My first thought was three words: Pulse Width Modulation. Feed the electromagnet with fast on-off pulses and modulate their width to change how strongly it attracts the magnets, depending on the output of the hall effect sensor.
Excellent idea. A theoretical idea I had was to suspend plasma around a metal core inside of a magnetic field in order to suspend the material indefinitely.
Now, theoretically speaking, is it possible to create such a thing in order to suspend a plasma like substance? I was considering a highly pressurized element about a hundred times more dense than osmium. The idea would be then to contain the element inside a magnetic chamber while being attached to a metallic core that could then be used to assist in suspending the element. The magnetic chamber would function as a shielding. A pin sized actuator could be installed, most likely created by another magnetic force which would allow for an exhaust or discharge in the event that it was needed. Now the idea goes further with plasma, not just a magnetic core, but creating an elemental compound. The compound itself would be suspended by the mixture rather than being held by a core. This new form of theoretical science could be exploited to create artificial gravity or a new system of propulsion for space travel.
How about using - Harmonic Resonance Frequencies - In a power system when the system's natural frequency corresponds to the frequency of a source of harmonic current, "HRF" Occurs. Surround the "Magnetic Chamber" in an array of "HRF."
Oscillations are inherent phenomena with on/off control, you will probably need to modify your circuit to have some form of PI or PID controller. It would be a good project to show how some of this concepts on a practical example like magnetic levitation or inverted pendulum.
@@martinda7446 I'm a German guy reading about an English guy reading about a Russian guy reading about an English guy reading about a German guy watching a German guy speaking English
A teacher from midschool designed something like that, and it worked great ! He used a lightbulb and a big LDR on the other side (half of the LDR whas shade because the ball floated before the lightbulb), to get an analoge reading of the height. Than the used a PID-regulator to control the coil. He made it to demonstrate each action of the PID regulator. Nice project ! :)
GreatScott! Hey Scott great project as always. You don't really need a PID, PI should be enough. Perhaps try with 2 coils on top of it and 2 hall sensors for better positioning in-between the coils.
The main problem is the two point regulator. It creates a limit cycle in most cases, which causes oscillation (e.g. iron with bi-metal switch). Using a simple P regulator may not help either because of the instable poles of the system. You definitely need at least a PD regulator in order to stabilize the levitation system.
DruMsfR3aK a digital signal is too crude i think, so analoge control of the coil is needed. PD regulator ? So if the magnets fall fast, the D will add will add a burst to compensate ? I think that will oscilate (magnet falls, Burst of coil, magnet rises again and proces repeats). And if the magnet hovers at the desired height, the D action does noting. The D action is usefull when you insert the magnet, to adjust the proces fast if the magnet is inserted. I think a I action is still needed to fine adjust to the perfect current trough the coil.
GreatScott! Those tremors you got are pretty common with electronics. Also i found the higher the voltage the bigger the tremors. Every time I wire 400V 3phase my hands are shaking if I hold the cable when electricity flows thru...
Hello Great Scott ! The reason why this doesn’t work is because a magmatic levitation device is a second order non-linear system that can’t be driven by turning ON and OFF the current . You’ll need a PID controller to drive the coil with the proportional current required to levitate. Big fan of your videos! Keep it up !
Another great video, always looking forward to the next one. I would suggest a different circuit which doesn't turn the electromagnet on and off but rather adjust the current flowing through the electromagnet such that it reduces/increases the current at a constant rate according to the hight of the floating magnet. You wouldn't get the oscillations and you'd be able to adjust the height of the magnet very accurately.
Is there a pre-made board sold commercially that controls electromagnets like this? I'm looking for one for my levitating sneaker display project but unfortunately I'm not too into the technical electronic stuff and not able to create a board like you did.
as long as I know, you need to know the transfer function of your system in order to have a better compensator in closed-loop controller. nice video and thanks!
Great video as always :) We made one maglev project 2 years back. Ours used an arduino UNO and anlaog hall effect sensors(one attached to the bottom of the electromagnet and other similar to how you did). That helped us to apply a simple position based PID loop (actually just PD was used). Also, before adding the "floating" magnet, we measured the full swing of our electromagnet (800 Copper Coils powered by bench power supply with iron nuts in core). We also had to account for magnetic hysteresis in final demo, but that was easy as it just needed a few adjustments of Kp and Kd. All that hard work brought as 5-10 seconds of levitation at max. It seems your commercial electromagnet has little to no magnetic retention, so you will definitely get better results with PID implementation.
I built this same device years ago (only using an optical sensor rather than a hall effect sensor) when I was an engineering student just for fun. I had the same problem- it would oscillate out of control, no matter how sensitively you tune the threshold. The problem, I realized after some simple computer simulations, is that you are not accounting for the momentum of the suspended object. If you are only triggering the electromagnet based on position, then you have no way to dampen oscillations that will inevitably build up. How I solved this was to create a differentiation circuit that would, in effect, tell me the velocity of the object and then modulated the electromagnet using both the position and differential. A simple PID controller. I don't recall if I used an integration circuit as well so it might have just been a PD controller. Worked like a charm and I had a stable levitating ball bearing.
What you are trying to achieve is exactly what I build as my 'Facharbeit' ~30 years ago. Using a photoelectric sensor and a PID (Proportional-Integral-Differential) regulator made out of of 741's those days it finally worked like a charm. I think the original schematic was described in a very old ELECTOR. BTW. I'm a big fan of your videos!
Thanks for the feedback. Glad that your project worked out fine. I will see whether I might do a basics video about PID an use this levitator as an example.
Not to take away from Great Scott because I too would like such a video.But in the meantime there is some great schematics info in some of Mr Carlson's lab videos if you're interested.But I recall them being just parts of his videos where he designed a project so it wasnt too in depth but still was some of the better stuff i've seen..
Scott, you need to have your magnet rotating to create centripetal force to ack as a gyroscope to balance the magnet at a single point in 3d space. You may want to change the vRef resistor R4 in the 1/2 voltage divider to a POT as you can then tune the circuit by changing the vRref. This way you can decrease the oscillations observed in the initial run of the experiment.
MagLev is a highly non-linear system. By using on-off state of control is insufficient, thus causing the oscillation. A conventional PID ocntroller isn't that encouraged to be implemented in this project too, except you apply model linearization beforehand. And I believe that the performance of the MOSFET isn't LTI (Linear time-invarient), thus suggest that first determine the LTI range of the MOSFET beforehand too. - A control system researcher, a Brian Douglas' huge fans -
I know this video is old, maybe someone can help me. I have bought an levitating Shoes Display but i lost the magnet that i habe to put in the shoe for floating. Can someone please tell me which magnet i can use instead? Thank you so much
Fun! As others have noted, a PID loop will solve this problem. And also don't directly control the on/off, use a PWM and have the PID control the on/off ratio of the PWM.
To make it work you could put a north polarity electromagnet on the top, a south one on the bottom and in the center place a magnet with the sane polarity as the electromagnets. Nice video.
Whole schematic is very simple, so you can expect problems everywhere. If you want to stabilize position of levitating magnet, I would recommend to use analog value from HAL sensor instead of comparator ON/OFF and try to use PID control.
I think if you use a microcontroller as an Arduino an make a PID control system using de analog part of the hall sensor it might work, with the PID you could vary the intensity of the magnetic field in the coil. Great Video!
ok lets just keep it simple: german(native): okay du kannst einfach den ausgang infrarotentfernungssensoren an den pwm duty cycle von nem mosfet treiber anschließen english(not one of my strenghs): you could just use IR distance "sensors" and control the output as an pwm signal duty cycle of a mosfet driver
Plz, show to make a levitating earth. Had this idea once but couldnt imagine how to do it with just an permanent ring magnet. Would love to see a solution. Keep up the great work!
They'd be worth huge money one day. Magnetic forces that are positive and negative connect, now positive and positive. They won't connect in fact rebel from each other. Put weight on one of the magnets while other magnet is in opposing position.
The solution is pretty simple - I tried the same and had the same problem - the hall sensor. It is far way to sensible and even if you try it with a PID controller it is still not working (like in your other video). I tried the H501 and it worked perfectly and very easy without and PID. But nice video :)
I think this video was rather expeditive. I'm not used to seeing you post what i would call an unfinished project really. I wouldn't mind at all if it was part of a longer video, but i'm surprised you didn't even tried anything else apart from slapping those 2 components together and finding it didn't work. I expect a bit more quality coming from you ! I really like your style and your videos nonetheless, hope the next video will be a bit more elaborated !
+Space Core Just because it was not shown in the video does not mean I did not try something else. I was running out of time and had to call it a day. Improvements for this project will follow in another video.
Worst cliffhanger ever! I was already trying to build it... Maybe the board is not fast enough to compensate the changes of height, as you proposed at the end of the video, but it was worth to try changing the position of the board itself. It could work on the side of the magnet, where maybe the variations of magnetic field are less intense. Probably it requires a lot more work to calibrate it though. Amazing video as always :)
Place the hall sensor just under the electromagnet. You are placing the sensor long away from the floating magnet. In this configuration the hall sensor cannot detect the slight change in position of the floating magnet from the magnetic field because of inverse square law. The sensor can see significant change in magnetic field only when there is large displacement of the magnet, but in that case the magnet is already at a long distance from the electromagnet which takes time to get it back to its intended position after switching on the electromagnet. This initiates oscillation. I have done it with similar setup and its nicely stable.
Hi Scott. I think you should place an RC filter in the gate of the MOSFET. The probem could be because the amplification of the closed loop system is too high, and you are using a simple P type compensator that is prone to oscillation. Adding an RC filter would introduce a pole in the system, which placed correctly would dampen the oscillation.
I think a better approach would be to change the comparator with a variable gain noninverting opamp, whose output would feed an RC circuit in the gate of the mosfet. This way the system should behave more linear.
Hey Scott, great video. Surely when you turn on the electromagnet it will disturb the magnetic field enormously and will create a huge hysteresis loop which causes the oscillation. Can you either: 1) add a variable feedback path around the comparator to offset the magnetic feedback, or 2) use a micro controller (or logic circuit) to pulse the electromagnet at high duty cycle and only read the comparator when the electromagnet is off and the Hall sensor is only influenced by the floating magnet?
Great video man! The magnet does not levitate properly because it’s a non linear system. To control it properly you have to linearize it in an operation point and design a linear control for it (maybe not PID). There’s a lot of information out there for this system and you would probably need a MCU for that. Best regards!
Very awesome video! I had always assumed that magnetic levitation just used static magnets, but now it makes sense that there needs to be a dynamic component to make it work correctly.
Great project ,you could place a current carrying loop ( orelectromagnetic sensors )on the sides that turn on the electromagnet as soon as the magnet goes through it.
Thanks for the video! And leaving it open source! I'll be building one to levitate a small flower or cactus, and another one with an induced "Edison" bulb e27 socket. 😊 any tips? I'm planning to put them on a platform.
I had a feeling that an electromagnet and voltage oscillation would be the solution and I don't even do electronics, I just enjoy interesting subjects and content on TH-cam.
I'm pretty sure the coils also need to be different phases. Like in the "Linear Induction Motor". If you want to see magnetic levitation, check out Professor Eric Laithwaite: The Circle of Magnetism - 1968 also "Magnetic River" Here on the "tube" --------th-cam.com/video/0tJfqMYHaQw-/w-d-xo.html----- The guy that invented the "Linear Induction Motor" has a series of videos from the 60's and 70's. He was considered the "Father of the Maglev".
I did a project just like this for university ! I think the problem you are having might be that your electromagnet also affects the HES. To fix this we used a substractor circuit to substract the influence of the electromagnet so that the hall effect sensor always gives an indication of the position of the magnets :) pm me if you want to talk about this
Could you please tell us how to achive revers engineering by measuring the components at 4:43 ? I mean how did you understand the circuit schematic just by measuring it ?
Use a multimeter in continuity mode(or make one yourself that flashes an LED or buzzer when something connects) to see which component connects to others. Keep a piece of paper next to you to sketch and later make a decent schematic. After that measure the resistors and other components for their values. For IC's,check datasheets for typical applications and see if there are similarities with your item.
What if the both magnet has a hole in the center and then try to put a stick in the hole and then position both magnet where both the same pole meet.Will it levitating then?
Hi i have a question i took apart a samsung plasma tv sound bar a d where you connect the wire to speakers before speaker there is a coil for both side left 2 speakers right 2 speakers strange is that while 4 wire was output to speaker bar 2 is left 2 is right and goes like this (2 whire red/white from board to bar connect to a coil then from coil to the 2 speakers from left same for the right.) also power output for sound specified by samsu g was 20watt max. So question why does a speaker need a extra coil?
What if you used the analog pin from the hall sensor and had that control the duty cycle of a pwm signal to the electomagnet. That would seem to keep it from turning on/off, but it would create smoother adjustments as the footing magnet moves
Did you consider an effect of your electromagnetic field on the sensor when you turn it on/off? Do you think it could be delay or that the electric magnet has just on/off states and maybe it would work better with analog output and pwm?
if you have a back emf diode across the mosfet it will make your electromagnet hold the magnetization for a couple ms after you turn current off. you call look it up, they specify this on datasheets of relays.
Try to add some PI compensation, may help you to solve this problem....the comparator simply have too big open loop gain and you need some feedback comp, try to add some RC between V- and the Vout of the OPAMP>
My first thought was three words: Pulse Width Modulation.
Feed the electromagnet with fast on-off pulses and modulate their width to change how strongly it attracts the magnets, depending on the output of the hall effect sensor.
Ok
Excellent idea. A theoretical idea I had was to suspend plasma around a metal core inside of a magnetic field in order to suspend the material indefinitely.
A sensor would be required to adjust for magnetic polarity.
Now, theoretically speaking, is it possible to create such a thing in order to suspend a plasma like substance? I was considering a highly pressurized element about a hundred times more dense than osmium. The idea would be then to contain the element inside a magnetic chamber while being attached to a metallic core that could then be used to assist in suspending the element. The magnetic chamber would function as a shielding. A pin sized actuator could be installed, most likely created by another magnetic force which would allow for an exhaust or discharge in the event that it was needed. Now the idea goes further with plasma, not just a magnetic core, but creating an elemental compound. The compound itself would be suspended by the mixture rather than being held by a core. This new form of theoretical science could be exploited to create artificial gravity or a new system of propulsion for space travel.
How about using - Harmonic Resonance Frequencies -
In a power system when the system's natural frequency corresponds to the frequency of a source of harmonic current, "HRF" Occurs.
Surround the "Magnetic Chamber" in an array of "HRF."
You have so many videos. I've been watching at least one daily for months and I'm still recommended unseen ones every day. Thank you for the content.
Oscillations are inherent phenomena with on/off control, you will probably need to modify your circuit to have some form of PI or PID controller. It would be a good project to show how some of this concepts on a practical example like magnetic levitation or inverted pendulum.
I'am a german guy watching a german guy speaking english 😂
the gramar was perfect but still sounds strange .
@@martinda7446 I'm a russian guy reading about english guy reading about german guy watching german guy speaking english in English
@@martinda7446 I'm a German guy reading about an English guy reading about a Russian guy reading about an English guy reading about a German guy watching a German guy speaking English
@@Max-fh7ij I'm an Indian. I'm simply looking at various foreigners discussing about other foreigners.
Omae va mou.. Shinderiu
A teacher from midschool designed something like that, and it worked great !
He used a lightbulb and a big LDR on the other side (half of the LDR whas shade because the ball floated before the lightbulb), to get an analoge reading of the height. Than the used a PID-regulator to control the coil.
He made it to demonstrate each action of the PID regulator.
Nice project ! :)
Thanks for the feedback. I will integrate a PID next. Maybe even make a basics video while I am at it.
GreatScott! Great :) i always enjoy watching your video's. A great source of insperation
GreatScott! Hey Scott great project as always. You don't really need a PID, PI should be enough. Perhaps try with 2 coils on top of it and 2 hall sensors for better positioning in-between the coils.
The main problem is the two point regulator. It creates a limit cycle in most cases, which causes oscillation (e.g. iron with bi-metal switch). Using a simple P regulator may not help either because of the instable poles of the system. You definitely need at least a PD regulator in order to stabilize the levitation system.
DruMsfR3aK a digital signal is too crude i think, so analoge control of the coil is needed. PD regulator ? So if the magnets fall fast, the D will add will add a burst to compensate ? I think that will oscilate (magnet falls, Burst of coil, magnet rises again and proces repeats). And if the magnet hovers at the desired height, the D action does noting. The D action is usefull when you insert the magnet, to adjust the proces fast if the magnet is inserted. I think a I action is still needed to fine adjust to the perfect current trough the coil.
0:10 greatscott's magnetic fidget spinner
th-cam.com/video/Dp72SwBDc0U/w-d-xo.html
Just stop
Igor Santo Andréa Visioli what language does he speaks?
I guess so!
His penmanship is the most amazing thing about the video.
I absolutely love Magnets and The Great Scott Channel! Better than anything on TV!
4:30 too much caffeine lately? ^^
+X99 Possible
xD
GreatScott! Those tremors you got are pretty common with electronics. Also i found the higher the voltage the bigger the tremors. Every time I wire 400V 3phase my hands are shaking if I hold the cable when electricity flows thru...
X99 mother board
@@greatscottlab why do you rarely reply
Hello Great Scott ! The reason why this doesn’t work is because a magmatic levitation device is a second order non-linear system that can’t be driven by turning ON and OFF the current . You’ll need a PID controller to drive the coil with the proportional current required to levitate. Big fan of your videos! Keep it up !
Keep this up. Very promising for the future of mankind.
Congratulations 700k subscribers!!!
Thanks
First time seeing cute writings with a LEFT HANDER
Another great video, always looking forward to the next one. I would suggest a different circuit which doesn't turn the electromagnet on and off but rather adjust the current flowing through the electromagnet such that it reduces/increases the current at a constant rate according to the hight of the floating magnet. You wouldn't get the oscillations and you'd be able to adjust the height of the magnet very accurately.
You may have failed but you came really close to succeeding and that was cool enough to see for this video.
Is there a pre-made board sold commercially that controls electromagnets like this? I'm looking for one for my levitating sneaker display project but unfortunately I'm not too into the technical electronic stuff and not able to create a board like you did.
BIG FAN FROM INDIA☺☺
Left handed! Dangerously creative 😜
This is a great video, it's extremely easy to follow and understand
Super class Brother
as long as I know, you need to know the transfer function of your system in order to have a better compensator in closed-loop controller. nice video and thanks!
Great video as always :)
We made one maglev project 2 years back. Ours used an arduino UNO and anlaog hall effect sensors(one attached to the bottom of the electromagnet and other similar to how you did). That helped us to apply a simple position based PID loop (actually just PD was used). Also, before adding the "floating" magnet, we measured the full swing of our electromagnet (800 Copper Coils powered by bench power supply with iron nuts in core). We also had to account for magnetic hysteresis in final demo, but that was easy as it just needed a few adjustments of Kp and Kd. All that hard work brought as 5-10 seconds of levitation at max.
It seems your commercial electromagnet has little to no magnetic retention, so you will definitely get better results with PID implementation.
I built this same device years ago (only using an optical sensor rather than a hall effect sensor) when I was an engineering student just for fun. I had the same problem- it would oscillate out of control, no matter how sensitively you tune the threshold. The problem, I realized after some simple computer simulations, is that you are not accounting for the momentum of the suspended object. If you are only triggering the electromagnet based on position, then you have no way to dampen oscillations that will inevitably build up. How I solved this was to create a differentiation circuit that would, in effect, tell me the velocity of the object and then modulated the electromagnet using both the position and differential. A simple PID controller. I don't recall if I used an integration circuit as well so it might have just been a PD controller. Worked like a charm and I had a stable levitating ball bearing.
Salute to person who watch this video in 2024 😂
Really good video. A bang bang controller. It would be interesting to see a video with a pid or optimal controller.
What you are trying to achieve is exactly what I build as my 'Facharbeit' ~30 years ago. Using a photoelectric sensor and a PID (Proportional-Integral-Differential) regulator made out of of 741's those days it finally worked like a charm. I think the original schematic was described in a very old ELECTOR. BTW. I'm a big fan of your videos!
Thanks for the feedback. Glad that your project worked out fine. I will see whether I might do a basics video about PID an use this levitator as an example.
Excellent video as usual
After watching this and reading the comments I feel as if my brain is the size of an ants
THE FORCE IS WITH YOU
Your handwriting is so good wow
You are my teacher iam watching you around 2 years from pakistan love you keepitup
I always love your videos and learn many things. I wish and would love if you keep making such videos.
Thanks, I will try to keep up my production :-)
Hey GreatScott please do a basics video on Schematic Diagrams !! Thanks!
It is on my to do list
Not to take away from Great Scott because I too would like such a video.But in the meantime there is some great schematics info in some of Mr Carlson's lab videos if you're interested.But I recall them being just parts of his videos where he designed a project so it wasnt too in depth but still was some of the better stuff i've seen..
Yes, I'm curios about GreatScott hand schematics "savoir faire"
Okay thanks, I'm going to check that out.
This video is freakin awesome!!!!Hasselhoff rules!!!! Thanks for the vid
I will soon become a patreon, just to support you man
Thanks mate
Scott, you need to have your magnet rotating to create centripetal force to ack as a gyroscope to balance the magnet at a single point in 3d space. You may want to change the vRef resistor R4 in the 1/2 voltage divider to a POT as you can then tune the circuit by changing the vRref. This way you can decrease the oscillations observed in the initial run of the experiment.
UFOs use same concept?
You should try changing the magnets strenght instead of just turning it off.✌🏻
MagLev is a highly non-linear system. By using on-off state of control is insufficient, thus causing the oscillation. A conventional PID ocntroller isn't that encouraged to be implemented in this project too, except you apply model linearization beforehand. And I believe that the performance of the MOSFET isn't LTI (Linear time-invarient), thus suggest that first determine the LTI range of the MOSFET beforehand too.
- A control system researcher, a Brian Douglas' huge fans -
I thought of the same idea
I know this video is old, maybe someone can help me. I have bought an levitating Shoes Display but i lost the magnet that i habe to put in the shoe for floating. Can someone please tell me which magnet i can use instead? Thank you so much
it should not be just "on" and "off":
i think voltage on EM should affinely depend on the voltage on a sensor.(some offset + some koefficient)
This is a great channel
Fun! As others have noted, a PID loop will solve this problem. And also don't directly control the on/off, use a PWM and have the PID control the on/off ratio of the PWM.
Woah 🤩🤩 thanks for this explanation
This is excellent please continue to experiment with magnetic levitation.
I will nominate you for Nobel prize and Oscar...
To make it work you could put a north polarity electromagnet on the top, a south one on the bottom and in the center place a magnet with the sane polarity as the electromagnets.
Nice video.
Whole schematic is very simple, so you can expect problems everywhere.
If you want to stabilize position of levitating magnet, I would recommend to use analog value from HAL sensor instead of comparator ON/OFF and try to use PID control.
I think if you use a microcontroller as an Arduino an make a PID control system using de analog part of the hall sensor it might work, with the PID you could vary the intensity of the magnetic field in the coil. Great Video!
ok lets just keep it simple:
german(native):
okay du kannst einfach den ausgang infrarotentfernungssensoren an den pwm duty cycle von nem mosfet treiber anschließen
english(not one of my strenghs):
you could just use IR distance "sensors" and control the output as an pwm signal duty cycle of a mosfet driver
Would be so interested in seeing you fixing this thing! Love your vids. Good job
you are brilliant man......... good job
No clue what he is saying but I like the blinking lights and the floaty stuff
Wow... i was just waiting for your video. After all it was sunday today.
Plz, show to make a levitating earth. Had this idea once but couldnt imagine how to do it with just an permanent ring magnet. Would love to see a solution. Keep up the great work!
This is just amazing. I've seen Hacker Lab's video with the hoverboard And wondered about how this works. And now this : )
the end of this series is probably "Transrapid DYI" :) great vid
Waiting for a video about isolation and safety. Also good video again.
Well safety is clearly a topic of +Electroboom.
raffmaxi it is lmao
Bros writing is magnificent
I know I'm not the only kid who saw magnets and immediately wanted to make them levitate like this
Geile Sache Alter..... Habe dich abonniert
all the best mate!
Do you keep all of the paper you write on during videos?
@calculatinGenius I would
They'd be worth huge money one day. Magnetic forces that are positive and negative connect, now positive and positive. They won't connect in fact rebel from each other. Put weight on one of the magnets while other magnet is in opposing position.
The solution is pretty simple - I tried the same and had the same problem - the hall sensor.
It is far way to sensible and even if you try it with a PID controller it is still not working (like in your other video). I tried the H501 and it worked perfectly and very easy without and PID. But nice video :)
We got ours to work with the same Hall Effect Sensor... which is EOL to this day... good luck finding them.
We used a PID controller.
I think this video was rather expeditive. I'm not used to seeing you post what i would call an unfinished project really. I wouldn't mind at all if it was part of a longer video, but i'm surprised you didn't even tried anything else apart from slapping those 2 components together and finding it didn't work. I expect a bit more quality coming from you ! I really like your style and your videos nonetheless, hope the next video will be a bit more elaborated !
+Space Core Just because it was not shown in the video does not mean I did not try something else. I was running out of time and had to call it a day. Improvements for this project will follow in another video.
@@greatscottlab where's that another video? 😐
Worst cliffhanger ever! I was already trying to build it...
Maybe the board is not fast enough to compensate the changes of height, as you proposed at the end of the video, but it was worth to try changing the position of the board itself. It could work on the side of the magnet, where maybe the variations of magnetic field are less intense. Probably it requires a lot more work to calibrate it though.
Amazing video as always :)
Place the hall sensor just under the electromagnet. You are placing the sensor long away from the floating magnet. In this configuration the hall sensor cannot detect the slight change in position of the floating magnet from the magnetic field because of inverse square law. The sensor can see significant change in magnetic field only when there is large displacement of the magnet, but in that case the magnet is already at a long distance from the electromagnet which takes time to get it back to its intended position after switching on the electromagnet. This initiates oscillation. I have done it with similar setup and its nicely stable.
Hi Scott. I think you should place an RC filter in the gate of the MOSFET. The probem could be because the amplification of the closed loop system is too high, and you are using a simple P type compensator that is prone to oscillation. Adding an RC filter would introduce a pole in the system, which placed correctly would dampen the oscillation.
I think a better approach would be to change the comparator with a variable gain noninverting opamp, whose output would feed an RC circuit in the gate of the mosfet. This way the system should behave more linear.
Fantastic work, man. :)
Thank you for this video. We will try this too to one of our tutorial video.
Hey Scott, great video.
Surely when you turn on the electromagnet it will disturb the magnetic field enormously and will create a huge hysteresis loop which causes the oscillation.
Can you either: 1) add a variable feedback path around the comparator to offset the magnetic feedback, or 2) use a micro controller (or logic circuit) to pulse the electromagnet at high duty cycle and only read the comparator when the electromagnet is off and the Hall sensor is only influenced by the floating magnet?
Great video man! The magnet does not levitate properly because it’s a non linear system. To control it properly you have to linearize it in an operation point and design a linear control for it (maybe not PID). There’s a lot of information out there for this system and you would probably need a MCU for that. Best regards!
Very awesome video! I had always assumed that magnetic levitation just used static magnets, but now it makes sense that there needs to be a dynamic component to make it work correctly.
Great project ,you could place a current carrying loop ( orelectromagnetic sensors )on the sides that turn on the electromagnet as soon as the magnet goes through it.
Finally! I have been waiting so much for this video!
Another amazing video! I learn so much from your videos. Thanks again!
Thanks for the video! And leaving it open source! I'll be building one to levitate a small flower or cactus, and another one with an induced "Edison" bulb e27 socket. 😊 any tips? I'm planning to put them on a platform.
Dude you're straight up awesome!!!
I had a feeling that an electromagnet and voltage oscillation would be the solution and I don't even do electronics, I just enjoy interesting subjects and content on TH-cam.
You need two coils in the top wound opposite each other and this will work either way from the top or bottom. That is how mag lev trains work.
I'm pretty sure the coils also need to be different phases.
Like in the "Linear Induction Motor".
If you want to see magnetic levitation, check out Professor Eric Laithwaite:
The Circle of Magnetism - 1968 also "Magnetic River" Here on the "tube"
--------th-cam.com/video/0tJfqMYHaQw-/w-d-xo.html-----
The guy that invented the "Linear Induction Motor" has a series of videos from the 60's and 70's. He was considered the "Father of the Maglev".
I think you might need a proper PID regulation with a PWM signal driving the MOSFET for this one to work. Great video.
air core too
Love your work your say real fact. Thanks
People liked the video so much that they turned their screen around and pushed the like button
I did a project just like this for university ! I think the problem you are having might be that your electromagnet also affects the HES. To fix this we used a substractor circuit to substract the influence of the electromagnet so that the hall effect sensor always gives an indication of the position of the magnets :) pm me if you want to talk about this
Cant wait for the next one.
Hey bro u have given an awesome idea for the minor project thanks. Always love your video.
Could you please tell us how to achive revers engineering by measuring the components at 4:43 ? I mean how did you understand the circuit schematic just by measuring it ?
Arduino Hocam By now this guy is like Neo's companions reading the Matrix
Use a multimeter in continuity mode(or make one yourself that flashes an LED or buzzer when something connects) to see which component connects to others. Keep a piece of paper next to you to sketch and later make a decent schematic. After that measure the resistors and other components for their values. For IC's,check datasheets for typical applications and see if there are similarities with your item.
www.google.co.in/amp/www.instructables.com/id/How-to-reverse-engineer-a-schematic-from-a-circuit/%3famp_page=true
lah
You check which lead goes where with, dozens of times.
Awesome try man ...stay creative
Yey
U make my idea posted on facebook for random q&a video!
Thank you!
What if the both magnet has a hole in the center and then try to put a stick in the hole and then position both magnet where both the same pole meet.Will it levitating then?
Hi i have a question i took apart a samsung plasma tv sound bar a d where you connect the wire to speakers before speaker there is a coil for both side left 2 speakers right 2 speakers strange is that while 4 wire was output to speaker bar 2 is left 2 is right and goes like this (2 whire red/white from board to bar connect to a coil then from coil to the 2 speakers from left same for the right.) also power output for sound specified by samsu g was 20watt max. So question why does a speaker need a extra coil?
Scott IS pretty great
What if you used the analog pin from the hall sensor and had that control the duty cycle of a pwm signal to the electomagnet.
That would seem to keep it from turning on/off, but it would create smoother adjustments as the footing magnet moves
8:35 I saw a spoiler for the next DIY or BUY... ;)
A French Gun Guy oh, un français
I like your videos!
Probably adding a PID control might improve it?
Hey Great Scott, nice video. I have a question. Why did you use a mosfet driver? Couldn't you drive the gate directly?
If you want your MOSFET to switch as fast as possible, you use a MOSFET driver.
would running the hallefect sensor as an anolog supply to control a pwm signal to drive the magnet be an answer ??..
Did you consider an effect of your electromagnetic field on the sensor when you turn it on/off? Do you think it could be delay or that the electric magnet has just on/off states and maybe it would work better with analog output and pwm?
The MF does not influence the sensor. But I think using PWM with a PID control should solve all the problems.
GreatScott! Really? Why not? Is the Hall sensor on the base then?
if you have a back emf diode across the mosfet it will make your electromagnet hold the magnetization for a couple ms after you turn current off. you call look it up, they specify this on datasheets of relays.
I liked very much this video, it was amazing, !
Try to add some PI compensation, may help you to solve this problem....the comparator simply have too big open loop gain and you need some feedback comp, try to add some RC between V- and the Vout of the OPAMP>