If a rod core cannot saturate, why choose powdered iron over ferrite?

[TimNJ]

Hi everyone,

Been thinking about this one for a while, but think I need some help:


A rod-core inductor is a "open" magnetic circuit, meaning part of the flux lines are through air, external to the core. These flux lines technically go out to infinity, but become increasingly weak as they get further away from the inductor. As I understand it, it is basically "impossible" to saturate a rod-core inductor, as the "air-gap" is so large (and thus the reluctance so high), that it would take an immense amount of current to saturate it.


If that's the case, what is the point of a powdered iron rod-core inductor? Most often, powdered iron is used in toroidal inductors which are closed magnetic circuits. These can definitely be saturated, so the advantages of a soft saturating material are obvious. If it's so hard to saturate a rod-core anyway, why use a soft-saturating material, as opposite to using a high permeability MnZn material? I see Micrometals Inc, and others, offer rod-core shapes, and I am mainly curious of the applications and/or reasons to go down that road.

I feel like I am oversimplifying my analysis, so maybe someone can help fill in the gaps.

Thanks!
Tim

[bson]

A closed loop core will better contain the magnetic field in the loop, while straight rod cores project magnetic fields along the rod axis.  Other inductors will happily pick up these stray fields, resulting in noise and crosstalk, in particular when two inductors have the same orientation.

Otherwise you'll find the same materials used in either.

[Marco]

As I understand it, it is basically "impossible" to saturate a rod-core inductor

This assumption is wrong.

[S. Petrukhin]

Probably, the frequency properties of the material are important here.

[TimNJ]

As I understand it, it is basically "impossible" to saturate a rod-core inductor

This assumption is wrong.

Thanks. I was sort of trying to imply that I knew my assumption was wrong with the quotes around impossible. A while ago, when I asked my ex-boss/mentor how to determine Isat for a rod-core design I did, he basically told me “don’t worry about it”.

The rod core geometry is harder to work with, i.e. to calculate the magnetic path reluctance, when compared to gapped ferrite cores or toroid cores. So, that’s probably why he kind of dismissed my inquiry.

Given that you can saturate a rod core, my best guess is that using powdered iron in this application might make sense when the currents are particularly high...or maybe something with very high momentary peak currents. Not sure!

[TimNJ]

Probably, the frequency properties of the material are important here.

How so? Any ideas? I also had a though that there may be differences in material stability, i.e. temperature stability.

[Berni]

Thanks. I was sort of trying to imply that I knew my assumption was wrong with the quotes around impossible. A while ago, when I asked my ex-boss/mentor how to determine Isat for a rod-core design I did, he basically told me “don’t worry about it”.

The rod core geometry is harder to work with, i.e. to calculate the magnetic path reluctance, when compared to gapped ferrite cores or toroid cores. So, that’s probably why he kind of dismissed my inquiry.

Given that you can saturate a rod core, my best guess is that using powdered iron in this application might make sense when the currents are particularly high...or maybe something with very high momentary peak currents. Not sure!

All of the field is still going trough the rod in the center of the inductor, at some point it can still overwhelm it and cause it to saturate. But yes as you said due to the large "air gap" on the ends takes a rather strong field to do.

Air gaps are not free lunch however. While they do drastically increase the saturation current of an inductor then also drastically decrease its inductance value, so you need more turns to get the inductance up and that makes the field even stronger and drops saturation current again.

What airgaps help with is increasing the energy storage capability of an inductor. The field concentrates itself into the air gap where saturation is not an issue, so most of the energy ends up stored inside the air gap rather than the magnetic core material. This is usefull when making inductors for switching DC/DC converters or transformers for typologies that require energy storage such as flyback.

However you still need the magnetic material to make the air gap small enough to a point where it doesn't cause the inductance to plummet to a unusably low value (And not cause massive interference to everything around it as others pointed out). If this material can sustain stronger fields before saturating means you can make the core size smaller and higher permeability can reduce the number of turns (reducing the size even more and saving on resistive losses). Powder iron is useful for this reason, but the magnetic losses of it start to become too high at high frequencies so various ferrite core blends become worth it. Perhaps the core needs to be bigger to get the same level of performance but it won't get scorching hot when operated at the higher frequencies. But the higher frequencies reduce the required minimum inductance in most switching converters, so a smaller inductor could be used.

Its all a tradeoff depending on the application.

[S. Petrukhin]

Probably, the frequency properties of the material are important here.

How so? Any ideas? I also had a though that there may be differences in material stability, i.e. temperature stability.

Different materials have different frequency characteristics. Iron cannot work at high frequencies, ferrite cannot work at low frequencies. More precisely, they can, but at the same time the losses become indecently large.

[T3sl4co1l]

Because cost, I suppose.  Extruded ferrite rods cheaper than pressed powder shapes?

Saturation isn't that far out. Like a 5A choke might saturate around 10A.  It's a bit gradual, because some flux lines are close in (short air path), some are long.  Sharper than powder core saturation.

Tim

[TimNJ]

Thanks. I was sort of trying to imply that I knew my assumption was wrong with the quotes around impossible. A while ago, when I asked my ex-boss/mentor how to determine Isat for a rod-core design I did, he basically told me “don’t worry about it”.

The rod core geometry is harder to work with, i.e. to calculate the magnetic path reluctance, when compared to gapped ferrite cores or toroid cores. So, that’s probably why he kind of dismissed my inquiry.

Given that you can saturate a rod core, my best guess is that using powdered iron in this application might make sense when the currents are particularly high...or maybe something with very high momentary peak currents. Not sure!

All of the field is still going trough the rod in the center of the inductor, at some point it can still overwhelm it and cause it to saturate. But yes as you said due to the large "air gap" on the ends takes a rather strong field to do.

Air gaps are not free lunch however. While they do drastically increase the saturation current of an inductor then also drastically decrease its inductance value, so you need more turns to get the inductance up and that makes the field even stronger and drops saturation current again.

What airgaps help with is increasing the energy storage capability of an inductor. The field concentrates itself into the air gap where saturation is not an issue, so most of the energy ends up stored inside the air gap rather than the magnetic core material. This is usefull when making inductors for switching DC/DC converters or transformers for typologies that require energy storage such as flyback.

However you still need the magnetic material to make the air gap small enough to a point where it doesn't cause the inductance to plummet to a unusably low value (And not cause massive interference to everything around it as others pointed out). If this material can sustain stronger fields before saturating means you can make the core size smaller and higher permeability can reduce the number of turns (reducing the size even more and saving on resistive losses). Powder iron is useful for this reason, but the magnetic losses of it start to become too high at high frequencies so various ferrite core blends become worth it. Perhaps the core needs to be bigger to get the same level of performance but it won't get scorching hot when operated at the higher frequencies. But the higher frequencies reduce the required minimum inductance in most switching converters, so a smaller inductor could be used.

Its all a tradeoff depending on the application.

Thanks for this. While I understand most of these ideas separately (i.e. toroidal vs rod core: closed vs open magnetic structure, iron powder/alloy core vs. gapped ferrite: soft saturating vs hard saturating), I'm still having a tough time figuring out why you'd really want to use an iron powder/alloy core in a rod core.

I tried this calculator: https://coil32.net/online-calculators/ferrite-rod-calculator.html

I used 60u to represent the iron-powder type, and 2,000u to represent a ferrite type. I tried a normal size I might use in an output LC filter, about 20mm height, 6mm diameter. I found the number of turns required for a 100nH, 1uH, and 4uH. What I found was: Regardless of what target inductance I wanted, the powdered iron version required only about 1 more turn than the ferrite version. To me, this sort of makes sense since the very large air gap is what dominates the path reluctance. The effective permeabilities were coming out just a tad higher with the ferrite versus the powdered iron. 

But that's only part of the story.

If I look at Micrometals -40 mix (powdered iron), it has a Bsat of 1.8T, compared to a MnZn ferrite with Bsat of 0.4T. If the number of turns (required for some target inductance) is roughly the same for both materials, and with same core cross-section, then the flux density in the core should be about the same, right? But, now the powdered iron version can handle 4x higher flux density...I think. So that's an advantage.

But what else? Hmm.

[Berni]

You can't just look at the Bsat values for a core material.

The advantage that ferrite alloy cores have is that they create less heat losses when being magnetized and demagnetized. The closer to Bsat you take it and the faster you magnetize the core the higher these losses are. So for a small power transformer these losses might only be around 10 to 100mW, so you don't care about the losses because you loose more in the resistance of the copper windings. But if you push the core higher and higher the losses climb and may get >1W or even >10W if pushed hard enough. If you keep pushing more power trough the grossly underspecified core it will start getting hotter and hotter, making smoke and then rapidly start loosing its magnetic properties, causing huge currents to flow trough the copper coils to maintain such a high amount of output power, causing the coils to get hot, smoke and the insulation on them to catch fire and short out, finally ending the life of the poor abused transformer.

So put simply once you take the core too close to saturation at too high of a frequency you essentially make an induction cooker for heating transformer cores. Using the apparently superior higher Bsat iron powder core at 50KHz might be perfectly fine, but if you try to use the same core in a transformer running at 800KHz the core will get burning hot and smoke. Even ferrite alloys can have a hard time dealing with such high operating frequency, so the manufacturer might specify an even lower Bmax when running above a certain freqency range. These Bmax can be something pathetic sounding like only 100mT, but at the same time having your DC/DC run at such a high frequency means you need much less turns on the transformer thus making the transformer smaller for the same power output. But at the same time this puts more strain on the transistors in the DC/DC.

Its typical in engineering that once you start to heavily optimize one parameter (like Bsat) the other parameters start to suffer as a consequence. Hence why you need to look at all the important parameters simultaneously and find the best compromise.

[TimNJ]

Thanks. I think I understand core (hysteresis) loss pretty well...As far as I know, Bsat does not vary with frequency, but the ~recommended~ Bmax (to avoid heating the core to failure) does. I think you made this distinction?

For application with large ΔB, AC flux swing, I'd lean towards ferrite to limit hysteresis losses. But, maybe for applications with high DC bias, like output LC filter inductor, smaller AC flux swing, think about powder iron/alloy cores. Softer saturation characterstic and higher Bsat means you can probably use a smaller one in this application.

This all makes sense for closed geometry cores like toroids...but rod core seems like a different animal, since the very large external air gap means that you have a low effective permeability anyway, regardless of core material.

For this particular question, still a little lost, but maybe I ought to just give up on it for now. For normal gapped or toroid geometries, I think I get it, I was mainly just curious about rod cores geometry. I appreciate all of your input. It was helpful.

[T3sl4co1l]

Low permeability means low B for a given H, meaning you probably won't hit Bsat on the higher materials before your wire melts into a puddle.

Can you imagine a 1.8T bar electromagnet?  With more attractive force than a NdFeB magnet?  Imagine how fast that would heat up, dang...

Tim

[Berni]

This all makes sense for closed geometry cores like toroids...but rod core seems like a different animal, since the very large external air gap means that you have a low effective permeability anyway, regardless of core material.

For this particular question, still a little lost, but maybe I ought to just give up on it for now. For normal gapped or toroid geometries, I think I get it, I was mainly just curious about rod cores geometry. I appreciate all of your input. It was helpful.

Its more that rod cores do not make sense for use as a power inductor/transformer.

The amount of inductance they can provide is just too small as a consequence of having that massive air gap. Sure they won't saturate but (as stated by T3sl4co1l) the copper windings also put a practical limit on the maximum current that can flow trough the inductor since at some point they also overheat. So if you create a core that saturates at 10x the current that the windings can take then that is a badly designed inductor. If the inductor had a more optimal core shape it could get the same inductance and same current capability using fewer turns of copper wire, making it smaller, lighter and cheaper.

If you look at commercially available power inductors from manufacturers such as TDK, Bourns, Coilcraft, Wurth..etc you will notice that the rated saturation current and the rated max current (offten for a reasonable 40°C temp rise) are always fairly close together. This is because they optimize the inductor design to offer the highest current capability using as little material as reasonably possible. Turns out the best core design for meeting this using copper wire coils is a closed core or a core with air gaps on the scale of millimeters. If you invent a wire material with 10x higher conductivity than copper then perhaps a rod core might make sense, but so far we don't have any such wire materials that are cheep or work at room temperature.

But there still are rod core inductors out there. One example is AM radio antennas are wire would around pen sized bar of ferrite. The ferrite used in that case is super high permeability to 'suck in' as much of the surrounding field as possible and pass it trough a small inductor, this makes the antenna much more compact, however it only works for reception, if you try to transmit using this antenna the core would just turn most of it to heat. Another example are tunable inductors that used to be very popular in RF products, here the benefit of a rod core is that its geometry allows the core to be easily slid in and out of a coil by a screw thread.

[T3sl4co1l]

I find reasonable results with rod inductors for power filtering; but I don't mind about EMI in an optimized design, I'm just putting the thing in a box anyway or whatever.  You never see them in that position, in commercial designs, and I think that's mostly for EMI reasons.  For low level filtering however, they're quite popular -- taking the edge off a flyback output, for example.

Commercially, barrel/spool/bobbin/however you want to call them, are maybe more popular (in the larger sizes that aren't practical to support in SMT types).  The shorter airgap path performs better, for the aforementioned reasons.

If you invent a wire material with 10x higher conductivity than copper then perhaps a rod core might make sense, but so far we don't have any such wire materials that are cheep or work at room temperature.

It's kind of fascinating that the best conductors we have (at room temperature, at this time) are best suited to a mu_r around 20.  The suitability is simply a matter of resistivity versus inductivity, and if we had just a few times better conductor, we wouldn't be very bothered about using cores at all, a lot of the time.  It's not a huge difference: it's like comparing the thermal conductivity of aluminum and copper, or copper and diamond.  Sadly, there's no "diamond" of electrical resistance at room temperature.

Now if we get superconductors involved, fields can get quite intense.  Pencil-fine wires carrying hundreds, even thousands of amperes.  Thousands of turns of that, on a solenoid or whatever, and you can store just... a lot of energy.  Up to 10, even 20T, given suitable cooling (maybe not HTS, that needs LHe I think).  20T corresponds to an energy density of 160 J/cm^3, or a pressure of 23k PSI (160 MPa) -- at these levels, coilguns become practical for example, perhaps competitive even!

IIRC, the plasma magnets in ITER will (are?) total some henries of inductance.

Apparently a few short-term electrical grid storage systems are operating now, using superconducting storage.

Tim

[virtualparticles]

Its typical in engineering that once you start to heavily optimize one parameter (like Bsat) the other parameters start to suffer as a consequence. Hence why you need to look at all the important parameters simultaneously and find the best compromise.

This is "Stones" Rule.

You just can't get what you want

and the corollary:

But if you try sometimes, you get what you need.