assume a coil of N turns was wound with a foil around the longest edge of a PLT43/28/4.1 ferrite plate. The inductance can be easily measured, but do you know of a reasonably accurate method to predict the saturation magnetic field strength of this configuration? I'm not sure I should follow the usual short gap heuristic in such an extreme case. There is suprisingly little in the Internet on this "EMI gun" layout.
Soft Ferrite Design Tool calculates that the E43/PLT43/3C95 configuration with a 1.3mm gap will start to saturate around NI=210A, so the plate itself should be better, but by how much?
This doesn't sound like a short gap situation - it's just a rod core with extreme mishapen dimensions.
I have used slab shapes to build low profile magnetics in the past. Unles your foil layers are less than 5, the use of foil to carry HF current is self-defeating. In a DC choke it's not such an issue.
The most that I can say about such shapes is that they are mechanically simple at low turns count and have a good surface area for heat removal. Leakage is usually pretty strong, despite the usually distributed nature of the gap, as if is outside the actual winding.
Certainly, but the flux lines need to close somehow, hence the word "gap". Even if its longer than the slab itself.
I'm more afraid of the induced interference and linking the flux to anything ferromagnetic nearby, possibly saturating it, so it's unlikely that this approach will ever materialize in my circuits. But I'd like to know how to estimate the saturation parameters out of pure curiosity, the gap "an unusually long short gap" approach is dubious to say the least.
Piotr, I think if you can measure inductance it should be simple to calculate the saturation current. If I'm doing this right... (1) L = V*s/I (2) B_slab = V*s/(N*A), A=cross-sectional area of slab
The 'gap' of a rod core is equal to its length. The core material's saturation flux density is a characteristic of the material, so you 'dial up' the required ampere turns to produce the flux desired to saturate it. This is not a saturated/nonsaturated condition, that can be switched between states - saturation occurs gradually with larger gaps, as any Hannah Curve chart will illustrate.
Permeability terms are dominated by the permeability of the free-space gap.
Isat = Bsat . lg / ( N . uo )
lg = length of gap = length of rod (Meters). N = number of turns uo = permeability of free space = 4 . pi . 10E-7 Isat = current required to 'saturate' the core (Amps) Bsat = flux density defined to 'saturate' the core material (Teslas)
As saturation approaches, lm (the length of the magnetic material flux path) begins to add to lg as their permeabilities approach equality. When completely saturated, Isat will be twice the 'saturaed' value, because the length of the 'gap' has effectively doubled.
Note that there are NO voltage, frequency or time units in this calculation, as you are calculating a static condition. For AC saturation the equation is modified by the inductance of the part, as it influences the time intervals required between achieving saturation of opposing polarity.
It's strange to me that the two equations connect, since you have defined "l" as the length of the rod, and "l" in the inductance formula is for the complete average magnetic flux path.
When the permeabilities of the two materials in the flux path differ by many orders of magnitude, it doesnt take much length of the lower permeability material to dominate calculations. L is, itself, similarly dominated by the lower permeability material dimension. In practical situations, L is simply an indicator of this dominance.
In this instance, keeping things simple is not just an affectation. If you examine the hannah curves of just about any ferrite, and join the points of x% inductance reduction for the multiple curves presented, you'll find that you always get the same result at roughly
10,000Amps/metre. For other non-ferrite magnetic materials, other materialy-constant figures can be extracted. Slab forms make interesting structures in a magnetic circuit, as they form a short circuit to available paths. Unfortunately the shape is restricted in it's availability, with intentionally low-loss material. It is more common in absorptive applications. I've constructed slab transformers on planar E forms in which the central leg of the E form served no useful purpose (save ~ mech support), but was just too expensive to remove.
With the introduction of composite low-loss materials in more malleable formulations, some of the physical restrictions of slab-like structures can be relieved.
Spiral winding core sandwiches or centerpole-free pot structures are natural derivatives and can be found in low profile microconverters and even inside integrated circuits.
An old WEMPEC paper illustrates some of the reasoning behind the avoidance of high-count foil layers in a slab-like structure.
formatting link
Though the windings have good thermal contact to the environment, you don't want eddy currents cooking them, intentionally. Each structure will have it's own optimum power frequency range.
No relation to V.E.Legg, who published extensive articles on toroidal core/winding characteristics (including air-core)in the Bell Systems journals during the 50s.
Thank you for your very detailed answer and the interesting paper!
I am considering this structure for a dI/dt limiting choke, so mostly DC until a short-circuit happens. The inductor buys time to shut the system down properly.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.