> If I were to use (strong rare earth) magnets as the cores, would that > significantly increase power transfer ? Magnets won't help. (Unless you have an oxygen free copper over-unity crystal dongle, alas). > I've recently discovered an > application where transfer of 20-30mA (@ 4-ish volts, ~100mW) > across about 6mm of air would be just super. Or is permanent > magnetism a no-no ? PM is relatively irrelevant. If you had a permeable core that could saturate then PM will make it saturate sooner on one side of the AC cycle but that's unlikely to have any/much relevance in this application. That power over that air gap is "easy enough" [tm] as long as coil area is unconstrained. Without delving into tables etc I'd guesstimate that a coil over 6mm diameter would do that easily at acceptable Q and that a few times larger would do so with aplomb and/or abandon. Resonance - series or parallel, is essential for any sort of distance at sensible efficincies. There are many tables / graphs / formulae available and I can dredge up off disk or point you to them if you don't find enough on web. I have previously cited the superb Microchip app note AN710 "Antenna Circuit Design for RFID Applications" >From my Stack Exchange RFID answer here see RFID coils usually line in a single plane - ie they are essentially 2D on shape but will hav 3D thickness. An extremely good guide to RFID aerial design is given in Microchip application note AN710 [ Antenna Circuit Design for RFID Applications](http://ww1.microchip.com/downloads/en/AppNotes/00710c.pdf) - (Maintenance page [here](http://www.microchip.com/stellent/idcplg?IdcService=3DSS_GET_PAGE&no= deId=3D1824&appnote=3Den011776) ) Also liable to be of significant assistance is Microchip AN680 [ microID=AE 13.56 MHz RFID System Design Guide](http://ww1.microchip.com/downloads/en/DeviceDoc/21299E.pdf) ____________________ Two "loop" aerials which are essentially 2D in shape will have a very distinct null when aligned orthogonally (ie at right angles). If the fixed antenna is on a flexible substrate then simply bending it in a curve will reduce the tendency to form sharp nulls, but will also reduce the peak response when the aerials are optimally aligned. In this application optimal alignment is unlikely to be important. _________ Phliips app note - 2D focus [mifare =AE (14443A) 13.56 MHz RFID Proximity Antenna](http://www.nxp.com/documents/application_note/AN78010.pdf) - 2002 __________________ Coupling - If the master and slave (=3D transmit-receive coil and tag) coils are flat coils that are essentially 2D with some thickness, then Optimum coupling occurs when - their planes are parallel, - they are on the same central axis and - close together. You will get decreased coupling if you move them - further apart along the central axis (planes still parallel) OR - if you slide one across its plane in any direction so they are not on the same axis OR - if you rotate one around its X and/or Y axis OR - Any combination of the above three. **If** the plane of your tag **coil** is almost perpendicular to the plane of your "pickup" coil you have near worst case orientation. ie the magnetic field vector is at right angles to the plane of the coil so to maximally couple the two coils the sens and receive coils need to be parallel. > > I've noticed that some of the inductors/coils I've pulled from boards > are wound on magnets, whilst some are magnetic (presumably soft > iron like power transformer cores) and others are ferrite > > Reading pages like this to get some ideas > > http://en.wikipedia.org/wiki/Resonant_inductive_oupling > > Not too worried about efficiency. The primary could be driven to > any necessarily high voltage at high frequency with an SMPS IC Russell McMahon --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .