L-DALI User Manual
123
LOYTEC
Version 5.2
LOYTEC electronics GmbH
In a math block adaptor with
n
inputs
v
1
,
v
2
, … ,
v
n
each output o
i
is calculated as a formula
depending on all inputs o
i
= f
i
(
v
1
,
v
2
, … ,
v
n
). Each output has two math formulae following
the same format as used in math objects (see Section 6.2):
Output value formula: This formula calculates the output value as a function of all input
values.
Output enable formula: This formula calculates an output enable (result > 0 is enable)
for the output. If the output is enabled, the output value will be written to the output. If
the output is disabled, the calculated output value is not written to the output.
In addition, each input slot can be configured whether it shall trigger the calculation or not.
Normally, any change in any input triggers the calculation of all outputs.
6.3.3
Automatic Generation and Templates
In a gateway application the systems engineer has a typical workflow: He will be confronted
with some network equipment of one technology that needs to be exposed to another
network technology. The task of generating the counterparts of data points in another
technology and connecting them is covered by the
smart auto-generate and connect
method. The existing data points are called
sources
and the generated data points are called
targets
.
In principle, the Configurator supports auto-generate for all source technologies but
generation is limited to select target technologies. Depending on availability on the device
model, the following technologies can be target for auto-generation:
CEA-709 (static NVs),
BACnet (server objects),
Registers,
Modbus (slave registers).
The target data point is generated with opposite direction and of the same class as the
source data point. Depending on the target technology, however, certain restrictions apply
on what can be generated. Typical issues are engineering units, state maps and data point
structures. The folder structure of the source data points is replicated for the target data
points.
For example, when generating matching counter parts to NVs, there are two types of NVs to
be considered: Simple NVs that hold only one value (scalar or enumeration), and structured
NVs, that consist of a number of fields. For simple NVs only one BACnet object per NV is
generated. For structured NVs, one BACnet object is generated for each structure member.
This method is called structure flattening. Some target technologies do support structures
and no flattening is applied. When generating an analog target, a data point with the best-
matching engineering unit is created. If the target allows arbitrary engineering units this will
be the same as the source engineering unit. If the target has only a limited number of
engineering units, the technology object with the best-matching unit is created. Multi-state
target data points are created with an equal number of states and compatible state IDs. For
example the CEA-709 state IDs are sorted and renumbered to start at ‘1’ in BACnet (i.e., a
‘-1’ of MOTOR_NUL in CEA-709 maps to a ‘1’ of MOTOR_NUL in BACnet). This is
necessary as the SNVT states ‘-1’ and ‘0’ cannot be represented in BACnet as a raw value,
because allowed BACnet multi-states start at 1.
The Configurator provides a preview dialog that shows, which target data points will be
created. Thus, the implicit generation rules are visible to the user. If the target technology
provides several options on what to generate, the user can change the default in this dialog.