Network configuration that could be used to connect a programmable controller system

I have been given the following question for part of my coursework,
a) Describe a communications network that could be used with a modern PLC system.
Guidance: Your work should include referenced evidence, and appropriate illustrations. The configuration type also needs to be supported with evidence that shows it is a real (commercial) network rather than one just in theory
The problem I am having is how can I show that a network is commercial? My original answer to this was basically as follows; line topology, rs485 communication standards with modbus protocol. I then referenced a company who use this exact configuration to connect their PLC controlled robot cells. However, the tutor says I have not linked my answer to a real commercial network. How else could I possibly show this is a "real" network and not just one in theory?
Thankyou in advance for your time

It seems like you're going down the right path. Can you provide the referenced evidence/illustrations that you showed your teacher?


Edit: I personally prefer Ethernet based networks, but Modbus over RS485 has been around forever, is widely supported, and isn't going away any time soon.

Thankyou for the reply mk42. This was one of 2 questions where I had to describe 2 different networks. The first one I described was similar to the one you suggested. Mine was tree topology using ethernet physical layer and profinet protocol. However this was again rebuffed as not a "real" network but just one in theory. Below is the company I referenced as evidence.
Evidence of commercial use – https://blog.robotiq.com and www.robotiq.com
Quotes from websites listed above – “We decided to standardize our products by using RS-485 back in 2013.” “Among these possibilities, the best network topology for RS-485 is a daisy chain (or line).” “Modbus is used as our communication protocol.”
I have questioned whether I am misunderstanding what he means as evidence that it is a real commercial network and asked for a pointer so I am just waiting for his reply.

Show a bill of materials with prices from a recognized control system vendor.
Show a quote for support services for this network.
Maybe you could do an interview with the "network admin" and get her to tell you why they love that network and what it allows them to achieve.
I think your tutor might be meaning, simply that company is not a control system network. So "Joe's Baked Goods use RS485 modbus to link their oven to their dough mixer. This allows them to turn the oven on and get it heated up just in time for the dough being ready."
Just guessing though, your assignment sounds vague. Control engineers love things that aren't clear. (Ha! No we don't, we are control freaks and love everything to be precise.)

https://www.rockwellautomation.com/...ustrial-networks/technical-data/overview.page

Check out the resilient Ethernet architecture white paper. There are lots of others too that are great.

Edit, just reread your post, not sure if this will help. That’s kinda an off the wall request.

I agree, the task is quite vague in its requirements so I have asked for some clarification. I think Australian might be on to something. Possibly I need to describe the application not just the network although it would be great if it actually stated this in the question. Thankyou everyone who has replied so far.

Also, your robotiq seems to be a product that has rs485, rather than a whole system.

Etech1 said:

I agree, the task is quite vague in its requirements so I have asked for some clarification. I think Australian might be on to something. Possibly I need to describe the application not just the network although it would be great if it actually stated this in the question. Thankyou everyone who has replied so far.

Click to expand...

How about I send you a picture of an I/O configuration for a project I am working on. EDIT : Actually No I can't do that, but I can make another one up along similar lines.

VFDs on Ethernet/IP, DOL Starters on Ethernet/IP, Other I/O on Flex I/O via Ethernet/IP adapters.

You might get bonus points for stating that the "/IP" after "Ethernet" has nothing to do with the "Internet Protocol", it actually stands for "Industrial Protocol".

Isn't that a good enough proof that such a network infrastructure exists, and is widely used in many factories all over the world, regardless of PLC manufacturer and type.

AustralIan said:

(Ha! No we don't, we are control freaks and love everything to be precise.)

Click to expand...

If the spec is vague and the customer REALLY DOESN'T CARE, most would be fine with that, I think. Almost never happens, though. Every so often, I see a project where the spec really is "it needs to make 1 of XYZ part every minute". Usually though, vague spec just means the customer is reserving is right to have an opinion later, and its our job to drag the opinion out of them before we need it.

Actually, that's a lie. It's the sales guy's job and the project manager's job and the design guy's job, but they either don't exist or don't do their jobs, so it falls to us anyway.

end rant.

Universities seem to like students to produce papers based on evidence.
Basically, none of the content is yours with exception of some parts of the conclusion, you start with a concept of what you are going to do, search through thousands of existing papers (if you can get access to them) for evidence, write that in your own words citing the author. so in essence you need evidence, evidence, evidence. This evidence should ideally be from existing papers that have been peer reviewed, reasonably current (sometimes difficult in some professions), full text and ideally from the country you live in. you have to watch out for plagiarism difficult as so many others have cited the original work & there are only so many ways you can write it using language pertinent to the profession (it seems that to use words like Cohort rather than group of people gets you more marks). Surprisingly, some of these so called technical words cannot be found in the dictionary and as many other have also used them, this makes it difficult to avoid plagiarism.
I have read many papers in my time & some seem to be full of unpronounceable words, virtually none of the authors own words. But the idea I suppose is to teach you the methods of research for evidence based best practice.
So you need to do lots of research and cite the evidence. There are many people here with experience, however, any ideas posted here need to be peer reviewed, accessible via your references and from trusted sources. I have 1 son, 2 step daughters who have or are currently studying at university so see the problems they are facing in producing papers/ dissertations.
Using technical language for example "people who live in areas of socioeconomic deprivation" will get you more marks than "people who live in poor areas". Other plus points are correct punctuation, linking the sections and so on. I have no idea if you are at university or doing some kind of apprenticeship but from what I have learnt from a number of apprentices I have worked with this seems to be the structure of both university & apprenticeship courses.

mk42 said:

If the spec is vague and the customer REALLY DOESN'T CARE, most would be fine with that, I think. Almost never happens, though. Every so often, I see a project where the spec really is "it needs to make 1 of XYZ part every minute". Usually though, vague spec just means the customer is reserving is right to have an opinion later, and its our job to drag the opinion out of them before we need it.

Actually, that's a lie. It's the sales guy's job and the project manager's job and the design guy's job, but they either don't exist or don't do their jobs, so it falls to us anyway.

end rant.

Click to expand...

I prefer that though to the customer that thinks they know best and spec's everything to the minute detail. We did a large project for a well known yoghurt manufacturer. They spec'd everything down to even the comments on each rung of the program!

There was some bizarre almost out of date hardware used as well, it was truly odd.

It took twice as long as it needed to and they paid dearly for it by around 30%.

I'd go really basic with your paper/report/assignment.




Illustrate how a an Allen Bradley PLC controls a Powerflex 750 series VFD over ethernet perhaps and describe in detail how to assign the VFD a static IP address (there's at least three ways, BOOTP, on the keypad of the HIM, and with Connected Components Workbench). You could reference Allen Bradley manuals which walk you through how to do this. There are a lot of AB PLCs controlling a lot of VFDs spinning a lot of motors out there.

I really think people should read the post this is an assignment, from my experience there are certain criteria for a report or assignment . 1 Follow the guidelines, 2, layout the report in the correct manner, 3. back up everything with reliable practiced based evidence from reliable sources (not saying here is not a reliable source but these are not peer reviewed). These types of assignments are often anywhere from a thousand words to five thousand +- 10%, this does not include certain sections, diagrams or tables and references. Also, Make sure you read the notes, often even the page margins & font are stipulated and incorrect layout may loose marks. Having worked with a number of apprentices & graduates and not knowing what level the OP is studying I do get concerned we could be leading them up the wrong path. If the OP could give us some indication of the level they are working to it would help.

Just looked on google and Danfoss have a paper on Modbus topology and there is a site called Modbus.org, these may help. Prolongcontrols.com have a system although may not be an actual working system it must have been used somewhere. just be cautious on copyright.

Hi parky,
I am working towards HNC electrical and electronics. The difficulty I have with this is that it is very difficult to find evidence or documentation for example a company actually using a configuration of profinet on a tree topology network. You are absolutely correct in saying that requirements are heavily evidence based which is quite frustrating when you can demonstrate understanding of a topic but cannot find any evidence to back it up even though you know it is right.