Manufacturing Test Lab Controls Set-Up

nikkih

Member
N
Join Date
Jan 2018
Location
Akron, OH
Posts
2
I recently moved from a wastewater facility that used AB PLCs and was in the process of switching from Wonderware to Ignition for SCADA. I'm currently in a manufacturing test lab that is looking to integrate PLC control, but currently has nothing. Everything is manually controlled and test results are taken by hand.

I'm a technician, with very little experience in programming, though I will be going to several different training courses once we decide what direction we're going. I will also have several of the manufacturing engineers available to assist me as I'm getting started.

Currently I'm trying to compare my hardware and software options, and am kind of stuck. We've looked at HBM eDAQ and QuantumX data acquisition hardware and will likely be using either multiple CompactLogix or a single ControlLogix PLC for controls.

I guess I'm just looking to see what is typical in this setting. Because the lab currently has nothing, the folks here don't really have the knowledge of what is needed. I would love to be able to do what we need on a ControlLogix PLC with Ignition, but that's because I'm familiar with it. I don't want to go down the path of what I know if I'm overlooking something big because it's a new environment.

Thanks!
 
The only reason not to do CLX and Ignition for this would be price, although your experience with CLX and Ignition (and inexperience with other solutions) should be factored into the cost of any other solution.

Can you afford CLX and Ignition? If so, lock it in and start tomorrow, rather than wasting time trying to hunt down hardware costs from suppliers. Just don't tell your dealer that you're not looking elsewhere ;-) The sooner it is up and running, the sooner you're making money!
 
Welcome to the forum!

If you know CLX and Ignition, then why not keep as much in there as possible? I have no experience with eDAQ or QuantumX, are they similar to NI DAQ (which I do have experience with)? Reasons for using this such as the sample rate is greater than a PLC scan cycle can handle? Do either of yours offer OPC suites?
 
nikkih said:
I recently moved from a wastewater facility that used AB PLCs and was in the process of switching from Wonderware to Ignition for SCADA. I'm currently in a manufacturing test lab that is looking to integrate PLC control, but currently has nothing. Everything is manually controlled and test results are taken by hand.

I'm a technician, with very little experience in programming, though I will be going to several different training courses once we decide what direction we're going. I will also have several of the manufacturing engineers available to assist me as I'm getting started.

Currently I'm trying to compare my hardware and software options, and am kind of stuck. We've looked at HBM eDAQ and QuantumX data acquisition hardware and will likely be using either multiple CompactLogix or a single ControlLogix PLC for controls.

I guess I'm just looking to see what is typical in this setting. Because the lab currently has nothing, the folks here don't really have the knowledge of what is needed. I would love to be able to do what we need on a ControlLogix PLC with Ignition, but that's because I'm familiar with it. I don't want to go down the path of what I know if I'm overlooking something big because it's a new environment.

Thanks!
Click to expand...

I'm not sure what kinds of instruments you're measuring with the data acquisition HW, but if you want it all in one place the Logix family should have analog input cards that cover a wide variety of sensors. Sending it all from the PLC to Ignition and creating a report there seems plausible to me.

However, I'm not too personally familiar with automation in test labs. I've seen a lot of places that use LabView (potentially with a CompactRIO as a controller/data acquisition unit), but those were typically less automated and more manual. I know you can do a fair bit with the cRIOs though; FIRST Robotics uses them for all the robots in the HS level competitions.
 
However, I'm not too personally familiar with automation in test labs. I've seen a lot of places that use LabView (potentially with a CompactRIO as a controller/data acquisition unit), but those were typically less automated and more manual.
Click to expand...

LabView combined with TestStand = automated. We use PXI not cRIO. Oh and this year we are testing 87M parts. As for the price? Sit down when you are opening an NI quote, not sure if the alternatives suggested differ. I am trained in LabView, TestStand, and have done work with their "scada" (NI lookout). Give me AB and IA any day...
 
the_msp said:
As for the price? Sit down when you are opening an NI quote
Click to expand...

That's been my (limited) experience as well. PLC type automation always feels expensive, but some of the alternatives are worse.
 
Take those manufacturing engineers out to lunch and ask what they're familiar with and how they would approach things.

I did this with a set of my junior engineers and got a lot of input about LabView and Arduino and Python and Raspberry Pi, all stuff they had experience with in college.

In our case, that allowed me to recruit one of them to program an Arduino-based microcontroller in a very limited scope application. All of them had touched LabView but admitted that their VIs were very small in scope and that large-scale LabView experience was an easy way to get pigeonholed in your career.

And it opened the door for me to explain to them why I thought ControlLogix and our various HMI platforms were the best business case overall, even though the hardware was more expensive.

But that's us: we don't need high-bandwidth or complex data analysis. We need simple raw data logs and timestamped events, not waveform analysis and image processing. I already have the ControlLogix experience, and our A-B distributor stocks hardware seven miles away.
 
Thanks for the input. I don't have experience with NI daq, but from a brief glance it looks pretty similar to the HBM hardware that we've been looking at.

So far going ControlLogix and Ignition looks like it's going to be the cheapest of our options. I just didn't want going with what is familiar to me to come back to bite me later if it doesn't fit what we need. I think the biggest thing is determining what resolution we really need as far as sample rate time is concerned. The HBM hardware samples between 20-100 kS/s (kHz) depending on the module. Seeing as we're currently getting, 1 or 2 sample values per test doing everything by hand, I don't think we really need that kind of resolution.
 
The reason you would use fast sampling daq would be to see the artifacts of the PLC scan time, network communications delays, and IO delays. but generally once you do it once you kind of know what they are. A good test is to put a function generator on the PLC analog input and make the PLC analog output (potentially with network in between the AI and AO) the same signal. put an oscilloscope on the AI and AO. increase the frequency until you see distortion. Now you know one of the limits on performance.

You could just use more AB hardware and ignition for the data acquisition side if sampling at 10 Hz is sufficient.
 
What exactly are you controlling? I would say a PLC is the way to go if the amount or type of devices would require one, otherwise it tends to be expensive and it's not just the PLC you have to buy. Racks, power supplies, industrial instrumentation, specific cards all cost good money because they're made for industrial settings and not a lab.

Likewise, a SCADA is interesting if you are controlling otherwise, other software packages may allow you to do the same for less.

If there is nothing or very little to control and price is a limitation, you can look at IO modules with EthernetIP or ModbusTCP output and link them directly to your data acquisition system.

Mind you as well, that at least for Control Logix software licenses are expensive... although I would assume these exist in the facility already.

The other thing I was considering is the traceability of the results, would a PLC system built in house conform to the standards of the industry and would the instruments be traceable to a calibrated source? Don't get me wrong, this is possible to do, but you should keep it in mind as well.
 

Similar Topics

C
The need for Manufacturing Maintenance Technicians throughout the US
Good Morning , I imagine all manufactures are having similar problems. The lack of Maintenance Technicians ( Mechanics ) . Could you share...
2
Replies
22
Views
3,636
janner_10
J
R
Free and open-source PLC driven warehouse/manufacturing simulation made with Unity
GitHub Link: https://github.com/Automation-Standard/Open-Industry-Project This is a little side project I've been working on that I felt would be...
Replies
14
Views
3,409
ryevdokimov
R
B
Small product manufacturing opportunities for inner cities and helping the homeless
Good Evening , This sound like a strange idea. This has been on my mind and heart for several years. We have a homeless and drug issue in...
Replies
4
Views
1,862
Baker in Virginia
B
D
PCB Manufacturing
Hello, I have recently started at a new company and currently in charge of setting up the tolerances for our AOI(s) to judge all of our many...
Replies
1
Views
1,071
I_Automation
I_Automation
A
Annual Shutdowns for Manufacturing Plant
Good afternoon all, What are your thoughts on having an annual shutdown vs not having an annual shutdown for a plant with big, heavy machinery...
Replies
13
Views
3,851
geniusintraining
geniusintraining
Back
Top Bottom