Quasar is a selective IB metal detector with metal recognition and direct signal processing. The VDI scale in Quasar is divided into 16 columns, with the ability to remove any of the columns from the search (Closing them with a mask) as well as audio multi-tone indication. In the latest firmware versions, the Quasar's operating frequency can be up to 17 kHz and depends on the search coil.
The Quasar metal detector circuit has an average level of complexity (The only component in short supply is the MCP3201, so there is already a metal detector circuitand Quasar AVR where, thanks to replacing the microcontroller, this problem was also eliminated). But the presence of a programmable microcontroller and coil for Quasar, as for any other selective metal detectors, creates some difficulties for radio amateurs. The manufacture of Quasar is carried out by people with experience in the manufacture of metal detectors. In general, the Quasar metal detector has medium difficulty level for making it yourself.
Download the Quasar metal detector diagram in pdf format —
The presence of an accessible screen makes the Quasar a very convenient and easily repeatable metal detector with metal recognition.
Working screen of the metal detectorQUASAR as follows:
Metal detector discrimination scaleQUASAR divided as follows:
The KVASAR metal detector is controlled by 6 buttons:
Firmware for the Quasar metal detector Version 1.4.5 (the latest firmware version for today) –
For flashing the microcontroller of the Quasar metal detector, programming fuses must be arranged as follows:
The developer of the Quasar metal detector gives a brief description of the search coil he made. Coil type DD with outer diameter 230 mm. TX – 40-45 turns with 0.5 mm wire and RX – 200 turns with 0.2 mm wire. The TX winding is connected to a metal detector with serial resonance, the capacitance of the capacitor is 0.3 mF, the resonant frequency is 8.192 kHz, the RX winding is connected to a metal detector with parallel resonance, and is adjusted to a frequency 1.5 - 2 kHz below the TX resonant frequency.
Below is diagram for connecting such a coil to the Quasar metal detector
Description of launching and setting up the KVASAR metal detector with oscillograms -
Description of the menu and settings in the Quasar metal detector -
Conclusion: The QUASAR metal detector has a simple circuit and inexpensive components (microcontroller, screen, etc.), which makes it very attractive for self-production. In operation, Quasar shows quite pleasant characteristics and good results, and can easily compete with entry-level branded metal detectors.
The metal detector project received its continuation in versions and QUASARAVR, Therefore, it’s worth starting to manufacture these particular versions of the metal detector, since the author has stopped releasing firmware updates for QUASAR!
When writing, materials were used from the author’s website - http://fandy.ucoz.org/
All questions about the Quasar metal detector can be asked in the comments to this article. And also write your review, wishes and suggestions for adding to this material.
And so, in the article we’ll talk a little about the Quasar Avr metal detector, look into history, analyze the characteristics and consider all aspects related to its creation.
Quasar AVR metal detector - this device is based on the principle of inductive balance, or, in simple terms, IB. Selective and has VDI, 16 discrimination sectors, each can be closed. All spare parts, according to the author, are available. Direct signal processing.
Quasar technical characteristics:
Overall, this is a fairly complex metal detector to assemble. It’s definitely not suitable for beginners; there’s a lot of hassle with both the board and the coil. But if you are a more or less experienced fighter, then reserve your strength and move forward.
In principle, there is nothing to say about the diagram; you need to look and figure it out yourself, all on your own and with your own hands. Here is a picture of it and below is a link to it.
Now let's talk about how to make a coil for a quasar. For a quasar, a DD sensor is used, which has the following parameters:
The connection diagram itself looks like this:
As you can see, TX is connected according to a circuit with a series resonance, and RX is connected with a parallel one. TX is adjusted to a frequency from 4.5 to 9 kHz, in RX to a lower frequency - from 1.5 to 2 kHz.
Let's talk a little about the settings of this device, what operational settings it has. This means that the Quasar Avr metal detector has the following settings:
We've finished setting up the quasar arm, now let's talk about the firmware.
In principle, there’s not even anything to tell here, download the latest version on the author’s website or from us, and flash it. Here are images of how the fuses should be.
In general, the Quasar metal detector has received very good reviews. One of the most popular homemade devices, in many respects it can be compared with mid- and high-level industrial metal detectors. An excellent device that competes with many devices. Its subsequent versions are further developed. Beginners should think about it, because a quasar is cheaper, even if you order from someone else, than the same ICQ 150 or 250, and its characteristics are many times better. So think, decide. But if you decide to buy, the main thing is to find a good seller and manufacturer, so as not to suffer and swear at the device later. A lot depends on the manufacturer and build quality. Look for reviews.
And finally, a little video of a cop working with a Quasar metal detector.
"Kvazar" is an IB metal detector with direct processing, developed on an affordable element base. A selective mode has been implemented with VDI display as a bar chart (signograph), and the ability to mask each of the 16 sectors. Sound indication - multi-tone. Ground response suppression is vectorial.
Forum with discussion of the device: http://md4u.ru/viewforum.php?f=95
General view of the layout
Device screen
VDI scale in degrees
Purpose of the buttons:
Firmware version 1.4.3
Firmware version 1.4.2
Firmware version 1.1.9 .
Firmware version 1.1.8 .
Firmware version 1.1.7 .
Firmware version 1.1.6 .
Firmware version 1.1.5.
Version 1.1.4 .
Version 1.1.3 .
Version 1.1.2 .
Version 1.1.1 .
Version 1.1.0 .
Version 1.0.8 .
Version 1.0.7 .
fuses settings for PonyProg:
fuses settings for SinaProg:
A DD sensor with the following parameters was used: outer diameter 230 mm, TX - 40-45 turns of 0.5 mm wire, RX - 200 turns of 0.2 mm wire. The TX circuit is connected according to a circuit with a series resonance, the approximate capacitance is 0.3 uF, in the prototype it was tuned to a frequency of 8.192 kHz, in general the device can operate at a frequency of 4.5 - 9 kHz. The RX circuit is connected according to a parallel resonance circuit, and is tuned to a frequency 1.5 - 2 kHz below the TX resonant frequency.
A metal detector is a means of finding car keys lost in the garden or sewer hatches under the leaves during autumn :)
This metal detector is called Quasar (Quasar), it was developed by Andrei Fedorov, but not without the help of members of the md4u.ru forum, who gave advice and reported errors during testing of new versions of the software.
Quasar is a metal detector with direct processing, working on the principle of induction balance. The main advantages of such metal detectors are the ability to detune from the ground, as well as the difference between metals in their resistance and ferromagnetic properties.
This metal detector can determine what metal lies underground, although not with 100% probability, but it can easily determine non-ferrous metals from ferrous ones, and in most cases which non-ferrous metal is located under its coil.
It can notify the owner about metal underground using sounds of different tonality (frequency), and display information on a sixteen-character two-row display in the form of a histogram. It has a bunch of settings, but first things first.
Be careful, there are a bit more pictures below.
In the current implementation we have:
The circuit is not complicated, there are no particularly scarce parts. You can download it
Let's start with the barbell. It remains with a simpler implementation of the "Volksturm sm+geb" metal detector. It was made from PVC pipes with adapters at 45 degrees. Before gluing, this design looked something like this:
After gluing we have a working stick:
The reel seat was made using plastic bolted connections, used in the same plumbing fixtures, which is then attached to the reel using epoxy glue and can be detached from the rod:
We made the armrest from the photodrum of a large A3 format copier :) That is, we attach a little grinder, a drill, to a rod and it turns out that the whole structure held up pretty well.
We wrap the handle with something soft, then close it with a large-diameter heat-shrink tube, warm it up, and you get a comfortable, ergonomic handle :)
We're almost done with the mechanics, we'll paint it later. We will not talk in detail about how the board was made; we will only dwell on the essential points. The Cradex Z5 case with dimensions 103*90*40 fits perfectly under the printed circuit board developed by one of the forum participants for microcircuits in DIP packages. Link to the board at the end of the article.
We buy parts, measure how suitable the board design is, and take electric capacitors from the low-ESR series.
Textolite was etched in ammonium persulfate. Poison quickly and beautifully. Just fill it with warm water, about 80 degrees.
Afterwards the display is soldered and turned on for the first time - testing.
If one line of dark rectangles is visible on the screen after power is applied, the screen is working and this is its self-test mode - when power is applied, but control commands have not yet been received (there has been no initialization).
You won't see some components on the board from the parts side, because... I couldn't find them in the DIP form factor. This is an adjustable zener diode TL431, a pair of filter capacitors and not beautiful wires in the area of the operational amplifier, because We couldn’t find the original one, we took a similar one, but it had a slightly different pinout - we had to be tricky :)
Let's start working with the body. You need to make several holes in it - for the screen, control buttons, coil connector and power connector. The case must also be insulated from moisture - otherwise the device may begin to malfunction or fail. For the convenience of cutting a hole for the screen, we took a screen with the same functionality, only with a blue filter, since our green one was already soldered to the board with a permanent connection.
It stood up perfectly, but :) When we tried to try it on for our screen, disappointment knew no bounds :) They turned out to have different sizes. I had to finish it.
In the end, everything worked out. Tried it on, connected it, it works :)
The upper front panel was recessed flush with the plastic so that it did not protrude, because Then it was all planned to be covered with film and a sticker. The screen itself was secured with a large amount of hot-melt adhesive. This type of connection has two advantages: water will not get inside and there are no bolted connections, which would then still have to be sealed.
They poured it with a regular heat gun, and where it didn’t warm up well, they helped with a hairdryer from a soldering station. At this moment, the screen itself may change color to bluish or some other color due to heating; the main thing here is not to overdo it. After cooling, the color returns to normal and everything works normally.
We made the board for the buttons ourselves, because... there was no suitable ready-made one for this building. There will be a file at the end of the article. The diodes in it are smd.
And so, all the holes are made, the button board, speaker, power connectors and coil connections are also sealed with hot-melt adhesive.
Regarding the design, we thought for a long time about what color to choose. We chose the black option.
The technology is simple. We print the picture and cut out a hole for the screen. They cut with a scalpel. Next, we glue the film under the screen of the design, then take a transparent, matte, self-adhesive film and glue the resulting pie onto the plastic, cut out the excess film and you’re done!
The block was attached to the rod using a piece of thick plexiglass, cut into strips and bent under the influence of local heating, screwed with one side to the box, the other to the “pipe holders” or whatever this crap is called...
By the way, later the two outer fastenings were removed, that is, the whole thing held up perfectly even on two fastenings.
So, after carrying out all these operations, we painted the bar and this is what came out:
Separately, it remains to talk about the coil. We can say that this is the most sensitive element and it must be assembled so that when searching for and touching all kinds of grass and other objects, it does not “microphone” and reacts only to the phase change caused by the metal under the sensor. We immediately wanted to make the coil as it should be, we wound the coils... By the way, all the wires were taken from an old CRT monitor. Its demagnetization loop fit perfectly under the transmitting TX coil, a thinner wire was found in another coil, the wire to the metal detector unit was taken from its non-detachable VGA cable, in general there were enough wires from there :)
After two coils have been wound, one of them (receiving, RX) must be wound into a screen made of foil or graphite. If it is foil, then it is necessary to make sure that there is no short-circuited turn from this screen; if it is graphite, then it is necessary that the resistance from the center to the edges of the coil is approximately 1 kOhm.
After selecting a resonant capacitor (the device, of course, adjusts itself, but we selected the frequency closer to 9 kHz), it was time to fill these coils in a mold with epoxy resin. And then a dispute broke out with the box and the Internet. The box says to dilute in a ratio of 1:5. One in five, damn it! Considering that we already had some experience working with epoxy, where the ratio of 10-12:100 was mentioned everywhere, some misunderstanding arose. But they decided to do as written, the manufacturer won’t write garbage on the box :) And they didn’t even decide to test it with a small volume of this resin. I want to go to the cops as soon as possible! In short, they started pouring it, then they changed their minds, because the proportions of resin and hardener were just right for 10-12:100, and then they forgot how much of what they had already poured... In general, they ruined the solution, but they tried to fill it in :)
And it didn’t even think about freezing. What to do? We pulled the coils out of the mold, cleaned them of all the resin, and another idea came to mind. After all, our CRT monitor is a kind of cornucopia for building a metal detector :) The stand from it was also useful. We take it, remove everything unnecessary, attach the coils, fill in epoxy in normal proportions, drill holes - ready!
All this showed its efficiency already in the first mine on the Sozh River:
As for the power supply of the metal detector - at the moment it comes from a regular 12 V lead battery, which you carry with you in your briefcase, but there is little buzz from this method. There are immediate plans to build a power supply on one 18650 element (about 2Ah at 3.7 V), make an indication of the charge level, charge from USB and a 3.7-7 converter, because It is from this voltage that the metal detector is powered. It would be possible to go up to 5 Volts, bypassing the stabilizer for the controller and ADC, but it is better to swing the coil at a higher voltage, then the sensitivity will be higher, but more on that in another article. It consumes about 100 mA at 7 V, so from one 18650 battery you can count on approximately 10 hours of operation. And the main thing is that this thing will be much lighter than a lead battery, which will allow it to be mounted together with the block on a rod.
The promised boards in lay format for the Quasar metal detector, as in this article.
All the best!
One day I had a chance to start assembling a Quasar. I’ll skip assembling the block, because... There’s nothing complicated there, I’ll stick with the reel. According to the Author's recommendations, a DD sensor was used with the following parameters: outer diameter 230 mm, in the transmitting coil TX - 40-45 turns of 0.5 mm wire, RX - 200 turns of 0.2 mm wire. The TX circuit is connected according to a circuit with a series resonance, the approximate capacitance is 0.3 uF. It was tuned to a frequency of 8.5 kHz, but in general the device can operate at a frequency of 4.5 - 9 kHz. The RX circuit is connected according to a parallel resonance circuit, and is tuned to a frequency 1.5 - 2 kHz below the TX resonant frequency.
Subtleties:
We wind the second coil in the opposite direction relative to the first (the current of the DD coil should flow in one direction), since we wind the coil on the same template, the second coil will turn out to be inverted and the current will flow in the opposite direction.
To obtain a minimum voltage in the future (after filling with epoxy, the coil will be slightly unbalanced), when winding the last turns of the transmitter coil, wind them with a margin so that you get a loop at the bottom of the “tuning turns” coils, and do not fill them. Now that everything is filled with resin, move this loop in different directions until you achieve the minimum voltage on the receiver coil, then you can finish pouring.
The coils were wound on mandrels and soaked in “88” glue. Coil stiffness
after drying, it is such that they can be bent to obtain the desired configuration.
The terminal of the receiving circuit that is closest to Cx (i.e., the internal terminal) should sit on ground. This nuance affects the quality of the balance.
If the detuning of the ground and ferrite is in the minus, re-solder the ends in places in the PX and everything will go to +.
When adjusting from the ground, the value should not be more than +80, roughly speaking it should be from 0 to +35, the adjustment from the ground affects the sensitivity. The less interference, the weaker the intuition.
If with foil, then shield only the RX (lower receiving) coil. TX - may not be shielded. If you shield both, then only through the insulation between them. If you cover
graphite, then completely without any gaps.
The sealed lead-in must be located in the TX field, the coil leads must not pass through Cx (the center is the intersection of the coils)
The auto balance button must be pressed with the sensor raised, and then lowered and raised.
For 5 kopecks of the USSR in black soil with a 25cm DD coil there should be a sensitivity of at least 25cm with clear discrimination. VDI can jump at 30cm, about 1-2 sectors with a figure of 3 segments (but there should be no gaps between segments, such gaps will exist if the target is made of black metal) and there should be a high sound signal on both sides of the swing, so you need to choose what to trust more: VDI or audio signal. In Quasar everything depends on the coil parameters...
