Flowserve DDC-100 Modbus Direct-to-Host - Manual (Page 21)

36 DDC-100 Direct-to-Host Programming Guide FCD LMAIM4019-00 FCD LMAIM4019-00 DDC-100 Direct-to-Host Programming Guide 37
The second level of monitoring Channel A/B fault occurs at the Host and provides communication
information for utilization by the Host. When a Host is configured with two network serial ports, one
port is designated as Channel A (Port 1) and the other port is designated as Channel B (Port 2).
During the course of polling, the Host will retrieve the field unit status register that contains the
field unit-defined status of Channel A/B fault described above. The Host may overwrite the retrieved
field unit Channel A/B faults (field unit Register 9 bits 10 and 11) in the Host’s memory to indicate
communication status between the Host serial ports (A and B) and each field unit.
Channel A and Channel B fault bits may equal a 1 or 0 depending on the Host’s requirement of a 1
or 0 to indicate successful communication. This provides an accurate indication of the communica-
tion status from both Host Serial ports to each field unit. Typically, this type of communication-fault
recording is consistent with Hosts containing “C”-based programs.
An equally effective method for recording the communication status between the Host’s two
network serial communication ports and networked field units is to create a separate memory loca-
tion in the Host for recording Channel A/B fault status. This memory location may be configured so
a register, say Register X, bits 0 and 1 are field unit 1 Channel A/B fault status, bits 2 and 3 are field
unit 2 Channel A/B fault status and so on.
Another permutation of this method is to configure the Host memory so Register X contains the
Channel A fault status for the first 16 field units and Register X+1 contains the Channel B fault
statuses for the first 16 field units and so on.
It should be remembered that a Host with only one serial communications port will not have the
ability to handle redundant loop network communications. Therefore, communication status will be
limited to only one channel.
NOTE: In the Direct-to-Host architecture, Channel A and B faults need to be set by the Host. Setting
these faults will allow the system integrator to establish field unit time-out, retry, and field unit
communication status.
4.3.2 Network Communication Examples
The following write-ups and examples indicate proven polling techniques. The programmer who
adheres to these various examples will find their projects easier to implement.
ALL EXAMPLES SHOWN DEPICT A NETWORK IN WHICH THE HOST IS OVERWRITING THE
RETRIEVED FIELD UNIT CHANNEL A/B FAULT BIT (FIELD UNIT REGISTER 9, BITS 10 AND 11).
The programmer has the choice of overwriting these communication fault bits or creating a
separate communication fault table in the Host’s memory as detailed in Section 4.3.1, Network
Communication Errors.
In Redundant Loop Mode, the Host provides communication redundancy to each configured field
unit on the network. The Host monitors the status of the communication path between port 1 and
each configured field unit and between port 2 and each configured field unit. Host port 1 commu-
nication status between port 1 and the addressed field unit is recorded in the field unit Channel A
Fault bit. Host port 2 communication status between port 2 and the addressed field unit is recorded
in the field unit Channel B Fault bit. Both Channel A and Channel B Fault bits are located in the field
unit Status Register 9, bits 10 and 11 in the Host memory table.
On a fault-free network where all configured field units are communicating, the Host will first poll
all field units via port 1, then poll all field units via port 2, back to port 1, and so on. As each field
unit is successfully polled, the respective Channel Fault bit is set to 0 in the Host’s field unit Status
Register. Remember the Host port 1 equals Channel A and the Host port 2 equals Channel B. (See
Example 1 in this section.)
If a field unit cannot be reached on a poll, the Host will set the corresponding Channel Fault bit to 1,
switch to the other port and attempt to communicate with the same field unit. If communications
with the field unit are not successful from the second port, the corresponding Channel Fault bit
will be set to 1, and the Host will resume polling on the original port. Once the Host has completed
polling all configured field units on the first port, the polling routine will switch to the other port and
repeat the above process.
Example
There are five field units on a network (redundant loop topology) and field unit number 3 has been
turned off. The Host is currently polling field units through Host port 1 (Channel A). Field unit
numbers 1 and 2 respond to the Host port 1 poll. Field unit number 3 does not respond to the port
1 poll causing the Host to set field unit 3 Channel A Fault bit to 1. The Host now changes to port 2
(Channel B) and polls field unit number 3. Field unit number 3 does not respond to the port 2 poll,
causing the Host to set the field unit 3 Channel B Fault bit to 1. Next, the Host changes back to port
1 and attempts to poll field unit number 4. This communication attempt is successful and the Host
now polls field unit number 5 through Host port 1. Field unit number 5 responds, completing the
port 1 poll.
Next, the Host repeats the process through Host port 2 (Channel B). Field units 1 and 2 respond,
field unit 3 does not respond, and the Host sets the field unit 3 Channel B Fault bit to 1. The Host
changes to port 1 (Channel A) and attempts to communicate with field unit 3. Field unit 3 does
not respond. The Host sets the field unit 3 Channel A Fault bit to 1, switches back to port 2, and
resumes polling the remainder of the configured field units. Once field units 4 and 5 have been
successfully polled via port 2, the Host then switches to port 1 and repeats the polling process. The
intermediate alternating port process described above continues until field unit 3 is powered on and
the communication fault clears. (See Example 2 in this section.)
Commands for field unit control should interrupt the polling process and be issued through the
current poll port. Once the field unit has acknowledged the command, the Host resumes the polling
process. In the event of a communication fault between the current poll port and a commanded
field unit, the Host should issue the command through the other communication port.
Redundant Loop Network Truth Table Summarizing the results of Examples 1 through 4:
Recorded in Host field unit Status Register Bits 10 and 11
Field
unit #
Example 1 Example 2 Example 3 Example 4
Ch. A Ch. B Ch. A Ch. B Ch. A Ch. B Ch. A Ch. B
1 0 0 0 0 0 1 0 1
2 0 0 0 0 0 1 0 1
3 0 0 1 1 0 1 0 1
4 0 0 0 0 1 0 0 1
5 0 0 0 0 1 0 0 1
Example 1: No faults
The Host is successfully communicating with each field unit and sets the bits for Channel A and B
Fault to 0. A value of 0 in Channel A and B Fault indicates successful communication.
Example 2: A field unit is off-line
The Host is successfully communicating with field units 1, 2, 4, and 5 via both ports. Field unit
number 3 is without power that causes the field unit 3 network board bypass relays to de-energize.
This de-energization of the bypass relays shorts the signal through the network board and isolates
field unit 3 from the DDC-100 Network.
Example 3: A break or short in the redundant loop
The Host is successfully communicating with field units 1, 2, and 3 via port 1, and field units 4
and 5 via port 2. When a field unit doesn’t communicate within a predetermined time-out period,
the Host sets the corresponding Channel Fault bit to a value of 1. This example indicates a wiring
problem between field units 3 and 4. This problem is typically a cable breakage, short, or improp-
erly terminated wire.