem. I made the GS5 Eduardo and chose its coil of 2 ohms and 412 uH. I chose a frequency between 1300 and 1500 Hz. Because more than this the 7912 regulator heats up. In this range its performance is better than all pi devices. I used 20 meters of enameled wire with a diameter of 0.5mm on a 30 cm core. With these specifications, it detects a coin with a diameter of 2.5 cm in the air at 35 cm. So far, no problem. But I made a coil with the same specifications of 2 ohms and 412 uH; but this time I increased its diameter to 40 cm. But this time, surprisingly, it did not detect the coin at a greater depth. I even increased the dimensions of the metal and used a large can of Lotfett brand soldering oil, which is 6.5 cm in diameter; but it still detected it at 55 cm, like a 30 cm loop. The depth should have increased as the loop got bigger, but this didn't happen. I don't understand why.

I'm planning to use a larger loop for more depth for the GS5 Eduardo. But I don't know if that's possible. But the sound of the GS5 Eduardo is unique. I love the beautiful sound it makes when it reveals the metal. The sound is smooth and high quality and very beautiful.

[gurdal]

Hello David222
This might be a classic question, but...
Does the GS5 discrimination work well?
Best regards,

Gurdal

[eduardo1979]

Hello friends. I have a strange problem. I made the GS5 Eduardo and chose its coil of 2 ohms and 412 uH. I chose a frequency between 1300 and 1500 Hz. Because more than this the 7912 regulator heats up. In this range its performance is better than all pi devices. I used 20 meters of enameled wire with a diameter of 0.5mm on a 30 cm core. With these specifications, it detects a coin with a diameter of 2.5 cm in the air at 35 cm. So far, no problem. But I made a coil with the same specifications of 2 ohms and 412 uH; but this time I increased its diameter to 40 cm. But this time, surprisingly, it did not detect the coin at a greater depth. I even increased the dimensions of the metal and used a large can of Lotfett brand soldering oil, which is 6.5 cm in diameter; but it still detected it at 55 cm, like a 30 cm loop. The depth should have increased as the loop got bigger, but this didn't happen. I don't understand why.

I'm planning to use a larger loop for more depth for the GS5 Eduardo. But I don't know if that's possible. But the sound of the GS5 Eduardo is unique. I love the beautiful sound it makes when it reveals the metal. The sound is smooth and high quality and very beautiful.

Hi David. With the new 40cm diameter coil, you'll probably need to change the damping resistor value, as this can sometimes happen. The NE5534 offset setting, the coaxial cable you're using, and the main sample delay setting can also influence the results. There are many factors to consider. Theoretically, with a larger coil, you should detect larger objects at greater depths than with a smaller diameter coil. The coaxial cable's capacitance is very important, as is the coil's. Try making a coil with Category 5 UTP cable (the kind used for internet networks), as it typically has lower capacitance than standard enameled copper cable. You should also try replacing R7 (2.2kΩ) with a 1kΩ resistor. This would give you a bit more depth and reduce thermal noise, but it might overload the preamplifier. If that happens, you should replace the 1M resistor R8 with a 470K resistor. This will maintain virtually the same gain, but reduce thermal noise and increase the preamp's speed.

[David222]

Hi David. With the new 40cm diameter coil, you'll probably need to change the damping resistor value, as this can sometimes happen. The NE5534 offset setting, the coaxial cable you're using, and the main sample delay setting can also influence the results. There are many factors to consider. Theoretically, with a larger coil, you should detect larger objects at greater depths than with a smaller diameter coil. The coaxial cable's capacitance is very important, as is the coil's. Try making a coil with Category 5 UTP cable (the kind used for internet networks), as it typically has lower capacitance than standard enameled copper cable. You should also try replacing R7 (2.2kΩ) with a 1kΩ resistor. This would give you a bit more depth and reduce thermal noise, but it might overload the preamplifier. If that happens, you should replace the 1M resistor R8 with a 470K resistor. This will maintain virtually the same gain, but reduce thermal noise and increase the preamp's speed.

Eduardo, I thank you. I will do this test as soon as possible.

[David222]

Hello David222
This might be a classic question, but...
Does the GS5 discrimination work well?
Best regards,

Gurdal

Hello. Yes, its differentiation works by changing the tone of voice.

[David222]

Hi David. With the new 40cm diameter coil, you'll probably need to change the damping resistor value, as this can sometimes happen. The NE5534 offset setting, the coaxial cable you're using, and the main sample delay setting can also influence the results. There are many factors to consider. Theoretically, with a larger coil, you should detect larger objects at greater depths than with a smaller diameter coil. The coaxial cable's capacitance is very important, as is the coil's. Try making a coil with Category 5 UTP cable (the kind used for internet networks), as it typically has lower capacitance than standard enameled copper cable. You should also try replacing R7 (2.2kΩ) with a 1kΩ resistor. This would give you a bit more depth and reduce thermal noise, but it might overload the preamplifier. If that happens, you should replace the 1M resistor R8 with a 470K resistor. This will maintain virtually the same gain, but reduce thermal noise and increase the preamp's speed.

Hi Eduardo. I followed your instructions and instead of the 2.2k resistor, I used two 2.2k resistors in parallel, which have good power and a value of 1.1k. You were right, it detects two cents more and the 7912 regulator runs cooler. I left the device on for 5 hours continuously and set the battery warning trimmer to 14.5 volts. As a result, the circuit was able to run for 5 hours continuously with 4 18650 and 2200 mAh batteries. After that, the battery warning light came on. And it took two hours for the batteries to fully recharge. Thank you. Do you know of another circuit that has the same quality but detects more depth?

[algan]

There are many versions of the GS-6 flying around the internet, but this is the correct one.

Hello, eduardo1979, do you have a video clip of how GS-6 works? Thank you.
-

[eduardo1979]

Hello, eduardo1979, do you have a video clip of how GS-6 works? Thank you.
-

Hi Algan.
I don't have any videos of the GS-6, but I found one on YouTube. Here's the link.
https://www.youtube.com/watch?v=RNjrlXEo_D0

[algan]

Hello, Eduardo, do you think this variant of yours with a ground balance of Surf-PI will be better than the original or if Surf-PI is made bipolar, will it be more efficient? Thank you.

[eduardo1979]

Hi Algan. A bipolar detector has its advantages over a unipolar detector, and its disadvantages as well. A bipolar detector naturally cancels out extraneous frequencies (EFEs), whereas a unipolar detector is somewhat more sensitive to low-conductivity targets, thanks to the low capacitance generated by its transmitter circuit, which only uses one MOSFET. More MOSFETs equal more parasitic capacitance, which equals less sensitivity to low-conductivity targets. There's always a compromise; perfection doesn't exist. There's a lot to discuss on this topic. A bipolar detector handles higher frequencies more easily than a unipolar detector without the MOSFET temperature rising too much. I've been working on a unipolar detector for quite some time, and after a lot of effort and a lot of money spent on very high-quality components, I've managed to detect a 0.09-gram gold nugget at a depth of 5 cm in highly mineralized soil without any problems with EFE or hot rocks. The coil I'm using is an 8-inch mono coil. Every detector has its advantages and disadvantages. It all depends on what you're looking for. My focus is on small gold nuggets.

[David222]

Hi Algan. A bipolar detector has its advantages over a unipolar detector, and its disadvantages as well. A bipolar detector naturally cancels out extraneous frequencies (EFEs), whereas a unipolar detector is somewhat more sensitive to low-conductivity targets, thanks to the low capacitance generated by its transmitter circuit, which only uses one MOSFET. More MOSFETs equal more parasitic capacitance, which equals less sensitivity to low-conductivity targets. There's always a compromise; perfection doesn't exist. There's a lot to discuss on this topic. A bipolar detector handles higher frequencies more easily than a unipolar detector without the MOSFET temperature rising too much. I've been working on a unipolar detector for quite some time, and after a lot of effort and a lot of money spent on very high-quality components, I've managed to detect a 0.09-gram gold nugget at a depth of 5 cm in highly mineralized soil without any problems with EFE or hot rocks. The coil I'm using is an 8-inch mono coil. Every detector has its advantages and disadvantages. It all depends on what you're looking for. My focus is on small gold nuggets.

Hi Eduardo. I have a question about GS5lite. How do you set the frequency to 2/3 khz!? If I do this, the 7912 regulator gets hot. But when I set it between 1/2 khz and 1/5 khz, the circuit works continuously for 5 hours without any problems. With a precise adjustment, I was able to detect a 3 cm coin at 40 cm in the air. This is the best setting. But when I raise and lower the coil on the ground, I have an extra beep. And when I activate the ground balance to 50%, the sound of the device increases and decreases. And it is not stable. I built the circuit exactly according to the instructions you gave. Without the slightest change. But I really can't solve the problem. I measure the frequency from the output of pin 3 of the 555 IC. Do you have a solution for this problem? Thanks.

[eduardo1979]

Hi Eduardo. I have a question about GS5lite. How do you set the frequency to 2/3 khz!? If I do this, the 7912 regulator gets hot. But when I set it between 1/2 khz and 1/5 khz, the circuit works continuously for 5 hours without any problems. With a precise adjustment, I was able to detect a 3 cm coin at 40 cm in the air. This is the best setting. But when I raise and lower the coil on the ground, I have an extra beep. And when I activate the ground balance to 50%, the sound of the device increases and decreases. And it is not stable. I built the circuit exactly according to the instructions you gave. Without the slightest change. But I really can't solve the problem. I measure the frequency from the output of pin 3 of the 555 IC. Do you have a solution for this problem? Thanks.

Hi David.
To increase the frequency without overheating the 7912 regulator, you should decrease the TX pulse width to 80µs or less. From what I can see, I think your search is more focused on coins than gold nuggets. For finding coins and relics, a frequency of 1 to 1.5 kHz is fine, and you could perhaps slightly increase the TX pulse width. If the 7912 gets hot, you should place a heatsink on it to help cool it down, but don't worry, it won't burn out easily, as it has internal protection if it reaches 150 degrees Celsius. Regarding the ground balance issue, there could be two factors. The first is that, by changing the frequency, the ground rejection point may have also changed. Therefore, you should add a 100 kΩ resistor in series with the 250 kΩ potentiometer to compensate for this change. Point number two could be the following: If, after making the changes in point one, the ground balance doesn't improve, it's very likely you have problems with the coil shielding or simply EFE issues. The EFE problem is solved by very precisely matching the resistors and capacitors of the two differential integrators, as well as their input resistors. The resistors should match within 0.1% and the capacitors within 1%. The low-pass filters at the integrator outputs should also match in pairs. If you want to fine-tune things even further, the J113 JFETs and the 470K resistors should also match.

[David222]

Hi David.
To increase the frequency without overheating the 7912 regulator, you should decrease the TX pulse width to 80µs or less. From what I can see, I think your search is more focused on coins than gold nuggets. For finding coins and relics, a frequency of 1 to 1.5 kHz is fine, and you could perhaps slightly increase the TX pulse width. If the 7912 gets hot, you should place a heatsink on it to help cool it down, but don't worry, it won't burn out easily, as it has internal protection if it reaches 150 degrees Celsius. Regarding the ground balance issue, there could be two factors. The first is that, by changing the frequency, the ground rejection point may have also changed. Therefore, you should add a 100 kΩ resistor in series with the 250 kΩ potentiometer to compensate for this change. Point number two could be the following: If, after making the changes in point one, the ground balance doesn't improve, it's very likely you have problems with the coil shielding or simply EFE issues. The EFE problem is solved by very precisely matching the resistors and capacitors of the two differential integrators, as well as their input resistors. The resistors should match within 0.1% and the capacitors within 1%. The low-pass filters at the integrator outputs should also match in pairs. If you want to fine-tune things even further, the J113 JFETs and the 470K resistors should also match.

thnak you as always