Owners manual

1: Introduction
Modbus Protocol User Guide 6
Extended Modbus System Example
Figure 1-1. Extended Modbus System Example
Figure 1-1 shows four specific styles of Modbus operations. Modbus/RTU devices
are traditionally split into two groups. (CoBox Modbus refers to an IAP Device
Server.)
Modbus slave devices generally are the workhorse devices. They perform their
tasks 24 hours a day, 365 days a year, for example, tasks such as flow metering,
temperature control, batch loading, or even running entire automated assembly lines.
The slave devices are not called “slaves” because they work all the time; they are
called slaves because as far as the data communications is concerned, they function
as passive servers. Modbus slave devices passively sit and wait for a remote
Modbus master device to ask them to report existing data values (Read) or accept
new data values (Write).
Modbus master devices generally are higher-level computers, devices in which
data and software are very important. The most common examples of Modbus
master devices are the “Human-Machine-Interface” (HMI) computers, which allow
human operators to monitor, adjust, and maintain the operations of the field devices.
Modbus master devices are clients that actively go out and “read” from and/or “write”
to remote Modbus slave devices to monitor or adjust slave behavior.
Modbus/TCP Master Talking to Modbus/TCP Slave
Devices A, B, E, and F are all new Modbus/TCP devices, which are improved over
Modbus/RTU (see more about Modbus/RTU limitations below). All four devices can
function concurrently as both Modbus master and Modbus slave. Both computers A
and B can treat controller E as a slave, polling data in real-time. Yet controller E can
also act as a master and poll data from controller F, which can in turn also act as a
master to write alarm data directly up to computers A and B to alert the operators to
the alarm condition. Traditional Modbus/RTU requires slave devices even with life
threatening alarm conditions to sit patiently and wait for a remote master to poll the
specific data that caused the alarm condition.