Hi Solarwind, Take a look on eBay for surplus gear and also www.minicircuits.com and Hittite Semiconductor. Up to 24GHz or so, equipment is quite common. It gets much more rare above that, mainly due to less commercial uses so there isn't much surplus available. Oscillators for these extremely high frequencies can either be vacuum-tube based (like Klystrons or Magnetrons) or transistor based. For transistors, special types are used, such as Gallium-Arsenide FETs (which are usually MESFETs, I think - a type of JFET which uses a metal-semiconductor junction for the gate) or Heterojunction Bipolar Transistors (BJTs made with two different types of semiconductor material) or High electron mobility transistors (HEMT) or pseudoHEMT (pHEMT). The common theme here is that you need much higher electron mobility in the semiconductor in order to have significant gain at these frequencies. If I remember correctly, electron mobility is the effective electron mass as it moves through a crystal lattice. Also, of course, parasitic capacitance and inductance are hugely important. For this reason, these circuits will often be built using hybrid modules (a machine assembled device containing several semiconductor devices connected together using a ceramic circuit board with tiny traces and very tiny wires to connect the traces to the semiconductor material). Also, you can use transmission-line techniques to make some of these parasitic elements less important. Look up stripline and microstripline. These are ways to make transmission lines on circuit boards, as well as make capacitors, inductors, transformers, antennas, and hybrid couplers all from just specially shaped traces on a PCB. At 100GHz, all of this needs to be very tiny and very precise. Microwave semiconductor oscillators often use resonator materials which are rarely seen at lower frequencies, like YIG (Yttrium Iron Garnet) - that is, a tiny little sphere or disk of dielectric material instead of an LC circuit. A transmission line is a structure which balances the stray capacitance and the stray inductance of conductors in such a way that they cancel and end up not negatively affecting your circuit. Sean On Sun, Apr 25, 2010 at 6:42 PM, solarwind wrote: > On Sun, Apr 25, 2010 at 6:33 PM, Michael Watterson wr= ote: >> solarwind wrote: >>> I've been looking at the Radio Amateurs of Canada band plans here: >>> http://www.rac.ca/en/rac/services/bandplans/allband.php and I see them >>> listing the 5 GHz - 248 GHz. Do people actually use these bands? >>> >> google Amateur microwave records >> http://www.microwavers.org/papers/Microwave-Distance-Records.pdf >> >> UK records >> =95 1.3GHz 2617km >> =95 2.3GHz 1083km >> =95 3.4GHz 980km >> =95 5.7GHz 1244km >> =95 10GHz 1275km >> =95 24GHz 391km >> =95 47GHz 203km >> =95 76GHz 79.6km >> =95 134GHz 17.7km >> =95 145GHz 1.29km >> =95 Light (red) 76km - free-space link >> >> 145 GHz world record is 79km >> 12 JAN 2003 CW FM07FM W2SZ/4 EM96WX WA1ZMS/4 79.6 145 WORLD > > So where can you get or how can you make an oscillator that oscillates > between 10 and 50 GHz (for example). > >> you see up to 47GHz is quite good :-) >>> What's the highest frequency that a radio amateur can practically use? >>> >> Ultraviolet :) > > Where do you get or how can you make an oscillator that oscillates at > such high frequencies? > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist