Read & Write Float in Modbus Python

Question

Sorry for my bad english. I try to write and read float in a plc register in python via Modbus using pyModbusTCP lib. This is my code that unfortunatly didn't go...

from pyModbusTCP.client import ModbusClient
from pyModbusTCP import utils

class FloatModbusClient(ModbusClient):
    def read_float(self, address, number=1):
        reg_l = self.read_holding_registers(address, number * 2)
        if reg_l:
            return [utils.decode_ieee(f) for f in utils.word_list_to_long(reg_l)]
        else:
            return None

    def write_float(self, address, floats_list):
        b32_l = [utils.encode_ieee(f) for f in floats_list]
        b16_l = utils.long_list_to_word(b32_l)
        return self.write_multiple_registers(address, b16_l)


c = FloatModbusClient(host=ip, port=porta, auto_open=True)
# write 10.0 at @0
c.write_float(registrow, [var]) 
print("write ok")

# read @0 to 9
float_l = c.read_float(registror)
print(float_l)
c.close()

Can someone help me?

Can you update the question with a bit more info - what does not work. The function aren't reading/writing correctly. there's an error, something else? — kidroca, Dec 29 '20 at 16:07
From the write_float method i expect that write in the register that i give it write a float number(var), it doesn't give me errror but it doesn't write anything. From the read_float i expect that read float number from specific register but it result me always 'None' — CarloMatto, Dec 29 '20 at 16:20

Jakub Szlaur · Accepted Answer · 2021-01-20T18:53:49.140

This is little bit tricky because all Modbus registers are only 16-bit unsigned data (those who are intended for sending numbers):

Object type      | Access     | Size    | Address Space
Coil             | Read-write | 1 bit   | 00001 - 09999
Discrete input   | Read-only  | 1 bit   | 10001 - 19999
Input register   | Read-only  | 16 bits | 30001 - 39999
Holding register | Read-write | 16 bits | 40001 - 49999

There for all the data you send needs to be unsigned 16-bit (in our case Int).
You have to handle the data transformation from Float to Int and from Int to Float both on Server side and Client side.
The code below is for Python program to be ran as Client.
In order to send Float data via Modbus you need to transform them via simple multiplication or division.

This is code that will allow you as a Client send positive Float numbers from Python to Modbus server:

from pyModbusTCP.client import ModbusClient

def send_single_float(client, reg_addr, reg_value, setting, by):
    if setting == "multiply":
        client.write_single_register(reg_addr = reg_addr, reg_value = int(reg_value*by))
    elif setting == "divide":
        if reg_value == 0:
            client.write_single_register(reg_addr = reg_addr, reg_value = int(reg_value))
        else:
            client.write_single_register(reg_addr = reg_addr, reg_value = int(reg_value/by))
    else:
        raise TypeError("Only string with following values can be set for 'Setting' attribute: ('multiply', 'divide')")

c = ModbusClient("localhost")
c.open()

my_float = 10.05

while True:
    send_single_float(c, 0, my_float, "multiply", 100)

You can use also this function to read Float numbers from Modbus:

def read_single_float(client, reg_addr, setting, by):
    if setting == "multiply":
        return client.read_holding_registers(reg_addr = reg_addr, reg_nb = 1)[0] * by
    elif setting == "divide":
        data = client.read_holding_registers(reg_addr = reg_addr, reg_nb = 1)[0]
        if data == 0:
            return data
        else:
            return data / by
    else:
        raise TypeError("Only string with following values can be set for 'Setting' attribute: ('multiply', 'divide')")

Read & Write Float in Modbus Python

1 Answers1