Wavelength Electronics LFI-3751, User Manual

Page: 25 / 132

25
2
Chapter 2 - Front Panel Operation
Autotune PID Operation
Temp
Time
Before Setpoint
Response Autotune
After Setpoint
Response Autotune
Temp
Time
Before Disturbance
Rejection Autotune
After Disturbance
Rejection Autotune
3. Choose an Autotune method
There are two Autotune optimization selections available: Setpoint
Response Tuning and Disturbance Rejection Tuning .
Setpoint Response Tuning is valuable when continuously stepping the
LFI-3751 from one temperature to another. Your system will go to
temperature quickly and settle in with little or no overshoot.
Disturbance Rejection Tuning is valuable when your thermal load is
exposed to varied environmental conditions, such as air currents or
ambient temperature fl uctuations. It also improves temperature stability in
applications where the device being temperature controlled experiences
signifi cant power or heat transients, such as those found in pulsed laser
diode applications.
When a system is optimized for Disturbance Rejection, it is not optimized
for Setpoint Response and vice versa. More simply, a thermal load that is
optimized to reject disturbances will often overshoot and ring when changing
from one setpoint temperature to the next.
Temp
Time
Before Autotune
After Autotune
Starts in opposite direction
because of right-hand zero
NOTE: Sometimes thermal loads behave in a manner referred to as a
“non-minimum phase system” or more simply, the initial response of the
system is to move in the opposite direction of the desired temperature
change. This effect is caused by a zero in the right-hand plane of the thermal
system. When the LFI-3751 performs a Setpoint Response Autotune, it
assumes the thermal load is a single pole system. The right-hand plane
zero is not compensated for and the load temperature will still overshoot.
In most cases where the thermal load exhibits this behavior, the Autotune
removes the majority of the overshoot and still improves the settling
time performance.
If your thermal load exhibits a non-minimum phase response, do not
optimize with Disturbance Rejection Autotuning.