> Are there "generic" specs for crystals?  I'm afraid that most of the
> crystals I have come from surplus dealer or salvage, and are terribly
> annoymous when it comes to detailed specifications...

Yes & no.
While crystals from different manufacturers can vary in small but 
significant ways, there are component combinations that are liable to work 
well enough a lot of the time.

Ultimately, if it's going to start OK and run OK then it will (usually) have 
a capacitor to ground at each end of the crystal, possibly a parallel 
resistor and possibly a series resistor. If you are extremely unlucky then 
maybe a parallel inductor. Finding the right values is the trick. Clean 
reliable oscillator startup and good waveform and good but not excessive 
operating amplitude are largely the aims. Oscillator amplitude at the output 
pin is typically 20% to nearly 100% of Vdd. More or less may cause problems. 
Oscillator input will be of somewhat lower amplitude than output.

The IC in use is a significant part of the equation. At lower frequencies 
something like 20 pF per leg to ground and no other resistors will often 
work OK. At higher frequencies some playing may be required. Looking at the 
signal level and waveform shape with an oscilloscope as you change values 
will give you a fair idea of what the IC/crystal combination "likes".  For 
crystal of low activity or reluctant oscillators, a parallel resistor, 
typically 1 to 10 Mohm can help. Many IC's have this internally. For overly 
active crystals a series resistor in the 100's of ohms to k ohms range can 
help reduce activity. In very rare cases a parallel inductor of 100 uH or so 
can help when all else fails. Ask me how I know :-). Capacitors affect not 
only running but also starting. There are some good crystal oscillator 
design guides on web but a scope and the above is probably enough.



        RM 

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist