High Frequency Electromagnet Possible?

[blainiac]

I was wanting to build a very high frequency electromagnet to test an idea and it requires something in the range of about 1-10 GHz, and I'm unable to find much information on what I'm attempting to do. I understand the impedance will be incredible, along with some other technical issues. Most of the stuff I've researched has to do with antennae and near/far field stuff, and I'm not sure if an electromagnet in this range will behave similarly.

There were two approaches I was thinking of. Looking into induction heaters and seeing if any are in the range somehow, or going with an LC tank circuit.

It boils down to this: Is there any way to build a 1-10 GHz electromagnet?

 

[berkeman]

(Moved from DIY to EE for now because it requires fairly sophisticated answers)

Well, the nH inductors on my 2.4GHz radio PCBA are basically little electromagnets, but the fields generated at the ends of those 0402 SMT inductors are pretty small.

Your application must involve pretty small volumes, right? What B-field magnitudes in what test volumes are you targeting? How much power are you willing to invest in this apparatus?

 

[blainiac]

Thank you for the response berkeman, it gives me hope this might be a possibility.

To the best of my knowledge I would say somewhere between 5 mT to 50 mT (something comparable to a bar magnet or even a small neodymium magnet if possible). The test volume would be about 3-4 cm³. If these field strengths are possible at a GHz frequency, I'm game for any power requirements!

This project means a lot so thank you.

 

[berkeman]

The test volume would be about 3-4 cm3.

Maybe use a small microwave oven?

 

[jim hardy]

Are not B and E fields inseparable ? At significant frequency they are best treated as radio waves.

Sounds to me like he needs a magnetron...
But, tinkering with microwave ovens is really dangerous.

I don't understand the purpose of a gigahertz electromagnet. Other than to heat something in its field...

 

[jim hardy]

Maybe use a small microwave oven?

Don't do what two guys at a sawmill did - they cut a hole through a microwave oven to use it for drying lumber. That endangers eyes .

 

[berkeman]

they cut a hole through a microwave oven to use it for drying lumber.

LOL

So @blainiac -- can you say a bit about what you are wanting to do? As you can see, working with these frequencies at these powers is not for amateurs with little experience. Maybe you could use simulation software like ANSYS or COMSOL to work on your idea in a safer way...?

https://www.researchgate.net/post/Is_COMSOL_multiphysics_or_ANSYS_more_powerful_for_simulating_electrochemical_processes

 

[blainiac]

LOL So @blainiac -- can you say a bit about what you are wanting to do? As you can see, working with these frequencies at these powers is not for amateurs with little experience. Maybe you could use simulation software like ANSYS or COMSOL to work on your idea in a safer way...? https://www.researchgate.net/post/Is_COMSOL_multiphysics_or_ANSYS_more_powerful_for_simulating_electrochemical_processes

Thank you Jim and berkeman for the replies and assistance, it's been helpful. I've been reading about mH inductors and some other LC tank designs but my background is programming. As for simulation 'software', I have been using an electrodynamics Java applet that simulates what I'm hoping to achieve as I had envisioned.

I'm wanting to see what happens when two electromagnets 90° out of phase and separated by a distance (d) of about 3 cm (1/4 of the wavelength of 2.45 GHz). The idea is that it takes a time of t=d/c for changes in the electromagnetic field to propagate to the other electromagnet. Each electromagnet will 'see' and react to the state of the other electromagnet as it was in the past (d/c). The idea was that at this frequency, phase, and distance, one electromagnet would always see the other as in phase, and the other electromagnet would see the other as 180° out of phase, leading to a net force with the momentum being conserved by the momentum being carried away via the electromagnetic field.

I thought of this years ago, and I would not be surprised if many other people have thought along similar lines. I'm sure if it were possible, it would have been done fifty years ago. I don't know if this will work or be anything more than a 'photon rocket' or directional antenna. I'm hoping that the force would be similar to two DC electromagnets separated at the distance, but I'm afraid it would be more antenna-like than electromagnet-like.

 

[jim hardy]

Sounds to me like you are tinkering with the idea underlying phased array antennas.
At your frequency , an electromagnet hasn't time to incite perceptible motion in familiar everyday objects we can see and feel
so my instinct is to look at your 'electromagnet' as an RF field source.

http://www.radartutorial.eu/06.antennas/Phased Array Antenna.en.html
has a nice animation .

 

[marcusl]

I don't get what you are proposing, much of it doesn't make physics sense. What is the level of your E&M education? Have you taken a senior- or graduate-level course?
Am I correct that you want to energize two coaxial wire rings that are λ/2 in diameter and λ/4 apart at 2.45 GHz to generate magnetic fields? There are a number of problems with just this part. First, the ring is over 3λ/2 in circumference, so you cannot support a constant current--the wire acts as a kind of transmission line that supports higher-order modes than the azimuthally symmetric one you want (mode 0). Second, each ring produces EM waves as was pointed out above, not magnetic fields. To get only B fields locally, the generating structure must be a tiny compared to a wavelength and you must be next to or inside the structure. If your ring were, say, λ/60 in diameter (~λ/20 in circumference), then your λ/4 distance to the next ring puts you in the far-field and, once again, it sees EM waves. Finally, you mention a net force. Assuming your coils are more like λ/50 apart to put them in each other's near field, any force changes direction through the cycle, so your average force is zero.

 

[sophiecentaur]

I'm wanting to see what happens when two electromagnets 90° out of phase and separated by a distance (d) of about 3 cm (1/4 of the wavelength of 2.45 GHz).

I wonder, from your use of the word "Electromagnet" in this context, rather than "inductor", what your level of knowledge can be. What you seem to be describing, albeit in unusual terms, is the basics of how an Induction Motor works. I can't help questioning why you want to do this at such a high Frequency, bearing in mind the added complexity and limited power of microwave circuits.

 

[berkeman]

I thought of this years ago, and I would not be surprised if many other people have thought along similar lines. I'm sure if it were possible, it would have been done fifty years ago. I don't know if this will work or be anything more than a 'photon rocket' or directional antenna. I'm hoping that the force would be similar to two DC electromagnets separated at the distance, but I'm afraid it would be more antenna-like than electromagnet-like.

Not exactly what you are asking about, but because you are curious and have an inquiring mind, you may enjoy this similar application: YIG magnetic oscillator in the GHz frequency range:

https://en.wikipedia.org/wiki/YIG_sphere

http://pw1.netcom.com/~dstraigh/yig.html

I worked with HP Spectrum Analyzers a lot when they were based on this technology. What a fascinating application of GHz magnetics!

 

[slow]

I was wanting to build a very high frequency electromagnet

When you coil a solenoid of $n$ turns with a wire that, without cutting, goes from the beginning to the end, the $n$ turns remain in series. If you cut the wire where one loop ends and another begins, you have $n$ individual turns, each with its pair of terminals.

What would happen if you connect in parallel the $n$ turns? One suppose you would have a much more powerful solenoidal magnetic field and the impedance of the set of turns, connected in parallel, would be $n^2$ times less than when putting the turns in series.

There is an additional detail, which is of particular interest in the alternate regime. With series turns there is a delay between the passage of the current through the first loop and the passage through the last. In parallel there is no delay. Each semicycle of the oscillation goes through all the turns simultaneously. That is, all the turns operate in phase. One suppose that, in high frequency, avoids the phase shift.

 

[sophiecentaur]

What would happen if you connect in parallel the nnn turns?

With nnn times the supply current you would get the same Field. The inductance would be different though so the rate of change of the field would be different at switch on. There is also the practical issue of the Impedance of the connecting cables.

 

[marcusl]

When you coil a solenoid of $n$ turns with a wire that, without cutting, goes from the beginning to the end, the $n$ turns remain in series. If you cut the wire where one loop ends and another begins, you have $n$ individual turns, each with its pair of terminals.

What would happen if you connect in parallel the $n$ turns? One suppose you would have a much more powerful solenoidal magnetic field and the impedance of the set of turns, connected in parallel, would be $n^2$ times less than when putting the turns in series.

There is an additional detail, which is of particular interest in the alternate regime. With series turns there is a delay between the passage of the current through the first loop and the passage through the last. In parallel there is no delay. Each semicycle of the oscillation goes through all the turns simultaneously. That is, all the turns operate in phase. One suppose that, in high frequency, avoids the phase shift.

You need to study E&M, Slow and blainiac. The quasi-static view of current through wires just doesn’t apply at the high frequencies being discussed here.