That's because I've been reading some of the excellent papers your buddy Charles Sullivan has put out. He also looked at using non-insulated stranded wire to save costs over Litz which he reckons might be useful in some not-so-demanding circumstances.
How do we access your rogue's gallery JN?
Publications from Dartmouth Magnetic Component and Power Electronics Research
I'm not sure. I thought everybody had easy access to galleries in this site. I'll try this:
Inductance tests - My Photo Gallery
jn
Thanks, that works. I'd heard you talking about your gallery before but never visited until now....
Thank you all for your feedback, very informative.
I'm checking with my winder to see the thinnest gauge wires he can handle while maintaining the overall diameter.
Will post back with result, just to make sure I'm not shooting myself in the leg.
Thanks again guys.
I'm checking with my winder to see the thinnest gauge wires he can handle while maintaining the overall diameter.
Will post back with result, just to make sure I'm not shooting myself in the leg.
Thanks again guys.
Multi-strand wire for tweeter circuits?
The graph posted above showing how losses in a coil start to climb rapidly above 10 kHz seems to suggest that there would be an advantage to using multi-stranded wire to wind inductors used in tweeter crossover circuits. I'm curious to hear opinions about that.
I have a lot of smaller gauge magnet wire and have thought about the idea of using it for small value coils. The downside of course is the higher resistance relative to large wire diameters as are normally used in crossover inductors. But, using two parallel strands halves the resistance, and so on. Would there be additional advantages to using a bundle of twisted small gauge magnet wire to wind air-core coils in the sub-0.2mH range? I think the trade off would be that the overall diameter of a bundle with the same solid cross-section as a larger single wire will be larger and thus the coil wound will be somewhat larger as well. So the bundle coil will have a bit greater length that the equivalent solid wire coil, but will it actually have lower resistance at frequencies above say 10 kHz?
The graph posted above showing how losses in a coil start to climb rapidly above 10 kHz seems to suggest that there would be an advantage to using multi-stranded wire to wind inductors used in tweeter crossover circuits. I'm curious to hear opinions about that.
I have a lot of smaller gauge magnet wire and have thought about the idea of using it for small value coils. The downside of course is the higher resistance relative to large wire diameters as are normally used in crossover inductors. But, using two parallel strands halves the resistance, and so on. Would there be additional advantages to using a bundle of twisted small gauge magnet wire to wind air-core coils in the sub-0.2mH range? I think the trade off would be that the overall diameter of a bundle with the same solid cross-section as a larger single wire will be larger and thus the coil wound will be somewhat larger as well. So the bundle coil will have a bit greater length that the equivalent solid wire coil, but will it actually have lower resistance at frequencies above say 10 kHz?
In a tweeter crossover, which is a highpass, the inductors are normally shunt elements. Not sure that losses matter as much there as when the inductors are series elements. The important aspect for the shunt element there is its impedance at low frequencies, to keep LF from the tweeter. Hence low DCR is crucial.
Down at 200uH or so, that's a relatively small air core and hence wouldn't need very thick wire anyway to get low enough loss.
Down at 200uH or so, that's a relatively small air core and hence wouldn't need very thick wire anyway to get low enough loss.
Last edited:
In principle, yes, but you have to see that in perspective: for a power converter constantly transferring, say 50W of power, losses of only 5% would be inacceptable as they would result in 2.5W heating.
For a crossover filter, the average power is minuscule, and all such losses will do is slightly increase the damping, resulting in a 0.5dB dip at most.
Thus, it is much less critical, but it will do no harm if you use multistrand conductor.
- Status
- Not open for further replies.