Eaton MTL838B-MBF Analogue Multiplexer Receiver

most appropriate action - either to repeat the original message, to try and diagnose

what has happened to the slave, to set alarms or to take whatever action is most

appropriate.

The data field

The data field transmits a number of hexadecimal values, each in the range 00 to FF.

In ASCII transmission mode this is made up of a pair of characters, in RTU it is a

single character.

A significant aspect of the communication between the master and it's slaves, that is

not defined by Modbus, is the encoding of numerical data. Modbus allows the

manufacturers of devices to determine which data encoding techniques are available

to users of the device. The encoding of data is discussed on page 13.

The data field is used to provide the slave with any additional information needed to

perform the function requested in the query. This would typically be a register address,

a register range or a value. With some functions, the data field is not required and will

not be included in the query.

If no errors occur, the data field of the response is used by the slave to pass data back

to the master.

If an error occurs, the data field is used to pass more information to the master relating

to the nature of the fault detected.

Byte count data

The responses to a number of queries require the slave to inform the master of the

number of data bytes that are being returned in the response, and this requires a

special implementation within the data field.

A typical example of this would be when the master has requested the slave to

communicate the status of a range of registers. The slave responds by repeating the

function code and it's own address, followed by the data field. The first byte of the data

field identifies the number of bytes that are being returned that contain the register

status information.

As was mentioned earlier, ASCII mode requires two 8-bit bytes to communicate a

single register content, compared to RTU which only requires a single 8-bit byte. This

difference is ignored when the byte count field is calculated, and the number of bytes

indicated is identical to the number of bytes communicated in RTU mode, but is half

the actual number of bytes communicated in ASCII mode.

The error check field

The error checking technique employed on the Modbus network depends on the

transmission mode selected. With ASCII the technique used is based on an LRC

(Longitudinal Redundancy Check) and with RTU a CRC (Cyclical Redundancy Check).

In both cases, the characters transmitted in the error check field are calculated by the

transmitting device and included in the resulting transmission. The receiving device

calculates what the error check field should contain, on receipt of the message, and

compares it with the error check field in the received message. If these two values do

not match exactly, then the receiving device knows that it has not received the

message correctly, and disregards it.

In both modes, parity checking can be optionally selected.

A fuller explanation of the error checking techniques used by Modbus is given in

Appendix A.

Data Encoding and Scaling

As has been mentioned earlier, an important area of the communication along the

network, that is not defined by the Modbus protocol, is the encoding of numerical data.

A related problem is the adoption of a scaling system for the data once it has been

encoded. (Note: this is an area which requires careful consideration by users of the

MTL838B-MBF.)

There is no problem here for manufacturers who are supplying complete systems,

based on the Modbus network, as they can select a data encoding and scaling system

appropriate to their needs. However, for manufacturers who are supplying products for

general use, there is no possibility that they will be able to determine which data

encoding system will be used by their customers, and they must allow the data

encoding technique to be user selectable.

Three data encoding techniques are the most popular - IEEE, 16-bit unsigned and 16

bit offset.

A further area of difficulty associated with the encoding of data is the way in which the

data is scaled - to provide a resolution of the measured value appropriate to the