Suggestions for optimizing code/control system/process facility

Guarado

Member
Join Date
Jul 2020
Location
Western Australia
Posts
3
First time posting to PLCtalk so be gentle with me. We just got done working on shortening a Mobile Spreader and the company has asked us to look into optimizing their facility. It is all Allen Bradley/Rockwell kit and the control system/machine is starting to get to the point where it needs to be updated. Because they have operated on a break fix mentality with minimal maintenance the machine faults a lot. I am a relatively inexperienced recent graduate so am looking for any suggestions no matter how simple/silly on how to improve their controls. This could be from common industry practices to technical solutions. Basically I’m looking for any help I can get to better understand PLC’s and how to make them function better. Thank you in advance

Edit: the sort of thing I’m looking for is switching to using AOI’s or having links to schematics on the HMI. I don’t mind if something you are thinking seems basic and everyone would have thought about it I would still like to hear it!
 
Last edited:
Welcome to the forum.


just my opinion, but no matter what you do to the plc program or hardware on any machine, you MUST keep up the machine maintenance!
machine parts wear out, cylinder seals age and leak, chains stretch/break. as the machine runs, the newness / tightness of the machine fades and maintenance must be kept up. as a result, cylinders take longer to get there, chains that stretch do not let the parts get there on time, parts that wear do not put the parts in process in the line to be seen by photos, proxes, limit switches. unless your plant realizes that, you are in for an uphill climb.
you need to go from start to end on the line, document everything that you find that needs repair, get a budget and add 30% (you will find more stuff during the repairs), start doing the repairs, and establish a pm plan for the line.

james
 
Last edited:
Simple. Start with WTF is it supposed to do? Draw up flowcharts from the very top level (parts fo from machine A to machine B...) Break it down until you have a good idea of what exactly is going on.

Then look at what is actually there. Draw up or find the plans for the plant. Rate each machine on a scale for reliability and quality of machine output.

Compare the two sections, what it needs to do and what is there. Plan on what needs to be scrapped, what can be upgraded/repaired and what can be left alone.

Then stick all this in a fancy report and hope that someone will pay for it.

Make three options, minimum viable, preferred and money is no object.
Make certain that the flaws with the minimum viable are highlighted. Make sure the money is no object has obvious more than needed features. The preferred option should be the best choice.

Remember at this point no one has commited to paying to improve it. They just want cost and what can you do. Is it going to save money because it doesnt break down? Will it run faster? Will it improve quality or quantity or output? Will it produce less waste?

Making each bit work is easy once it broken down.

And on a final note: if you make it run faster, things like the relays cycle more and might break. Make sure you have spares, and have calculated how much life they will have remaining.
 
Check the age of the control system components like PLC, HMI, SCADA, IO cards. Find out if the current components are still in support by the supplier, losing support soon, or outdated. If the only source for replacement parts is E-BAy that is a major issue that should be addressed.
This is usually a hard sell as many companies do not see the need to replace something that is working but if the components are no longer supported by the vendor then it should be replaced to keep the system reliable.
 
optimizing their facility

What does that even mean? Far too vague to form any sort of plan.

What is causing the breakdowns? I highly doubt the PLC is - you can have the best control system in the world, but if the mechanics attached to it are a bag of *****, then you would of achieved nothing.

Bunging a few AOIs in the PLC code is not going to achieve anything on an old knackered machine.

They need to guide you to their problems and you can then offer solutions.

If they are willing to pay, spend a few days there as an operator and really get the crux of the problems.

Beware of getting involved with old temperamental stuff, you could end up inheriting a nightmare, unless it a bone fide upgrade and retro fit.

(in customers eyes, you were the last person to touch it, so therefore everything is your fault!)
 
As you study the PLC program to get an idea of how things work, look for areas of the program where someone has made modifications as work-arounds for deficiencies in the machine or to bypass failed components. The hacks aren't always easy to spot, but if you find yourself asking "why would they do that the way they did", be suspicious.
Then fix the deficiencies and eliminate the work-arounds.
 

Similar Topics

I'd like people's opinions on which manufacturer of hardware they prefer. We are an aerospace manufacturing facility that does little to no SCADA...
Replies
10
Views
483
I am setting up control for Hypochlorite dosing. The easy part is the dosing calculation for flow pacing but I would also like to setup trimming...
Replies
8
Views
895
I'm working on an application and have hit a bit of a snag. I need to find an industrial touchscreen display that can: - Tolerate down to -40C...
Replies
7
Views
718
Hello Everyone, I am reaching out to this knowledgeable community with a peculiar problem I've been experiencing with a PLC 5/40E. This issue has...
Replies
16
Views
2,209
Hi, I have never used modbus communication in Codesys 2.3, any tips and tricks are much appreciated. The system is a Wago 750-881 that controls...
Replies
3
Views
1,766
Back
Top Bottom