[time-nuts] Strange oscillations
Richard (Rick) Karlquist (N6RK)
richard at karlquist.com
Sat Sep 4 12:35:06 EDT 2004
> Proportional control does not have any hysteresis. The thermistor
> and the oven control should also be free from hysteresis. You are
> correct that you need hysteresis to have a limit cycle. What you
> have is a limit cycle because the amplitude is relatively stable.
I did some production engineering on the HP10811 while working
at HP; (I'm now with Agilent). In that oscillator, we used the
largest integrator capacitor that fit into the available space.
I think it was something like 1 or 2 uF. Given this capacitor,
there is a maximum practical integrator gain. If you exceed
this gain, the 10811 oven will oscillate: the current slowly
ramps up for a while then slowly ramps down. I think the period
is typically a minute or two. The oscillations increase in
amplitude until they reach a limit due to zero current or
maximum current. (The 10811 can, by design, never draw more
current than a 47 ohm resistor would draw). The 10811 oven is
definitely proportional and definitely doesn't have hystersis.
I also helped to design the E1938A oscillator, which has digital
control. It's oven dynamics are vastly superior to the 10811
because of relieving the capacitor size constraint, among
other things. We had to slam the environmental chambers from
-40? to +80?C as fast as they could go to measure the oven response,
which was miniscule.
Rick Karlquist N6RK
More information about the time-nuts