+------------------------------------------------+ | | | 50K potentiometer + | + | | \ | | \ +-----/\/\/\/\/\---+---------+----------| >0----+---| >0---- OUTPUT A | | | / | / | | | + + +-------+ ----- ----- | 68pF | capacitor GNDThis ckt will top out around 12 MHz (for faster still, reduce the cap value and be more careful with layout). Beware, the 74HC family works from 2 to 6 volt supplies; unlike the old RCA family, these fellers poop out and die at VCC=6.5V.
-- -- Mark Johnson MIPS Computer Systems, 930 E. Arques M/S 2-02, Sunnyvale, CA 94086 (408) 524-8308 mark@mips.com {or ...!decwrl!mips!mark} > Here is one of my favorites. It should work from less than 1 Hz to several >MHz, depending on the chip used and the resistor and capacitor. The order of >magnitude of the frequency will be 1/RC. Stability and spectral purity is not >very good, but it makes a decent digital system clock. > > R > _____/\/\/\_____ > | | > | |\ | > | | \ | > | | \ | > | | \ | > | | \ | > |-----| O---------> Output > | | / > | | / > __|__ | / 74C14, 74HC14, etc. Schmitt Inv > C _____ | / > | |/ Not 74C04, 74HC04, etc. > __|__ > / / / A TTL 7414 can also be used but > component values are more critical. The problem with this circuit is that it has quite high sideband noise (that said, I use it in almost all low stability applications I need. It is great for multiplexing LED sticks for example).For slightly more dosh, this one works great, and it gives TTL levels up to 25 MHz.It ain't as complex as it looks. ______ +5V | \ / \ 'LS00 / ___ 1/4 'HC04 |_____| \ |\ _________ | | | & |O----> output _______| \ | |<----Co-ax cable |/ |__| | | | O-----------------/\/\/\/\--------| T1 |___/ | | / | |_________| R3 | |\ | |/ | __|__ | | | R1 | ___ ___|___ | |___/\/\/\___| _ / \ __|__ | | 1N914 / \ ___ | \ ----- _ | / R2 | | \ | | __ / __|__ | || X|| | ___ |---|| L||---| _ | || 1|| | | -- | __|__ __|__ __ __ C1 __ __ C2 | | | | __|__ __|__ ___ ___ _ _ R1 = 10M R2 = 100k R3 = 1k R4 = 1k all metal film ( I use 0.6W,1% simply because they are the cheapest I can get) C1 = 20pF C2 = 20pf trimmer (frequency adjust) XL1 = crystal,100kHz to 28MHz (higher frequency may be dodgy)The co-ax cable is simply a way of moving RF around the PCB.It can probably be omitted for certain applications.The frequency variability is not very high (a few ppm) but can be divided with logic to give a very wide range of frequencies.I would recommend the 74LS292 and 74LS294 programmable frequency dividers.The '292 divides the input frequency by 2 to the power n,where n is a number from 2 to 31. Alternatively you could use cascaded 74167 BCD rate multipliers to generate a very accurately settable output in the Hz to MHz range,but here you're talking about frequency generators,and that might be a little complex.It's still a very feasible option though.