YO! I'm still waiting to HEART a comment! in the meantime, ALUMAGNETIC FOREVER! Paramagnetic, diamagnetic or antimagnetic properties of non-ferromagnetic... sorry, ALUMAGNETIC metal are very week magnetic properties observable in DC fields, which I don't care about. They are overshadowed by MUCH greater Eddy Current magnetic forces in AC. ALUMAGNETIC focuses on great AC forces of whatever that is not ferromagnetic!
05:16: The core splits itself apart because the flux lines pass through each laminated layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other
Mehdi, you did a fantastic job, covering 2 types of metal detectors (energy theft and BFO) in just 20 minutes, with no loss of fingers or eyesight. Well done, and thanks for the shout-out. As for the cross-coupled transmitter, I've used that in a metal detector design (White's TRX) and did not use any bias resistors on the transistors. You should not need them. On the transformer plates, they push apart because the eddy current generated in each plate (across the thin cross-section) interacts with the coil's magnetic field to produce a force which is the cross-product of the two, and is perpendicular to both. This means that at the surfaces of the plates, the forces oppose each other and try to push the plates apart. Hard to explain in words, easy to illustrate. Any decent EM book explains it. A dangerously impressive example of this force can be found in coin shrinkers, where a sufficiently strong EM field creates enough radial force to literally shrink a coin. If you were to do a video on coin shrinkers, I would not dare to wager that you have 10 fingers at the end. Edit: I couldn't remember the name of the force so I had to look it up: Lorentz force, where F = J x B (cross product).
i believe what you're describing should be the Laplace force. the Lorentz force for a continuous charge distribution, which i assume you mean from the use of *J* , should also include a term ρ *E* , where ρ is the charge density of the volume in question.
@@matthewbertrand4139 The general equation for the Lorentz force is f = ρE + JxB, but E=0 inside the conducting plates, so we're left with the eddy currents interacting with the B-field. I think the term "Laplace force" is synonymous with Lorentz force for cases of current flowing through a conductor, like we have with the plates.
@@mmmk1414 No, not even close. Mehdi's circuit is continuous wave BFO, something that was popular in the 1960s and 70s. The GPZ uses a form of pulse induction called constant-current PI.
9:44 you can use a cheap half bridge gate driver (such as IRS2153D) with the bootstrap pin tied to ground. i've used it in this arrangement for induction heating. this way you get quick transitions & dead time without wasting power. great video!
For 4:56, due to the core having multiple individual pieces, each producing its own magnetic field in the same direction, it makes them repel each other. Maybe that's why core pieces are tightly stuck together to avoid vibration due to repelling.
@@KnightsWithoutATable it doesnt matter once you break apart lets say a permanent magnet, it wont stick together ever again because it will form its own magnetic field around itself
WOW! This video was incredible. It started out with your usual style, but going more into it, you demonstrated how educated you are and became on the topic and in the field of electronics. I love how you were able to walk through every step in the process of designing this circuit: Research, testing, refining, etc. To make something resembling what you see in stores all around the world. And still made it digestible enough for me and a lot of other people understand. You show all the complex stuff, yet explain it simply, catering to both audiences. I absolutely love your content Mehdi! ❤️🇮🇷❤️🇪🇬❤️
I love all of your videos, but for some reason this one really stood out. Watching you develop different circuits, showing them and then improving on them taught me a lot on the development process. Thanks!
Mehdi, this was one of your best videos in a long time!!!! Is so amazing to see the electronic circuit forming before our eyes and see all the logic behind it and the reason for each component!!!! Do more content like this one
I love how you've shown the process here. A complicated electrical circuit is, fundamentally, a bunch of sub-circuits stuck together. It sort of reminds me of a project I did in undergrad, and I solved each step in the signal chain by reading The Art of Electronics to get the signal processing right!
I love how when I was a kid being fascinated by his knowledge in electronics and thinking he's a man of magic and now that I'm studying my bachelor's in electronics understanding everything he says and recognising circuits. When I look back I see I've grown a lot and will also be a magic man to future generations
@@Stoic_Persistence his profile picture is just him(or someone else) wearing a backpack. I really don't want to assume you mean his skin color because that would mean you were either joking or stupid.
05:16 Imagine your transformer or inductor as a stack of really thin sheets, like a deck of cards. These sheets are there to guide the invisible magnetic lines that make everything work. Now, these lines don't like to hang out too close together, and that's where the magic happens. There's a small gap between each sheet, and when the magnetic lines pass through, it's like a bunch of tiny magnets trying to push each other away. Think of it like having a bunch of magnets on a table, all facing the same way. You push them close, and instead of sticking together, they go "Nope, I don't want to be near you!" That's what happens between these layers in your device. This pushing away is a smart move by designers. It helps to avoid wasting energy and makes your device work more efficiently. It's like turning a potential mess into a well-choreographed dance, where each sheet knows its role, keeping things smooth and efficient. And that, my friend, is the secret behind why your device splits itself apart but still does its job beautifully! 🚀
4:57 the magnetic flux is concentrated in the iron core. The separate plates are aligned with the field axis, and so adjacent plates will always have the same polarity and repel. Like lining up a set of bar magnets next to each other in the same orientation
I had the same idea, but I didn't offer to comment because I couldn't think of the right terms, I'm not a master at this kind of stuff but I like learning about it
In general I did well in school, and in school I found physics to be quite easy. Chemistry took some effort to follow. Electricity just seems like a totally foreign language comprised of multiple languages at the same time lol. I always disregarded it as something I could live without. Cut to me in my 30s now really invested in computer tech and watching LTT videos and broadening my interest in tech and I find myself needing to understand electricity more and more. Not to mention it seems to be a useful thing to know when it comes to other things like cars and homes. So I'm glad you're as entertaining as you are. I've decided I'm just going to watch a few videos of yours every day and hope I just start picking things up over time lol. Because otherwise I have no real idea where to start that won't either bore me to death or be overwhelming... tbf this one got close to overwhelming in the last 3rd as you were talking about converting all of those signals in order to make the data presentable to the human ear lol.
This came at the best time, lol I've been struggling to build a metal detector for a land mine detecting robot, its honestly looking like ground penetrating radar would be easier to build 16:17 Wow, that's actually impressive its clicking at that distance with such a small coil. 17:35 Yeah, that's one of the problems we had, and if you're actually using it in the real world, you gotta deal with changes in soil (i.e. mineral deposits) which can throw off the coil's dynamics. What I found works good is using a microcontroller. You can use the PWM for a fairly stable signal. That's how I'm running mine, just a simple on/off circuit powered by an IGBT used to run a car ignition coil, then we just probe a leg of the coil with an analog pin. I'm still using low voltage, though. 28v, you might want to throw a capacitor in line and add a clamping diode after it or something The cool thing with using a microcontroller is that you can do the filtering with the controller and have it run a decay function that slowly increases or decreases coil sensitivity to account for changes in the environment. The big problem we're having is trying to figure out what is a piece of metal and what is just environmental noise, also the fact a lot of landmines have very little metal in them at all. I do think we are going to go with GPR though, it's just better able to detect objects underground, but before we go that far I want to try one more time with the pulse inductance and maybe a VLF
HI I had the same problem on a metal detector project a few years ago : the sensing coil had to fit in the palm of a hand. The electrical circuit was very similar to this one (pair of oscillators, listening to the beat frequency) and it was very sensitive to human body proximity, including to the hand ! We got around by going towards a completely different circuit, the pulsed induction you mentionned. The time it takes for the voltage spike (following a sudden current cut in the coil) to decline indicates the presence of a metal, and it's much more selective than the beat frequency method, i.e. very insensitive to surrounding moisture, and even to small deformations of the coil
GPR doesn't work well on wet ground etc. A metal detector is your better option. A good rule of thumb is that a metal detector can only see into the ground about the same as the diameter of the sensing head. This is why a lot of stuff used for UXO (UneXploded Ordinance) is so big. Land mines of the antipersonnel type are about the size of a can of cat food and have almost no metal in them. They use trained pouch rats to sniff them out. The critter is light enough not to set them off.
@kensmith5694 GPR should work just fine in water logged soils. It measures the difference in the refractive index of materials, and we don't need to sense very deeply, just a foot or so. The main issue is a lot of the landmines (e.g., the butterfly mines) are just sitting on top of the ground, so we'd have to deal with soil reflection. Another idea we're toying with is to just give our robot a hammer or spike and tell him good luck. Realistically, we can make a dumb robot for under $100 to just go trigger mines. It'd still be cheaper than what it costs now per mine, but at a certain point, you're better off just driving an RC car through the minefield
Now this is what I call quality content! Really love the in depth design process. It gave me the "why use this and not that" and the "if that is unavailable, this can also work" insight, which I find very rarely explained on other channels. THANK YOU, MEHDI!!!
I honestly wish Mehdi was one of my college professors for EE. Even though I graduated with a BSc in EE I still struggle with circuit design and control systems. I learned more from Mehdi videos than the entire last 2 years in college :)
I'm just starting college, planning on going for EE. I'm finding Mehdi's videos fascinating even though I only half understand it. Any tips for a new EE student?
@@galacticviper4453 In one word: RUN! I went to school studying EE while having a full time job and a wife with 2 kids. The EE program ASSUMES you're totally OK with not having a life whatsoever and that it is "Reasonable" to work 14-hour days studying, assignments, etc. And no I am NOT exaggerating. Furthermore, the first 2 years of EE is a walk in the park. The third year gets a bit challenging but still manageable with some grit. The 4th year though is totally insane. The instructors switch into the "we're not here to educate you but we're here to grade you and try our best to make you fail. YOU have to prove yourself worthy to breathe the same Oxygen in the room we occupy. Oh, and did we mention that while you were reading this sentence there is a 10% penalty on your assignment because you turned it in 2 minutes late?". By the way, I am in no means exaggerating. It is THAT bad! Now what's the good news? The good news is that if you go through it and make it out and graduate you have "proven" yourself to be an engineer and therefore can qualify for entry level positions that start somewhere in the 65K to 100K range. Not too impressive when compared to some entry-level IT jobs, especially in data mining and AI that can have a STARTING range upwards of 125K. You do the math. And btw, I really do mean i t. IT is a PHENOMENAL amount of work. Legalized slavery that is, to finish an EE program! :)
This feels like the classic electroboom type of content which I love. It's chaotic enough, it's got simple educational parts for us non-EEs, it's got some damn massive schematics that give me flashbacks for those that understand them and there's a complete working project at the end, love it
Yeah, I got the part about the need to multiply frequences to make the change when metal is brought closer noticeable but the rest needed some EE studying I guess XD.
Usually, long term use is just beefing up the mechanical connections, adding weather proofing (if required) and generally making the circuit hard to kill or to be injured by. He could probably pack the circuit into a project box, make sure the solder joints are top notch, and have an actual legit metal detector.
Fascinating, i never knew how those clicks and tones were generated, being a computer guy i just figured it was just a digital tone generator mapped to some signal and not the actual signal (if that makes sense)
Yeah, I thought exactly the same - the clicks were generated by a microprocessor when it gets a matching, programmed signal. Apparently it's more simple and complicated at the time (for me at least).
I wouldn't be surprised if thats how modern or more expensive detectors work, but the old WW2 ones were probably very similar to what Medhi has built here.
Modern tech, being at such a high level interface, obfuscates what the computer's actually doing behind the scenes, which is the reason I love looking back at analog circuitry and the computers of the past where you can see a direct correlation between all components.
Okay, I thought I understood how metal detectors worked - to the point of being confident when I sell them at work. But now... Man, you just opened up a lot to me. There are people who just want to know how the thing works, and there are people who want to "KNOW" how it works. Now I can explain so much better.
Now this is a very cool and informative video. The Core is repelling because of laminations between the slits of core material. two slits have the same direction of eddy current causing the magnetic field to be generated to repel each other. Its the same as one piece of mage is solid but when it breaks the two pieces repel each other.
That's somewhat similar to what I proposed, but i don't think eddy current has anything to do with it as the core is split up like that specifically to avoid eddy current, aka the core doesn't act like a short-circuited coil. But i do believe it's similar to the broken magnet thing.
Electroboom 101 is entering circuit logic already (he has videos on relays, and relay logic) and I wouldn't doubt further episodes in that series will eventually dive into such topics as analog math, digital math, signal processing and others. It just seems to take Mehdi ages to produce those videos. Probably because of all the extra research, fact-checking, error corrections, and editing he needs to do.
I love the practical explanation of the circuit and how it generated sound along with the oscilloscope display. I know there was product placement involved but it was also very educational and instructive. Thanks for the video Mehdi!
@ElectroBOOM my deepest gratitude to you, bro. I've been trying to get my 9yo son to watch more educational videos but nothing got his attention until I show him your 'electric guitar' video. I was surprised coz not only he watched the whole video, he even watched a few more of your videos! Though he was mostly laughing his ass off at your 'accidents', he did ask a few technical questions & I hope his scientific curiosity will grow. Thank you again & keep up what you're doing - the way you're doing it! 👍👏
I like your sense of humor, your integrity in owning when you make a mistake and your selflessness in promoting other UA-cam pages that also get it right. It shows you will always be about the facts
Hi Mehdi. The 24 amp maximum current of the IRFZ24N Seems massive especially if you're planning on a PP9 type, battery powered portable metal detector. I'm therefore taking my clue from this as I'm unfamiliar with the MOSFET and Schottky type of astable multivibrator you've used in this example. So I'm hoping for a Mehdi induced heartfelt like with my idea of what you're hoping for. I Do go on to mention how I learned because the learning curve was such a massive moment especially after being at college and coming no where. I hopefully explain why I'd be employing such an experimentalist approach and, if I'm wrong I was wrong but not for the want of having an estimated guess. Anyway. A two transistor BJT astable multivibrator is possibly the answer. TIP41 or TIP42 transistors should supply enough current through the 100uh + the series of the 115uh search coils. I hindsight the resistor supplying the emmiter follower regulator seems to verge around the 1 amp current region so I'm more inclined to believe this. Ditch the Schottky diodes and I'd replace them with an experimental value of around 220nf then the 150 ohms might be used in series with the search coils (in case of over load) and in their original place I'd insert something like a 2.2k ohm resistor to be used for the Resistor in the CR timing network. OR leave the 150 ohm resistors where they are to act as a voltage limit stop resistors and cut the connection between the 100uf switch spike surprising capacitors positive terminal to the 150 ohm resistors shared common connection and insert a 10k pot to change the voltage and therefore speed the 220nf capacitors charge up times. Therefore giving experimental frequency change. I'd choose this experimental method because I'm not to good with theory especially with frequencies. I have been a self taught electronics enthusiast since the age of 13 finally teaching myself digital fundamentals whilst I designed and constructed a Radio Controlled up / down step encoder into a Resistor divider DAC utilizing a toy car Remote control and receiver chip. 1 "Nudge Turn" of the remote would notch 1 up / down control on the encoder and therefore Digitally control the amount of resistors in the resistor adder and give Analogue control for the timing function of the mains cycle for the 240 volt ac bulb to dim or brighten up. Obviously timed to coincide with the correct part of the phase during the 50hz cycle. Cheers and TTFN. ;)
The peak current rating of the MOSFET will not impact the battery drain. However the voltage dropped across the FET when saturated is less than the BJT, so you end up with more voltage across the coil and therefore more current through the coil, so for the same supply the MOSFET will yield a better field. The MOSFETs can handle more current for the same package type (TO220) in this case, because it does not get as hot because of less voltage drop across the transistor as previously mentioned.
@@NCmountainview Absolutely understood but I mentioned the BJT astable as I'm more familiar with its turn on / off characteristics having made many LED flip flops I'm also more familiar with the 0.7 volt saturation voltages of the BJTs being a lower sensitivity than that of the FETs which MAY cause weird oscillator results than I could positively say I knew about as I've never built a variable frequency MOSFET astable. I'm not saying the FETs would use more current, this would be limited by the one amp rating of the main regulators feed resistor. I also realise the exceptional sensitivity of the gate junction of a MOSFET and with that realization I can also theorise the BJTs tendency to give a rounder top to their waveform especially when the transistor base goes below its saturation voltage, this does cause more heat but one amp is within the rating for the type and possibly achievable without adding a heatsink. (If not bung a pair of BU508s in there.) LOL. The sinusoidal wave shape at the waves peak must give a better all round smoother electro magnetic RF flow.?
@@Stuck_Farmer The BJT does not have a saturation voltage of 0.7V, as that is the ideal junction voltage and usually associated with the base emitter region in forward bias. The saturation voltage is measured across the collector emitter and varies depending on the load, but it is generally larger than the channel voltage for a FET under the same load. There is some merit to your point, as FETs can have issues with linear response because of the transconductance profile. However, the reality is that you can design stable oscillators with either. In fact the higher input impedance of the FET can make the oscillator much more stable when under load.
Having built metal detectors with my dad as a kid, the only thing I would change is do a flat coil like an antenna instead of a regular coil. But that is more difficult to wind up. Another thing you can do to make it extremely sensitive is to have two coils, one above the other running half off phase signals and sum them in the end to essentially make an EM interferometer.
@@locinolacolino1302 Instead of measuring the changes of one coil's frequency and peak, you're measuring the difference between two. After amplifying, I could detect 0.01% difference between the two. But it's super finicky, the coils have to be perfectly aligned, the frequencies have to be on spot. It's a fun experiment, but probably not worth it for practical use. Those professional detectors that have the coils shaped like an 8 use two of them already, but they work by being a coupled oscillator. You can detect coins 30cm away with those easily.
@@locinolacolino1302 As they hinted at this design is based on interferometry. When there is no metal present the two 180 degree out of phase signals destructively interfere to produce a constant potential. In essence your output signal is the result of recombining those two waves, when nothing is interfering with the field they will destructively interfere and cancel out, when there is however the two coils being in different physical locations will cause them to experience a different change in frequency thus recombining the two will produce an interference pattern that can be detected.
@@chinnapank No, it has been 20 years. But if you look up for two phase shift oscillators leading to an amplifier, it is pretty much the same thing. Just need to adapt for the coils and voltages you will be using.
Mehdi, you did a fantastic job, covering 2 types of metal detectors (energy theft and BFO) in just 20 minutes, with no loss of fingers or eyesight. Well done, and thanks for the shout-out. As for the cross-coupled transmitter, I've used that in a metal detector design (White's TRX) and did not use any bias resistors on the transistors. You should not need them. On the transformer plates, they push apart because the eddy current generated in each plate (across the thin cross-section) interacts with the coil's magnetic field to produce a force which is the cross-product of the two, and is perpendicular to both. This means that at the surfaces of the plates, the forces oppose each other and try to push the plates apart. Hard to explain
You were going crazy with your engineering skills on this one! It's so cool to hear the technical stuff 😊 thanks for another great and highly entertaining video!
It's actually crazy how much this kind of field of work relates to music/sound design with basic waveforms. Even the terms used are the same. Math really does connect everything.
5:15 like you said in the eddy current video "opposing fields repel each other" and in that case for each sheet of iron of the magnetic core are magnetically polarised about the same direction as the other sheets since all sheets changes through them the same magnetic flux and with all sheets having the same magnetic poles and the strong changing magnetic flux of the primary transformer, that's why the sheets repel each other hard and that's the one of the reasons why manufacturers glue the iron sheets altogether to make sure they won't strongly repel each other and another key benefit is to isolate each sheet from the other thus the total resistance of the iron sheet is higher hence reduced eddy currents and less magnetic losses in the iron core
I absolutely love watching your videos, it is fun, educational and makes me realize how happy I am now that I made it into my final year of studying electrical engineering and only have 1 semester left.
Nice! I just finished my PhD in EE, and although the work I do is mostly theory and simulation, it was love of circuit building and tinkering with electronics that pushed me into the program initially. EE is a huge field with a lot of really cool opportunities!
The joy I gain watching you demonstrate how not to do things is the gift that keeps on giving. Seriously, I worked as an electrician for a decade and it never fails to make me laugh when shit pops. Thank you 😀
There are numerous solutions for the gate bias resistors: 1. At low frequency, don't worry about it. Scope the gate waveform: if the rising edge is fast enough, you're done. At 25kHz, probably some kohms is acceptable. 2. Add a buffer. I've done this before, add a complementary emitter follower (just a simple class C one will do, no biasing components) between the pull-up / diode connection, and the gate. Use a pull-down resistor at the gate, to bias the follower (this is necessary for startup, because this is a linear amplifier at its heart -- it won't start up at the right frequency, or at all, if there is too much gain or distortion in the feedback path!). I've done this before, on, let me get it out here -- I had 2 x FDP33N25 for the inverter, and 2N3904/6s for the buffers. Oh, I didn't use a pull-up resistor with the diode, I used a PNP current source, fancy; well, about 470Ω pull-up would be equivalent. It ran over 600kHz with good gate waveforms as I recall, so it doesn't take much as you can see. That suggests 27k would be enough for IRFZ24N down at 25kHz! Call it 10k for good measure. Maybe 1k or less without the buffer circuit. 3. Use different devices in a related circuit. The original Baxandall* circuit, I believe used a feedback winding, with bipolar transistors. This separates bias and feedback from the output voltage, giving more freedom to design the gain and drive strength. Example circuit in following comment. *The common "ZVS" oscillator configuration is actually due to Baxandall, the same one of audio tone control fame; published back in the 50s or 60s when transistors were new, though I don't have the exact citation handy unfortunately. It's commonly called "Royer" by the amateur/HV community, but this is erroneous -- Royer refers to a non-resonant (specifically, saturable core commutated) oscillator. (Occasionally even professionals make this error: for example, Jim Williams (of Linear Technology) in AN49; the correct citation is even given (Royer, et al; in the title, "in saturable core circuits" -- but these are resonant, non-saturating circuits!). So, go figure, even the best of us make mistakes.)
This is the best diy tutorial video you've made. Just loved how the analog electronics working without any microcontroller rubbish! Such a beauty, Joy forever!
Agreed that it is fun to see an analog only design but an MCU could've allowed for fine grained identification of metals just by switching between frequencies automatically and by not being annoying to hear. This is actually super impressive because you don't need more than a single MCU to do that with the way he designed it
@@amogusenjoyer yes, you're right. I'm a microcontroller person and if I had to design a metal detector like this, I would've used a microcontroller. My analog electronics skill is - well, questionable. This video sparked my interest in analog electronics again.
Amazing. Its really motivating mehdi put together this very complex circuits and concepts and show that its not that complicated. Makes me feel like i can too build and learn these awesome things.
1. Replace Bipolar Transistor with MOSFET: Use a MOSFET (e.g., IRF54ON) for the switching element to improve efficiency and power handling. 2. Add a Gate Driver: For optimal control of the MOSFET, consider using a gate driver IC (e.g., TC4420) to ensure fast switching times and reliable operation. 3. Implement Current Limiting: Incorporate a current-limiting circuit to protect the components and prevent excessive currents during faults. 4. Add Protection Elements: Include a fuse and surge suppressor to safeguard the circuit from damage.
Electricity is an absolute mystery to me. You take so much time in your content to explain, and probably from your viewpoint, simplify. I'm so happy that you have accrued such a massive following. You deserve it because you are a great educator. I have accepted that my limitations will never allow me to fully appreciate your content. Every video leaves me baffled, though. Love this channel for the learning and the laughs, but I don't envisage a time that I will ever understand what is going on. I like clever people, but I'm frustrated that I get left behind.
I think it's mostly a matter of studying the subject. It's very difficult to understand any of this without studying at least some basic electronics. It's like playing an instrument, you gotta practice in order to play songs.
to understand these videos you really just need to understand the basic concepts of electricity and magnetism which is easy. There are channels on youtube that'll help like, ' Physics videos by Kugene Khutoryansky '. Do not limit yourself.
You most definitely can understand this, there's just some prerequisites. Don't gauge your abilities based on a youtube video. This video assumes prior knowledge, he didn't explain every detail. If you're not familiar with circuit theory, this would be like trying to understand differential equations before learning arithmetic. If you take it one step at a time it's very easy. Knowledge is free online, you can get all the books and papers they teach in college for free from beginner to postgraduate level. I believe in you, believe in yourself.
I was studying a metal detector to modify mine last night. Good thing you posted it and really helpful. thank you professor boom. i'd like to see its final form for next video 😅
Hey Mehdi, for your question at 5:10, is the answer that the pieces of the core become magnetised in such a way that the poles are facing the same way causing them to repel each other but they are held together by the coil in the centre.
though I had some problem with sea water It also ua-cam.com/users/postUgkxa-FNYUOM93a388gi9a4brtSCEVmrHgJH land for finding any things very easily. (thought it would work as normal due to it being water proof within certain parts of the detector), on dry land and sand worked well. My first one, so still have lots to learn
a good portion of my mining geophysics course dealt with metal detection at various scales/depths. it is nice to see this stuff discussed at a more accessible level.
And this is the most simple type of metal detector, it falls in the BFO type (beat frequency oscillator). Other more precise and accurate like pulse induction or induction balance can become way more complicated
I am by no means new to this channel. But I can barely watch these videos with him so bravely handling... Everything that he does, lol. When I saw that lightbulb I already knew. My heartrate was already skyrocketing, haha!
9:28 I suggest you put a free-wheeling diode in parallel to the 100uF capacitor. This is just to avoid the Collector terminal of KSB596Y from going too negative by the inductors when the current through the PNP is throttled by the 2N2904 (when trying to limit the current).
Finally some awesome proper electric design content we needed this Everybody love to understand how things works and how to make them using basic electrical knowledge
9:46 Perhaps an exhaust and interior cooling fan may work? It’ll use more power, but keep it cool. Another idea is that you can just use a pump and tubes full of coolant and have it flowing around the system. And the final idea is that you can just replace those parts with your preferred WORKING parts that will stay cool and do the same thing with more efficiency. Plz heart
The transformer core repels because the magnetic fields generated are of the same polarity. Each piece of iron is acting like a magnet and a bunch of them stuck together is like putting together magnets with the same polarity. On a side note, can you make a video on flux gate magnetometer?
If you want to reduce power consumption overall in this circuit, my suggestion would be to use some sort of class B or C amplifier and place your entire circuit on a clock. The coil doesn't need to be hot all of the time, it just needs to be on frequent enough to find the stuff underneath of it. So make a dial to tune the polling rate of the coil to match your walking speed, this would reduce the constant power consumption of the resistors in your circuit at 9:26 to just peak power consumption allowing you to use lower rated components, as well as reducing the overall power consumption of the device.
Well that was illuminating. Especially when the light bulb blasted out of the holder! But seriously, this was a great illustration of how metal detector oscillators can be tuned and filtered. I think I finally understand how the "discriminator" on my ancient Radio Shack Micronta metal detector works. Since the circuitry is no longer a mysterious "black box" to me, I think I can use it more effectively.
The best metal detector model you can use is the Double D model with one coil being the emitter and the other coil being the receiver which can detect the phase shift and how much positive or negative displacement it has can tell the type of metal from silver and gold to combre, aluminum and iron, maybe you can even tell the difference between iron and steel (I'm not sure about that)
The core splits itself apart because the skibidi flux lines pass through each laminated alpha layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other. Very skibidi sigma.
Another way to deal with the small frequency difference, instead of going all the way down the rabbit hole to make pulses: if you can tune a second oscillator close enough and nudge it by the difference, you can use the frequency beating to hear a result.
1:03 I can always count on your videos to lift my spirits when I'm a bit down / lonely etc. Thanks so much for putting your body on the line to bring us a smile 😅 ✌️
“Don’t be afraid, for I am with you. Don’t be discouraged, for I am your God. I will strengthen you and help you. I will hold you up with my victorious right hand” (Isaiah 41:10)
@@locinolacolino1302 You just had to bring religion into this, huh? Seriously? People like you are making the rest of the Christians look bad by pushing your religion. What do you think you can change on people when hundreds before you haven't succeeded?
@@fidelcatsro6948 haha yes i completely agree with you my friend, i absolutely LOVE cats & have always had one or more in my life. I very quickly establish a deep connection with any cat, and feel that i can understand whats going on in their head.. Easiest way to explain it is, I believe im an empath, with humans, and animals n such. Its like, if i briefly look at someone, even someone i dont know, i instantly start getting a read on them. Its not so much thoughts, but feelings. I start feeling what theyre feeling, but on a deeper level, like i can feel what theyre thinking (no not like a mind reader, i cant hear their thoughts or anything).... its more like a sense / feeling. Its super strange / difficult to explain, this is the first time ive ever attempted, and its to some rando on youtube 😅 So please excuse my weird comment, i kinda went of on a tangent 🤪🤡 But to answer your message properly & to the point: I don't kill cats, I love them to bits. Its just my gamertag from my early teen years. I accidentally hit one in my car when i was like 17, broke my heart. But my nobhead mates started calling me cat killer, so i owned it, and used it as my unique gamertag. The NAC stands for Nelson Airsoft Corpse, my regional Airsoft club here in New Zealand (google it if u dunno what it is), essentially its paintball but with high powered gas & electric BB guns, FAR more accurate. Im not gonna slag off paintball tho, lots of people love it, its just not for me. Why the heck did I write so much!?!? (boredom) 😂 😺😸😺 ✌
A good alternative to your MOSFET based ZVS oscillator could be built from small signal BJT's or CMOS inverters in a similar circuit -- with inputs and outputs cross-coupled, but with the tuned circuit on the _input side_ where currents are low. Basically, you just build a standard bistable multivibrator and then put a paralell RLC tank between the bases of the BJT's. In effect, by biasing the flip-flop between it's stable states you're holding it in unstable equilibrium, which can be viewed as a symmetrical region of negative differential reisitance between the nodes which cancels the resistance of the tank circuit, sustaining oscillation. Tecnically it's still a ZVS circuit and it operates in much the same way as the power version, but that's not the point.
Too bad this video didn't come out half a year ago when I started learning to design and make videos on my own DIY metal detector to originally find lost keys in the snow. I eventually decided to build a more serious project designed around pulse induction PI detector type after trying other designs and schematics but this video gives a great introduction to the principals that are used in beat frequency oscillator BFO detectors which are the simplest to wrap ones head around how it works. VLF dual coil types can detect a bit better at depth and discriminate metal types while PI mono coil types have no discrimination but detect the deepest of all three types because of a different and more complex detection method used. Professional units are always VLF or PI types, but the BFO types like shown in this video were widely used in the early days because of their simplicity and low cost of design.
Damn this channel almost has 6 Million subs. I remember first discovering this channel where it only has a few thousand subs. Good on ya man! I dunno anything about electricity but I always find your channel entertaining!
Watched this guy since I was like 16 which was 9 years ago which is crazy to me, but he’s got me learning electrical engineering it’s crazy. Gonna edit now, seems like this would be similar to a geiger counter, no? I am watching this cause I’m getting old so I picked up a metal detector and now it’s my new hobby, it runs on 18vs so I wanna put a drill battery on it.
Since you're kinda already introduced it, maybe a video on lock-in amplifiers is an idea :-) Measuring tiny signals is an art in itself, talking from experience in my own field of applied superconductivity
I guess the same electromagnetic field is created in the core. So each north field faces another north field and vice-versa for south. So they should repel each other... Am I right ? 🤔
Don't think so, the field created should be in the same direction through the entire core, which points south of each slice to north of neighboring slice. Feel like it's got something more to do with all the eddy currents going in the same direction
@@pocarski No. Thats because if you have the layers like that |||||||||||, then the coil isnt wound in a way, so that the magnetic field is like this: N|S|N|S|N|S, but actually: N|N|N|N|N| S |S|S |S|S| So each layer repells each other. Thats also the reason, why these flat magnets in hard drives will repell each other at the breaking point, when you break them. Thats because they are magnetized in the same way, so that they act as a very short, but large area magnet.
Idk how much you've helped me to understand my electronics subjects by implementing it with basic components and god tier explanations. Forever grateful! ❤️
The reason the core ends up splitting is because the magnetic flux passes through each of those laminated layers. Since there's a gap between each layer its like they become magnets that dont want to be near each other so they end up pushing apart.
I think if the metal plates have a slight gap and a magnetic field is applied, then the system stabilizes itself by pushing the plates apart. Like the before the magnetic field, the plates are together. After the magnetic field, they push apart so there must be some sort of north north and south south alignments happening. It would be interesting to see where the metal plated would end up going if they were free to move around! Can you make that happen? I think they would push apart then flip to align with the field lines
Also, with my reasoning here, the reason that the metal atoms dont fly apart and into a field line, is because the atomic forces are more powerful at that scale
9:53 Based from my experiences and tests, to reduce the generated heat to a Transistor, On your schematics try changing the shottky diode with a zener diode(to stable the input voltage or current going in the base-emitter of it). And also by adding a non-polarized capacitor(like 104) parallel to each resistor to your mosfet(is to protect your Mosfet from high frequency...) Try it it might give changes. Just try and use right value...
My answer to the iron core question: The electromagnet generated by the coil when current passes through it creates many thin magnets (the iron sheets) which are oriented with the same polarity (for example at one point in time, all have the north pole pointing upwards, then downwards, depending on current direction). That's why all the sheets repel each other.
A trick we used to do back when we were doing Class-A and -AB audio amps was to separate the ceramic wire wound emitter resistors and put them on a heatsink.
5:44The core splits itself apart because the flux lines pass through each laminated layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other
YO! I'm still waiting to HEART a comment! in the meantime, ALUMAGNETIC FOREVER! Paramagnetic, diamagnetic or antimagnetic properties of non-ferromagnetic... sorry, ALUMAGNETIC metal are very week magnetic properties observable in DC fields, which I don't care about. They are overshadowed by MUCH greater Eddy Current magnetic forces in AC. ALUMAGNETIC focuses on great AC forces of whatever that is not ferromagnetic!
k nerd
do mine then
heart me
Because it's AC...?
Check mine out hope it answers your question
05:16: The core splits itself apart because the flux lines pass through each laminated layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other
I gave the same answer 2 days earlier and never got a heart :(
@@RavenLuni you can have my heart ❤
@@RavenLuni❣
CHAT GPTTTT
Yeah what mobs chan said
Mehdi, you did a fantastic job, covering 2 types of metal detectors (energy theft and BFO) in just 20 minutes, with no loss of fingers or eyesight. Well done, and thanks for the shout-out. As for the cross-coupled transmitter, I've used that in a metal detector design (White's TRX) and did not use any bias resistors on the transistors. You should not need them.
On the transformer plates, they push apart because the eddy current generated in each plate (across the thin cross-section) interacts with the coil's magnetic field to produce a force which is the cross-product of the two, and is perpendicular to both. This means that at the surfaces of the plates, the forces oppose each other and try to push the plates apart. Hard to explain in words, easy to illustrate. Any decent EM book explains it. A dangerously impressive example of this force can be found in coin shrinkers, where a sufficiently strong EM field creates enough radial force to literally shrink a coin. If you were to do a video on coin shrinkers, I would not dare to wager that you have 10 fingers at the end.
Edit: I couldn't remember the name of the force so I had to look it up: Lorentz force, where F = J x B (cross product).
That coin shrinker sounds cool and scary at the same time.
i believe what you're describing should be the Laplace force. the Lorentz force for a continuous charge distribution, which i assume you mean from the use of *J* , should also include a term ρ *E* , where ρ is the charge density of the volume in question.
@@matthewbertrand4139 The general equation for the Lorentz force is f = ρE + JxB, but E=0 inside the conducting plates, so we're left with the eddy currents interacting with the B-field. I think the term "Laplace force" is synonymous with Lorentz force for cases of current flowing through a conductor, like we have with the plates.
Do you think that the GPZ7000 using zvt is somewhat applying that same concept of a design like Mahid did with his zvs ?
@@mmmk1414 No, not even close. Mehdi's circuit is continuous wave BFO, something that was popular in the 1960s and 70s. The GPZ uses a form of pulse induction called constant-current PI.
9:44 you can use a cheap half bridge gate driver (such as IRS2153D) with the bootstrap pin tied to ground. i've used it in this arrangement for induction heating. this way you get quick transitions & dead time without wasting power. great video!
Maybe 120 is a little overkill
It’s crazy how smart Electroboom is but still keeps it funny with the fails and stuff.
He looks like the wizard from the smurfs
@@Cayden-m4y 🤣🤣🤣 he kinda does now that I think of it 🤣
Failing is a big part of being smart. It's just not often the part you see.
Mr beast
A true expert knows how to suck intentionally
For 4:56, due to the core having multiple individual pieces, each producing its own magnetic field in the same direction, it makes them repel each other. Maybe that's why core pieces are tightly stuck together to avoid vibration due to repelling.
Thats definitely got it. Here before heart. :)
@@Dvplexx No. They are lined up in the right direction to to stick together, so that isn't it.
@@KnightsWithoutATable it doesnt matter once you break apart lets say a permanent magnet, it wont stick together ever again because it will form its own magnetic field around itself
In addition, each piece produces its own magnetic field because the eddy currents that are generating the magnetic field won’t cross an air gap.
@@SethPentolope Magnetic fields do cross air gaps.
WOW! This video was incredible. It started out with your usual style, but going more into it, you demonstrated how educated you are and became on the topic and in the field of electronics.
I love how you were able to walk through every step in the process of designing this circuit: Research, testing, refining, etc. To make something resembling what you see in stores all around the world. And still made it digestible enough for me and a lot of other people understand. You show all the complex stuff, yet explain it simply, catering to both audiences.
I absolutely love your content Mehdi! ❤️🇮🇷❤️🇪🇬❤️
0:57 if you're lucky enough to stab some metal,
Elecroboom : oh sh*t
Me : your light bulb will blow right off.
Does this count as one of electroboom’s hair-raising electrifying shocking adventures?
Yes😂
Hello fellow electrifier you are electreofing my elecroballs
My hair was standing on end the entire time!
Haha, nice throwback! 😂
Because the core is made of plates that are isolated from each other
I love all of your videos, but for some reason this one really stood out. Watching you develop different circuits, showing them and then improving on them taught me a lot on the development process. Thanks!
Mehdi, this was one of your best videos in a long time!!!! Is so amazing to see the electronic circuit forming before our eyes and see all the logic behind it and the reason for each component!!!! Do more content like this one
Comedic and electrical genius. A rare combination of talent.
I love how you've shown the process here. A complicated electrical circuit is, fundamentally, a bunch of sub-circuits stuck together. It sort of reminds me of a project I did in undergrad, and I solved each step in the signal chain by reading The Art of Electronics to get the signal processing right!
"The Art of Electronics"... oh it's been a long time I didn't hear this book title. Brings me back to the time I spent in the library.
2:55 yum, high voltage
I love how when I was a kid being fascinated by his knowledge in electronics and thinking he's a man of magic and now that I'm studying my bachelor's in electronics understanding everything he says and recognising circuits. When I look back I see I've grown a lot and will also be a magic man to future generations
Please go abroad for better opportunities. Unless u wanna work in the IT field
@@manan-543 How in the hell would you know what country he lives in or what opportunities are available there?
@@jayesgazebobecause of his profile picture
@@Stoic_Persistence his profile picture is just him(or someone else) wearing a backpack. I really don't want to assume you mean his skin color because that would mean you were either joking or stupid.
05:16 Imagine your transformer or inductor as a stack of really thin sheets, like a deck of cards. These sheets are there to guide the invisible magnetic lines that make everything work.
Now, these lines don't like to hang out too close together, and that's where the magic happens. There's a small gap between each sheet, and when the magnetic lines pass through, it's like a bunch of tiny magnets trying to push each other away.
Think of it like having a bunch of magnets on a table, all facing the same way. You push them close, and instead of sticking together, they go "Nope, I don't want to be near you!" That's what happens between these layers in your device.
This pushing away is a smart move by designers. It helps to avoid wasting energy and makes your device work more efficiently. It's like turning a potential mess into a well-choreographed dance, where each sheet knows its role, keeping things smooth and efficient. And that, my friend, is the secret behind why your device splits itself apart but still does its job beautifully! 🚀
0:57 lamp catapult
yes😂😂
It looks epic on 0.25x speed
4:57 the magnetic flux is concentrated in the iron core. The separate plates are aligned with the field axis, and so adjacent plates will always have the same polarity and repel. Like lining up a set of bar magnets next to each other in the same orientation
I had the same idea, but I didn't offer to comment because I couldn't think of the right terms, I'm not a master at this kind of stuff but I like learning about it
Was about to write the same - at least that is what my mind would deem the most plausible ^^
15:49
Mehdi you made a Geiger counter 😂😂😂
In general I did well in school, and in school I found physics to be quite easy. Chemistry took some effort to follow. Electricity just seems like a totally foreign language comprised of multiple languages at the same time lol. I always disregarded it as something I could live without. Cut to me in my 30s now really invested in computer tech and watching LTT videos and broadening my interest in tech and I find myself needing to understand electricity more and more. Not to mention it seems to be a useful thing to know when it comes to other things like cars and homes.
So I'm glad you're as entertaining as you are. I've decided I'm just going to watch a few videos of yours every day and hope I just start picking things up over time lol. Because otherwise I have no real idea where to start that won't either bore me to death or be overwhelming... tbf this one got close to overwhelming in the last 3rd as you were talking about converting all of those signals in order to make the data presentable to the human ear lol.
This came at the best time, lol
I've been struggling to build a metal detector for a land mine detecting robot, its honestly looking like ground penetrating radar would be easier to build
16:17
Wow, that's actually impressive its clicking at that distance with such a small coil.
17:35
Yeah, that's one of the problems we had, and if you're actually using it in the real world, you gotta deal with changes in soil (i.e. mineral deposits) which can throw off the coil's dynamics.
What I found works good is using a microcontroller. You can use the PWM for a fairly stable signal. That's how I'm running mine, just a simple on/off circuit powered by an IGBT used to run a car ignition coil, then we just probe a leg of the coil with an analog pin. I'm still using low voltage, though. 28v, you might want to throw a capacitor in line and add a clamping diode after it or something
The cool thing with using a microcontroller is that you can do the filtering with the controller and have it run a decay function that slowly increases or decreases coil sensitivity to account for changes in the environment.
The big problem we're having is trying to figure out what is a piece of metal and what is just environmental noise, also the fact a lot of landmines have very little metal in them at all.
I do think we are going to go with GPR though, it's just better able to detect objects underground, but before we go that far I want to try one more time with the pulse inductance and maybe a VLF
Just throw 240VAC through it and you should detect things meters apart lmao
HI
I had the same problem on a metal detector project a few years ago : the sensing coil had to fit in the palm of a hand. The electrical circuit was very similar to this one (pair of oscillators, listening to the beat frequency) and it was very sensitive to human body proximity, including to the hand ! We got around by going towards a completely different circuit, the pulsed induction you mentionned. The time it takes for the voltage spike (following a sudden current cut in the coil) to decline indicates the presence of a metal, and it's much more selective than the beat frequency method, i.e. very insensitive to surrounding moisture, and even to small deformations of the coil
GPR doesn't work well on wet ground etc. A metal detector is your better option.
A good rule of thumb is that a metal detector can only see into the ground about the same as the diameter of the sensing head.
This is why a lot of stuff used for UXO (UneXploded Ordinance) is so big.
Land mines of the antipersonnel type are about the size of a can of cat food and have almost no metal in them. They use trained pouch rats to sniff them out. The critter is light enough not to set them off.
@kensmith5694
GPR should work just fine in water logged soils. It measures the difference in the refractive index of materials, and we don't need to sense very deeply, just a foot or so.
The main issue is a lot of the landmines (e.g., the butterfly mines) are just sitting on top of the ground, so we'd have to deal with soil reflection.
Another idea we're toying with is to just give our robot a hammer or spike and tell him good luck.
Realistically, we can make a dumb robot for under $100 to just go trigger mines. It'd still be cheaper than what it costs now per mine, but at a certain point, you're better off just driving an RC car through the minefield
@@noahw4623 Microwaves don't travel through mud.
Some guy in Africa made a thing that gets blown by the wind to stomp across a field.
Sometimes, the explosions make you forget how intelligent this man is 🤓
I think, the explosions not killing him, because it was all planned, make him look even more intelligent
he can plan ways to make the explosions without seriously injuring him _and_ make it seem accidental. A genius with an oscar.
should be wearing safety glasses
Well. It’s electro boom. It has to have explosions.
@@tomservo5007 Barring that...at least a Safety Tie? But that requires being around potentially fast spinning objects...
Now this is what I call quality content! Really love the in depth design process. It gave me the "why use this and not that" and the "if that is unavailable, this can also work" insight, which I find very rarely explained on other channels. THANK YOU, MEHDI!!!
I honestly wish Mehdi was one of my college professors for EE. Even though I graduated with a BSc in EE I still struggle with circuit design and control systems. I learned more from Mehdi videos than the entire last 2 years in college :)
I'm just starting college, planning on going for EE.
I'm finding Mehdi's videos fascinating even though I only half understand it.
Any tips for a new EE student?
@@galacticviper4453 In one word: RUN! I went to school studying EE while having a full time job and a wife with 2 kids. The EE program ASSUMES you're totally OK with not having a life whatsoever and that it is "Reasonable" to work 14-hour days studying, assignments, etc. And no I am NOT exaggerating. Furthermore, the first 2 years of EE is a walk in the park. The third year gets a bit challenging but still manageable with some grit. The 4th year though is totally insane. The instructors switch into the "we're not here to educate you but we're here to grade you and try our best to make you fail. YOU have to prove yourself worthy to breathe the same Oxygen in the room we occupy. Oh, and did we mention that while you were reading this sentence there is a 10% penalty on your assignment because you turned it in 2 minutes late?". By the way, I am in no means exaggerating. It is THAT bad! Now what's the good news? The good news is that if you go through it and make it out and graduate you have "proven" yourself to be an engineer and therefore can qualify for entry level positions that start somewhere in the 65K to 100K range. Not too impressive when compared to some entry-level IT jobs, especially in data mining and AI that can have a STARTING range upwards of 125K. You do the math. And btw, I really do mean i t. IT is a PHENOMENAL amount of work. Legalized slavery that is, to finish an EE program! :)
@@galacticviper4453go to trade school instead, save your time and money
This feels like the classic electroboom type of content which I love. It's chaotic enough, it's got simple educational parts for us non-EEs, it's got some damn massive schematics that give me flashbacks for those that understand them and there's a complete working project at the end, love it
Yeah, I got the part about the need to multiply frequences to make the change when metal is brought closer noticeable but the rest needed some EE studying I guess XD.
Bro I’m so tired I didn’t understand the most simplest thing you said
1:28 Detector showed us that Mehdi is made out of metal.
So that explains why Mehdi gets schocked faster than the average human
Real iron man
This is actually quite impressive. I don't know how well this will hold up to long-term use, but as a proof of concept it's remarkable. Thank you :)
Usually, long term use is just beefing up the mechanical connections, adding weather proofing (if required) and generally making the circuit hard to kill or to be injured by. He could probably pack the circuit into a project box, make sure the solder joints are top notch, and have an actual legit metal detector.
You sir deserve your own Netflix series. I have been entertained all day watching your videos. You are genuinely likeable.
@@designofperformance Broke boy
@@Cyno_hari Nah dude he's right, if you have Netflix you either have a ton of money to spare or you don't know how to properly handle your finances
Fascinating, i never knew how those clicks and tones were generated, being a computer guy i just figured it was just a digital tone generator mapped to some signal and not the actual signal (if that makes sense)
Makes sense
Yeah, I thought exactly the same - the clicks were generated by a microprocessor when it gets a matching, programmed signal. Apparently it's more simple and complicated at the time (for me at least).
Same. It's satisfying to learn something that was completely out of my imagination radar.
I wouldn't be surprised if thats how modern or more expensive detectors work, but the old WW2 ones were probably very similar to what Medhi has built here.
Modern tech, being at such a high level interface, obfuscates what the computer's actually doing behind the scenes, which is the reason I love looking back at analog circuitry and the computers of the past where you can see a direct correlation between all components.
Okay, I thought I understood how metal detectors worked - to the point of being confident when I sell them at work. But now... Man, you just opened up a lot to me. There are people who just want to know how the thing works, and there are people who want to "KNOW" how it works. Now I can explain so much better.
That's awesome! They'll be amazed by your knowledge.
so what makes the difference between a 5k metal detector and a 500 detector then their signal goes deeper they have a larger battery ?
6:55 hertz😂
_hærtz_
Aluminum is paramagnetic, but the funny part is that now there's a term on urban dictionary alumagnetic
Now this is a very cool and informative video. The Core is repelling because of laminations between the slits of core material. two slits have the same direction of eddy current causing the magnetic field to be generated to repel each other. Its the same as one piece of mage is solid but when it breaks the two pieces repel each other.
That's somewhat similar to what I proposed, but i don't think eddy current has anything to do with it as the core is split up like that specifically to avoid eddy current, aka the core doesn't act like a short-circuited coil. But i do believe it's similar to the broken magnet thing.
Really loved this video. Nice to see the analog logic and math. Would love a series on this…
Electroboom 101 is entering circuit logic already (he has videos on relays, and relay logic) and I wouldn't doubt further episodes in that series will eventually dive into such topics as analog math, digital math, signal processing and others.
It just seems to take Mehdi ages to produce those videos. Probably because of all the extra research, fact-checking, error corrections, and editing he needs to do.
There are more than enough extant channels that cover the theory in detail. Like EEVBlog. We come for the boom and the doubletakes. And the unibrow.
@@Mr.Sparks.173 I hope Medhi makes a video on logic gates that use LED transistors!
I hope this series includes Medhi making a Hall sensor or explaining why it is not used in a metal detector.
I love the practical explanation of the circuit and how it generated sound along with the oscilloscope display. I know there was product placement involved but it was also very educational and instructive. Thanks for the video Mehdi!
@ElectroBOOM my deepest gratitude to you, bro. I've been trying to get my 9yo son to watch more educational videos but nothing got his attention until I show him your 'electric guitar' video. I was surprised coz not only he watched the whole video, he even watched a few more of your videos!
Though he was mostly laughing his ass off at your 'accidents', he did ask a few technical questions & I hope his scientific curiosity will grow.
Thank you again & keep up what you're doing - the way you're doing it! 👍👏
I like your sense of humor, your integrity in owning when you make a mistake and your selflessness in promoting other UA-cam pages that also get it right. It shows you will always be about the facts
Hi Mehdi.
The 24 amp maximum current of the IRFZ24N Seems massive especially if you're planning on a PP9 type, battery powered portable metal detector.
I'm therefore taking my clue from this as I'm unfamiliar with the MOSFET and Schottky type of astable multivibrator you've used in this example.
So I'm hoping for a Mehdi induced heartfelt like with my idea of what you're hoping for.
I Do go on to mention how I learned because the learning curve was such a massive moment especially after being at college and coming no where.
I hopefully explain why I'd be employing such an experimentalist approach and, if I'm wrong I was wrong but not for the want of having an estimated guess.
Anyway.
A two transistor BJT astable multivibrator is possibly the answer.
TIP41 or TIP42 transistors should supply enough current through the 100uh + the series of the 115uh search coils.
I hindsight the resistor supplying the emmiter follower regulator seems to verge around the 1 amp current region so I'm more inclined to believe this.
Ditch the Schottky diodes and I'd replace them with an experimental value of around 220nf then the 150 ohms might be used in series with the search coils
(in case of over load)
and in their original place I'd insert something like a 2.2k ohm resistor to be used for the Resistor in the CR timing network.
OR leave the 150 ohm resistors where they are to act as a voltage limit stop resistors and cut the connection between the 100uf switch spike surprising capacitors positive terminal to the 150 ohm resistors shared common connection and insert a 10k pot to change the voltage and therefore speed the 220nf capacitors charge up times.
Therefore giving experimental frequency change.
I'd choose this experimental method because I'm not to good with theory especially with frequencies.
I have been a self taught electronics enthusiast since the age of 13 finally teaching myself digital fundamentals whilst I designed and constructed a Radio Controlled up / down step encoder into a Resistor divider DAC utilizing a toy car Remote control and receiver chip.
1 "Nudge Turn" of the remote would notch 1 up / down control on the encoder and therefore Digitally control the amount of resistors in the resistor adder and give Analogue control for the timing function of the mains cycle for the 240 volt ac bulb to dim or brighten up.
Obviously timed to coincide with the correct part of the phase during the 50hz cycle.
Cheers and TTFN. ;)
The peak current rating of the MOSFET will not impact the battery drain. However the voltage dropped across the FET when saturated is less than the BJT, so you end up with more voltage across the coil and therefore more current through the coil, so for the same supply the MOSFET will yield a better field.
The MOSFETs can handle more current for the same package type (TO220) in this case, because it does not get as hot because of less voltage drop across the transistor as previously mentioned.
@@NCmountainview Absolutely understood but I mentioned the BJT astable as I'm more familiar with its turn
on / off characteristics having made many LED flip flops I'm also more familiar with the 0.7 volt saturation voltages of the BJTs being a lower sensitivity than that of the FETs which MAY cause weird oscillator results than I could positively say I knew about as I've never built a variable frequency MOSFET astable.
I'm not saying the FETs would use more current, this would be limited by the one amp rating of the main regulators feed resistor.
I also realise the exceptional sensitivity of the gate junction of a MOSFET and with that realization
I can also theorise the BJTs tendency to give a rounder top to their waveform especially when the transistor base goes below its saturation voltage, this does cause more heat but one amp is within the rating for the type and possibly achievable without adding a heatsink. (If not bung a pair of BU508s in there.) LOL.
The sinusoidal wave shape at the waves peak must give a better all round smoother electro magnetic RF flow.?
@@Stuck_Farmer The BJT does not have a saturation voltage of 0.7V, as that is the ideal junction voltage and usually associated with the base emitter region in forward bias. The saturation voltage is measured across the collector emitter and varies depending on the load, but it is generally larger than the channel voltage for a FET under the same load.
There is some merit to your point, as FETs can have issues with linear response because of the transconductance profile. However, the reality is that you can design stable oscillators with either. In fact the higher input impedance of the FET can make the oscillator much more stable when under load.
Having built metal detectors with my dad as a kid, the only thing I would change is do a flat coil like an antenna instead of a regular coil. But that is more difficult to wind up. Another thing you can do to make it extremely sensitive is to have two coils, one above the other running half off phase signals and sum them in the end to essentially make an EM interferometer.
How does having two coils out of phase make it more sensitive?
@@locinolacolino1302 Instead of measuring the changes of one coil's frequency and peak, you're measuring the difference between two. After amplifying, I could detect 0.01% difference between the two. But it's super finicky, the coils have to be perfectly aligned, the frequencies have to be on spot. It's a fun experiment, but probably not worth it for practical use.
Those professional detectors that have the coils shaped like an 8 use two of them already, but they work by being a coupled oscillator. You can detect coins 30cm away with those easily.
@@locinolacolino1302 As they hinted at this design is based on interferometry. When there is no metal present the two 180 degree out of phase signals destructively interfere to produce a constant potential. In essence your output signal is the result of recombining those two waves, when nothing is interfering with the field they will destructively interfere and cancel out, when there is however the two coils being in different physical locations will cause them to experience a different change in frequency thus recombining the two will produce an interference pattern that can be detected.
Do you have a circuit
@@chinnapank No, it has been 20 years. But if you look up for two phase shift oscillators leading to an amplifier, it is pretty much the same thing. Just need to adapt for the coils and voltages you will be using.
Mehdi, you did a fantastic job, covering 2 types of metal detectors (energy theft and BFO) in just 20 minutes, with no loss of fingers or eyesight. Well done, and thanks for the shout-out. As for the cross-coupled transmitter, I've used that in a metal detector design (White's TRX) and did not use any bias resistors on the transistors. You should not need them.
On the transformer plates, they push apart because the eddy current generated in each plate (across the thin cross-section) interacts with the coil's magnetic field to produce a force which is the cross-product of the two, and is perpendicular to both. This means that at the surfaces of the plates, the forces oppose each other and try to push the plates apart. Hard to explain
You were going crazy with your engineering skills on this one! It's so cool to hear the technical stuff 😊 thanks for another great and highly entertaining video!
😂😂😂 this video is funy
I’d like to see him make a voltage controlled oscillator kinda like the Original Moog Synth.
Heck maybe he can make a synthesizer
I was thinking theromin
He should go full custom Eurorack!
lookmomnocomputer is your guy.
Of course he can make a synth! It's just an oscillator!
It's actually crazy how much this kind of field of work relates to music/sound design with basic waveforms. Even the terms used are the same. Math really does connect everything.
HE WAS FIRST
5:15 like you said in the eddy current video "opposing fields repel each other" and in that case for each sheet of iron of the magnetic core are magnetically polarised about the same direction as the other sheets since all sheets changes through them the same magnetic flux and with all sheets having the same magnetic poles and the strong changing magnetic flux of the primary transformer, that's why the sheets repel each other hard and that's the one of the reasons why manufacturers glue the iron sheets altogether to make sure they won't strongly repel each other and another key benefit is to isolate each sheet from the other thus the total resistance of the iron sheet is higher hence reduced eddy currents and less magnetic losses in the iron core
I absolutely love watching your videos, it is fun, educational and makes me realize how happy I am now that I made it into my final year of studying electrical engineering and only have 1 semester left.
Nice! I just finished my PhD in EE, and although the work I do is mostly theory and simulation, it was love of circuit building and tinkering with electronics that pushed me into the program initially. EE is a huge field with a lot of really cool opportunities!
The joy I gain watching you demonstrate how not to do things is the gift that keeps on giving. Seriously, I worked as an electrician for a decade and it never fails to make me laugh when shit pops. Thank you 😀
There are numerous solutions for the gate bias resistors:
1. At low frequency, don't worry about it. Scope the gate waveform: if the rising edge is fast enough, you're done. At 25kHz, probably some kohms is acceptable.
2. Add a buffer. I've done this before, add a complementary emitter follower (just a simple class C one will do, no biasing components) between the pull-up / diode connection, and the gate. Use a pull-down resistor at the gate, to bias the follower (this is necessary for startup, because this is a linear amplifier at its heart -- it won't start up at the right frequency, or at all, if there is too much gain or distortion in the feedback path!). I've done this before, on, let me get it out here -- I had 2 x FDP33N25 for the inverter, and 2N3904/6s for the buffers. Oh, I didn't use a pull-up resistor with the diode, I used a PNP current source, fancy; well, about 470Ω pull-up would be equivalent. It ran over 600kHz with good gate waveforms as I recall, so it doesn't take much as you can see. That suggests 27k would be enough for IRFZ24N down at 25kHz! Call it 10k for good measure. Maybe 1k or less without the buffer circuit.
3. Use different devices in a related circuit. The original Baxandall* circuit, I believe used a feedback winding, with bipolar transistors. This separates bias and feedback from the output voltage, giving more freedom to design the gain and drive strength. Example circuit in following comment.
*The common "ZVS" oscillator configuration is actually due to Baxandall, the same one of audio tone control fame; published back in the 50s or 60s when transistors were new, though I don't have the exact citation handy unfortunately. It's commonly called "Royer" by the amateur/HV community, but this is erroneous -- Royer refers to a non-resonant (specifically, saturable core commutated) oscillator. (Occasionally even professionals make this error: for example, Jim Williams (of Linear Technology) in AN49; the correct citation is even given (Royer, et al; in the title, "in saturable core circuits" -- but these are resonant, non-saturating circuits!). So, go figure, even the best of us make mistakes.)
!@QSDRD
I like, when people copy paste stuff from Wiki.
@@tuff_loverikr
Man typed out a whole dream tweet
A smart Iranian. We love you, compatriot🇮🇷😎✌
He was proud be Iranian so he moved Canada 😂
@@lazyicebearbrainless fool it’s not a safe place in Iran
16:14 Why does it sound like Geiger Counter
This is the best diy tutorial video you've made. Just loved how the analog electronics working without any microcontroller rubbish! Such a beauty, Joy forever!
Agreed that it is fun to see an analog only design but an MCU could've allowed for fine grained identification of metals just by switching between frequencies automatically and by not being annoying to hear. This is actually super impressive because you don't need more than a single MCU to do that with the way he designed it
@@amogusenjoyer yes, you're right. I'm a microcontroller person and if I had to design a metal detector like this, I would've used a microcontroller. My analog electronics skill is - well, questionable.
This video sparked my interest in analog electronics again.
Amazing. Its really motivating mehdi put together this very complex circuits and concepts and show that its not that complicated. Makes me feel like i can too build and learn these awesome things.
1. Replace Bipolar Transistor with MOSFET: Use a MOSFET (e.g., IRF54ON) for the switching element to improve efficiency and power handling.
2. Add a Gate Driver: For optimal control of the
MOSFET, consider using a gate driver IC (e.g., TC4420) to ensure fast switching times and reliable operation.
3. Implement Current Limiting: Incorporate a current-limiting circuit to protect the components and prevent excessive currents during faults.
4. Add Protection Elements: Include a fuse and surge suppressor to safeguard the circuit from damage.
Mehdi is no longer just an electrical engineer, he is also a sound design master!
whod of thought sound is created by coiled copper wiring too LUL
he should collab with Andrew Huang to make a synthesizer. Ultimate collab and challenge
@@ThunderBlastvideo a great idea honestly
The landmine bit killed me 🤣🤣👌
@@papagrounds oh god I had tears in my eyes XDDD
Electricity is an absolute mystery to me.
You take so much time in your content to explain, and probably from your viewpoint, simplify.
I'm so happy that you have accrued such a massive following. You deserve it because you are a great educator.
I have accepted that my limitations will never allow me to fully appreciate your content.
Every video leaves me baffled, though.
Love this channel for the learning and the laughs, but I don't envisage a time that I will ever understand what is going on.
I like clever people, but I'm frustrated that I get left behind.
I think it's mostly a matter of studying the subject. It's very difficult to understand any of this without studying at least some basic electronics. It's like playing an instrument, you gotta practice in order to play songs.
to understand these videos you really just need to understand the basic concepts of electricity and magnetism which is easy. There are channels on youtube that'll help like, ' Physics videos by Kugene Khutoryansky '. Do not limit yourself.
You most definitely can understand this, there's just some prerequisites. Don't gauge your abilities based on a youtube video. This video assumes prior knowledge, he didn't explain every detail. If you're not familiar with circuit theory, this would be like trying to understand differential equations before learning arithmetic. If you take it one step at a time it's very easy. Knowledge is free online, you can get all the books and papers they teach in college for free from beginner to postgraduate level. I believe in you, believe in yourself.
I was studying a metal detector to modify mine last night. Good thing you posted it and really helpful. thank you professor boom. i'd like to see its final form for next video 😅
In one word: Amazing! The explanation and performance of metal detection made things very clear. People have learned how they work, me too!
15:30 doing redstone i guess
Hey Mehdi, for your question at 5:10, is the answer that the pieces of the core become magnetised in such a way that the poles are facing the same way causing them to repel each other but they are held together by the coil in the centre.
One suggestion boss, use switch , not teeth 3:00
2:55
Gotta have that risk
though I had some problem with sea water It also ua-cam.com/users/postUgkxa-FNYUOM93a388gi9a4brtSCEVmrHgJH land for finding any things very easily. (thought it would work as normal due to it being water proof within certain parts of the detector), on dry land and sand worked well. My first one, so still have lots to learn
a good portion of my mining geophysics course dealt with metal detection at various scales/depths. it is nice to see this stuff discussed at a more accessible level.
Man, I had zero idea that metal detectors are this complicated
Not more than my life
i mean this is very interesting since its going to deeper dive on those detectors
And this is the most simple type of metal detector, it falls in the BFO type (beat frequency oscillator). Other more precise and accurate like pulse induction or induction balance can become way more complicated
or using coils
@@bhabok20 At least life has a logical destination, Jesus, who makes our lives simple, but the field of magnetometers is a never ending rabbit hole.
I am by no means new to this channel.
But I can barely watch these videos with him so bravely handling... Everything that he does, lol. When I saw that lightbulb I already knew. My heartrate was already skyrocketing, haha!
9:28 I suggest you put a free-wheeling diode in parallel to the 100uF capacitor. This is just to avoid the Collector terminal of KSB596Y from going too negative by the inductors when the current through the PNP is throttled by the 2N2904 (when trying to limit the current).
Finally some awesome proper electric design content we needed this
Everybody love to understand how things works and how to make them using basic electrical knowledge
As a metal detecting enthusiast, this is my favorite video, Mehdi!
9:46 Perhaps an exhaust and interior cooling fan may work? It’ll use more power, but keep it cool. Another idea is that you can just use a pump and tubes full of coolant and have it flowing around the system. And the final idea is that you can just replace those parts with your preferred WORKING parts that will stay cool and do the same thing with more efficiency. Plz heart
The transformer core repels because the magnetic fields generated are of the same polarity. Each piece of iron is acting like a magnet and a bunch of them stuck together is like putting together magnets with the same polarity.
On a side note, can you make a video on flux gate magnetometer?
If you want to reduce power consumption overall in this circuit, my suggestion would be to use some sort of class B or C amplifier and place your entire circuit on a clock. The coil doesn't need to be hot all of the time, it just needs to be on frequent enough to find the stuff underneath of it. So make a dial to tune the polling rate of the coil to match your walking speed, this would reduce the constant power consumption of the resistors in your circuit at 9:26 to just peak power consumption allowing you to use lower rated components, as well as reducing the overall power consumption of the device.
It is not even two minutes in and you almost lost a finger there 1:06
Electroboom never die
Bro legit has plot armor
That part got me free plane tickets to Ohio to roll on the floor
@nicolaedavid2243
He just respawns to the nearest spawn point😂.
I can't.......I can't stop😂😂😂😂😂😂😂😂
Well that was illuminating. Especially when the light bulb blasted out of the holder! But seriously, this was a great illustration of how metal detector oscillators can be tuned and filtered. I think I finally understand how the "discriminator" on my ancient Radio Shack Micronta metal detector works. Since the circuitry is no longer a mysterious "black box" to me, I think I can use it more effectively.
The best metal detector model you can use is the Double D model with one coil being the emitter and the other coil being the receiver which can detect the phase shift and how much positive or negative displacement it has can tell the type of metal from silver and gold to combre, aluminum and iron, maybe you can even tell the difference between iron and steel (I'm not sure about that)
i rlly love how he teaches science whilst putting alot of entertainment into his vidoes.
The core splits itself apart because the skibidi flux lines pass through each laminated alpha layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other. Very skibidi sigma.
I never said this???
Another way to deal with the small frequency difference, instead of going all the way down the rabbit hole to make pulses: if you can tune a second oscillator close enough and nudge it by the difference, you can use the frequency beating to hear a result.
1:03 I can always count on your videos to lift my spirits when I'm a bit down / lonely etc. Thanks so much for putting your body on the line to bring us a smile 😅
✌️
“Don’t be afraid, for I am with you. Don’t be discouraged, for I am your God. I will strengthen you and help you. I will hold you up with my victorious right hand” (Isaiah 41:10)
@@locinolacolino1302 You just had to bring religion into this, huh? Seriously? People like you are making the rest of the Christians look bad by pushing your religion. What do you think you can change on people when hundreds before you haven't succeeded?
dont kill cats! 🐱
@@fidelcatsro6948 Nobody said anything about cats??
@@fidelcatsro6948 haha yes i completely agree with you my friend, i absolutely LOVE cats & have always had one or more in my life. I very quickly establish a deep connection with any cat, and feel that i can understand whats going on in their head.. Easiest way to explain it is, I believe im an empath, with humans, and animals n such. Its like, if i briefly look at someone, even someone i dont know, i instantly start getting a read on them. Its not so much thoughts, but feelings. I start feeling what theyre feeling, but on a deeper level, like i can feel what theyre thinking (no not like a mind reader, i cant hear their thoughts or anything).... its more like a sense / feeling. Its super strange / difficult to explain, this is the first time ive ever attempted, and its to some rando on youtube 😅
So please excuse my weird comment, i kinda went of on a tangent 🤪🤡
But to answer your message properly & to the point: I don't kill cats, I love them to bits. Its just my gamertag from my early teen years. I accidentally hit one in my car when i was like 17, broke my heart. But my nobhead mates started calling me cat killer, so i owned it, and used it as my unique gamertag. The NAC stands for Nelson Airsoft Corpse, my regional Airsoft club here in New Zealand (google it if u dunno what it is), essentially its paintball but with high powered gas & electric BB guns, FAR more accurate. Im not gonna slag off paintball tho, lots of people love it, its just not for me.
Why the heck did I write so much!?!? (boredom) 😂
😺😸😺
✌
A good alternative to your MOSFET based ZVS oscillator could be built from small signal BJT's or CMOS inverters in a similar circuit -- with inputs and outputs cross-coupled, but with the tuned circuit on the _input side_ where currents are low. Basically, you just build a standard bistable multivibrator and then put a paralell RLC tank between the bases of the BJT's. In effect, by biasing the flip-flop between it's stable states you're holding it in unstable equilibrium, which can be viewed as a symmetrical region of negative differential reisitance between the nodes which cancels the resistance of the tank circuit, sustaining oscillation. Tecnically it's still a ZVS circuit and it operates in much the same way as the power version, but that's not the point.
9:08 just right click and click add to dictionary
Defeats the hole purpose ngl 😂
That's the old Mehdi in first 2 minutes 😂😂
Always love your informative sessions❤
Too bad this video didn't come out half a year ago when I started learning to design and make videos on my own DIY metal detector to originally find lost keys in the snow. I eventually decided to build a more serious project designed around pulse induction PI detector type after trying other designs and schematics but this video gives a great introduction to the principals that are used in beat frequency oscillator BFO detectors which are the simplest to wrap ones head around how it works. VLF dual coil types can detect a bit better at depth and discriminate metal types while PI mono coil types have no discrimination but detect the deepest of all three types because of a different and more complex detection method used. Professional units are always VLF or PI types, but the BFO types like shown in this video were widely used in the early days because of their simplicity and low cost of design.
I built a Pirat detector circuit but it didn't work.
9:45 Maybe, if you put more resistors in parallel, the current would be split between them and their load would be smaller, creating less heat
Full-Bridge Rectifier
Damn this channel almost has 6 Million subs. I remember first discovering this channel where it only has a few thousand subs. Good on ya man! I dunno anything about electricity but I always find your channel entertaining!
Same here
Watched this guy since I was like 16 which was 9 years ago which is crazy to me, but he’s got me learning electrical engineering it’s crazy.
Gonna edit now, seems like this would be similar to a geiger counter, no?
I am watching this cause I’m getting old so I picked up a metal detector and now it’s my new hobby, it runs on 18vs so I wanna put a drill battery on it.
Since you're kinda already introduced it, maybe a video on lock-in amplifiers is an idea :-) Measuring tiny signals is an art in itself, talking from experience in my own field of applied superconductivity
I guess the same electromagnetic field is created in the core. So each north field faces another north field and vice-versa for south. So they should repel each other... Am I right ? 🤔
too early
What is TEH link? (2:30) 😅😅😅
Don't think so, the field created should be in the same direction through the entire core, which points south of each slice to north of neighboring slice. Feel like it's got something more to do with all the eddy currents going in the same direction
Yes, I think you are correct 😅
@@pocarski
No. Thats because if you have the layers like that |||||||||||, then the coil isnt wound in a way, so that the magnetic field is like this: N|S|N|S|N|S, but actually:
N|N|N|N|N|
S |S|S |S|S|
So each layer repells each other.
Thats also the reason, why these flat magnets in hard drives will repell each other at the breaking point, when you break them. Thats because they are magnetized in the same way, so that they act as a very short, but large area magnet.
Idk how much you've helped me to understand my electronics subjects by implementing it with basic components and god tier explanations. Forever grateful! ❤️
The reason the core ends up splitting is because the magnetic flux passes through each of those laminated layers. Since there's a gap between each layer its like they become magnets that dont want to be near each other so they end up pushing apart.
13:17 That’s what Fourier said❤
Who's Fourier?
@@Nsodnoajdjksl Joseph Fourier look it up
@@VoidsDemise5443 nah, I don't need any AI search, I rather need human search.
I think if the metal plates have a slight gap and a magnetic field is applied, then the system stabilizes itself by pushing the plates apart. Like the before the magnetic field, the plates are together. After the magnetic field, they push apart so there must be some sort of north north and south south alignments happening. It would be interesting to see where the metal plated would end up going if they were free to move around! Can you make that happen? I think they would push apart then flip to align with the field lines
Also, with my reasoning here, the reason that the metal atoms dont fly apart and into a field line, is because the atomic forces are more powerful at that scale
I like how he knows "yep, this is gonna shock me"
But his brain goes "ooh wonder if it still hurts the same"
9:53 Based from my experiences and tests, to reduce the generated heat to a Transistor, On your schematics try changing the shottky diode with a zener diode(to stable the input voltage or current going in the base-emitter of it). And also by adding a non-polarized capacitor(like 104) parallel to each resistor to your mosfet(is to protect your Mosfet from high frequency...) Try it it might give changes. Just try and use right value...
4:19 I laughed way too hard at this... hehh ... Hertz!
0:49 THIS metal detector looks like a *gun!!*
4:38 You mean diamagnetic?
My answer to the iron core question:
The electromagnet generated by the coil when current passes through it creates many thin magnets (the iron sheets) which are oriented with the same polarity (for example at one point in time, all have the north pole pointing upwards, then downwards, depending on current direction). That's why all the sheets repel each other.
thinking about the comically long potentiometer from 12:37
is that in the bundle?
A trick we used to do back when we were doing Class-A and -AB audio amps was to separate the ceramic wire wound emitter resistors and put them on a heatsink.
5:44The core splits itself apart because the flux lines pass through each laminated layer. Since there is a gap between each layer, you end up having the said layers behave like like-oriented magnets, thus they repel each other
4:12 diamagnetic