26 Brochure Amorphous C-Cores
26 Brochure Amorphous C-Cores
26 Brochure Amorphous C-Cores
| High saturation flux density | Low losses | Small size | Flexible due to individual air gaps
SEKELS GmbH, Dieselstr. 6, 61239 Ober-Mrlen; www.sekels.com Page 1 of 16
Applications
C-cores made from amorphous Fe-based alloys offer an interesting combination of a high saturation flux density and low magnetization losses. The main applications for amorphous c-cores are storage chokes or power factor correction (pfc) chokes in the frequency range of about 20 to 50 kHz. The relatively low losses enable the design of size-optimized solutions.
Page 2 of 16
Standard Series
We offer selected types of the American AMCC series as well as customer-specific sizes. Please ask for bobbins and other standard types.
type
length a [mm]
height f [mm] 30 30 25 40 45 35 40
Page 3 of 16
type
lFe [cm]
Acu, 50% [cm] 3,3 4,2 7,0 7,0 7,0 10,4 9,8
Effective surface O [cm] 172 216 303 356 373 464 533
The amorphous strip is fairly brittle after annealing. Please use protecting means like cloves and glasses when handling the cores. Slight core flaking of the inner and outer strip layer is not possible to avoid and no quality criteria. Magnetic properties are not influenced by broken outer and inner layers.
Page 4 of 16
B(H) characteristic of a typical amorphous c-core, w/o air gap and with increasing air gaps to demonstrate the influence. An optimized air gap corresponds with a size optimized design.
Increasing the air gap decreases the permeability and allows higher currents (field strengths). Higher number of turns are necessary to achieve the required inductivity. Bigger air gaps will also negatively influence copper and core losses by interactions of the stray field.
Page 5 of 16
Storage energy
Close to the edge designs close to the optimum storage energy minimize size and costs. This is achieved by the best possible use of the winding area and the magnetic performance of the material. The energy storage capacity EChoke is determined by the inductivity L of the choke and the square of the effective current Ieff. The storage capacity is adequate to the work which is necessary to magnetize the choke:
The relation is valid as long as the choke or the core is not saturated. For storage chokes and power correction chokes (PFC) the following correlation between LI, the basic data of the core and the electrical and magnetic application data can be used:
Assuming typical data for the current density Seff and the maximum induction Bmax allows the calculation of typical LI values for a specific core dimension which can serve as a starting point for further refinement. The values for the rated current IN,eff and the effective permeability in table 4 have been determined by using the approximation formulae described in the following chapter. The data assume a current ripple of 10% of the rated current and a maximum inductivity drop at the maximum current of about 20 %. Adjustments are of course necessary for differing application data.
Richtwerte fr Bmax = 1,3 T, L = 0,5 mH Typ LI2eff [VAs] AMCC 20 AMCC 32 AMCC 50 AMCC 80 AMCC 100 AMCC 125 SU 75b 0,20 0,29 0,43 0,59 0,65 0,81 1,0 IN,eff [A] 20 24 29 34 35 40 45 N (gerundet) 44 44 53 39 37 44 33 eff (ca.) 133 127 108 127 130 115 142 lp [mm] 2,0 2,5 3,75 3,0 3,0 4,25 3,25 Seff [A/mm] 2,7 2,5 2,2 1,9 1,85 1,7 1,5
Page 7 of 16
max
The magnetic field is determined by the no. of turns multiplied by the current and divided by the magnetic path. The corresponding induction is determined by the effective permeability and thus by the air gap. As the inductivity drops when approaching the saturation induction of the material a careful selection of the effective permeability is required to neither overrun the core nor to waste costs by using only a part of the available performance. Figure 6 shows the typical saturation behavior for amorphous c-cores for typical air gaps. When approaching saturation the permeability drops, reaching a value of close to 1 for high field strengths. This dependency of permeability from the field strength is shown in the blue curve, leading to an (artificial) reduction of the induction value in the diagram when plotting B = 0 * (H) * H. Of course the real induction in the core is approaching the saturation Bs. In a first approximation the maximum blue curve indicates a permeability drop of about 20 %. The corresponding induction is about 1,15 T at room temperature and about 1,1 T at 130 C.
Generalized and approximated characterisation of the saturation behavior of amorphous c-cores. The effective permeability is determined by the air gap(s), and is dropping after a constant region when approaching saturation.
eff is the effective permeability determined by the air gap(s) in the linear region.
Page 8 of 16
The second limitation for the number of turns is the available effective winding area ACu of a concrete core size and the possible current density depending from the absolute current and the cooling options:
Typical current densities are about 1 A/mm for high currents and free convection, and about 5 10 A/mm for small currents and or additional cooling measures. The ideal number of turns for each core size neither over-stresses the core magnetically nor thermally. These basic correlations are shown in figure 7. Estimation of the permissible thermal and magnetic currents in a PFC choke. Calculations have been made for core type AMCC 125, inductivity 0,6 mH, frequency 20 kHz, 20 % current ripple and a temperature raise of 75 K.
Using these correlations it is possible to calculate for each core size the required effective permeability for a given current density S:
For Bmax it is either possible to use the estimation from the previous chapter (1,3 T), or smaller values if a more linear characteristic is desired. Please note that the lower Bmax is chosen the lower the effective permeability will be with disadvantages especially with respect to total losses. Which current density to start with? The only way is to use a reasonable value from experience, start the design by calculating eff, and check then the possible induction value (and later the temperature raise). The number of turns is then:
The air gap can be roughly approximated from the effective permeability. The following correlation has been developed from measurements with core size AMCC 125. Other sizes show bigger deviations from this approximation. Empirical testing is necessary anyway.
With just the current density to be estimated as a starting point it is possible to calculate with a few iterations a possible design. Of course the chosen type of wire (litze wire, massive round wire, copper bars) has not only an influence on the effective winding area (by different filling factors), but more on the total losses due to skin and proximity effects. For chokes with air gaps the stray flux of air gap strongly influences both core and copper losses. The approximations in the next chapter shall help to roughly calculate and consider these effects. However the physical mechanisms are fairly complex and it should be clear that it is only a rough approach.
Page 10 of 16
The correction factor KProx depends on the frequency, the type of wire, number of turns and the concrete winding design (one or more layers). Typical values in the frequency range of about 20 kHz are about 2 - 3. In the literature an approximation formula for the core losses of amorphous c-cores can be found, which delivers reasonable results in the frequency range of 10 - 30 kHz for typical current ripples of 10 - 30 %. However the influence of an air gap is not considered in this formula:
Please note to use f in kHz. The ripple induction Bripple can be calculated by:
The stray field of the air gap increases both core and copper losses. The main parameter is of course the size of the air gap, but as already mentioned also the used winding parameters. For cores consisting of two halves (one air gap at each leg) the following rough estimation can be used:
; KL 1
Page 11 of 16
The correction factor KL has been developed for litze wire and effective permeabilities in the range of 50 - 250. For other configurations the total losses may strongly deviate from the approximation. In the literature the following estimation for the temperature raise of chokes with amorphous ccores can be found:
O in this estimation is the surface of an (hypothetical) cubic casing of the choke without the ground face. Estimated values can be found in table 3. As already mentioned a few times experimental testing is essential. The tolerances of the approximations may add to significant deviations. Furthermore the concrete cooling conditions play an important role. Figure 8 shows as an example the thermal image of a choke with upright copper bars with just dc-current and free convection. In this case the critical range (even without the current ripple) is reach at about 50 A, about 10 % lower than predicted by the approximations.
Thermal image of choke with upright copper bars of 7,1x2 mm and a current of 52 A for the purpose of checking the design.
Figure 8: Thermal image of AMCC choke with upright copper bar design (example)
Page 12 of 16
A 51 57 71 71 71 84
B 47 53 67 67 67 79,2
C 12 14 17 17 18 20
D 24 28 36 35,5 35,5 40
E 44 49 49 63 70 54
Page 13 of 16
Mechanical design
The following remarks shall help to realize a suitable and mechanically stable design. Amorphous c-cores are impregnated with an epoxy-type of glue for mechanical stabilization after annealing. Thus they are a composite of about 80 % metal and about 20 % plastic material. A mechanical stable design needs to consider thermal movement due to different thermal expansion coefficients as well as reversible softening of the epoxy system at higher temperatures. Both effects can influence the air gap and change the inductivity. Typically a (or several) metallic clamping strap is used for mechanic stabilization. A clamping force of about 0,8 N/mm is recommended. The air gap filler should be bonded with the core.
Example of a choke with upright type of copper bars, standard bobbins, clamping strap and metallic dedendum angle. Core size SU 75b, copper bars 10x2,5 mm, 52 turns.
Page 14 of 16
Symbol
Description Magnetic flux density (Induction) in Tesla (1 T = 10 000 Gau = 1000 mT = 1000 000 T = 1 000 000 000 nT) Magnetic field strength (1 A/m = 4/1000 Oerstedt) Effective relative magnetic permeability of a magnetic core with air gap
B H eff 0
Vs/Am
Saturation induction at high field strengths and eff 1 Curie-Temperature (disappearance of the spontaneous magnetization due to thermal agitation) Saturation magnetostriction (relative length and volume change) Specific electrical resistance Specific material density Effective magnetic cross-section of a core Mean magnetic path length of a core Stacking factor = relation of effective cross-section to geometric cross-section Effective copper cross section of winding area Mean length of a copper winding Surface of a (hypothetical) cubic casing of the choke without the ground face Energy storage capacity, adequate to the work which is necessary to magnetize the choke Inductivity of a choke. A inductance of one H (Henry) is given, if a voltage of one volt is induced while altering the current by one ampere per second Maximum peak current of all (superimposed) currents Thermally effective value of all (superimposed) currents Number of turns of copper winding
A/m m W W
Current density (effective) Total air gap of a c-core Copper losses Core (or iron) losses Correction factors for copper and total losses Number Pi ( 3,14159)
Page 15 of 16
About us
SEKELS GmbH develops, produces and trades technical products which are mostly related with magnetism. With a team of about 25 employees, more than half of them being physicists or engineers, SEKELS presently serves more than 500 customers worldwide. Since more than 25 years we are familiar with amorphous and nanocrystalline alloys and their applications. Our service includes technical consultation, comprehensive stock keeping and worldwide logistics. SEKELS develops, designs and produces customer-specific laminations and core packages, magnetic shielding and shielding systems, inductive components and magnet systems - from prototyping to series deliveries. We are DIN EN ISO 9001:2008 certified.
SEKELS GmbH Dieselstrasse 6 61239 Ober-Mrlen Germany Tel.: +49 (0) 6002 9379-0 Fax: +49 (0) 6002 9379-79
mail@sekels.de www.sekels.de
All statements, information and data given herein are believed to be accurate and reliable, but are presented without guarantee, warranty or responsibility of any kind, expressed or implied on our part. Published by SEKELS GmbH, Germany. All rights reserved.
Page 16 of 16