Gossen MetraWatt R2600 - Page 34

GOSSEN-METRAWATT GMBH 34
Performing an oscillation test
If it is not possible to perform a trial run, e.g. where adjacent control loops strongly influence the actual value, or where an active switching output is
required to hold the actual value (working point of cooling), or where certain reasons require optimizing to the set point, the control parameters can be
established by means of a continuous oscillation. However, the calculated values for tu could eventually be very inaccurate in this case.
– Get started as described above. It is possible to perform the test without a recorder, if the actual value is watched on the display and the times on a timer.
– Set dbnd = 0 for three-state, split range and step controllers.
– Activate the control elements in automatic mode, eventually start a recorder. Record several oscillations until they are alike.
– Measure out the oscillation width x
ss
.
– Record the On time T
I
and the Off time T
II
of the switching output I and/or the continuous output of the oscillations.
Evaluation of the oscillation test
Parameter values
tu
1)
0.3 • (T
I
+ T
II
) 0.2 • (T
I
+ T
II
– 2tY)
tc tu / 12 tY / 100
Pb I x
ss
• 100 %
MBU
x
ss
• T
II
• 100 %
MBU (T
I
+ T
II
)
x
ss
• 200 %
MBU
x
ss
• T
II
• 200 %
MBU (T
I
+ T
II
)
x
ss
• 50 %
MBU
Pb II – Pb I • (T
I
/ T
II
) – Pb I • (T
I
/ T
II
) –
Parameter Two-state controller 3-state controller Contin. controller Split range controller Step controller
T
I
T
II
x
ss