Hi folks. While I was searching around for small inductors to use in an output lowpass filter I came across an interesting dual toroid from J.W. Miller magnetics that has two windings on the same core and can be wired up in a dual differential mode. Here's a link to the datasheet: http://www.jwmiller.com/pdf2/pm3600.pdf
My question is this: instead of using two inductors with separate cores for each half of a bridged output filter, would it be possible to use two windings on a single core in dual differential mode? If so, has this been done already and by whom?
Theoretically the inductors in both halves of a bridged output filter should have the exact same amount of current flowing through them (although in opposite directions) at any given time, so why wouldn't it be beneficial to have their fluxes linked through a common core? It seems like that would give a more symmetrical output current waveform through both halves of the bridge, and that surely can't be a bad thing, right? Wouldn't that also suppress noise better than two separate inductors?
I didn't really grasp electromagnetics as well as I should have in college and that's probably quite evident to those of you reading this who know that subject well.
As such, I would appreciate anything anyone has to say about this topic! Thanks.
Brian
My question is this: instead of using two inductors with separate cores for each half of a bridged output filter, would it be possible to use two windings on a single core in dual differential mode? If so, has this been done already and by whom?
Theoretically the inductors in both halves of a bridged output filter should have the exact same amount of current flowing through them (although in opposite directions) at any given time, so why wouldn't it be beneficial to have their fluxes linked through a common core? It seems like that would give a more symmetrical output current waveform through both halves of the bridge, and that surely can't be a bad thing, right? Wouldn't that also suppress noise better than two separate inductors?
I didn't really grasp electromagnetics as well as I should have in college and that's probably quite evident to those of you reading this who know that subject well.
As such, I would appreciate anything anyone has to say about this topic! Thanks.Brian
Hi Brian
The inventor of the patent ref. here
http://www.freepatentsonline.com/6281767.pdf
claims a few ways of utilising such coils. If one has the appropriate cores in his disposition for to implement the drawing, it would be interesting to report back results. As far as i can understand, this is for PWM with fixed frequency (i.e. not for Tripath chips).
Regards
George
The inventor of the patent ref. here
http://www.freepatentsonline.com/6281767.pdf
claims a few ways of utilising such coils. If one has the appropriate cores in his disposition for to implement the drawing, it would be interesting to report back results. As far as i can understand, this is for PWM with fixed frequency (i.e. not for Tripath chips).
Regards
George
Those are interesting coils, I had not seen them. They seem to be surface mount. Tiny! The highest current for the 10uH is 2.7 amps, not bad.
Miller don't say what they use as a core. We'd want to know that it is good up to a few MHz. Must inductors SMPS are not much good past 1Mhz.
The common mode choke seems like a good idea, Miller also sell some larger ones. Worth looking into.
Miller don't say what they use as a core. We'd want to know that it is good up to a few MHz. Must inductors SMPS are not much good past 1Mhz.
The common mode choke seems like a good idea, Miller also sell some larger ones. Worth looking into.
George, thanks for the link to the patent. That is quite interesting. It could still be applied to Tripath applications because they use do use PWM, but they add dither to the carrier signal and also vary the switching frequency depending on the level of the input signal.
Michael, you're about those inductors. Miller doesn't seem to tell you much about any of their inductors. It would sure be nice if they did though! Perhaps I'll send them an email to see if they can tell me anything else about the core.
The inductors I was looking for need to be about 15uH and would be demodulating the signal from a fixed 250kHz switching frequency. They really only need to be rated for about 1A and a surface mount toroid would be desirable.
Anyone think something like PE-53601 or PE-53632 from Pulse would work well for my application?
(clicking on the link might automatically download the file to your downloads folder)
http://www.pulseeng.com/pdf/PC2005_37.pdf
Michael, you're about those inductors. Miller doesn't seem to tell you much about any of their inductors. It would sure be nice if they did though! Perhaps I'll send them an email to see if they can tell me anything else about the core.
The inductors I was looking for need to be about 15uH and would be demodulating the signal from a fixed 250kHz switching frequency. They really only need to be rated for about 1A and a surface mount toroid would be desirable.
Anyone think something like PE-53601 or PE-53632 from Pulse would work well for my application?
(clicking on the link might automatically download the file to your downloads folder)
http://www.pulseeng.com/pdf/PC2005_37.pdf
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.