Best practices for effective automation applications, Part 3: Automation Q&A session

Automation application insights

• Two system integrators answered questions from the webcast on many topics such as some of the aspects involved in the case study mentioned and more detail on the lessons learned.
• The Q&A session also discussed the most overlooked aspects of automation and where companies should start in their automation process.

No matter the automation technologies being considered or applied, universal rules apply, including how to decide what should be automated. Whether just starting or a seasoned expert, heed this advice when applying automation. Automation can do quite a bit, but do not over-reach automation’s capabilities. Automation cannot cure overly complex operations. Get the right design first, then apply the right automation. Real-world examples will help fill skills gaps with smart manufacturing. This has been edited for clarity.

Control Engineering content manager, Mark T. Hoske, moderated the webcast “Best practices for effective automation applications.” The Aug. 16, 2022, course, archived as a webcast for a year, aims to answer the following questions about optimizing automation applications:

• Assess the process or processes under consideration for automation.

• Determine if a redesign is needed prior to automating.

• Consider key ways automation can influence and enhance process design.

• Identify steps for an automation implementation.

• Examine lessons learned and best practices for effective automation applications.

Chris Clayton, advanced manufacturing engineer for Applied Manufacturing Technologies, and Kevin Tom, lead automation engineer at E Tech Group, offer insights on automation effectiveness through their experiences and a case study example. This has been edited for clarity.

Questions and answers about automation projects

Mark Hoske: We know we need a system integrator to help with a major upcoming automation project. When should the integrator be brought in, ideally?

Chris Clayton: That’s easy from my perspective, is the earlier, the better. As long as everyone can be on the same page from day one, you’re going to deal with a lot less headaches down the road.

Kevin Tom: I think especially if you’ve already identified that this is an automation project. I think you already know that automation is needed, it’s great to bring automation in as early in on the design phase as you can. Great.

Hoske: In the case study presented, was the relationship between the set point offsets quadratic in nature?

Tom: Not necessarily. I think for all we know the relationship was something more exponential or cubic or something else. The quadratic model was just that. It was a model. We based it on empirical data that we gathered the running system many, many times to understand the relationship between those two setpoint.

Hoske: Also in the case study, what process was used for determining which model worked best?

Tom: That was a lot of trial and error. We’d run those TCUs with LN2 cooling, with the varying set point offsets, offset point offsets as well as the varying set points themselves. Every time we ran a test load, it’d take hours and hours for the sequence to initiate and it’d take us several more hours to feel confident that yeah, we’d reached set point stably. We’re constantly running out of liquid nitrogen and we had to call for refills from the vendor almost daily, and once we finally got some data about what set point offsets worked for what set points, we had to determine which model fit the data best. Early on it looked like the linear model would work, but one of the confounding factors in all this was that for each of the TCU to LN2 heat changer pairs, they had a different correlation profile, so though the linear model was working for some of them, we opted to use the quadratic model for consistency, and once we got more data, we found that it worked well in all cases.

Setting goals for automation and what to automate

Hoske: Switching back to discrete automation, would you target the highest volume part number, the most profitable as opposed to the entire product line, or the process most easily to automate to gain buy-in by management? How do you choose?

Clayton: I think the initial, that goal-setting or target-setting process early on during your data collection phase, that’s where you need to get on the same page with management. You tend to get a good idea of what their appetite is for automation. A lot of times, especially if it’s a smaller company just dipping their toes in again, they may want to go with a small cell somewhere out of the way, not necessarily right in the middle of their production floor that they can use to test or get comfortable with automation.

Typically, that would work on maybe the more profitable part if the idea is, “Hey, if this small test cell is successful, let’s blow it up into an actual full-sized production cell.”

I think it depends on management. You can go any one of those ways. Some companies are gung-ho, they want to automate no matter what, so you target the highest volume part number and try to get as much of that product mix in the automation as possible.

Hoske: Kevin, does the same thing apply to the process area?

Tom: Yeah, I would say that’s true too. Yeah, I don’t think there would be a difference there between discrete end process sites for that.

Overlooked benefits of automation

Hoske: In your opinion, what’s generally the most overlooked return from implementing an automation solution?

Clayton: In my mind, that’s people. Sometimes automation or robotics gets a bad rap for stealing jobs or taking human’s jobs away. In my experience, from my perspective, robots are freeing up the human potential to do what humans are good at, and that’s use their brain and think and be creative. It’s not to sit on an automation line necessarily and do repetitive tasks. That’s what automation is for. The freeing up of human capital, ingenuity and creativity, that is by far the biggest return outside of the monetary typical payback that you’ll get from automation.

Hoske: Kevin, on the case study, if automation had been involved earlier, what might they have said to avoid the design flaws that were seen later?

Tom: We’ve seen similar situations before, and I think we could have provided more of that control systems operations perspective on this part of the design. One solution maybe we could have proposed is get rid of that TCU bypass valve and have the TCU control the LN2 control valve directly, and this would be a much simpler design without that risk of creating that low flow condition and the freezing. The limitation is you can’t simultaneously provide multiple TCUs with LN2 two cooling from the same heat exchanger, but I guess an important question that you’d want to ask your end user is how often is that actually needed? Does the need for that actually outweigh the design, materials, insulation and automation costs associated with creating that functionality?

If that’s the case, you might want to consider designing more of an LN2 cooling distribution loop with drops for each TCU or something like that. Of course, that’ll cost more because you need a dedicated pump and additional control loops, so in the end, I think it’s a balancing act of priorities. There’s no one answer, but what’s important is having that conversation to make sure that the customer’s getting what they want in the end, a system that works and meets their needs.

Where should a company begin to automate?

Hoske: For a company that has little or no experience with automation, where should we start?

Clayton: Systems integrators are a great resource as well as automation consultant. I may be biased, but that’s what I do as an automation consultant is guide a customer or client through that process of collaboratively figuring out whether or not automation makes sense, so I’d start either of those places.

Tom: Yeah, we’ve seen a lot of companies that don’t have an automation department, they don’t have any automation engineers on staff, and so they come to us and say, “Help us build a URS. Help us figure out what it is that we want and what we need.”

We’ve seen so many of these and we are really good at helping collaborate and figure out what it is the customer really needs to automate and what parts are not as important.

Addressing automation upgrades, how to overcome challenges

Hoske: What are some of the challenges that come up when upgrading older or aged or obsolete automation systems?

Clayton: I think reliability would probably land at the top of that list. Obviously, older systems have a tendency to break down, have older parts, and potentially parts or replacement parts that are harder to find, so potentially more expensive just maintaining that system in the long run. That entirely depends on what type of automation and how old it is, but that’s the most frequent problem I see with the older systems.

Tom: I think my take on this would be a lot of times you have these older obsolete automation systems and the customer wants to upgrade something that’s very similar, but at this point there’s nothing similar left. An upgrade is needed when there are no more replacement parts and there’s nothing that could have a one-to-one kind of relationship for replacements. Sometimes, you just have to move over the software onto new hardware.

What you end up having to do is really understand everything about their system and look through all their documentation so that you can do your best and cross your fingers that everything will work out when you move over. That requires a lot of testing to make sure that everything still works the same, and everything is still qualified.

Edited by Chris Vavra, web content manager, CFE Media and Technology, cvavra@cfemedia.com.

Do you have experience and expertise with the topics mentioned in this content? You should consider contributing to our CFE Media editorial team and getting the recognition you and your company deserve. Click here to start this process.