The magnetron's output is non-ionizing, which means there is no risk of having cells' DNA restructured by the radiation. So, no radiation sickness or anything like that. The biggest hazard is thermal -- The RF can cause localized burns that are difficult to detect since they may affect internal organs where there aren't many nerve endings. Having said this, the risk is pretty low. Pilots in WW2 used to stand in front of radar antennae intentionally to keep warm, but it caused eye problems.

The transformer is designed to saturate in an attempt to keep the current constant. The magnetron won't oscillate until a certain voltage is reached. When this voltage is reached, it will continue to oscillate while the voltage drops to a certain low point. The square wave is the magnetron firing and the transformer saturating and keeping the current somewhat regulated.

Loving your videos! The frequency of oscillation is actually a function of the LC tank circuit each of the cavities makes. So they are tuned to 2.45GHz. The magnets are there to provide centripetal force to the electrons so they spin around in a helical pattern to impart energy into each of the cavities as they spin around. Because of mutual coupling, placing an antenna in any one cavity will couple the energy of all the cavities at the same time. Again great video!

You just explained in the first minute what several videos I've watched failed to clearly explain.

Very interesting measurements. Microwave ovens are indeed potentially one of the deadliest consumer products found in the home, especially when it comes to servicing. Thanks for sharing.

Awesome video as always, Ben! Can't wait to see where you're going with the magnetron experiments. One quick safety tip that was given to me by an old Television tech: Be careful of outlet strips on the front edge of your workbench. If you get hit with high voltage, the ground/neutral can arc over to metal zippers etc. I don't know if 4Kv will do it - but he swore up and down that he'd gotten the shock of a lifetime from a color TV (I'm guessing 15Kv or so)

You could measure the voltage on both sides of the capacitor. Then use the dv/dt to calculate the current on the cap during the transmit part of the phase.

@Applied Science: Thank you so much... Its so clear... I realize your explanations are easily understood because they reflect reality. I've been watching other videos and reading articles trying to understand the operation of microwaves, but many others present ideas that are confusing and sometimes conterdicting (and other times plain false), and I realize its probably because its purely theoretical for them, they haven't put their ideas to the test, they haven't needed to use that knowlege to build something with it; that "knowlege" just sits there, and so it might as well be false, and it will still sit there. But you explain mostly what you've seen with your own eyes, or what you understand because you use it, either directly or as building blocks for other experiments or projects. I thank you for being so generous and sharing your projects and your ideas.

Very interesting project, i will be looking forward to the next video for sure!

Did you ever get around to hooking it up to a capacitor bank?

If I understand it correctly, one half-period is clipped by the forward voltage of the diode and the other half-period is clipped by the magnetron itself. A slight bow can be seen on the scope in the second half-period indicating fairly low differential resistance there.

Another excellent highly informative video. I had the same current probe as you (until recently) and your right it's not very accurate. I also noticed the same time delay on measurement. Keep up the good work! :D

"I wanted to play with a magnetron... So I took my microwave apart :D" Win!

"i wanted to play with a magnetron" so a average tuesday?

The power supply is more complex then you diagram. Look carefully at the transformer. You will note a magnetic shunt between the primary side (where the 120VAC comes in, and also the filament 4V winding is) and the secondary side. This type of supply is a "ferroresonant regulator". The cap does more then double the voltage; it is in series resonance (at 60 Hz) with the leakage inductance (created by the magnetic shunt) of the secondary. Because of this resonance, the output voltage would build up "to infinity", (like a tesla coil) if it were not for "some limiting mechanism". That limiting mechanism is the saturation of the steel in the magnetic shunt. So the power supply is a sinusoidal (half-wave) current source. It must be, as a magnatron is an "obnoxious" load, like a gas tube, with negative resistance. This regulator is used in mercury vapor and sodium vapor streetlights, and was used in old (heavy!) computer regulated supplies. To work properly, only the frequency needs be correct; line voltage can vary a lot. (Topaz "constant voltage transformers" also work by the same principle.) If you try to push more current through a magnetron then the cathode can emit, the current abruptly cuts off (in nanoseconds!) resulting in a huge spike. For this reason, the diode has a special "soft" characteristic to damp this. (At turn on, the cathode is cold, so no current flows. As it heats up, current starts to flow at the start of current sine wave, where the current is low. When the current exceeds the emission of the (not yet totally hot) cathode, and current suddenly cuts off. As the cathode gets hotter, conduction time gets longer until suddenly cathode is hot enough to allow conduction the complete half-cycle. Listen carefully as oven is turned on. You will hear a "ba-zap" sound like Darth Vader's laser sword being powered up. That sound is produced by the changing phase shift of the cutoff, then abruptly stops when cathode is hot enough. If you put scope in long record mode, you should be able to see the changing current cutoff, if the EMI spike doesn't hide it.) I mention this because I think magnetron will "stall" if you try to "shove" spike of current through. Even if it doesn't, the cathode will limit the current because of its limited thermionic emission. A huge electric field will build up at the cathode, maybe leading to destructive arc (blowing off the thorium at the surface of the thoriated tungsten filament). If magnetron "works", it may arc over at the output probe. Radar tubes are typically in pressurized container, some with SF6, up to the point where the horn antenna is flared enough to prevent flashover. Newer microwave ovens have switching supply. I don't know the details how they work, other then having resonant transformer (with litz wire primary) and 2 diodes in a voltage doubler configuration. I have photos of newer supplies.

If I'm not mistaken, microwave ovens are rated by their output power, not their input power. That looks like an 1100 watt microwave (Sharp R308N or similar). Based on Ben's measurement of 1.33 kW power to the magnetron, that would make it 83% efficient, which is about what you would expect for a magnetron.

Sweet! After years of taking apart and fixing MOs, I finally understand how the cap/diode circuit works. It's nothing more than a voltage doubler!

Your voltage graph should be roughly an offset sine-wave. I don't see how it becomes nearly a square wave unless you have clipping somewhere.

I'm curious about your further microwave experiments. I can't wait for the next video. :)

Could the electrons that circulate through the cavity of the magnetron, since they're moving as if in a coil of wire (even though they're moving through a vacuum) act as an inductor, causing the current to lag the voltage?

What happens when you use the "power level" setting on the microwave? Does the it pulse less frequently, or reduce the power of the pulses?

+Applied Science Part of the time delay between voltage and current may be due to the fact that many electrons return to the filament (after having lost energy by creating microwave photons). They make the filament hotter by hitting it. Hotter filaments emit more electrons (more current), but as a nonlinear regenerative process, it takes a while to get going. The overall process causes some undesired Frequency Modulation in the 2.45 GHz output. Also, when the HV section starts to flow real current, that robs some power from the filament winding. It would be interesting to see how things would behave if the filament was powered independently.

This project sounds really interesting.

I once presented a small research on microwave cavity resonators and simulating them. After the presentation, a member of the "jury" told me that I'd get 3x the power out of a microwave oven magnetron if I cooled it. Run cold water trough the vanes?

Really like your videos. Perhaps a caution to others would be a good idea. I worked on these for several years, and they can be LEATHAL. I was at a makers tutorial and we heard that at the last course someone was killed! Lots of power here. Regards... Eric

Hi nice video.i want to ask one question .do we have different magnetron rating is it all same or ...?I noticed from articles that they have a filament which is mostly supplied 3.2V ac. Do we have any other such parameters ?I wonder what really makes the microwave output power higher?is it the high voltage of the filament voltage ?

Or rather more simply, a 10r shunt, with a resistor, battery and led across it. An optocoupler would work fine too, but air gaps are good.

Excuse me,can the power of magnetron decreased to 10~50W through by the changing of voltage ?

that is the coolest scope ever

Thanks for explaining that, I've always wondered how it worked :)

Did you look at 117VAC current draw with the magnetron disconnected? Just curious about the power transformer magnetizing current.

Why is the range so weak on such powerful magnetrons? I see videos of people putting cans on the end but it still seemed limited to around a foot in front of it. It would be good if you could make a beam so you could then use it for some point to point wireless powered devices.

Which is the easiest way to get about 25KV from a magnetron output?

Does the tube have a collector plate and grid?

What is the benefit of putting 1 ohm resistor between test point and HV probe?

Um... In high voltage circuit there is a 2100V 1uF capacitor in series. This is half-wave voltage doubler. During half wave it is charged ( until 2100V , about 2.2J) and then release 4.4 joule at another half wave..... therefore the HV side consumes 4.4 x 60= 264J per second in case of 60Hz... So magnetron HV wattage is 264W! And mean current is 264w/2100V=0.125A This is my opinion. Is it right?

May I ask on what do you intend to do with a several kW pulse microwave device? DIY EMP maybe? :D

would it be possible to use 2 MOT to produce the high voltage needed instead of voltage doublers? also, would you do an episode of focusing the microwave through a parabolic dish of some sort?

- Darling: Did you see the microwave? - AS: I don’t know… I think we’ll have to buy another one (just in case something goes wrong) (parenthesis whispering to himself) LOL 😂

I believe the2 ms. The current is space charge limited as the electrons just spin around until the voltage is high enough for the electron frequency to match the resonant frequency of the RF structure. At this point energy is coupled out of the beam, orbits decay and electrons hit the anode, removing themselves from the electron cloud. Current flows.

can we adjust the output frequency of the electric current from a magnetron, sir?

i have a 12kJ 16kV cap. I wonder if breakdown would occur at those voltages.

I don't know exactly what this project you're planning is going to be, but if it involves high power cap discharge through a magnetron, I know it's going to be... Interesting!

So... The lag between voltage and current is probably correct. You noted that you thought that it might be an artifact of your probe, but that lag is partly why cavity magnetrons are not used in high power communications. Next, they have harmonic overtones... 1/2 wave, 1/3 wave, 1/4 wave etc.... and each of those produce a disruptive tone which can actually reach resonance before you hit the frequency that the magnetron is supposed to be tuned to. This is something you probably need to know, but you're probably also aware of. As a quick note, if you can find a tone not absorbed by water, sugar, and oils, that might be safer to work with, but if you're trying to flash cook nm. (aka only helpful if you're making an EMdrive) Thirdly, Cavity magnetrons are voltage controlled. The calculation is based on the distance of the anode from the cathode but it also involves the magnetic field. If you want better control over the magnetron for more precision you should actually remove the permanent magnets and use an electromagnetic coil (like a solenoid). The magnetic current through the permanent magnets controls the electron bunching through the magnetron; if replaced with an electromagnetic coil, the voltage through the electromagnet could be used to control the frequency. Finally... Especially you with your awesome lab, could actually make a cavity magnetron tuned to a frequency that isn't lossy (aka not water and fat). I've been kinda working on it but i figured it'd be a bit before i made a video about it..... here's some math that I've noted... Frequency = (((Magnetic permability of free space)(dielectric constant of material in the not metal parts of cavity magnetron)(circumference of 1 resonant cavity)(length of magnetron tube * distance of the resonant cavity from the cathode))) / (Length of magnetron tube * distance or break in circumference of resonant cavity) If you notice, the length of the magnetron can actually be dropped out and you can simply look at the magnetron in it's entirety as a circular cross section. I suspect that the break in the resonant cavity should also appear as a subtraction term in the circumference of the resonant cavity, but i might be wrong. Any way... i spent the last few weeks incidentally looking into microwaves and thought i might be helpful. Don't cook yourself with a stray microwave emitter, and I can send better mathematical formula if this helped at all. If you made a magnetron, there's a few other things i can throw up like pi strapping, resonance conditions, and some things that effect efficiency.

I've posted a video in response that shows an interesting artefact of the 50% duty cycle of microwave ovens. I filmed a CD and some steel wool getting zapped at 600 and 1200fps and there is a very clear pulsing to the sparks and flames as the microwave cavity is alternately filled and emptied of microwaves (as they were absorbed by the CD, steel wool)

what engineer category do you come in ???? i was just curious to know :D

Good idea!

Hehe opening shot: pointing at high-voltage components with a graphite-filled stick - good plan :-D

If positive output is connected with MOV body and cabinet why not it's dangerous to touch the cabinet ( even at negative voltage at output ! ) Or it is because ... ! The output is isolated from mains so touching the cabinet can't close the circuit or something similar ??

I've always wanted to do experiments with magnetrons but i don't want to risk playing with radiation. Hopefully with your videos you can talk a little bit about safely working with a magnetron. I have acquired probably 8 microwaves over the last 2 or 3 years, but I use the transformer for a homemade arc welder.

So when are we going to see some magnetron projects?

Frequency determined by size of cavities ?

Could you build a "maser" or something?

Measuring the current in the high voltage side is quite simple. Get a floating V-F (voltage-to-frequency) converter -- power from a battery and let it float on the HV. Have it drive an LED into a length of fiber optic, build a simple receiver on the other end, measure with a scope or a freq. meter. The V-F converter can be based on an op-amp and 555, like in Feb 21, 2008 EDN Design Ideas, or use a V-F chip. A bare 555 will do V-T conversion, most freq meters and scopes measure pulse length too.

Sweet Scope!

i like the video, it helps me understand a lot.i have one thing to ask ,normally most of the time as i observed most of the magnetrons are supplied with a 3 to 4V Ac and up to 10 A current flowing through the filament. what will happen if the supplied voltage to the filament is increased or decreased? what makes the microwave output power high or low ?is is the filament voltage variation or the high voltage side variation ?

It's worth mentioning there are hazardous materials in those tubes, possibly deadly so dispose of them properly

Ah, that should be interesting... One more question - will the oil diffusion pump be capable of delivering the required vacuum levels? Anyways, good luck ;-)

WHY?

Ooh, LeCroy, nice!

From whipped chocolate making to pulsed energy weapons :-D But jokes aside, it's really interesting to know what you gonna do with that. I guess I just have to wait for the next video..

Too late, fail... Anyway, great video. Looking forward to the next high-power RF experiment!

I would guess that the current lagging the voltage is the power factor caused by the large transformer acting as an inductor.. its interesting that the capacitor is oversized quite nicely and that it isn't dropping voltage near the end of the waveform, I would have thought manufacturers would have put the smallest cap in that would only just get the job done.

Projects in mind, stopping russian tanks GPS with a re-tuned microwave oven. If its off tuned from water it might be safe with the door open. Or a horn. Safe is relative.

Maybe you can measure the HV current by clamping the clampmeter on both filament wires?

I love the "How it works" at 4:00! Since the capacitor is salvageable from the microwave oven, there isn't any specialized components to get! Nice! I wonder what you are going to do with it. Build a microwave gun like the military? Too weak...

I had a few questions about magnetrons 1) How do you calculate the amount of electrons emitted from the cathode? 2) How the amount of electrons emitted from cahtode and the voltage between the anode and cathode affect the power output of the magnetron? Youre help is greatly appreciated.

And if you're a ham radio operator, you're probably making a transmitter with the magnetron. :)

4:15 Did you mean during the positive half of the cycle its doing nothing but charging the capacitor?

Use an optocoupler to measure currents in dangerous areas.

According to what you said instead of permanent magnets one can use electromagnet. With changing voltage on the electromagnet the frequency should change as well. A way of FM modulation. Or it works differently than I think.

Surprised you didn't just unbolt the transformer and diode from the chassis and insert your shunt there; then ground voltage is maintained for safety and you can measure current via ohm's law. As to the average power, if the peak is 1A and the waveform is roughly sinusoidal, then you should have average of 707mA by RMS. So you have 2.8kW half the time (halfwave rectification), thus 1.4kW average input. I'm not seeing anything out of the ordinary here.

Why is your voltage output a square wave? It should never be a perfect square since the start of the negative 60Hz cycle voltage should be just the peak of 2200v. In the middle of the negative cycle, it will not be 4400V eithrr because the capacitor already discharged some of its energy. Towards the end of the 60Hz negative cycle, the voltage will diminish close to zero, or if the capacitor is big enough, to 1800v to 2000v range. When you have load, ie you have food or water in the oven, the capacitor will never be big enough

i feel so stupid when i keep watching this guys vids.

Hookup your ultracap bank to the magnetron to do some cool plasma experiments!

Happened to me. I was a kid and my dad was repairing a TV. It was not plugged in, but I guess some capacitors were still charged, because when I touched the anode, (the part looking like a suction cup, I thought it was rubber, but apparently I touched some bare metal) all the muscles in my body suddenly went sore. I dunno what was the entry/exit point though. I think I was barefoot on the cement floor in our basement, so it must have been the ground.

"Then the wife yelled at me so I told her 'shutup woman, this is for science!'"

imma chargin my microwave!

I would disect my microwave (unless it broke,) but I have three of them I've collected from my sister's apartments/dorms over the years in my garage...

Hi, the magnetron of my microwave it´s broken , so i need to replace it . The magnetron is a Galanz M24FB-210A G0208o12 GAL02 GEMC . My questions are: 1. Can i replace it with another model, or it must be identical to the original one? 2. What is the GAL02 Number represents? Thank you !

Only God can help the amount of harmonics and power factor that these devices are putting on the grid

Get a second microwave, pull the power supply and connect it to the 1st magnetron with opposite phasing with 2 extra HV diodes to double the output power. Get a small squirrel cage blower and mount it against the magnetism to increase the cooling. Or water cool it..

The vacuum tube itself is a rather good capacitor ;-)

For a fixed magnetic filled, the I V curve is actually much steeper than you have drawn. There is a threshold of electron angular velocity below which the tube will not oscillate, therefore no energy or velocity is taken from the electron cloud and the current is limited by space charge to essentially zero current. If you detected RF power you will see perfect alignment with RF and cathode current. Once the tube is making power the impedance is very low and current is limited by the leakage inductance and resistance of the transformer.

ha i just got a magnetron and a 2,1kv 1µf cap from a microwave someone threw yesterday, sadly the diode was busted and thered or the microwave was missing..... anyway this video came just in time XD

To our younger scientists and electronic engineers I'd like to point out that working with magnetrons is serious and is extremely dangerous, embolisms (microwaves can and do evaporate subcutaneous fluids) and cancers (DNA and RNA damage) come immediately to mind in addition to burns and very possible fatal electrocutions. As soon as you start work on a microwave your are on the wild side of dangerous explorations. Inform yourself of all precautions before looking under the hood.

Discharge a cap bank into the magnetron? The FCC is gonna come knockin at your door. :)

Anyone that enjoyed this video will also like the video on microwaves that the Engineer guy just released, I would highly recommend subscribing to him as well. Chanel: engineerguyvideo It's amazing that these two videos came out within a day of each other!

I just hope your not trying to make another neural stimulator using a microwave transformer to power it... just the way you talked about discharging voltage when it hits a certain point and charging a bank if capacitors.. oh dear.

May be your should not =be in a habit of pointing with a pencil... or it could give you a shock through the lead.

i feel so sleepy in your class.. can you summarize it??

Great video. You also forgot to mention... Do not try this at home!

Until you get blinded.