Core material for high power high frequency coil/solenoid

  • Thread starter artis
  • Start date
265
36
I read the limits of typical ferrite cores and it seems that most sources claim that they tend to work somewhere up to 1 Mhz. Surely they work great under that frequency, but is there any material that could provide a core material for a wide frequency range starting from few tens of Khz up to tens if not hundreds of Mhz?
The problem is I need a high strength B field at high frequency, and making a coil with air as the core material then demands extreme kA currents through the coil in order to achieve decent (up to 1T field strengths) I was wondering can this be somehow resolved by adding a core material to increase the permeability a bit so to decrease the required current through the coil itself?

Any ideas folks?

Thanks.
 
220
126
Most all RF inductors are air or ferrite. There are about a million different types of ferrite to choose from; some are designed for RF applications. However, you will still have some core loss to deal with.
 

berkeman

Mentor
55,087
5,316
I need a high strength B field at high frequency
What is the application?
(up to 1T field strengths)
Yikes, it's going to be hard to find a material that doesn't saturate at such high fields.

How big of a volume do you need to fill with this high-power RF EM field?
 

berkeman

Mentor
55,087
5,316
And is this related to your previous thread?

 
102
54
No way. Not even with an air core.
 
265
36
Yes it is related but only a little bit. I am not building a klystron.
I need the field to cover a circular area so it can be thought of as a 2d surface through which flux must go through , the surface can be put as close to the solenoid/coil as physically possible, but the problem as you know is that for a core material I can route the B field and provide the field with a higher permeability path so I need less current, sure you say core loss etc but that is a minimal problem compared to achieving decent field strength with an air coil that requires tremendous amounts of current, technically I could arrange and supply the kiloamp current but then cooling etc becomes a problem.

Well are there any known materials that are used in similar applications where both high frequency and high field strengths are required?

Also I can't seem to find any decent calculator or formula for calculating how much current would actually be necessary and what shape of coil could achieve the field strengths I need for the case if it was an air core coil.


my area can vary but for example let's suppose 700cm2 area.

I have had ideas that maybe i could make this coil from multiple smaller coils in parallel , the ends at one side (the side of my surface) could be attached to a capacitor plate while the other ends tied together in a large wire, why capacitor ? because the surface that I need the field to run through is also flat so i could form a variable cap (plate separation) which would determine the frequency and the current running in the loop would pass through the coils creating the necessary field (hopefully) ?
what do you think?


@bobob , No way what ?
 
265
36
Just to bump this thread, is there any material (that I am not aware of) that could be used as a core material that has high magnetic permeability and is still effective at very high frequency?
I am checking google and wiki but it is hard to judge by those tables, the highest frequency limits are given for ordinary ferrites which are about 1 Mhz but then I read that there are core materials for RD transformers up to 100+ Mhz, also the typical ferrites have rather low permeability of about 400-600 μ
 

berkeman

Mentor
55,087
5,316
Be sure to check the saturation flux for those materials you are reading about. I'm pretty sure they all saturate way below what you are wanting to generate...
 
265
36
yes that is another problem , the saturation limit. but since the frequencies involved here would be high shouldn't that somehow lessen the saturation problem ?

Then again what is the average B field strength within a powerful smps ferrite core under operation ?
 

berkeman

Mentor
55,087
5,316
what is the average B field strength within a powerful smps ferrite core under operation ?
Just look up some datasheets. Here is a list of part numbers from one SMPS transformer manufacturer with part numbers. Search the part numbers on their website to get the datasheets, which should have the saturation flux values listed...

 
356
148
I read the limits of typical ferrite cores and it seems that most sources claim that they tend to work somewhere up to 1 Mhz. Surely they work great under that frequency, but is there any material that could provide a core material for a wide frequency range starting from few tens of Khz up to tens if not hundreds of Mhz?
The problem is I need a high strength B field at high frequency, and making a coil with air as the core material then demands extreme kA currents through the coil in order to achieve decent (up to 1T field strengths) I was wondering can this be somehow resolved by adding a core material to increase the permeability a bit so to decrease the required current through the coil itself?

Any ideas folks?

Thanks.
Unfortunately most ferrites saturate at B max ~500-600mT, this is reasonable if you think about the proportion of amperian loops available in pure Fe vs Fe2O3. Ferroxcube has materials that work out to ~100MHz (EMI suppression materials).

Is the purpose of this coil energy storage or the 1T flux density?

I think you may find that air is the only plausible choice for the flux density you are looking for if that is the goal.
 
356
148
Also if the field doesn't need to be continuous you may be able to make the kA currents fairly "easily" by building an LC tank, charge the cap, connect the coil and let it ring!
 
265
36
well the field should need to be continuous but it's an AC field, so if you think pulse operation then no. although what would pulse mode change , the saturation limit is still the same isn't it ?

the goal is simple, have as high as practically possible flux through a 2D surface.
Air sure does the job but at the cost of enormous thick pipe coil, water cooled at best and I have to somehow supply the coil with kiloamps of high frequency current, quite a task.
 

anorlunda

Mentor
Insights Author
Gold Member
6,824
3,753
the goal is simple, have as high as practically possible flux through a 2D surface.
Air sure does the job but at the cost of enormous thick pipe coil, water cooled at best and I have to somehow supply the coil with kiloamps of high frequency current, quite a task.
What in the world are you trying to make?
 
356
148
well the field should need to be continuous but it's an AC field, so if you think pulse operation then no. although what would pulse mode change , the saturation limit is still the same isn't it ?
Yes off course its an AC field! 😉

The pulse operation won't change the saturation limit of ferrites or the currents needed to achieve a certain flux density, what it does is allow you to create those currents relatively speaking quite easily compared to continuous operation. With Mhz of continuous operation you'd probably still want an LC tank so you could drive it with a square wave rather than having to have some monstrous RF amplifier. 1Mhz square wave drive might even be in the realm of diy if you know what you are doing.

Keep in mind if you do put in ferrites that will get you to the 500-600mT quite easily, however above that you'll be basically back to air core and quite nonlinear currents. So I would think the drive requirements for a air cored coils while higher currents are probably a bit more manageable.


the goal is simple, have as high as practically possible flux through a 2D surface.
Air sure does the job but at the cost of enormous thick pipe coil, water cooled at best and I have to somehow supply the coil with kiloamps of high frequency current, quite a task.
Unfortunately this is reality. Even in electric machines air gap flux density is not generally above 1T and thats using a lot of really good magnetic material (Bsat ~1.6-1.7T) to get there (this flux density is what basically limits the performance of machines).

So the aim is flux normal to a 2D surface? Ie the interior field of a solenoid?
 
265
36
@anorlunda asked the question , also you @essenmein , well yes I have also noticed that ordinary transformers at low frequencies as well as motors/generators have the flux somewhere around 1T in their cores and airgaps, 1T would also suffice for me the problem is getting that 1T at those frequencies.
yes the flux should simply cut a given circular 2D area.

as for what i'm building , a high frequency self resonant or driven (depends on the setup) generator, in the self resonant case I could use the B field from the generator to power the field coil, but as long as and this is important the field coil represents a very low impedance because the generator also has very low impedance and high current but low voltage , so technically I could pump kiloamps of current continuously through the field coils (assuming they are thick enough and cooled enough for such operation) but I would need the coils to hava matching low impedance othervise I have no chance of getting that current through.

sure air sounds better because if at those RF frequencies all the available materials have drawbacks in terms of saturation etc,
but looking purely from formulas the permeability drop that air represents is drastic, my output voltage would be about 4/5 volts if I could get 1T flux in the airgap, getting that with a material that has permeability around 500-1000 would be way easier but with air I would literally have to pump 10kA through a coil to get the field and that is if my math is correct if not it might be more. Drastic.
basically a superconducting field could would be needed.
 
356
148
sure air sounds better because if at those RF frequencies all the available materials have drawbacks in terms of saturation etc,
but looking purely from formulas the permeability drop that air represents is drastic, my output voltage would be about 4/5 volts if I could get 1T flux in the airgap, getting that with a material that has permeability around 500-1000 would be way easier but with air I would literally have to pump 10kA through a coil to get the field and that is if my math is correct if not it might be more. Drastic.
basically a superconducting field could would be needed.
Sounds to me like you should add more turns!

Mag field is fundamentally driven by A-t, if you need 10kA with 4-5V, you could achieve the same A-t with 40-50V and 1kA or 4-500V with 100A, by simply increasing turns count by 10 or 100.

Since you are at 1Mhz, skin depth is what a few 10's of um? build your coils out of copper tube (like refrigeration line), run non conductive coolant down the tube.
 
265
36
1Mhz is the lower limit , the upper might be tens of Mhz and higher, at such frequencies I can't make many turns or I will run into higher impedance
Ideally I could run large currents through the coils but they have to have very low impedance for a low voltage high current source to be able to "drive" them

well higher voltage at those frequencies might not be so easy to get.
then I would have to put my output through a transformer and i'm afraid there I run into additional problems.
 
356
148
1Mhz is the lower limit , the upper might be tens of Mhz and higher, at such frequencies I can't make many turns or I will run into higher impedance
Ideally I could run large currents through the coils but they have to have very low impedance for a low voltage high current source to be able to "drive" them

well higher voltage at those frequencies might not be so easy to get.
then I would have to put my output through a transformer and i'm afraid there I run into additional problems.
With such a range of frequencies you'll likely need a few different coils.

Do you have a source in mind? The lower 1Mhz might be doable with MOSFET bridge operating in quasi resonant mode (at 1Mhz minimizing switching loss will be a high priority), but much above that I think you'll be into RF amplifier land and I don't know much about those...

What are you doing with this field? More importantly are you removing energy from it? Where I'm going is I wonder if building a high Q resonant tank and then having an exciter coil so you could build up energy slowly in a large resonating system, and if you are not removing energy from it, at that point you don't have to make kilo amps, you just have to over come the system losses. You'd be limited in frequency range though, you'd have to make variable capacitors, which might not be that hard.
 

Baluncore

Science Advisor
6,529
1,977
is there any material (that I am not aware of) that could be used as a core material that has high magnetic permeability and is still effective at very high frequency?
High frequency performance is not so much chemical composition as it is the dimension of the insulated particles that make up the core. The magnetic field moves rapidly through the binding insulation. Skin effect then decides velocity and depth of penetration into the magnetic particles. Deeper penetration into the larger magnetic particles results in losses because the magnetic energy cannot be recovered before the field has reversed.

Your magnet coil will need few turns to self resonate as a tank circuit at higher MHz. That will require a very high circulating current at resonance to get the field required.

You may notice that the dielectric constant of a ferrite core affects the coil self capacitance. You may then take advantage of the fact that the dielectric constant of a ferrite core material is a function of frequency. You will need to make a test capacitor with your selected ferrite as dielectric, then measure the capacitance at different frequencies.
 

Baluncore

Science Advisor
6,529
1,977
To physically survive a 1T field you will need to consider the topology of a bitter magnet plate and make your RF tank circuit like that.
 
265
36
@essenmein I believe is almost getting my idea. Yes in fact no energy is being taken from the exciting B field much like in an ordinary alternator or generator where the exciting field is just there but the energy is taken from the rotational inertia and prime mover of the rotor. So the current only has to overcome the losses presented by the coil.

As for semiconductors, I don't need them I think, because the very generator itself is a LC tank circuit, so in theory I can use the large B field of the generator to power simply another LC circuit which would be my excitation field coil, providing a variable capacitor would not be a problem in order to control frequency in a given range, as already stated the largest concern for me is to have the coil with a very low impedance (preferably in mili to micro Ohm range) because my output of the generator is low voltage high current so in order to utilize a wide range of frequencies I would need the coil to have very low impedance in that range of frequencies otherwise current and resulting B field would be affected/limited.
So high current supply in my case is not a problem unless the impedance gets high.


Thanks @Baluncore, I have read about Bitter magnets before , now they have high current capabilyt sure I see why that is the case but due to the large flat parallel surfaces isn't the capacitance higher than in an ordinary coil? what I'm saying is isn't the impedance of such a flat coil higher at the same frequency as with an ordinary coil? Also is the B field concentrated only in the circular opening in the middle of the Bitter coil or does the field exists also in the space of the copper itself? Because if it is in the middle mostly then I would need a very large diameter coil which would seem impractical
 

Baluncore

Science Advisor
6,529
1,977
... but due to the large flat parallel surfaces isn't the capacitance higher than in an ordinary coil?
Yes. The capacitance depends on the overlap and separation between plates. But you can make a transmission line resonator rather than a coil and capacitor.

The magnetic field is concentrated in the centre hole. You can have a strong field or a large area.

For the fields you are demanding a normal wire coil will be forced outward and stretched by the field. How will you cool the coil?
 
356
148
I don't know for sure but it looks like bitter magnets are mainly DC?

As mentioned the inter winding capacitance would make this difficult to work at higher frequency.

Looks like basically a helical coil made of individual plates? Sort of like how some planar transformers are build?
243232
 
265
36
Well the bitter coil is basically a very thick flattened wire wound in the shape of a coil with holes to allow for cooling to run through. but as already mentioned the capacitance would be very great compared to other coil types.
Also the reported field strengths with bitter magnets seem huge (record about 45T) but I think @essenmein is correct in the assumption that it is a static DC field with static DC current through the coil? maybe low frequency AC, also the problem is that it seems the field is concentrated in a rather small circular area essentially the middle of the coil, which doesn't suffice for my application sadly.

@Baluncore can you say more about the transmission line resonator ? I am googling it as we speak but any additional input would be nice.
 

Want to reply to this thread?

"Core material for high power high frequency coil/solenoid" You must log in or register to reply here.

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top