UM-0085-B09
DT80 Range User Manual
Page 77
RG
There may be fewer than expected samples in the first sample period after starting a schedule. This is because, by
default, schedule execution is synchronised to midnight (see
Time Triggers — Synchronizing to Midnight (P60)
) so a one
minute schedule will always execute on minute boundaries.
If insufficient statistical samples have been taken at the time when the report schedule runs then an error message
returned and the data value is flagged as "not yet set". The
SD
and
INT
options require a minimum of two samples, the
others require at least one sample. To minimise the chance of this condition occurring:
•
switch off the synchronise to midnight flag (
/s
), or
•
ensure that the reporting schedule period is substantially longer than the statistical schedule period.
Statistical Functions
Average (AV)
The average or mean is the sum of all the channel readings divided by the number of readings. It is very useful in
reducing sensor noise.
Standard Deviation (SD)
Standard deviation is a measure of the variability of the data about the average or mean. The variation may be due to
electrical noise or process changes. The units of standard deviation are the same as the channel reading.
Maximum and Minimum (MX and MN)
The maximum and minimum of a set of channel readings can be reported with the
MX
and
MN
channel options.
The time at which these occurred can be reported with the
TMX
and
TMN
options, the date with
DMX
and
DMN
, and the
combined date/time ("instant") with the IMX and IMN channel options.
For example:
RS1M RA30M 1TK(AV) RB1D 1TK(MX)(TMX)(MN)(TMN)
1TK 24.2 degC
1TK 21.9 degC
1TK 19.0 degC
1TK 33.9 degC (Max)
1TK 15:10:00.000 (Tmx)
1TK 12.9 degC (Min)
1TK 04:33:00.000 (Tmn)
The above job measures the temperature once a minute (
RS1M
). Every 30 minutes the average for the 30 minute period
is returned by the A schedule. Once a day (at midnight), the daily min/max temperatures are returned, along with the
times at which they occurred.
Integration (INT)
The integration channel option returns the integral (area under the curve) with respect to time in seconds using a
trapezoidal approximation. The units of an integration are those of the original reading multiplied by seconds.
In the following example a sensor returns a voltage that is proportional to the flow rate (0-1000mV = 0-0.2 l/s):
BEGIN
RS100T
S5=0,0.2,0,1000"litres"
1CV=0
RA2S 1V(S5,INT,+=1CV,W) 1CV("Fuel Used",FF3)
END
Fuel used 0.012 litres
Fuel used 0.104 litres
Every 100ms, the voltage output from the sensor is measured, scaled by span
S5
(yielding a value in litre/s) and the
integral is progressively accumulated (yielding a value in litres). This is then accumulated in
1CV
(yielding the total
number of litres used since the schedule started), which is reported every 2 seconds.
The differences between the
INT
and
IB
options (both of which calculate integrals):
•
The
IB
option uses two points only (the current value and the previous value) and calculates the area under the
curve using a single trapezoid. It does not involve the statistical schedule.
•
INT
is a statistical option. It calculates the integral using a trapezoid for each sample point measured by the
statistical schedule.