Télécharger Imprimer la page

CARLO GAVAZZI WM30-96 Manuel De L'utilisateur page 14

Masquer les pouces Voir aussi pour WM30-96:

Publicité

CARLO GAVAZZI
A u t o m a t i o n C o m p o n e n t s
DIGITAL FILTER PROGRAMMING EXAMPLES
Example 1
How to stabilize the value of the VL-N variable displayed on the dis-
play, fluctuating from 222V and 228V.
The parameters of the digital filter have to be programmed as follows:
FILTER S: the variable has fluctuations within the mean value whose
amplitude is equal to ±0,75% of the full scale rated value of the variable
itself (obtained by the following calculation: (228-222)/ 2= ±3V, then
±3*100/400V= ±0,75% where 400V is the phase-neutral rated value of an
AV5 input). The "range" parameter, representing the action range of the
digital filter, is to be programmed to a value which must be slightly high-
er than the percentage amplitude of the fluctuation: ex. 1.0%.
FILTER CO: if the new value measured by the instrument is within the
action range of the filter, the new displayed value is obtained by adding
algebrically the previous value to the variation divided by the filtering
coefficient. As a consequence, a value higher than this coefficient implies
a longer settling time and therefore a better stability. You generally obtain
the best result by setting the filtering coefficient to a value equal to at
least 10 times the range parameter value.
In the following example: 1,0*10=10, the stability of the filtering coeffi-
cient can be improved by increasing the filtering coefficient, the allowed
values are included within 1 and 255.
P P R R O O G G R R A A M M M M I I N N G G W
WHAT IS THE ACTION OF THE DIGITAL FILTER PARAMETERS ON THE MEASURE?
No filter action
The first filter parameter is
in figure on left side (each small square is one digit). Until the measured value (red curve in figure) is within this band, the filter is active;
as soon as the value is external, the filter is deactivated and a new band will be active around the new value.
The range of the fluctuation (in digit) is a good starting value for such parameters.
The suggestion to set this parameter is to look at the size of the fluctuation (in digit) and use this value.
The second parameter is
displayed values (black in figure). There is not a theoretical rule to define this parameter, it is to be set on the field: however a rough
suggestion is to start with the same value of the
The digital filter affects the values retransmitted both via serial communication and analogue output.
Example 2
How to stabilize the value of the displayed System Active Power
(W∑), fluctuating between 300kW and 320kW (the load is connected
to the instrument by means of a 300/5A CT and a direct measure of
the voltage).
The parameters of the digital filter must be programmed as follows:
FILTER S: the variable has fluctuations within the mean value whose
amplitude is equal to ±2,78% of the full scale rated value of this variable.
This value is obtained by the following calculation: (320-300)/ 2= ±10kW,
then ±10*100/360kW= ±2,78%, where 360kW is the rated value of the
System Active Power of an AV5 input, at the above mentioned CT and VT
ratios and obtained by means of the following formula: "VLN * VT * IN *
CT * 3" where VLN = rated input voltage (400V for the AV5 input), VT= pri-
mary/secondary ratio of the voltage transformer being used, IN = rated
current (5A for the AV5 type input), CT = primary/secondary ratio of the
voltage transformer being used (in this example "400*1*5*60*3=360kW).
The RANGE parameter, representing the digital filtering coefficient action
range, is to be programmed to a value which must be slightly higher than
the percentage of the fluctuation: eg. 3.0%.
FILTER CO: if the new value acquired by the instrument is within the fil-
tering action range, the new displayed value is obtained by adding alge-
brically the previous value to the variation divided by the filtering coeffi-
cient. As a consequence, a value higher than this coefficient implies an
higher settling time and therefore a better stability. Generally speaking the
best result is obtained setting the filtering coefficient to a value equal to
at least 10 times the value of the range parameters. In the example:
3.0*10=30. In order to improve the stability you can increase the filtering
coefficient, the admitted values are included within 1 and 255.
W M M 3 3 0 0 - - 9 9 6 6
FILTER S
and defines the operating range of the filter. This operating range is represented as a yellow band
FILTER CO
and represents the filtering coefficient. The higher is
FILTER S
coefficient and then increase it until the desired stability is reached.
FILTER CO
, the smoother is the curve of the
Example 3.
It's necessary to stabilize the value of the displayed variable AL 1
(phase current 1), fluctuating within 470V and 486V.
To be able to manage the alarm function and activation and deactivation
of the relay, this value is not to be subject to continuous fluctuations. In
this example we have considered using a 500/5A CT. The parameters of
the digital filter is to be programmed as follows:
FILTER S: the variable has fluctuations within the mean value whose
amplitude is equal to ±1,60% of the full scale rated value of this variable
(obtained by means of the calculation: (486-470)/ 2= ±8A, then
±8*100/500A= ±1,60% where 500A is the value referred to the primary of
the transformer being used). The "range" parameter, which represents the
action range of the digital filter, is to be programmed to a value slightly
higher than the pourcentage amplitude of the fluctuation: for example
2.0%.
FILTER CO: if the new value acquired by the instrument is within the fil-
tering action range, the new displayed value is calculated algebrically
adding to the previous value the variation divided by the filtering coeffi-
cient. As a consequence, a higher value of this coefficient implies a high-
er settling time and therefore a better stability. Generally speaking, the
best result is obtained setting the filtering coefficient at a value equal to
at least 10 times the value of the range parameter. In the example:
2.0*10=20. To improve the stability you can increase the filtering coeffi-
cient, the admitted values are within 1 and 255.
13

Publicité

loading