In a message dated 2007/09/16, gmvoeth@........... writes:

> What ever happened to the old fashioned oscillators like colpits and 
> heartly and stuff like that.
> I understand if you take great effort to properly isolate the oscillators 
> they are
> quite stable.

Hi Geoff,

       It depends on whether you want a highly stable but variable oscillator 
or a fixed frequency. Stable variable oscillators may be required in radio 
receivers. The change from thermionic valves to transistors in the 1960s greatly 
reduced the thermal drift problems.

       For driving detector circuits you need a fixed frequency. The simple 
cheap way of doing this is to use an AT cut Quartz crystal and step down 
counters. We are looking for ppm frequency and amplitude stability for seismic 
detector systems. The detector may have a range of several mm, but we need 
nanometre resolution and stability.

       If you want a very high stability sine wave, you can generate it 
directly with a oscillator / binary counter like a CD4060 as a step sine 
approximation and then use a low pass filter to smooth off the step corners. I use a 
2.4576 M Hz crystal which steps down in factors of 2 to 19,200, 9,600, 4,800, 
2,400 Hz etc to choice.

       There are problems with seismic instrumentation which need to be 
solved / compensated, but there are also problems which you don't need to have!

       Regards,

       Chris Chapman   
In a me=
ssage dated 2007/09/16, gmvoeth@........... writes:



What ever happened to the old f=
ashioned oscillators like colpits and heartly and stuff like that.

I understand if you take great effort to properly isolate the oscillators th=
ey are

quite stable.




Hi Geoff,



       It depends on whether you want a highly=
 stable but variable oscillator or a fixed frequency. Stable variable oscill=
ators may be required in radio receivers. The change from thermionic valves=20=
to transistors in the 1960s greatly reduced the thermal drift problems.



       For driving detector circuits you need=20=
a fixed frequency. The simple cheap way of doing this is to use an AT cut Qu=
artz crystal and step down counters. We are looking for ppm frequency and am=
plitude stability for seismic detector systems. The detector may have a rang=
e of several mm, but we need nanometre resolution and stability.



       If you want a very high stability sine=20=
wave, you can generate it directly with a oscillator / binary counter like a=
 CD4060 as a step sine approximation and then use a low pass filter to smoot=
h off the step corners. I use a 2.4576 M Hz crystal which steps down in fact=
ors of 2 to 19,200, 9,600, 4,800, 2,400 Hz etc to choice.



       There are problems with seismic instrum=
entation which need to be solved / compensated, but there are also problems=20=
which you don't need to have!



       Regards,



       Chris Chapman