I recently took apart a microwave, and I managed to get a magnetron out of it, as well as a transformer that appears to be the power source for the magnetron, but looks a little small for the advertised amount of power this magnetron can handle (~1 kW). Also, the magnetron does not have any leads for supplying the filament voltage, just for the HV. I would like to get that magnetron running, I plan to put on a waveguide and do a bit of experimentation with microwaves.

1) Does the magnetron have something built in to step down a fraction of the power recieved to the 3.3V required to heat the filament?

2) Is it possible to run the magnetron simply by powering the microwave transformer with mains and hooking up the hv output to the input of the magnetron, or do I need a microwave capacitor? I was not able to locate a capacitor in the microwave, I think that might be because it is an inverter-style microwave.

  • 16
    \$\begingroup\$ darwinawards.com \$\endgroup\$
    – EM Fields
    Commented Sep 15, 2014 at 0:25
  • \$\begingroup\$ If you are saying what I think you are, then my response is: There was not an HV capacitor in this microwave, like I said it is an inverter-style which at least in this case appears to use transistors to regulate power instead of an HV cap. I have taken apart another microwave before this one and seen the capacitor, so I know what they look like and would not be dumb enough to touch one (hopefully). Also, this microwave has not been in operation for quite some time, and so the cap would have discharged long by now, especially if it had a bleeder resistor. \$\endgroup\$
    – user16871
    Commented Sep 15, 2014 at 0:56
  • \$\begingroup\$ @user16871, OK sorry I'll delete my comment. Do be careful though. \$\endgroup\$ Commented Sep 15, 2014 at 13:51
  • 2
    \$\begingroup\$ You do realize, that when I can buy one for 70 dollars in the store, that the manufacturer has already answered your question. \$\endgroup\$
    – gbarry
    Commented Nov 14, 2014 at 17:52

3 Answers 3


I believe the way microwave oven magnetrons are wired is that there are two coils on the transformer more or less in series, one to supply the HV and one to supply filament. The filament doubles as the cathode and it has some series inductors to keep the RF from leaking out the power terminals. The HV return from the anode could just be the metal case as it is less than an amp. An inverter-driven microwave likely has the same setup, however the transformer will be smaller as the switching frequency will be much higher (60 Hz vs several kHz). The inverter should have essentially the same connection to the magnetron as the older style single transformer solution. Also, you won't be able to run the inverter transformer by itself, you'll also need the rest of the inverter drive and control electronics. I'm not sure if it will work without the front panel/user interface; it may be required to turn the inverter on and possibly select the power level in some way.

However, turning on a magnetron outside of a case is very dangerous. Not only could you kill yourself with the high voltage, the high power microwaves could give you serious burns as well as possibly causing interference and damage to electronic deivces. The magnetron could also overheat due to reflected power if it is not properly coupled to a waveguide.

Edit: Looks like the inverter microwaves operate on exactly the same principle as the older ones, just at a much higher frequency. You'll either need to keep the front panel or figure out what signal it sends to the inverter to turn it on.

Panasonic inverter schematic

Image from http://www.electronicspoint.com/threads/microwave-inverter.234684/


Acknowledging that this is an old question...

  1. Does the magnetron have something built in to step down a fraction of the power received to the 3.3V required to heat the filament?

Sort of... The power circuit generates the low power DC for the heater and the high power AC for the tube and the magnetron separates them. On the bottom of the magnetron, just below the input terminals, there is usually a square crimped-on cover. With that cover removed you should see two heavy gauge leads that run to a pair of fat RF chokes. These filter out the high voltage AC and leave the low voltage DC required for the filament heater.

  1. Is it possible to run the magnetron simply by powering the microwave transformer with mains and hooking up the hv output to the input of the magnetron, or do I need a microwave capacitor? I was not able to locate a capacitor in the microwave, I think that might be because it is an inverter-style microwave.

No. If your goal is to eliminate the microwave control circuitry you'll need to keep several components. For a non-inverter microwave that would be at least a fuse, transformer capacitor, and diode. Most microwaves have a schematic in them. Pay attention to the relay circuit that turns the magnetron drive circuit on and off. Connecting AC mains power to the magnetron will very likely destroy it in very short order.

This link can answer many of your questions. Do be careful, an operating magnetron is like a 1,000 Watt Lightbulb. Close up or as a directed beam it can burn you and make you blind faster than you can blink. Additionally, it can destroy electronics at obscene distances. I have yet to see one equipped with an undo button.

  • \$\begingroup\$ @PeterR.McMahon This is better posted as an answer instead of a comment. I know they won't work off Mains voltages, I was quoting the original question. Thanks. \$\endgroup\$ Commented Aug 27, 2020 at 13:29

It is an inverter type. The transformer will not work on the mains. It needs a much higher frequency and the current needs to be controlled, using pulse width modulation, which is done by the control circuit. The magnetron acts like a 4000V zener and needs to be run off a current source or high impedance source. According to the diagram, there is a small extra secondary on the transformer to generate the 3.3V for the filament, which is also the cathode (known as a directly heated cathode). The 2 inductors block any RF coming out through the cathode. In one magnetron, they appeared to be tuned, as the turns were separated, which leads me to believe that they may have a second purpose: to smooth the cathode current & improve efficiency by preventing current surges as the spoked electron cloud inside switched from 1 set of vanes (which are now negative, to the next ones, which are positive). They have nothing to do with generating the 3.3V, which is between the lugs. Click on the address below to see how the magnetron works. You can use a mains frequency transformer from another microwave. The mains frequency circuit has a high leakage transformer which has magnetic shunts with air gaps, between the primary & secondary, so some of the flux can leak across the gap, bypassing the secondary, making it act like it has an inductor in series. This gives it a high impedance output so it can accommodate the square wave across the magnetron. The capacitor will have an odd looking value like 1.19uF. As well as forming a voltage doubler with the diode, it is designed to resonate with the secondary to bring the voltage back up, while retaining the high impedance. Magnetron Working Principle

enter image description here

enter image description here

  • \$\begingroup\$ Source of diagram: my own experiments. \$\endgroup\$ Commented Sep 16, 2020 at 3:10
  • \$\begingroup\$ Could you explain more about the High Leakage transformer ? \$\endgroup\$
    – EJE
    Commented Sep 1, 2022 at 18:21
  • \$\begingroup\$ @EJE The high leakage transformer has magnetic shunts (laminated steel blocks) between the center part of the core and the outer parts, with an air gap, and are placed between the primary and secondary. With no load, the flux takes the easier path through the secondary. But when loaded, the secondary current opposes the changes in the flux, and tries to average it out to zero. The primary would normally force it through, but, with the shunts, the flux is able to bypass the secondary via the air gaps, allowing the secondary voltage to drop under load like there is an inductor in series. \$\endgroup\$ Commented Sep 10, 2022 at 6:44
  • \$\begingroup\$ Thank you for the clarification. I agree with @Transistor comments, as the Image should be used as a ‘Supplement’ on this original answer, including some textual clarification as you provided. Few curiosities, probably of no major magnetic significance: Transformer’s main magnetic core is vertically laminated. The magnetic shunts were represented as horizontal lamination. (1) Was is just a graphic representation or it really happens? (2) Being somehow ‘loose’ in that gap, don’t they move? Or both shunt halves are interconnected (then being horizontal laminated makes more sense)? Thanks again. \$\endgroup\$
    – EJE
    Commented Sep 10, 2022 at 12:52
  • 1
    \$\begingroup\$ @EJE That is how it is constructed. The magnetic shunts are slid in from the front or back, with insulation packed around them. If they were stacked the same way as the core, they would have to be glued together before inserting. As long as the flux goes along the laminations, it doesn't matter how they are stacked. They are not interconnected, just held in place by the insulation, and maybe, varnish as well. The air gaps can be on the inside, outside, or a bit of both. Electric arc welders used the same type of transformer, but with shunts that are wound in or out, to adjust the current. \$\endgroup\$ Commented Sep 12, 2022 at 7:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.