I am building a simple non regulated dual rail (+/- 50V) PSU that feature a CLC fiter to reduce output ripple, and I am trying to figure out the best positioning for the inductors.
Each one of the two rails has an inductor in series, and the current that flows through through one of the two inductors has alwalys opposite direction compared to the current that flows through the other inductor.
I1 = -I2
Since the two inductors are (almost) identical, the magnetization of the two cores has the same magnitude compared one to each other, but in opposite directions.
M1 = -M2
Thus, if I place the 2 inductors very close to each other with the cores in the same axis but rotated 180 degrees one to each other, the lines of dispersed flux of one inductor should partially concatenate with the windings of the other inductor and vice-versa. I guess this should increase the resulting inductance L of each one of the two inductors since the magnetization of each core is higher that if the 2 inductors were apart or on different planes.
M1' = M1 + M21
M2' = M2 + M21
Does it make any sense?
Each one of the two rails has an inductor in series, and the current that flows through through one of the two inductors has alwalys opposite direction compared to the current that flows through the other inductor.
I1 = -I2
Since the two inductors are (almost) identical, the magnetization of the two cores has the same magnitude compared one to each other, but in opposite directions.
M1 = -M2
Thus, if I place the 2 inductors very close to each other with the cores in the same axis but rotated 180 degrees one to each other, the lines of dispersed flux of one inductor should partially concatenate with the windings of the other inductor and vice-versa. I guess this should increase the resulting inductance L of each one of the two inductors since the magnetization of each core is higher that if the 2 inductors were apart or on different planes.
M1' = M1 + M21
M2' = M2 + M21
Does it make any sense?
Last edited:
I doubt there would be any noticeable change in L, unless you were butting a magnetic leg of one choke up next to the winding or magnetic gap (if it were on an external leg) of the other choke. You would likely see more change in L from any small change in normal operating DC or AC current, as the L is likely declining noticeably at the nominal DC operating current and operating with some minor BH loop out around a BH point heading towards the saturation region.
It would also be interesting to know how you would go about measuring any change in L at the given operating conditions.
It would also be interesting to know how you would go about measuring any change in L at the given operating conditions.
Last edited:
Inductors cross-talk is a well known phenomena. My idea was to take advantage of cross-talk between the two cores to increase AC suppression, since they are magnetized at the same instant by currents of the same amplitude in opposite directions.
I don't need necessarily need to measure the effect to benefit from it. If I can have more AC rejection by choosing on orientation of the cores rather than another, why not doing it?
- Status
- Not open for further replies.