Inductor with flux path for high inductance at low load

Abstract

A magnetic component has a variable inductance over a range of DC bias currents. The component includes a bobbin with a coil positioned around a passageway between first and second end flanges. First and second E-cores have respective middle legs positioned in the passageway with end surfaces of the middle legs juxtaposed within the passageway and spaced apart by a first magnetic gap. An I-bar is positioned in the passageway parallel to and spaced apart from respective first longitudinal surfaces of the middle legs to form a second magnetic gap between the I-bar and the longitudinal surface of the middle leg of the first E-core and to form a third magnetic gap between the I-bar and the longitudinal surface of the middle leg of the second E-core. The magnetic component provides higher inductances for lower bias currents and provides lower inductances for higher bias currents.

Claims

1. A magnetic component comprising: a bobbin having a first end flange, a second end flange and a passageway through the bobbin from the first end flange to the second end flange; at least one coil positioned around the passageway between the first end flange and the second end flange; a first E-core and a second E-core, each E-core having a respective main body, a respective middle leg, a respective first outer leg and a respective second outer leg, the legs of each E-core extending from the respective main body to respective end surfaces, the middle legs of the two E-cores positioned in the passageway of the bobbin with the respective end surfaces of the middle legs juxtaposed within the passageway and spaced apart by a first magnetic gap, each middle leg having a respective first longitudinal surface perpendicular to the respective end surface, the first and second outer legs of the two E-cores positioned outside the bobbin with the end surface of the first outer leg of the first E-core engaging the end surface of the first outer leg of the second E-core and with the end surface of the second outer leg of the first E-core engaging the second outer leg of the second E-core; and a first I-bar positioned in the passageway in alignment with the middle leg of the first E-core and in alignment with the middle leg of the second E-core, the first I-bar spanning the first magnetic gap with a first portion of the first I-bar parallel to and spaced apart from the first longitudinal surface of the middle leg of the first E-core to form a second magnetic gap between the first I-bar and the longitudinal surface of the middle leg of the first E-core with at least a portion of the second magnetic gap positioned within the passageway, and with a second portion of the first I-bar parallel to and spaced apart from the first longitudinal surface of the middle leg of the second E-core to form a third magnetic gap between the first I-bar and the longitudinal surface of the middle leg of the second E-core with at least a portion of the third magnetic gap positioned within the passageway.

2. The magnetic component of claim 1, further including a spacer positioned between the I-bar and the longitudinal surface of the middle leg of the first E-core, the spacer having a thickness that defines the second magnetic gap.

3. The magnetic component of claim 2, wherein the spacer is also positioned between the I-bar and the longitudinal surface of the middle leg of the second E-core.

4. The magnetic component of claim 1, further comprising: a respective second longitudinal surface of each middle leg, each respective second longitudinal surface parallel to the respective first longitudinal surface; and a second I-bar, the second I-bar parallel to and spaced apart from the second longitudinal surface of the middle leg of the first E-core by a fourth magnetic gap, the second I-bar parallel to and spaced apart from the second longitudinal surface of the middle leg of the second E-core by a fifth magnetic gap.

5. The magnetic component of claim 4, wherein the fourth and fifth magnetic gap have substantially the same length as the second and third magnetic gap.

6. The magnetic component of claim 1, wherein the middle leg of each of the first E core and the second E-core has a height substantially equal to a height of each of the first outer leg and the second outer leg.

7. A method for constructing a magnetic component comprising: positioning at least one coil onto a bobbin, the bobbin having a first end flange, a second end flange and a passageway through the bobbin from the first end flange to the second end flange, the at least one coil positioned around the passageway of the bobbin between the first end flange and the second end flange; providing a first E-core and a second E-core, each E-core having a body portion, a respective first outer leg, a respective second outer leg and a respective middle leg, each leg of each core having a respective end surface, each middle leg having a respective longitudinal surface extending from the body portion to the respective end surface of the middle leg; inserting the middle leg of the first E-core into a first end of the passageway proximate to the first end flange, and inserting the middle leg of the second E-core into a second end of the passageway proximate to the second end flange, the middle legs positioned in the passageway with the end surfaces spaced apart from each other to form a first magnetic gap, the outer legs positioned outside the passageway with the end surface of the first outer leg of the first E-core engaging the end surface of the first outer leg of the second E-core and with the end surface of the second outer leg of the first E-core engaging the end surface of the second outer leg of the second E-core; and positioning a first I-bar in the passageway in alignment with the middle legs of the two E-cores, the first I-bar spanning the first magnetic gap, the first I-bar having a first portion positioned parallel to and spaced apart from the longitudinal surface of the middle leg of the first I-core to form a second magnetic gap between the first I-bar and the longitudinal surface of the middle leg of the first E-core with at least a portion of the second magnetic gap positioned within the passageway, the first I-bar having a second portion positioned parallel to and spaced apart from the longitudinal surface of the middle leg of the second I-core to form a third magnetic gap between the first I-bar and the longitudinal surface of the middle leg of the second E-core with at least a portion of the third magnetic gap positioned within the passageway.

8. The method of claim 7, further comprising positioning a spacer between the first I bar and the longitudinal surface of the middle leg of the first E-core, the spacer having a thickness that defines the second magnetic gap.

9. The method of claim 8, wherein the spacer is also positioned between the first I bar and the longitudinal surface of the middle leg of the second E-core.

10. The method of claim 7, further comprising: positioning a second I-bar into the passageway, the second I-bar positioned parallel to and spaced apart from a second longitudinal surface of the middle leg of the first E-core by a fourth magnetic gap, the second I-bar positioned parallel to and spaced apart from the second longitudinal surface of the middle leg of the second E-core by a fifth magnetic gap.

11. The method of claim 10, wherein the fourth and fifth magnetic gap have substantially the same length as the second and third magnetic gap.

12. The method of claim 7, wherein the middle leg of each of the first E core and the second E-core has a height substantially equal to a height of each of the first outer leg and the second outer leg.

13. A method for controlling the inductance of a magnetic component to provide a first range of inductances over a first range of DC bias currents and to provide a second range of inductances over a second range of DC bias currents, the method comprising: providing a magnetic component by positioning at least one coil around a passageway of a bobbin, inserting a respective middle leg of a first E-core into the passageway from a first end of the passageway, the middle leg of the first E-core having a respective end surface and a respective longitudinal surface, the longitudinal surface of middle leg of the first E-core perpendicular to the end surface of the middle leg of the first E-core, each of the first outer leg and the second outer leg of the first E-core parallel to and spaced apart from the middle leg of the first E-core, each of the first outer leg and the second outer leg of the first E-core having a respective end surface, inserting a respective middle leg of a second E-core into the passageway from a second end of the passageway, the second middle leg having a respective end surface and a respective longitudinal surface, the longitudinal surface of the middle leg of the second E-core perpendicular to the end surface of the middle leg of the second E-core, the end surface of the middle leg of the second E-core parallel to and spaced apart from the end surface of the middle leg of the first E-core by a first magnetic gap, each of the first outer leg and the second outer leg of the second E-core parallel to and spaced apart from the middle leg of the second E-core, each of the first outer leg and the second outer leg of the second E-core having a respective end surface, the end surface of the first outer leg of the first E-core engaging the end surface of the first outer leg of the second E-core, the send surface of the second outer leg of the first E-core engaging the end surface of the second outer leg of the second E-core, and inserting a first I-bar into the passageway in alignment with the middle leg of the first E-core and the middle leg of the second E-core, the first I-bar spanning the first magnetic gap, the first I-bar having a first portion with a longitudinal surface parallel to and spaced apart from the longitudinal surface of the middle leg of the first E-core by a second magnetic gap with at least a portion of the second magnetic gap positioned within the passageway, the first I-bar having a second portion with a longitudinal surface parallel to and spaced apart from the longitudinal surface of the middle leg of the second E-core by a third magnetic gap with at least a portion of the third magnetic gap positioned within the passageway; applying a first DC bias current to the at least one coil, the first DC bias current having a first magnitude in a first range of current magnitudes, the first range of current magnitudes selected to be less than a current magnitude that saturates a magnetic path through the second magnetic gap, the third magnetic gap and the I-bar, the magnetic component having a first range of inductances when the DC bias current has a magnitude in the first range of current magnitudes; and applying a second DC bias current to the at least one coil, the second DC bias current having a magnitude in a second range of current magnitudes, the second range of current magnitudes selected to be at least sufficient to cause the magnetic path through the second magnetic gap, the third magnetic gap and the I-bar to saturate, the magnetic component having a second range of inductances when the DC bias current has a magnitude in the second range of current magnitudes, wherein each inductance in the second range of inductances is less than inductances in the first range of inductances.

14. The method of claim 13 further comprising positioning a spacer between the first I bar and the longitudinal surface of the first middle leg, the spacer having a thickness that defines the second magnetic gap.

15. The method of claim 14, wherein the spacer is also positioned between the first I bar and the longitudinal surface of the second middle leg.

16. The method of claim 13, further comprising: positioning a second I-bar into the passageway, the second I-bar positioned parallel to and spaced apart from a second longitudinal surface of the first middle leg by a fourth magnetic gap, the second I-bar positioned parallel to and spaced apart from the second longitudinal surface of the second middle leg by a fifth magnetic gap.

17. The method of claim 16, wherein the fourth and fifth magnetic gap have substantially the same length as the second and third magnetic gap.

18. The method of claim 13, wherein the middle leg of each of the first E core and the second E-core has a height substantially equal to a height of each of the first outer leg and the second outer leg.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

(1) FIG. 1 illustrates a perspective view of a conventional magnetic device having a bobbin and two E-cores.

(2) FIG. 2 illustrates an exploded perspective view of the magnetic device of FIG. 1.

(3) FIG. 3 illustrates a plan cross-sectional view of the magnetic device of FIG. 1 taken along the line 3-3 in FIG. 1.

(4) FIG. 4 illustrates a perspective view of the second E-core of FIGS. 1-3.

(5) FIG. 5 illustrates a perspective view of the bobbin of FIGS. 1-3.

(6) FIG. 6 illustrates a top plan view of the first and second E-cores of FIGS. 1-3.

(7) FIG. 7 illustrates an enlarged plan cross-sectional view of the magnetic device of FIG. 1 taken along the line 7-7 in FIG. 1.

(8) FIG. 8 illustrates a graph of the DC bias characteristics of the magnetic component of FIGS. 1-7.

(9) FIG. 9 illustrates a perspective view of a magnetic component having a bobbin, an E-core with a conventional end surface of the middle leg and an E-core having a stepped end surface.

(10) FIG. 10 illustrates an exploded perspective view of the magnetic device of FIG. 5.

(11) FIG. 11 illustrates a perspective view of the step-gap E-core of FIGS. 9-10.

(12) FIG. 12 illustrates a top plan view of the step-gap E-core of FIG. 11.

(13) FIG. 13 illustrates a plan cross-sectional view of the magnetic device of FIG. 9 taken along the line 13-13 in FIG. 9.

(14) FIG. 14 illustrates a graph of the DC bias characteristics of the magnetic component of FIGS. 9-13 in comparison with the DC bias characteristics of the magnetic component of FIGS. 1-7.

(15) FIG. 15 illustrates a perspective view of a magnetic component having a bobbin, two E-cores and an I-bar extending along the lengths of the middle legs of the two E-cores.

(16) FIG. 16 illustrates an exploded perspective view of the magnetic device of FIG. 15.

(17) FIG. 17 illustrates an elevational cross-sectional view of the magnetic device of FIG. 15 taken along the line 17-17 in FIG. 15.

(18) FIG. 18 illustrates an enlarged elevational cross-sectional view of the magnetic device of FIG. 15 taken within the dashed area 18 of FIG. 17.

(19) FIG. 19 illustrates a graph of the DC bias characteristics of the magnetic component of FIGS. 15-18 in comparison with the DC bias characteristics of the magnetic component of FIGS. 1-7 and the DC bias characteristics of the magnetic component of FIGS. 9-13.

(20) FIG. 20 illustrates a perspective view of a magnetic component having a bobbin, two E-cores, a first I-bar extending along the lengths of the upper surfaces of the middle legs of the two E-cores, and a second I-bar extending along the lengths of the lower surfaces of the middle legs of the two E-cores

(21) FIG. 21 illustrates an exploded perspective view of the magnetic device of FIG. 20.

(22) FIG. 22 illustrates an elevational cross-sectional view of the magnetic device of FIG. 20 taken along the line 22-22 in FIG. 20.

(23) FIG. 23 illustrates an enlarged elevational cross-sectional view of the magnetic device of FIG. 20 taken within the dashed area 23 of FIG. 22.

(24) FIG. 24 illustrates an enlarged elevational cross-sectional view of the magnetic device of FIG. 20 taken within the dashed area 24 of FIG. 22.

(25) FIG. 25 illustrates a graph of the DC bias characteristics of the magnetic component of FIGS. 20-24 in comparison with the DC bias characteristics of the magnetic component of FIGS. 1-7, the DC bias characteristics of the magnetic component of FIGS. 9-13, and the DC bias characteristics of the magnetic component of FIGS. 15-18.

DETAILED DESCRIPTION OF THE INVENTION

(26) In the following description, a reference to a gap, and air gap, or a magnetic gap is a reference to a discontinuity in the magnetically permeable material forming a core. The gap may be a filled space or an unfilled space between adjacent magnetically permeable materials. References herein to the gap length, are used to refer to the distance between two surfaces that form the boundaries of a gap. The term gap distance may also be used to represent the distance between the boundary surfaces of the gap. The boundary surfaces of a gap may have lengths and widths that define the area or the cross-section of the gap; however, the term gap length is used only to refer to the distance between boundary surfaces.

(27) FIGS. 15-18 illustrate an inductor 900 configured in accordance with an embodiment of the present invention. The inductor comprises the first E-core 110 and the second E-core 112, which correspond to the two like-numbered E-cores of FIGS. 1-7. Accordingly, corresponding features of the two E-cores of FIGS. 15-18 are numbered in like manner. As shown in FIG. 16, the middle legs 150, 160 of the E-cores have the common width W1 between the respective first side surface 180 and the respective second side surface 182. The middle legs have the common height H1 between the respective lower surface 184 and the respective upper surface 186.

(28) The inductor 900 includes a bobbin 920 having a passageway 922. The other features of the bobbin generally correspond to the features of the bobbin 120 of FIGS. 1-7 and are numbered accordingly. The passageway has the width W2 between a first inner side surface 930 and a second inner side surface 932. In the illustrated embodiment, the width W2 of passageway may be the same as or slightly greater than the width W1 of the middle legs 150, 160 of the E-cores 110, 112 as previously described. The passageway has a height H3 between a lower inner surface 934 and an upper inner surface 936. As described below, the height H3 of the passageway in FIGS. 15-18 is greater than the common height H1 of the middle legs of the E-cores.

(29) The additional height of the passageway 922 is provided to accommodate an I-bar 940. The I-bar comprises a ferrite material and is configured as a rectangular parallelepiped having a first side surface 950, a second side surface 952, a lower surface 954, an upper surface 956, a first end surface 960 and a second end surface 962.

(30) The I-bar 940 has a length L4 between the first end surface 960 and the second end surface 962. In the illustrated embodiment, the length L4 of the I-bar is approximately the same as the length L1 of the two combined E-cores 110, 112. In other embodiments, the length L4 of the I-bar may be greater than or less than the length L1. For example, the length L4 may be less than the length L1.

(31) The I-bar 940 has a width W4 between the first side surface 950 and the second side surface 952. In the illustrated embodiment, the width W4 of the I-bar is approximately the same as the width W1 of the middle legs 150, 160 of the two E-cores 110, 112. In other embodiments, the width W4 may differ from the width W1. For example, the width W4 may be narrower than the width W1.

(32) The I-bar 940 has a height H4 between the lower surface 954 and the upper surface 956. The height H4 of the I-bar is selected such that when the I-bar is positioned in the passageway 922 of the bobbin 920, the I-bar fits between the upper surfaces 186 of the middle legs and the upper inner surface 936 of the passageway. In particular, the height H3 of the passageway 922 is greater than the common height H1 of the middle legs by a distance slightly greater than the height H4 of the I-bar. The I-bar may also be positioned below the lower surfaces 184 of the middle legs of the E-cores.

(33) The slight difference in the total height (H1+H4) of the middle legs 150, 160 and the I-bar 940 and the height H3 of the passageway 922 allows a spacer 970 to be inserted between the upper surfaces of the middle legs of the E-cores and the lower surface of the I-bar. For example, the spacer may have a height H5 between a lower surface 972 and an upper surface 974. When installed as illustrated in FIGS. 15-18, the total height (H1+H4+H5) of the middle legs, the I-bar and the spacer is approximately equal to the height H3 of the passageway. In the illustrated embodiment, the spacer has a length and a width corresponding to the length L4 and the width W4 of the I-bar. Although illustrated as having a length and width corresponding to the length and width of the I-bar, the spacer may have smaller dimensions. For example, the spacer may be segmented, with segments positioned at locations selected to displace the lower surface of the I-bar away from the upper surfaces of the middle legs of the E-core. In the illustrated embodiment, the spacer may comprise a polyester film having a thickness of approximately 0.05 millimeter. The spacer may also comprise a thin layer of tape adhered to the lower surface of the I-bar. The thickness of the spacer may be varied to increase or decrease the distance between the parallel lower surface of the I-bar and the upper surfaces of the two middle legs.

(34) As illustrated in FIG. 18, the spacer 970 forms a first thin magnetic gap 980 between the lower surface 954 of the I-bar 940 and the upper surface 186 of the middle leg 150 of the first E-core 110. The spacer forms a second thin magnetic gap 982 between the lower surface of the I-bar and the upper surface 186 of the middle leg 160 of the second E-core 112. Each thin gap has a height between the adjacent parallel surfaces that is much shorter (in the direction perpendicular to the adjacent, spaced-apart surfaces) than the conventional gap 230 between the end surfaces 210, 220, respectively, of the middle legs 150, 160 of the two E-cores 110, 122. In the illustrated embodiment, the height of the spacer determines the height of the gap, and the gap has the height H5. The spacing between adjacent parallel surfaces is referred to herein as the gap distance or gap height of the thin gaps. For example, the gap distance of the single large gap 230 between the end surfaces of the middle legs may be 0.25 millimeters in comparison to the 0.05 millimeter gap distance of each of the thin gaps 980, 982 between the I-bar and the upper surfaces of the middle legs. Each thin gap also has a much larger surface area than the conventional gap formed between the end surfaces of the two middle legs. The larger surface areas defining the two thin gaps and the shorter gap distances of the two thin gaps compared to the conventional gap cause the magnetic reluctance of the magnetic path through the two thin gaps and the I-bar to be much lower than the magnetic reluctance of the magnetic path through the conventional gap with the smaller gap area and the larger gap distance.

(35) The inductance of the inductor 900 is affected by the two thin gaps 980, 982 as illustrated by a curve 1210 of the DC bias characteristics of the inductor shown on a graph 1200 in FIG. 19. The previous curve 410 of the DC bias characteristics of the conventional inductor 100 and the previous curve 810 of the DC bias characteristics of the step-gap inductor 500 are also shown for comparison.

(36) As illustrated by the curve 1210, a low DC bias currents, the I-bar 940 in combination with the two much thinner gaps 980, 982 between the I-bar 940 and the middle legs 150, 152 of the two E-cores 110, 112, provides a low reluctance magnetic path in parallel with the magnetic path through the much larger air gap 230 between the end surfaces 210, 220, respectively, of the middle legs of the two E-cores. The low reluctance path causes the inductor 900 to have a much higher inductance at low DC bias currents. For example, the total inductance peaks at approximately 10.2 millihenries at a DC bias of approximately 0.05 ampere. As the DC bias current increases above 0.05 ampere, the magnetic path through the I-bar and the two thin gaps begins to saturate, which causes a corresponding increase in the reluctance in the parallel magnetic path through the I-bar.

(37) As the DC bias current continues to increase above 0.5 ampere, the reluctance in the parallel magnetic path continues to increase, which causes the inductance contribution of the parallel magnetic path through the I-bar 940 to continue to decrease at a greater rate. For example, the total inductance decreases to approximately 3.8 millihenries at a DC bias current of approximately 0.25 ampere. The total inductance continues to decrease at a lower rate as the DC bias current increases. At a DC bias current of approximately 0.7 ampere, the parallel magnetic path through the thin gaps 980, 982 and the I-bar is almost fully saturated, and the total inductance is determined almost entirely by the much larger gap 230 between the end surfaces 210, 220, respectively, of the middle legs 150, 160 of the two E-cores 110, 122. This effect is represented by the portion of the DC bias characteristics curve 1210 of the inductor 900 that follows the curve 410 of the conventional inductor when the DC bias exceeds approximately 0.7 ampere. In FIG. 19, the dashed line of the curve 1210 are offset from the solid line by a small amount at currents above 0.7 ampere to allow the dashed line to be seen; however, the dashed line may be coincident with the solid line at currents above 0.7 ampere.

(38) As illustrated by the curve 1210 of the graph 1200 of FIG. 19, the inductor 900 of FIGS. 15-18 provides a combination of a high inductance at light loads (low DC bias currents) and a lower, approximately constant inductance at heavy loads. The lower inductance at the higher DC bias currents can be easily set by adjusting the length of the larger gap 230 between the end surfaces 210, 220, respectively, of the middle legs 150, 160 of the two E-cores 110, 122. The higher inductance at the lower DC bias currents can be set by adjusting the common thickness of the thin gaps 980, 982 (e.g., by selecting the thickness of the spacer 970). The thicknesses of the thin gaps may also be controlled by other techniques for spacing the lower surface 954 of the I-bar 940 apart from the upper surface 166 of the middle leg 150 of the first E-core 110 and the upper surface 186 of the middle leg 160 of the second E-core 112. For example, the illustrated continuous spacer of polyester film may be replaced with multiple spacers at selected locations between the juxtaposed surfaces. The inductance provided by the thin gaps may also be adjusted by adjusting the areas of the thin gaps. For example, the areas of the thin gaps may be decreased by reducing the length or the width or both the length and the width of the I-bar and thereby reducing the area of overlap between the I-bar and the upper surfaces of the middle legs of the E-cores.

(39) FIGS. 20-24 illustrate an inductor 1300 configured in accordance with another embodiment of the present invention. The inductor 1300 is similar to the inductor 900 of FIGS. 15-18 and includes the first E-core 110 and the second E-core 112 as described above. The inductor 1300 includes a bobbin 1320 with a passageway 1322 as previously described. The passageway has a width W6 between a first inner side surface 1330 and a second inner side surface 1332. The width is selected to be approximately the same as, or slightly greater than, the width W1 of the middle legs 150, 160 of the two E-cores. The passageway has a height H6 between a lower inner surface 1334 and an upper inner surface 1336.

(40) The inductor 1300 further includes a first I-bar 1340 and a second I-bar 1342. Each I-bar has a respective lower surface 1350, a respective upper surface 1352, a respective first side surface 1354, a respective second side surface 1356, a respective first end surface 1360 and a respective second end surface 1362. In the illustrated embodiment, each I-bar has a height H7 between the upper and lower surfaces, a width W7 between the first and second side surfaces and a length L7 between the first and second end surfaces. The height, width and length may correspond to the height width and length of the I-bar 940 of FIGS. 15-18; however, one or more of the dimensions (e.g., the height) may be different from the previously described embodiment.

(41) The inductor further includes a first spacer 1370 and a second spacer 1372. Each spacer has a respective lower surface 1380 and a respective upper surface 1382. Because of the thinness of the spacers, the respective end surfaces and side surfaces are not numbered. Each spacer has a respective height H8 between the lower surface and the upper surface. In the illustrated embodiment, each spacer has a length L7 and a width W7 corresponding to the length and width of the I-bars; however, the length and width may differ in other embodiments.

(42) The height H6 of the passageway 1322 is selected to accommodate the combined common height H1 of the middle legs 150, 160 of the two E-cores 110, 112, the combined heights (2H7) of the first I-bar 1340 and the second I-bar 1342, and the combined heights (2H8) of the first spacer 1370 and the second spacer 1372 (e.g., H6=H1+(2H7)+(2H8)).

(43) As shown in FIGS. 20-24, the inductor 1300 is assembled by positioning the lower surface 1350 of the first I-bar 1340 on the lower inner surface 1334 of the passageway 1320. The lower surface 1380 of the first spacer 1370 is positioned on the upper surface 1352 of the first I-bar. The middle legs 150, 160 of the two E cores 110, 112 are positioned in the passageway with the respective lower surfaces 184 of each middle leg positioned on the upper surface 1382 of the first spacer. The lower surface 1380 of the second spacer 1372 is positioned on the respective upper surfaces 186 of the middle legs of the two E-cores. The lower surface 1350 of the second I-bar 1342 is positioned on the upper surface 1382 of the second spacer.

(44) When the assembly of the inductor 1300 is completed, the five components fit within the passageway as shown in the cross-sectional views in FIGS. 15, 22, 23 and 24. As illustrated, the two I-bars 1340, 1342 and the two spacers 1370, 1372 form four thin magnetic gaps with respect to the middle legs 150, 160 of the two E-cores 110, 112. The first spacer 1370 forms a first thin magnetic gap 1500 between the upper surface 1352 of the first I-bar and the lower surface 184 of the middle leg of the first E-core 110. The first spacer also forms a second thin magnetic gap 1502 between the upper surface of the first I-bar and the lower surface 184 of the middle leg of the second E-core 112. The second spacer 1372 forms a third thin magnetic gap 1504 between the upper surface 186 of the middle leg of the first E-core and the lower surface 1350 of the second I-bar 1342. The second spacer also forms a fourth thin magnetic gap 1506 between the upper surface 186 of the middle leg of the second E-core and the lower surface of the second I-bar.

(45) The inductor 1300 of FIGS. 20-24 operates in a similar manner to the inductor 900 of FIGS. 15-18 as illustrated by a curve 1610 of the DC bias characteristics of the inductor shown on a graph 1600 in FIG. 25. At low DC bias currents, the first thin gap 1500, the second thin gap 1502 and the first I-bar 1340 form a first low-reluctance magnetic path in parallel with the magnetic path through the much larger air gap 230 between the end surfaces 210, 220, respectively, of the middle legs of the two E-cores 110, 112. Also, at low DC bias currents, the third thin gap 1504, the fourth thin gap 1506 and the second I-bar 1342 form a second low-reluctance magnetic path in parallel with the magnetic path through the much larger air gap between the end surfaces of the middle legs of the two E-cores. The two low-reluctance parallel magnetic paths cause the inductor 1300 to have a much higher total inductance at low DC bias currents. For example, the total inductance is approximately 12 millihenries at a DC bias of approximately 0 ampere. When the DC bias current increases to approximately 0.4 ampere, the magnetic paths through the two I-bars and the two thin gaps associated with each I-bar begin to saturate. The saturations of the two paths cause a corresponding increase in the reluctance in the magnetic paths, which results in a decrease of the additional inductance provided by each path. The total inductance continues to decrease to approximately 4.0 millihenries as the current increases to approximately 0.6 ampere. At DC bias currents above approximately 0.6 ampere, the total inductance remains substantially constant at approximately 4.0 millihenries with the magnetic path through the larger air gap between the ends of the two middle legs of the E-cores providing a much greater portion of the inductance.

(46) The inductor 900 and the inductor 1300 have a number of advantages. For example, unlike the inductor 500 having a step-gap core, the inductor 900 and the inductor 1300 require only a single gap length between the end surfaces of the middle legs and are therefore much easier to manufacture. The parallel magnetic paths provided by the I-bars positioned across the air gap between the end surfaces of the middle legs increases the maximum inductance at light loads (e.g., low DC bias currents). The maximum inductance at light loads is easy to adjust by varying the spacing between the surfaces of the middle legs of the E-cores and the surface of the single I-bar or between the surfaces of the middle legs and the surfaces of the two I-bars.

(47) In the illustrated embodiment, the four thin gaps 1500, 1502, 1504, 1506 have substantially the same gap lengths. In alternative embodiments, the gap lengths 1500, 1502 between the first I-bar 1340 and the middle legs 150, 160 may differ from the gap lengths 1504, 1506 between the second I-bar 1342 and the middle legs. The magnetic path incorporating the thinner pair of gaps will saturate at lower DC bias currents causing an initial decrease in the inductance over a first current range. The magnetic path through the thicker pair of gaps will saturate at higher DC bias currents causing a second decrease in the inductance over a second current range. The two current ranges may overlap such or may be spaced apart. For example, if the two current range overlap, the inductance may initially begin to decrease at a first rate over the first current range and then decrease at a second rate when the DC bias current reaches the second current range. If the two current ranges do not overlap, the inductance may initially decrease to a first level over the first current range, remain approximately constant over an interim range of currents, and then decrease further over the second current range. As discussed above, the gap lengths and the gap areas may be adjusted to determine the ranges of currents over which the inductances vary.

(48) The previous detailed description has been provided for the purposes of illustration and description. Thus, although there have been described particular embodiments of the present invention of a new and useful Inductor with Flux Path for High Inductance at Low Load, it is not intended that such references be construed as limitations upon the scope of this invention except as set forth in the following claims.