Modbus

Don Diego

Member
Join Date
Apr 2017
Location
Tema
Posts
22
Hi folks ,
I am an Electrical Engineering Technician technician based in Ghana - Africa.
Which is the most prevalent industrial protocol on the planet? Modbus or Hart?
I want to know the usefulness of the modbus protocol beyond controlling vfds and ON/OFF control /monitoring of field devices.
Can modbus be used to for instance gauge the volume of liquid flowing through a transmitter and how?

I am really impressed with this blog!

Facebook wisesamlafo
 
Last edited:
Modbus!! RTU or TCP - RTU is probably more popular. Many devices have Modbus available - most flow transmitters I see are 4-20ma though.
 
My observation - and it's by no means universal - is that Modbus is used extensively for communications between higher-level control devices like drives, I/O, and other "smart" transmitters, whereas HART is quite common with process measurement devices like flow meters, pressure sensors, etc.

My opinion, which is even less universal, is that Hart will in the long term probably lose more and more market share to Ethernet-based technology like Modbus TCP/EthernetIP/Profinet. Of course, Hart will always have it's place, and I think it'll live longer than I will, but I wouldn't be investing the farm in Hart these days.
 
My observation - and it's by no means universal - is that Modbus is used extensively for communications between higher-level control devices like drives, I/O, and other "smart" transmitters, whereas HART is quite common with process measurement devices like flow meters, pressure sensors, etc.

My opinion, which is even less universal, is that Hart will in the long term probably lose more and more market share to Ethernet-based technology like Modbus TCP/EthernetIP/Profinet. Of course, Hart will always have it's place, and I think it'll live longer than I will, but I wouldn't be investing the farm in Hart these days.

I think one of the advantages that HART has is that it is very cheap to implement. There is a lot more cost in putting an ethernet port on something than there is in adding a fancy 4-20ma output.

I don't see HART taking marketshare from EIP/Profinet/ModbusTCP, but I do see it continuing to grow compared to a simple analog device.
 
HART was designed to configure and troubleshoot an instrument. It was not designed to transfer either lots of process data or transfer it quickly.

There are some techniques for getting more than one process variable from a HART device (Triloop break-out box) or for networking multiple HART devices (HART multiplexor), but the transfer rates tend to be slow and the networking tends to somewhat costly. And when using a multiplexor, one still ends up having to deal with Modbus to get the data out of the mulitplexor, because Modbus is truly the universal data fetch protocol.

The implementation of HART functionality beyond the configuration stage is miserable. There are hundreds of HART instruments that fail to allow the user to select a secondary or tertiary variable to get out of the box via HART. You get the primary variable, but there is no secondary or tertiary and there's no means to configure any of the background variables to be a secondary or a tertiary variable. All that horsepower wasted.

Getting more than one HART variable through a DCS or PLC front end AI card is a very costly exercise for the hardware. It's always the extreme high end AI card.

Modbus has its problems, too, for instance, I read on control.com the other day that one vendor implemented only Function 23 (read and write in one transaction) on their slave/server; which although that function code is defined in the Modbus spec, I am unaware of any Modbus Master/client vendor every having implemented FC23. So it's the wild, wild west out there when dealing with Modbus. Some implementations are great, most are 'workable', a few outright awful, like the FC23.

HART will remain around because instruments will always need configuration. And HART works for that. And it works in hazardous areas. But it just is not very good as a data collection protocol.

If you need to collect data, at even moderate data rates, consider Modbus first. HART is not only slow, but HMI software has no drivers to get HART, like they do for Modbus.
 
There is always a primary variable. If it's a pressure transmitter, the primary variable is pressure. But there is provision for 3 other variables.

the four register are numbered:
1. primary
2. secondary
3. tertiary
4. quatenary

Pressure transmitters frequently monitor the pressure body temperature and make that temperature available as a secondary variable.

Coriolis meters typically use all 4 variables:
1. mass flow
2. volumetric flow
3. density
4. process temperature

Historically, the variables are pre-assigned by the manufacturer. With increased processing power and revisions to the HART protocol, some manufacturers provided the ability to assign a measurement or a diagnostic status to a variable, but that trend is slow in spreading through the market.

There are breakout boxes that talk HART to the transmitter, get the variables and 'break the signals out' (make the variables) back into analog signals or make them available as Modbus slave values.




ailable from a HART device, if the manufacturer givesignal on top of a 4-20mA always transmits o
 
I think one of the advantages that HART has is that it is very cheap to implement. There is a lot more cost in putting an ethernet port on something than there is in adding a fancy 4-20ma output.

I don't see HART taking marketshare from EIP/Profinet/ModbusTCP, but I do see it continuing to grow compared to a simple analog device.

Agreed - but as danw said, the cost saving in the instrument is partially outweighed by the cost increase at the PLC/DCS end. Endress & Hauser are doing a lot of Ethernet/IP-capable instruments now, because with one (albeit expensive) instrument, you can get a ton of information. What's more, you can get it very quickly with no extra hardware at the PLC end, it's infinitely re-configurable to do more and more powerful things with a simple firmware update, and the user can configure it without needing any expensive hardware or software - it's configured via a web browser, and the AOP/EDS files to add it to the PLC can be downloaded from the device itself via that web browser. That sort of power and flexibility is something that HART just simply can't compete with.

I mean, you're right - HART will always have it's place, and as I say, I think it'll outlive me by a long way. And while I wouldn't expect HART to start taking market share from Modbus/Ethernet/etc, I'd be surprised if ethernet-based comms didn't start eating into HART's market share quite considerably.

I think Dan's comments hit the nail on the head:
...HART will remain around because instruments will always need configuration. And HART works for that. And it works in hazardous areas. But it just is not very good as a data collection protocol.
 
The only Hart stuff we use is when it is speced in by the end customer. Power plants love the stuff. In a hydraulic system where we would usually use a couple of switches to signal low oil warning and shutdown, they specify radar level transmitters with HART.

Oh well, the customer is always right, right.
 

Similar Topics

Hello, I need to access the bits from the fire system control panel through RS-485. So, I used Read Var block to read from the panel...
Replies
0
Views
50
I am having a problem communicating my PLC with Drive via Modbus connection. I start by opening the port and there is no problem, but then when I...
Replies
5
Views
73
Does anyone have an example project of the cm ptp ET200 SP HA with 410-5H DCS (PCs7 9.1 SP1) for MODBUS MASTER/SLAVE communication ?
Replies
2
Views
68
Hi, Do you have any trick to make a software working when the 30 days demo version is expired? I've tried to uninstall/clean the registers etc...
Replies
6
Views
244
Hello, I am new to Codesys, and am trying to learn about it for a project we're developing. I've got a couple questions, but first a little...
Replies
1
Views
73
Back
Top Bottom